Investigations on Zinc Resorption using

in vitro Intestinal Models

vorgelegt von:

Dipl.-LMChem. Maria Henrietta Maares

von der Fakultät III – Prozesswissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktorin der Naturwissenschaften

- Dr. rer. nat. -

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr. Eckhard Flöter

Gutachter: Prof. Dr. Dr. Hajo Haase

Gutachterin: Prof. Dr. Anna Kipp

Tag der wissenschaftlichen Aussprache: 29.03.2019

Berlin 2019

Table of Contents

Table of Contents

Declaration .................................................................................................................................. I

Summary ................................................................................................................................ III

Zusammenfassung ...................................................................................................................... V

Abbreviations ........................................................................................................................... VII

List of Tables .............................................................................................................................. XI

List of Figures ............................................................................................................................. XI

Chapter 1. Introduction ............................................................................................................. 1

Chapter 2. Literature Review .................................................................................................... 3

2.1 The Intestinal Tract ..............................................................................................................3

2.2 Zinc – Role in the Organism .............................................................................................. 11

2.3 In vitro Studies on Intestinal Zinc Resorption ................................................................... 27

Chapter 3. Objectives and Structure of Thesis ........................................................................ 37

Chapter 4. Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc

Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport ... 39

4.1 Introduction ...................................................................................................................... 40

4.2 Experimental ..................................................................................................................... 41

4.3 Results ............................................................................................................................... 44

4.4 Discussion and Conclusion ................................................................................................ 54

4.5 Conflict of Interest ............................................................................................................ 56

4.6 Funding ............................................................................................................................. 57

4.7 References ........................................................................................................................ 57

Chapter 5. The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption

in the Caco-2/HT-29-MTX Co-culture Model, ........................................................ 61

5.1 Introduction ...................................................................................................................... 62

5.2 Methods ............................................................................................................................ 64

5.3 Results ............................................................................................................................... 68

5.4 Discussion ......................................................................................................................... 76

5.5 Conclusion ......................................................................................................................... 79

5.6 Conflict of Interest ............................................................................................................ 79

5.7 Acknowledgements .......................................................................................................... 79

5.8 References ........................................................................................................................ 79

Table of Contents

Chapter 6. In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins ..................... 85

6.1 Introduction ...................................................................................................................... 86

6.2 Results ............................................................................................................................... 88

6.3 Discussion ......................................................................................................................... 99

6.4 Materials and Methods ................................................................................................... 103

6.5 Conclusion ....................................................................................................................... 106

6.6 Author Contributions ...................................................................................................... 107

6.7 Funding ........................................................................................................................... 107

6.8 Acknowledgements ........................................................................................................ 107

6.9 Conflicts of Interest ......................................................................................................... 107

6.10 References ...................................................................................................................... 107

Chapter 7. General Discussion .............................................................................................. 113

Chapter 8. References ........................................................................................................... 127

Appendix VIII

A. Supplemental Material of Chapter 4 ............................................................................... VIII

B. Supplemental Material of Chapter 5 .................................................................................. X

C. Supplemental Material of Chapter 6 ................................................................................ XV

D. Experimental conditions for Instrumental Zinc Quantification ........................................ XX

E. Supplemental Results of Zinc Resorption Studies with in vitro Intestinal Models .......... XXI

F. Application of in vitro Caco-2 Monocultures .................................................................. XXV

G. Author contributions .................................................................................................... XXXII

List of Publications............................................................................................................... XXXIV

Acknowledgements ............................................................................................................ XXXVII

Declaration

I

Declaration

This cumulative dissertation comprises three scientific studies (Chapter 4-6), which were

prepared as independent manuscript and published in peer-reviewed journals. As the

manuscripts were composed in cooperation with co-authors, they are written in first person

plural (detailed listing of author contributions in Appendix G). The publications in Chapter 4-

6 are the final versions of the articles post-refereeing, thus the articles include contributions

of co-authors as well as of referees as outcomes of the peer-review process.

As leading author of the three manuscripts (Chapter 4-6), I developed the concept of the

studies with support of my supervisors Prof. Dr. Dr. Hajo Haase and Dr. Claudia Keil. I

executed most of the experiments, conducted the analysis and wrote the text of the

manuscripts. Experiments carried out by co-authors were performed as mentioned in

Appendix G. Published data or methods used in the manuscripts as well as results or

statements provided by others were cited as indicated by references in the respective

manuscripts.

I hereby declare that this thesis is the result of my own independent work, except where

otherwise stated. Other sources are acknowledged by explicit references.

Berlin, 28.01.2019

Maria Maares

Declaration

II

This thesis is based on three peer-reviewed publications1:

Chapter 4:

Maria Maares, Claudia Keil, Susanne Thomsen, Dorothee Günzel, Burkhard Wiesner, Hajo

Haase. "Characterization of Caco-2 cells stably expressing the protein-based zinc

probe eCalwy-5 as a model system for investigating intestinal zinc transport."

Journal of Trace Elements in Medicine and Biology 2018. 49: 296-304, DOI:

10.1016/j.jtemb.2018.01.004.

https://doi.org/10.1016/j.jtemb.2018.01.004

https://www.sciencedirect.com/science/article/pii/S0946672X17309033?via%3Dihub

Chapter 5:

Maria Maares, Ayşe Duman, Claudia Keil, Tanja Schwerdtle, Hajo Haase. "The impact of

apical and basolateral albumin on intestinal zinc resorption in the Caco-2/HT-29-

MTX co-culture model." Metallomics 2018. 10(7): 979-991, DOI:

10.1039/C8MT00064F.

https://doi.org/10.1039/C8MT00064F

https://pubs.rsc.org/en/Content/ArticleLanding/2018/MT/C8MT00064F#!divAbstract

Chapter 6:

Maria Maares, Claudia Keil, Jenny Koza, Sophia Straubing, Tanja Schwerdtle, Hajo Haase.

"In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins." International

Journal of Molecular Science 2018. 19(9): 2662, DOI: 10.3390/ijms19092662.

https://doi.org/10.3390/ijms19092662

https://www.mdpi.com/1422-0067/19/9/2662

1 The publications in Chapter 4-6 are the accepted versions of the articles post-refereeing.

Summary

III

Summary

The essential trace element zinc is mainly resorbed in the small intestine, where the luminal

available cation is absorbed by enterocytes of the intestinal epithelium and transported into

the blood circulation. In the past four decades substantial progress in the overall

understanding of zinc homeostasis was made by elucidating that the intestinal zinc

resorption regulates the systemic homeostasis of the metal. This was further complemented

by the discovery of the intestinal zinc transporters as well as the zinc binding protein

metallothionein and their regulatory role for zinc resorption. Despite these

accomplishments, molecular parameters that control intestinal zinc resorption are still

scarce and dietary factors affecting its luminal availability for the intestinal epithelium have

to be further scrutinized. For this, in vitro intestinal models provide a standardized and

versatile microenvironment to analyze zinc uptake into enterocytes and its subsequent

transport into the blood.

Hence, this thesis aimed to investigate intestinal zinc resorption with innovative in vitro

intestinal cell models. For this, a three-dimensional in vitro model to analyze zinc transport

via the intestinal epithelium had to be developed, which is closer to the in vivo situation than

the already existing ones. Herein, luminal and basolateral factors as well as cellular

composition should be optimized investigating their impact on zinc resorption. Moreover,

the scope of this thesis was to establish intestinal model systems to investigate cellular zinc

uptake with chemical- and protein-based fluorescent zinc sensors in addition to conventional

analytical approaches (such as inductively-coupled plasma mass spectrometry (ICP-MS) and

atomic absorption spectrometry (AAS)).

For investigation of zinc transport in an experimental setting closer to the physiological

environment in vivo, a three-dimensional co-culture of Caco-2 cells and the mucin-producing

goblet cell line HT-29-MTX was established. This Caco-2/HT-29-MTX model improves various

disadvantages of conventional Caco-2 monocultures, including the lack of a mucus layer

covering the cell monolayer as well as optimized luminal and basolateral buffer composition.

More precisely, the luminal and basolateral medium composition was adapted with regard

to its future application for zinc resorption studies, excluding apical addition of proteins

(such as fetal calf serum (FCS)) which would severely impact zinc availability and does not

represent the luminal situation in vivo, and including albumin in the basolateral

compartment to resemble the blood serum in vivo.

In fact, zinc transport studies in both improved Caco-2 monocultures and co-cultures

demonstrated that basolateral albumin is certainly important for investigating the in vitro

zinc resorption and acts as a basolateral zinc acceptor increasing cellular zinc release into the

basolateral compartment. Interestingly, the optimized in vitro model Caco-2/HT-29-MTX

showed enhanced net absorption of apically applied physiological zinc concentrations (25-

100 µM) compared to conventional Caco-2 monocultures. Lastly, the amounts of actually

transported zinc with this in vitro model are quite similar to the estimated amounts

transported in vivo.

Summary

IV

Furthermore, the present work suggests that the mucus layer plays a beneficial role in

intestinal zinc absorption, making it an integral part of intestinal zinc resorption. More

specific, mucins bind zinc with physiologically relevant affinity and provide several binding

sites for the metal. Consequently these glycoproteins buffer the available zinc concentration

for intestinal cells, as demonstrated in short-term zinc uptake experiments. The presence of

mucins increased zinc absorption and yielded higher transport of the cation to the

basolateral side of three-dimensional models, suggesting that this physical barrier even

facilitates zinc resorption and act as a zinc delivery system to the underlying epithelium.

In addition, the intestinal cell line Caco-2 was stably transfected with the zinc Förster

resonance energy transfer (FRET)-biosensor eCalwy and characterized regarding enterocyte-

specific properties and its maintenance of intracellular zinc homeostasis compared to Caco-2

wildtype cells; generating a well characterized intestinal model system for future

investigations of zinc uptake in enterocytes. In fact, applying the low molecular weight

sensor Zinpyr-1 and Caco-2-eCalwy cells to analyze free zinc in intestinal cells, even small

changes in cellular zinc can be determined which is of particular interest for short-term zinc

uptake. Moreover, these sensors provide the promising option to investigate spatial

distribution of zinc upon its uptake into enterocytes and to illuminate the zinc transfer in

enterocytes throughout the intestinal resorption process. To this end, involvement of two

different cellular free zinc pools in the maintenance of enterocytes’ zinc homeostasis during

zinc resorption could be illuminated.

In conclusion, findings of this thesis indicate that mucins assist apical zinc uptake and what is

more, basolateral albumin increases enterocytes’ zinc release to the blood side. This

contributes profoundly to our knowledge of the in vitro and in vivo zinc uptake and transport

processes on the apical mucosal membrane as well as on the serosal side of enterocytes.

Consequently, combining these luminal and basolateral factors in the three-dimensional in

vitro model Caco-2/HT-29-MTX, this in vitro model represents a suitable platform to

investigate intestinal zinc transport as well as further elucidate molecular mechanisms that

regulate intestinal zinc resorption. By applying Caco-2-eCalwy clones and the low molecular

weight sensor Zinpyr-1 in Caco-2 the cellular distribution of the essential metal upon its

absorption and its transfer during its resorption can be additionally examined.

Zusammenfassung

V

Zusammenfassung

Das essentielle Spurenelement Zink wird hauptsächlich im Dünndarm resorbiert. Hier wird

luminal verfügbares ionisches Zink von Enterozyten des intestinalen Epithels aufgenommen

und an das Blut abgegeben. Die Forschung der letzten vier Jahrzehnte hat maßgeblich zum

Verständnis der Zink-Homöostase beigetragen; bedeutende Fortschritte waren dabei vor

allem die Aufklärung der regulatorischen Rolle der intestinalen Zinkresorption in der

Aufrechterhaltung der Homöostase des Mikronährstoffes und die Zink-abhängige Expression

der intestinalen Zinktransporter und des Zink-bindenden Proteins Metallothionein. Dennoch

gibt es weiterhin Forschungsbedarf bezüglich der molekularen Parameter der Regulation der

intestinalen Zinkresorption sowie nahrungsbedingten Faktoren, die die Verfügbarkeit des

Kations im intestinalen Lumen beeinflussen. In vitro Intestinalmodelle stellen hierfür eine

standardisierte und vielseitig einsetzbare Mikro-Umgebungen dar, in denen sowohl die

Aufnahme des Metalls in die Enterozyten als auch dessen Transport über das intestinale

Epithel in die Blutzirkulation detaillierter analysiert werden können.

Das Ziel dieser Arbeit war daher, die intestinale Zinkresorption mit geeigneten in vitro

Intestinalmodellen zu untersuchen. Hierfür sollte zum einen ein drei-dimensionales in vitro

Modell für die Analyse des intestinalen Zinktransportes entwickelt werden, dass die in vivo

Situation im Darm so nah wie möglich abbildet. Dabei sollten luminale und basolaterale

Faktoren sowie die zelluläre Zusammensetzung optimiert und deren Einfluss auf die

Zinkresorption näher aufgeklärt werden. Des Weiteren sollten intestinale Modellsysteme

etabliert werden, mit denen die zelluläre Zinkaufnahme, zusätzlich zu der Anwendung von

konventionellen analytischen Methoden, wie zum Beispiel der Massenspektrometrie mit

induktiv gekoppeltem Plasma (ICP-MS) und Atomabsorptionsspektrometrie (AAS), mit

chemischen und Protein-basierten Zink-Fluoreszenzsonden analysiert werden können.

Zur Untersuchung des Zinktransportes wurde eine Ko-kultur aus Caco-2 Zellen und der

Muzin-produzierenden Becherzelllinie HT-29-MTX etabliert. Das Caco-2/HT-29-MTX Modell

entspricht der physiologischen Situation in vivo besser als konventionelle Caco-2

Monokulturen, da es eine Mukusschicht beinhaltet und die apikale und basolaterale

Pufferzusammensetzung an eine ideale Umgebung für Zinkresorptionsstudien angepasst ist.

Hierbei wird zum einen auf den apikalen Zusatz von Proteinen (wie fötales Kälberserum

(FKS)) verzichtet, da diese das zellulär verfügbare Zink stark beeinflussen würden und nicht

der in vivo Situation im Lumen entsprechen, zum anderen muss der basolateralen Seite des

Modells Albumin zugesetzt werden, um die Umgebung im Blutserum in vivo darzustellen.

Demgemäß zeigten Zinktransportstudien mit Caco-2 Mono- und Ko-kulturen, dass

Serumalbumin einen wichtigen Faktor für die in vitro Zinkresorption darstellt, indem es als

basolateraler Zinkakzeptor fungiert und die zelluläre Zinkabgabe in das basolaterale

Kompartiment erhöht. Zinktransportstudien mit dem optimierten in vitro Intestinalmodell

dieser Arbeit ergaben dabei nicht nur eine erhöhte fraktionelle Resorption nach apikaler

Zugabe von physiologischer Zinkkonzentration, sondern zeigten zusätzlich, dass die absolut

Zusammenfassung

VI

transportierten Zinkmengen vergleichbar mit den geschätzten Werten in der

Humanresorption in vivo sind.

Zudem legt diese Arbeit dar, dass der intestinale Mukusschicht eine förderliche und unter

Umständen regulatorische Rolle in der Zinkresorption hat. Die Ergebnisse dieser Arbeit

belegen, dass Muzine eine Vielzahl an möglichen Zinkbindungsstellen mit einer physiologisch

relevanten Affinität für das Kation enthalten und dadurch die verfügbare Zinkkonzentration

für die darunterliegenden Intestinalzellen puffern. In drei-dimensionalen Zellkulturstudien

nahm die Zinkaufnahme in Anwesenheit von Muzinen in das intestinale Epithel zu und führte

zu einem erhöhten Zinktransport zur basolateralen Seite. Diese Ergebnisse deuten darauf

hin, dass die Mukusschicht die intestinale Zinkresorption erleichtert und somit als eine Art

Zink-Transportsystem für das darunterliegende intestinale Epithel darstellt.

Des Weiteren wurde in dieser Arbeit die intestinale Zelllinie Caco-2 mit der Förster

Resonanzenergietransfer (FRET)-basierten Zinksonde eCalwy stabil transfiziert und

hinsichtlich Enterozyten-spezifischer Eigenschaften und der Aufrechterhaltung der

Zinkhomöostase mit dem Caco-2 Wildtyp verglichen. So wurde ein gut charakterisiertes

Modellsystem geschaffen, dass für zukünftige Studien der Zinkaufnahme in Enterozyten

angewendet werden kann. Die Anwendung der niedermolekularen Sonden Zinpyr-1 in Caco-

2 und des Caco-2-eCalwy Modells in dieser Arbeit verdeutlicht, dass diese eine geeignete

Methode darstellen, um bereits kleine Änderungen des zellulären Zinkgehaltes nach

Zinkaufnahme nachzuverfolgen. Das ist besonders relevant bei der Analyse von Kurzzeit-

Zinkaufnahmen. Diese Sonden liefern weiterhin die ideale Möglichkeit, auch die

intrazelluläre Verteilung des freien Zinks in Enterozyten nach dessen Aufnahme aufzuklären

und den Zinktransfer durch den Enterozyten während des Resorptionsprozesses zu

untersuchen. Auf diese Weise konnte gezeigt werden, dass zwei verschiedene zelluläre

Zinkpools in der Aufrechterhaltung der Zinkhomöostase in Enterozyten während der

Zinkaufnahme beteiligt sind.

Zusammenfassend zeigen die Ergebnisse dieser Arbeit, dass die intestinale Mukusschicht

und basolaterales Serumalbumin wichtige Faktoren für die Zinkresorption darstellen, was zur

Aufklärung der Prozesse an der apikalen und basolateralen Membran der Enterozyten

während der intestinalen Zinkresorption beiträgt. Das Caco-2/HT-29-MTX Model kombiniert

diese luminalen und basolateralen Faktoren und stellt somit ein geeignetes in vitro

Intestinalmodell zur Verfügung, mit dem der intestinale Zinktransport sowie molekulare

Regulationsmechanismen der Zinkresorption weiter aufgeklärt werden können. Durch die

Anwendung von Caco-2-eCalwy Klonen und Zinpyr-1 in Caco-2 Zellen können zudem bereits

kleine Änderungen des intrazellulären Zinkgehaltes nach dessen Aufnahme in Enterozyten

erfasst werden und die zelluläre Verteilung des Metalls während des Resorptionsprozesses

aufgeklärt werden.

Abbreviations

VII

Abbreviations

3D three-dimensional

λem emission wavelength

λex excitation wavelength

ALP alkaline phosphatase

ANOVA analysis of variance

BB brush border

BCA bicinchoninic acid

BRET bioluminescence resonance energy transfer

BSA bovine serum albumin

Calwy CFP-Atox1-linker-WD4-YFP

cDNA complementary deoxyribonucleic acid

CLSM confocal laser scanning microscopy

DAPI 4′,6-diamidin-2-phenylindole

DGE German Society for Nutrition; ger. Deutsche Gesellschaft für Ernährung

DMEM Dulbecco’s Modified Eagles Medium

DMT-1 divalent metal transporter

ECACC European Collection of Authenticated Cell Cultures

EDTA ethylene-diamine-tetra-acetic acid

EFSA European Food Safety Authority

FAAS flame atomic absorption spectrometry

FCS fetal calf serum

FD (FITC)-Dextran

FITC fluorescein isothiocyanate

FLIM fluorescence lifetime imaging microscopy

F fluorescence signal

Abbreviations

VIII

Fmax maximal fluorescence signal

Fmin minimal fluorescence signal

FRET Förster resonance energy transfer

Fuc fucose

Gal galactose

GalNAc N-acetylgalactosamine

GlcNAc N-Acetylgluosamine

HD high density

HBSS Hanks' Balanced Salt Solution

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HSA human serum albumin

ICP-MS inductively-coupled plasma mass spectrometry

ICP-OES inductively-coupled plasma optical emission spectrometry

IZiNCG International Zinc Nutrition Consultative Group

JAM junctional adhesion molecule

Km half saturation constant

KHB Krebs-Henseleit buffer

L lysosomes

LC lethal concentration

LIM Lin-11, Isl-1, Mec-3

LMW low molecular weight

M mitochondria

M cell microfold cells

mRNA messenger ribonucleic acid

MT metallothionein

MTF-1 metal regulatory transcription factor 1

MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

Abbreviations

IX

MUC mucin apoprotein

N nuclei

n.a. not available

NAC N-acetylcysteine

NEAA, non-essential amino acids

NeuAc N-Acetylneuraminic acid

NRU neutral red uptake

Nx nexus

PAR 4-(2-pyridylazo)resorcinol

Papp apparent permeability

PAS periodic acid Schiff

PBMC peripheral blood mononuclear cells

PBS phosphate buffered saline

PC polycarbonate

PE polyethylene

PES polyester

PET photo-induced electron transfer

pNPP p-nitrophenyl phosphate

PTS proline, threonine, serine

qPCR quantitative real time polymerase chain reaction (PCR)

Ref reference

RING really interesting new gene

SD standard deviation

SEM scanning electron microscopy (in Chapter 4)

standard error of mean (in Chapter 5 and 6)

SLC solute carrier

SRB sulforhodamine B

Abbreviations

X

TBS Tris(hydroxymethyl)aminomethane-buffered saline

TBST Tris-buffered saline with Tween 20

TEER transepithelial electrical resistance

TEM transmission electron microscopy

TJ tight junction protein

Tris Tris(hydroxylmethyl)aminomethan

TPEN N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine

vWF von Willebrand factor

WHO World Health Organization

WST water soluble tetrazolium

ZE zinc excess

ZD zinc deficiency

Zincon 2-carboxy-2′-hydroxy-5′-sulfoformazylbenzene monosodium salt

ZIP Zrt-, Irt-like protein

Zn zinc

ZnT zinc transporter

ZO zonula occludens

List of Tables

XI

List of Tables

Table 2.1 Expression profile of secreted and transmembrane gastrointestinal mucin .......... 10

Table 2.2 Total human body zinc ............................................................................................. 11

Table 2.3 Recommended daily allowance for dietary zinc intake for selected life-stages ...... 21

Table 2.4 Zinc and phytate content, as well as phytate : zinc-molar ratios of foods adapted from ........................................................................................................... 22

Table 2.5 Zinc transport studies using in vitro intestinal models ............................................. 30

Table 4.1: Oligonucleotide sequences used for qPCR ............................................................... 43

Table 5.1. Oligonucleotide sequences used for qPCR ............................................................... 66

Table 7.1 Main Results of in vitro zinc transport studies from this thesis ............................. 121

Table 7.2 Total amounts of resorbed zinc in vivo and in the Caco-2/HT-29-MTX model of this thesis ............................................................................................................ 124

List of Figures

Figure 2.1 Human gastrointestinal tract ...................................................................................... 3

Figure 2.2 Structure and composition of the small intestinal epithelium ................................... 6

Figure 2.3 Molecular structure of mucins and intestinal mucin O-glycan types ......................... 8

Figure 2.4 Cellular zinc homeostasis .......................................................................................... 13

Figure 2.5 Zinc buffering and muffling role of metallothioneins ............................................... 14

Figure 2.6 Regulation of intestinal zinc resorption .................................................................... 17

Figure 2.7 Schematic representation of the three-dimensional in vitro intestinal model Caco-2 ............................................................................................................ 28

Figure 2.8 Application of in vitro intestinal models to study intestinal zinc transport ............. 34

Figure 2.9 Chemical- and protein-based fluorescent sensors ................................................... 35

Figure 4.1: Alkaline phosphatase (ALP) in Caco-2-WT and Caco-2-eCalwy cells. ....................... 45

Figure 4.2: Transmission electron micrographs. ......................................................................... 46

Figure 4.3: Scanning electron microscope images. ..................................................................... 47

Figure 4.4: Localization of tight junction proteins. ..................................................................... 48

Figure 4.5: Zinc-toxicity on differentiated Caco-2-WT and -eCalwy. .......................................... 49

Figure 4.6: Effect of zinc on cellular protein levels. .................................................................... 50

Figure 4.7: Zinc homeostasis in Caco-2-WT and -eCalwy. .......................................................... 51

Figure 4.8: Life cell imaging. ........................................................................................................ 52

Figure 4.9: Two photon microscopy. .......................................................................................... 53

Figure 5.1: Impact of serum and Zinpyr-1 on short-term zinc uptake. ....................................... 68

Figure 5.2: Effect of albumin digestion on intestinal zinc resorption. ........................................ 69

Figure 5.3: Impact of serum on zinc cytotoxicity. ....................................................................... 70

Figure 5.4: Influence of serum on zinc uptake in Caco-2 cells after incubation for 24 h. .......... 71

Figure 5.5: Impact of zinc concentration and serum on gene expression of proteins involved in cellular zinc homeostasis. ....................................................................... 72

Figure 5.6: Effect of serum albumin as a basolateral zinc acceptor on intestinal zinc resorption in a Caco-2/HT-29-MTX co-culture. ........................................................ 74

Figure 5.7: Impact of basolateral albumin concentration on cellular zinc uptake and transport. .................................................................................................................. 75

Figure 6.1: Effect of mucins on zinc availability for 4-(2-pyridylazo)resorcinol (PAR). ............... 88

List of Figures

XII

Figure 6.2: Zinc binding properties of gastrointestinal mucins. ................................................. 89

Figure 6.3: Zinc binding affinity of gastrointestinal mucins. ....................................................... 90

Figure 6.4: Effect of mucin depletion on zinc-resorption in HT-29-MTX. ................................... 92

Figure 6.5: Impact of zinc-depleted mucins on zinc uptake by enterocytes. ............................. 93

Figure 6.6: Impact of extracellular mucins on zinc uptake in Caco-2 cells measured with FAAS. ................................................................................................................. 94

Figure 6.7: Effect of mucin zinc saturation on zinc uptake by enterocytes. ............................... 95

Figure 6.8: Comparison of zinc resorption in Caco-2 monocultures and Caco-2/HT-29-MTX co- cultures. .................................................................................................... 97

Figure 6.9: Zinc transport rates in Caco-2 monocultures and Caco-2/HT-29-MTX co-cultures...................................................................................................................... 98

Figure 7.1 The role of the intestinal mucus layer as a luminal factor of intestinal zinc resorption ................................................................................................................ 117

Introduction

1

Chapter 1. Introduction

The essential trace element zinc plays a key role for several important biological processes as

it is required in more than 3000 metalloproteins in the human body [1]. To compensate the

daily endogenous zinc loss and to maintain the human body zinc homeostasis, the

micronutrient has to be replenished on a daily basis by dietary intake [2]. Human body zinc

homeostasis is in fact generally regulated by its resorption in the intestine [3]. Here zinc

resorption occurs mainly in the duodenum and proximal jejunum of the small intestine [4,5],

where the metal is absorbed by enterocytes of the intestinal epithelium and transported

into the blood circulation [3]. In this process, zinc transporter on the apical and basolateral

membrane of enterocytes are engaged, regulating cellular and body zinc homeostasis

together with the cellular zinc binding protein metallothionein [6,7]. Despite this knowledge

and ongoing research, a deeper understanding of the molecular processes that regulate zinc

resorption via the intestinal epithelium are still scarce.

A severe zinc deficiency is particularly manifested in an impaired immune system [8,9],

placing zinc deficiency among the ten highest risks for human health for people from

developing countries with high morbidity rates [10]. Deprivation of this essential cation is

mostly related directly to an inadequate resorption in the intestine [11,12], either due to

insufficient zinc intake, inadequate bioavailability from the diet, or malabsorption diseases,

and is to date affecting about one third of the world’s population [13]. Absorption in the

intestine is influenced by various dietary factors, among them inhibitory components,

decreasing luminal bioavailability of the cation as well as beneficial ingredients that enhance

its absorption by enterocytes [11]. Consequently zinc resorption is not only dependent on an

adequate dietary intake but highly dependent by its intestinal availability from the diet’s

ingredients.

To further illuminate the impact of these factors on zinc absorption by the intestinal

epithelium remains of topic in research [14,15]. Herein, in vivo human studies using (stable)

isotope techniques are still the main standard [16]. During the past 50 years though,

attempts to establish suitable three-dimensional in vitro models to mimic processes in vivo

got more attention. This is mainly due to high costs and ethical standards of animal studies

and the benefits of in vitro models providing a microenvironment benefitting studies of

cellular processes on a molecular level [17,18].

In fact, in vitro intestinal models provide a promising and standardized platform to analyze

molecular mechanisms of enterocytes’ zinc transport. Investigating zinc uptake and

transport via intestinal epithelium using three-dimensional models can further identify

dietary or physiological factors that impact zinc resorption. What is more, application of

chemical and protein based fluorescent zinc sensors in in vitro enterocytes extends the

investigation of cellular zinc content to its (sub-) cellular zinc pools and offers the great

opportunity to additionally track its intracellular distribution after its uptake into

enterocytes. Yet, suitable in vitro intestinal models are still needed to further study zinc

transport and its bioavailability from different (food derived) matrices in the intestine.

Introduction

2

It is crucial that these models always represent the in vivo situation as close as possible with

respect to their cellular composition as well as apical and basolateral buffer constituents,

simulating intestinal epithelium as well as its luminal and serosal environment in vivo.

Particularly buffer and medium components are known to severely impact the zinc

speciation affecting the actual free zinc concentration that is available for cells [19,20].

Hence, it is of great importance to consider these effects when using in vitro models.

Although some previous three-dimensional in vitro intestinal models studying zinc resorption

adapted the apical buffer composition, basolateral components were rarely acknowledged.

The intestinal epithelium in vivo is covered by a mucus layer produced and secreted by

goblet cells protecting the underlying epithelium against dehydration, physical damage and

most importantly pathogens [21]. Moreover, this physical barrier is essential for absorption

of nutrients [22] and was suggested to play an important role for zinc uptake by the

intestinal mucosa [23,24]. However, neither mucins nor goblet cells were included in

hitherto existing in vitro models to investigate intestinal zinc transport.

This thesis aimed to develop and apply different in vitro intestinal models to study intestinal

zinc uptake and resorption. Herein, using three-dimensional in vitro intestinal models, the

influence of cellular composition as well as luminal and basolateral factors on zinc resorption

via the intestinal epithelium was evaluated. Moreover, luminal and basolateral factors of

intestinal zinc resorption were scrutinized, demonstrating that basolateral albumin acts as a

zinc acceptor and enhances the zinc resorption. What is more, these findings indicated that

the intestinal mucus layer facilitates the zinc uptake into enterocytes and acts as a zinc

delivery system for the intestinal epithelium. Lastly, by applying chemical and genetically

encoded fluorescent zinc sensors in enterocytes, already small changes of cellular zinc upon

its absorption were tracked and the involvement of two different cellular free zinc pools in

the maintenance of enterocytes’ zinc homeostasis during zinc absorption could be

illuminated, making them suitable model systems to further study the metal’s distribution

and transfer throughout the resorption process.

Literature Review

3

Chapter 2. Literature Review

2.1 The Intestinal Tract

The intestinal tract is the main organ of nutrient absorption in the human body. Here, food

components, already partly digested in the mouth and stomach, are further degraded into

small fragments for their uptake and transfer by enterocytes, the absorptive cells of the

intestinal epithelium. The intestine consists of the small intestine which is the major site for

digestion and resorption [25], and the large intestine, the so called colon (Figure 2.1). The

intestines main role besides absorption of nutrients is the active absorption and secretion of

electrolytes and water as well as the protection against pathogens [26,27].

Figure 2.1 Human gastrointestinal tract

Schematic representation of the human gastrointestinal tract. The digestion already starts in the mouth (not shown), where the food is mechanically broken and in parts digested by salivary enzymes. Swallowed food subsequently enters the stomach. In this acidic environment (pH = 2) proteins are mainly digested by pepsin. The main digestion and absorption of nutrients takes place in the small intestine (duodenum (length = 20-32 cm), jejunum (length = 100-250 cm), ileum (length = 200-350 cm)) [25]. For this, intestinal liquid contains enzymes, mucins, hormones, pancreatic and bile secretions, the latter is released from the gall bladder into the duodenal lumen, and is complemented by various enzymes at the brush border of the intestinal epithelium. Subsequently, the chyme enters the large intestine (ascending colon - descending colon) for further digestion and absorption of not yet absorbed nutrients and is finally excreted with the feces [25,28].

Generally, the small intestine is divided into three functional regions: the duodenum, the

jejunum and the ileum. Although these segments are structurally similar, they differ

regarding their functionality. In detail, the intestinal epithelium in duodenal and jejunal

regions produces vast amounts of brush border digestive enzymes [26], whereby 90% of

absorption takes place in duodenum and proximal jejunum [25,29]. Both duodenum and

Literature Review

4

jejunum are major sites for absorption of water-soluble vitamins, iron, calcium and zinc

[29,30]. In the jejunum the majority of monosaccharides, amino acids as well as fatty acids

are resorbed, whereas bile salts and vitamin B12 are mostly absorbed in the ileum [29,30]. In

the colon, metabolites produced by the intestinal microbiota, such as short chain fatty acids

[30], are absorbed [25]. Otherwise, mostly water and electrolyte absorption as well as

hydrogen carbonate and potassium secretion is maintained in this intestinal segment [25].

The gastrointestinal liquid is mainly comprised of water, electrolytes, enzymes, mucins, and

other bioactive substances which are mostly secreted by the exocrine glandular cells or

goblet cells throughout the gastrointestinal tract. In the intestinal tract, about 3 L fluid per

day is secreted into the intestinal lumen following a gradient of osmosis, while 6.8–7.2 L are

reabsorbed daily [28]. Aside of mucus secretion by goblet cells throughout the entire

intestinal tract, intestinal secretions are mostly maintained by specialized glands in the

duodenum as well as the ileum and complemented by pancreatic and bile secretions that

contain digestive enzymes and bile acids [28]. More precisely, duodenal secretions are

produced by specialized gland cells, mainly controlling the gradual pH increase from the

acidic environment in stomach (pH = 2) to the intestinal pH of 6–7.4 [31] by secreting

bicarbonate [28]. In the Ileum these secretions contain enzymes, mucins, and hormones

secreted by cells in the crypt of Lieberkühn, including enterocytes, goblet cells,

enteroendocrine cells and antimicrobial Paneth cells [28,29]. The latter are also limiting

bacterial growth in the small intestinal lumen that is additionally supported by peristaltic

movements constantly flushing the small intestinal lumen and keeping it bacterial free and

sterile. The colon on the other hand provides the perfect habitat for commensal bacteria,

which amongst other functions contribute to the digestion of nutrients [22].

Literature Review

5

2.1.1 Intestinal epithelium

The intestinal tract can be described as a tube, lined with a monolayer of intestinal cells

forming the intestinal epithelium which lays on the lamina propria [26,28] (Figure 2.2A). This

intestinal mucosa consisting of the intestinal epithelium, the lamina propria, and the

muscularis mucosa, is bolstered by the underlying submucosa and muscularis externa

[26,29]. The latter is formed by circular and longitudinal muscles which are responsible for

peristaltic movements and contradictions facilitating degradation and transport of food

along the gastrointestinal tract [29]. Furthermore, the mucosal and submucosal layers of the

small intestine are highly organized by folds, so called plica circularis, as well as villi and

microvilli, increasing the absorptive area about more than 100-fold from ca. 0.33 m² [32] to

30–40 m² [33]. Consequently, the intestinal tract has a small volume (~3 L) [28], but a very

large surface area.

In general the epithelium is divided into the apical side which is turned towards the

intestinal lumen and the basolateral, serosal side [29] (Figure 2.2C-D). In detail, apical and

basolateral membranes of epithelial cells are structurally and functionally different,

guaranteeing biased partitioning of membrane proteins that results in polarization of the

epithelium [26]. This polarization is very important for transcellular transport of nutrients

and is therefore additionally maintained by tight junction as well as adherens junction

proteins and desmosomes [29,34-36]. In fact, intestinal tight junction proteins, particularly

claudin and occludin, provide a sealed and intact epithelium that is additionally supported by

cell-cell adhesion molecules like cadherin assembling the adherens junction between

adjacent cells [36,37]. Moreover, scaffolding proteins like zonula occludins-1 (ZO-1) anchor

tight junction proteins to the cytoskeleton (Figure 2.2D) [36]. Of note, the integrity of the

intestinal barrier can be experimentally monitored by impedance measurements. The

transepithelial electrical resistance of an intact human intestinal barrier is reported to be 25–

60 Ω cm2 [18,38].

The intestinal epithelium is composed of different cell types with specific functions [26]. The

columnar shaped enterocytes represent 80% of intestinal epithelial cells [27]. In general

these cells are divided into non-absorptive enterocytes, mainly found in the upper part of

the crypt, and absorptive enterocytes, which are mainly located in the middle of the villi and

whose primary role is the absorption of nutrients [40]. To accomplish their functions,

absorptive enterocytes are structurally specialized by apically expressing microvilli,

approximately 3000–7000 per cell [25] and 1 µm in length. These microvilli form the

characteristic brush-like shape of enterocytes’ apical membrane and are important for

nutrient absorption, as they increase the cellular surface as well as the amount of present

enzymes [29].

Literature Review

6

Figure 2.2 Structure and composition of the small intestinal epithelium

(A) Layers of the small intestinal tract, including the intestinal epithelium, submucosa and muscularis externa (adapted from [28]). (B) The mucosal and submucosal layers are folded into the so called plica circularis which has a surface covered by villi (adapted from [28]). (C) Shown is a close-up of two villi, which are covered by intestinal cells, including absorptive enterocytes, goblet cells as well as Paneth cells and stem cells in the intestinal crypts. Additionally, the intestinal epithelium is covered by a mucus layer [39]. (D) The intestinal barrier is sealed by tight junction (TJ) and adherens junction proteins as well as desmosomes. Depicted are major TJ proteins: claudin, occludin and junctional adhesion molecules (JAM) as well as the cell-cell adhesion molecule E-cadherin. Additionally, zonola occludens-1 (ZO-1) anchors to the actin cytoskeleton [35,36].

The most abundant cells after the enterocytes are goblet cells, which mainly produce and

secrete mucins [29], but are also discussed to release antimicrobial proteins [39].

Furthermore, the intestinal epithelium consists of enteroendocrine cells, Paneth and

microfold (M) cells. Enteroendocrine cells produce peptide hormones that regulate

gastrointestinal functions, thereby linking central and enteric neuroendocrine systems [39].

Antimicrobial Paneth cells and M cells mediate the intestinal immune system [41]. In detail,

while Paneth cells secrete antimicrobial substances into the lumen [41], M cells are thought

to rapidly take up pathogens or antigens and deliver them to dendritic cells in the underlying

mucosa [39,42].

Specific distribution of these cell types, from the crypt up to the tip of the villus, results in

specific functions of the respective areas. Whereas stem cells, enteroendocrine cells and

Paneth cells are primarily localized in the crypt, enterocytes and goblet cells reside alongside

the villus. [29] Stem cells differentiate into specialized cell types of the intestinal epithelium

Literature Review

7

[27]. In this process, except regarding their differentiation into Paneth cells, stem cells

migrate from the crypt toward the villus until they reach the villus tip. Here, self-renewal of

the intestinal epithelium takes place, detaching and releasing adult cells of the villus into the

lumen [26,27]. This self-renewal occurs every 3-4 days, whereas a complete turnover of the

intestinal epithelium takes place on a weekly basis [27].

2.1.2 Mucus Layer

2.1.2.1 Composition and Structure

The mucus layer covers the whole gastrointestinal tract, protecting the underlying

epithelium against mechanical damage from the chyme as well as the luminal pH and serves

as an additional diffusion barrier for commensal bacteria, pathogens and macromolecules

[43-45]. Additionally, this physical barrier is essential for the intestinal resorption of

nutrients [22] and was shown to play a beneficial role in the enterocytes’ uptake of metals,

like iron and lead [24,46,47]. Although the organization of the small intestinal mucus is not

yet well characterized [48], it consists of a single, discontinuous, and loosely bound layer

that enables the absorption of nutrients [21]. In the colon and stomach on the other hand,

this loosely bound barrier is complemented by an additional adherent mucus layer [22].

Additionally, the mucus provides a pH gradient increasing from the gastrointestinal lumen up

to pH 7 at the epithelium. This is due to bicarbonate secretion by secretory cells facilitating

the protective function of this physical barrier, particularly in the stomach. [49]

In rats, the stomach mucus layer was analyzed to be 200–280 µm in total, decreasing down

to 170 µm and 125 µm in the duodenum and jejunum, respectively. Subsequently, the

mucus layer is increasing again in downstream direction reaching a thickness of 480 µm in

the ileum and 830 µm in the colon [50]. The inner distal colonic mucus layer is firmly

attached to the underlying epithelia cells [48] and impermeable to bacteria [48,51], whereas

the outer layer is a perfect habitat for commensal bacteria [45]. However, mucus layer

thickness in the human gastrointestinal tract was proposed to be even thicker, but has yet

not been possible to measure.

On a daily basis up to 10 L mucus are secreted into the gastrointestinal tract (mainly by

specialized mucin-producing cells), recycled and excreted with the feces [52]. Interestingly,

the turnover rate of the mucus layer in the intestine is suggested to occur even faster than

that of the intestinal epithelium [53,54], particularly preventing the invasion of the inner

mucus layer by bacteria and contact with the intestinal epithelium in the colon [53].

The mucus layer is mainly composed of ~95 % water, ~5-10% mucins, ~0.5-1% salts, ~1-2%

lipids, ~0.5% non-mucin proteins [43]. The latter include antimicrobial components mainly

produced by Paneth cells [41], which play an important role in keeping the small intestine

bacteria free and sterile [22].

Literature Review

8

Figure 2.3 Molecular structure of mucins and intestinal mucin O-glycan types

(A) The intestinal mucus layer is mainly constituted of water, ions, lipids and 5-10% highly glycosylated proteins: mucins [21]. These glycoproteins are produced by goblet cells which are either anchored into their apical membrane (not shown) or secreted densely packed in granules, resulting in a viscoelastic gel which covers the intestinal epithelium [55]. (B) Molecular structure of a mucin monomer which is mainly composed of proline, threonine, serine (PTS) tandem repeats and O-linked oligosaccharides (adapted from [21]). (C) Common mucin type O-glycans found in the intestinal tract (core 1–4) [55] (adapted from [55]).

Mucins are the major structural compound of the mucus layer (Figure 2.3), maintaining its

physicochemical properties and being essential for its viscoelasticity [43]. They are highly

glycosylated proteins with a molecular weight of 0.5–20 MDa [21], wherein intestinal mucins

are reported to have an approximate molecular mass of 2.5 MDa [56,57] Their network-like

structure, depicted in Figure 2.3A, is mainly derived by proline, threonine, serine (PTS)

tandem repeats containing O-linked oligosaccharides, building the main structure of mucin

monomers. Further, PTS tandem repeats are intermitted by less-glycosylated regions,

containing some N-glycans and cysteine-rich regions, like von Willebrand factor (vWF) D and

C domains and cysteine-knots. The latter were shown to be involved in the dimerization of

these mucin monomers [21,58,59]. In detail, mucins consist of about 80% carbohydrates

[22], mainly N-acetylgalactosamine (GalNAc), N-acetylglucosamine (GlcNAc), fucose (Fuc),

galactose (Gal) and N-acetylneuraminic acid (NeuAc or sialic acid) [58] (Figure 2.3C). More

precisely, oligosaccharides are linear or branched and can contain up to 20 monosaccharides

Literature Review

9

[21,60]. The intestinal glycan paradigm is grouped in core- and peripheral regions, wherein

the core-region mainly consists of GalNAc binding one or two additional monosaccharides.

This is additionally prolonged by Gal, GalNAc, NeuAc, GlcNAc, fucose or even sulfates

forming the peripheral region [55,61]. However, the human O-glycosylation pattern was

shown to be highly diverse in the gastrointestinal tract [62] and alterations in glycosylation

are discussed to be associated with gastrointestinal diseases [55,63].

Nomenclature of mucins is defined by their mucin apoprotein (MUC), leading to 21 mucin

families based on the number of so far identified muc-genes [43]. In addition,

gastrointestinal mucins are classified into two main mucin-types: secreted, gel forming

mucins (MUC2, MUC5AC, MUC5B and MUC6) and transmembrane bound mucins (MUC1,

MUC3, MUC4, MUC12, MUC13, MUC16, and MUC17) [45,64]. Transmembrane mucins are

anchored into the apical cell membrane, covering the surface of epithelial cells, such as

enterocytes. These firmly attached glycoproteins are part of the glycocalix [22], which was

estimated to be around 0.5-1 µm thick [45]. Besides providing protection for the cells,

transmembrane mucins are probably engaged in cell surface sensing and signaling [65],

whereas secreted mucins are pivotal for forming the network-like structure of the mucus

layer [45].

Expression profile of muc-genes, encoding for MUC apoproteins differs greatly between the

distinct parts in the gastrointestinal tract (Table 2.1) [55]. Herein, the main gel-forming

mucin produced in the intestinal tract is MUC2, whereas the gastric and colonic mucus layer

is mainly structured by MUC5AC alone and MUC2 and MUC5AC, respectively [57,64,66]. Yet,

diversity of gastrointestinal mucins is not only influenced by the variate expression profile of

the MUC apoprotein, but above all influenced by its highly diverse O-glycan paradigm [62],

mostly dependent on the individual glycosyltransferase profile [55]. Regardless, the human

small intestine mainly contains core-3 O-glycan, whereas core-1 and 2 were found in the

duodenum and stomach and core-3 and -4 structures are most prominent in the human

colon [55,67] (Figure 2.3C, p. 8).

Literature Review

10

Table 2.1 Expression profile of secreted and transmembrane gastrointestinal mucin [55]

Mucin type muc-gene Localization

Transmembrane

MUC1 stomach, duodenum, colon

MUC3A/B jejunum, ileum, colon

MUC4 stomach, colon

MUC12 stomach, colon

MUC13 small intestine, colon

MUC15 small intestine, colon

MUC17 stomach, small intestine (duodenum), colon

MUC20 colon

MUC21 colon

Secreted

MUC2 small intestine (jejunum, ileum), colon

MUC5AC stomach

MUC5B colon

MUC 6 stomach (glands), duodenum

Mucin synthesis and secretion is generally inherited by specialized mucus producing cells

throughout the gastrointestinal epithelium. In the intestine this is maintained by goblet cells,

whereas gastric foveolar mucous cells mainly secrete mucins in the stomach [68]. More

precisely, mucin apoproteins are modified co- and post-translationally. In the rough

endoplasmic reticulum, C-mannosylation, N-glycosylation and dimerization of their C-

terminal ends takes place [64,69]. The most important step in mucin synthesis, the O-

glycosylation, occurs in the Golgi during transit to the cell surface [70]. This process is

maintained by O-glycosyltransferases, stepwise adding single monosaccharides.

Subsequently, dimers at their N-termini are multimerized and mucins densely packed into

granules due to high calcium concentration and low pH [21,55,64]. These granules are either

stored or directly secreted by exocytosis, which occurs continuously or induced by

extracellular stimuli [71]. The exocytosis of these granules results in hydration of the packed

mucin by which it rapidly expands in size yielding a viscoelastic gel, covering the underlying

epithelium [55] (Figure 2.3). In contrast to secreted mucins, transmembrane mucins are

cleaved into two subunits in the endoplasmic reticulum before the O-glycosylation in the

Golgi and their subsequent transport to the apical cell membrane: one that is anchored into

the apical membrane of the cells and the N- and O-glycosylated extracellular side [64].

Literature Review

11

2.2 Zinc – Role in the Organism

Zinc is an essential trace element and the most abundant micronutrient in the human body

after iron [72,73]. Here, more than 3000 metalloproteins [1] require the bivalent cation for

catalytic, structural and regulatory functions [7]. In detail, zinc is crucial for gene expression,

influences the activity of metalloenzymes and provides a major structural component in zinc

fingers and zinc finger-containing domains such as LIM (Lin-11, Isl-1, Mec-3) and RING (really

interesting new gene) domains [74,75]. Consequently, zinc is essential for various biological

processes in the cell such as differentiation, apoptosis, and proliferation, influencing growth

and development [76]. Moreover, in the past two decades, the knowledge about its

importance as a signaling molecule increased [77] particularly in the immune system and as

a neuro-modulator in synaptic vesicles [78]. Body zinc status is therefore especially critical

for the immune system [8,9] and brain function [79]. Hence, to maintain these processes, it

is of particular significance to guarantee a stable zinc homeostasis.

2.2.1 Zinc Homeostasis

The human body consists of 2–3 g zinc from which the highest concentration can be found in

bone and skeletal muscle (~ 86%), followed by skin (6%) and liver (5%) [80]. Detailed

distribution of zinc in the human body is summarized in Table 2.2. Plasma zinc levels in

healthy individuals vary from 12–16 µM [81-83] from which 60% is bound to albumin, 30% to

α-macroglobulin, and 10% to transferrin [84], corresponding to less than 1% of whole body

zinc in serum. In fact, serum represents the rapidly exchangeable zinc pool, distributing zinc

within the body to guarantee biological processes which require this micronutrient. In

contrast, skeletal muscle and bone comprises zinc with a lower turnover and slower

availability for the systemic zinc homoeostasis. [85].

Table 2.2 Total human body zinc [80]

Tissue Approximate Zinc Concentration

[µg/g wet weight]

Total Zinc

Content [g]*

Proportion of

human body zinc [%]

Skeletal muscle 51 1.53 ~57

Bone 100 0.77 29

Skin 32 0.16 6

Liver 58 0.13 5

Brain 11 0.04 1.5

Kidneys 55 0.02 0.7

Heart 23 0.01 0.4

Hair 150 <0.01 ~0.1

Blood plasma 1 <0.01 ~0.1

*calculated for a 70 kg male.

Literature Review

12

There is no zinc storage in the human body, thus zinc has to be continuously replenished by

dietary intake [2]. The main regulatory mechanisms for human zinc homeostasis are

resorption and excretion of the trace element [3] and the small intestine, pancreas and liver

play a central role in its maintenance [3]. Of note, endogenous zinc is continuously excreted

into the intestinal lumen, from which parts are reabsorbed [3] while the remainder, varying

between 0.82.7 mg zinc per day, is excreted with feces [86-89]. Thus the close interplay of

resorption of exogenous zinc and excretion and reabsorption of endogenous zinc provides a

stable balance of body zinc homeostasis. In this manner, body zinc homeostasis can be

maintained over a wide range of exogenous zinc intake [3,11,90-92], balancing its systemic

content between very small (2.8 mg zinc/d) and high (40 mg zinc/d) amounts of dietary zinc

intake [90].This was also observed in zinc deficient states, where fecal and urinal zinc losses

rapidly decreased adapting to the low zinc supply [86,93,94]. Only when these processes fail

to sustain zinc-requiring processes, the plasma zinc pool is mobilized followed by reduction

of the less exchangeable zinc from tissues like liver, testes and bone [7,95]. Consequently,

plasma zinc is declining during zinc deficiency [90,94]. Anyhow, the plasma zinc level itself is

not a reliable biomarker for body zinc status [7], as it also changes during inflammation [9],

stress or even after a meal [7].

2.2.1.1 Cellular Zinc Homeostasis

On the cellular level, zinc homeostasis comprises three main cellular zinc pools: zinc bound

to proteins, stored in vesicles, also called zincosomes [96], and cytoplasmic free zinc (Figure

2.4). The latter is only complexed by small molecule ligands [97] and was suggested to be the

biological active form of the metal [98]. In fact, this mobile zinc species is either in transit

through the cell, being “re-distributed”, or serves as a signaling molecule [99]. Therefore, the

cytoplasmic free zinc concentration has to be tightly regulated [98] and is buffered to a

picomolar level [97] being either transported out of the cell, sequestered into vesicles or

bound to proteins such as metallothioneins (MTs) [99] (Figure 2.4).

As depicted in Figure 2.4, cellular zinc is regulated by two main zinc transporter families: the

zinc transporter (ZnT)-family (solute carrier (SLC)30A), exporting zinc or sequestering it into

organelles or vesicles [96,101,102], and members of the Zrt-, Irt-like protein (ZIP)-family

(SLC39A), which transport the metal from outside the cell, vesicles or organelles into the

cytosol [103].

Literature Review

13

Figure 2.4 Cellular zinc homeostasis

There are three main cellular zinc pools: cytoplasmic free zinc which is only complexed by low molecular weight ligands, protein-bound zinc, depicted as metallothionein (MT)-bound zinc, and zinc stored in vesicles. The cellular zinc homeostasis is maintained by three main groups of proteins: the zinc transporter (ZnT)- and the Zrt-, Irt-like protein (ZIP)-family as well as the zinc binding metallothioneins. They regulate the cytoplasmic free zinc concentration and provide its distribution into organelles, such as the endoplasmic reticulum, Golgi complex, mitochondria, and nucleus. Adapted from [100] and [101].

In contrast to the reported total zinc concentration in eukaryotic cells of several hundred

micromolar, the intracellular free zinc pool represents only transients of the cellular zinc

content [97]. A stable intracellular free zinc concentration is crucial to maintain the

intracellular zinc homeostasis, as increase of cellular free zinc above a certain level was

shown to be cytotoxic in several cell lines [98,104]. This is controlled by an elaborate zinc

buffering and muffling system, which guarantees sufficient available zinc for re-distribution

in the cell while keeping the amount of cytoplasmic buffering proteins as low as possible

[101]. The term “muffling” includes all processes that encompass changes of cellular free

zinc other than steady-state fluctuations [101]. In addition to the thermodynamic buffering

of zinc ions, this concept introduces a time-dependent component to the equilibrium of

cellular zinc homeostasis [105]. Three main protein families are discussed to be involved in

this system: on the one hand the aforementioned zinc transporters, regulating export and

import of cytoplasmic zinc and therefore taking part in muffling cellular free ion content. On

the other hand metallothioneins, the main zinc binding proteins in the cytoplasm which

buffer this zinc pool and muffle its transfer to proteins, such as transporters [106] (Figure

2.5A).

Metallothioneins are relatively small proteins (6 kDa) and incorporate seven zinc-binding

sites with a wide range of affinities [107,108]. This protein-family was suggested to bind

Literature Review

14

about 5-15% of the cellular zinc pool [109] and although the zinc-MT-complex is

thermodynamically stable it releases the cation quickly [110], mainly mediating its re-

distribution within the cell, e.g., transfer to proteins and zinc transporters, by a redox-active

mechanism [111] (Figure 2.5). Aside of zinc, MTs also store the essential trace element

copper and bind toxic metals such as cadmium playing an important role in their

detoxification [109].

Figure 2.5 Zinc buffering and muffling role of metallothioneins

(A) Zinc buffering and muffling role of metallothioneins (MTs). MTs and other ligands (such as proteins) bind free zinc and thereby buffer its cytoplasmic concentration. Additionally to zinc transporters, MTs represent zinc muffling moieties, decreasing free zinc content in the cytoplasm by transferring the bound cation to transporters, sequestering it into organelles, vesicles or outside the cell. Notable, free zinc itself can also be transported into organelles, where in this process the zinc transporter (ZnT) solely undertakes the muffling [101]. Moreover, MTs re-distribute intracellular zinc by transferring it to other ligands, such as metalloproteins [112]. (B) This zinc transfer is based on a redox-active mechanism. Oxidation of the sulfur-ligand by oxidants such as glutathione disulfide, selenium compounds or reactive species like nitric monoxide releases zinc to other zinc binding proteins. In addition, the oxidized MT form thionin can also be reduced to its apo-protein thionein. This redox-cycle requires redox-couples like glutathione/glutathione disulfide. [113] Moreover, zinc itself induces mt expression by binding to the metal regulatory transcription factor 1 (MTF-1) [114] (B adapted from [111]).

The importance of MTs for cellular zinc homeostasis is additionally supported by the fact

that these proteins are ubiquitously distributed in the human body. There are four known

MT genes (MT-1–MT-4) encoding eleven functional human MT-isoforms. While MT-1 and

MT-2 are expressed in all body cell types, MT-3 and -4 were mainly found in brain or

epithelial tissues, respectively [106,115]. The relevance of MTs as the only proteins

mediating zinc trafficking, however, is limited by the fact that MT knockout mice (for MT-1

and -2 genes) were indeed more sensitive to additional dietary zinc, but were still viable and

reproductive [116,117]. Furthermore, experimental modeling of MTs as mufflers indicated

that metallothioneins are possibly not the only proteins mediating its transfer to

Literature Review

15

transporters [118]. Hence, these findings imply that there must be other proteins

maintaining zinc trafficking through the cell.

2.2.2 Intestinal Zinc Resorption

Zinc is absorbed throughout the whole small intestine [5,119]; though the major site of

intestinal human zinc resorption still is controversial. In animal studies using rats the highest

resorption rate was reported to occur both in the duodenum and ileum [120-122] or only in

the ileum [123] or jejunum [5,124], respectively. In vivo studies investigating the actual site

of zinc resorption in humans are scarce. However, using small intestine perfusion techniques

in healthy individuals, the major resorption sites in the human intestine were revealed to be

both the duodenum [4] and jejunum [5]. Hence, it is most likely that zinc absorption mainly

occurs in the duodenum and proximal jejunum as these are the sites in the intestine that

dietary zinc passes first [125].

In detail, zinc uptake takes place at the intestinal brush border membrane, where it is

transported from the lumen into absorptive cells of the epithelium: the enterocytes. The

following excretion of the cation at the basolateral side of enterocytes releases it into the

portal blood, where it is mostly bound to albumin, distributing the metal in the body [3,126].

Interestingly, while several in vitro studies showed transport from the basolateral to the

luminal site of the intestinal epithelium [127-129], this was not yet observed in humans

[130]. Additionally, only a rather low mucosal zinc secretion into the lumen was reported in

vivo using perfused rat intestines and physiological serum zinc concentrations [131].

Although, the recent discovery of a bidirectional zinc transporter (ZnT-5B) on the apical

membrane of enterocytes [132] suggests that this could possibly represent an additional

regulatory mechanism of cellular and body zinc homeostasis [133,134].

Zinc resorption kinetics were described by carrier-mediated and saturable processes

[5,125,130,135], where the zinc uptake at the apical membrane of the intestinal mucosa

seems to be the rate limiting step [131]. Saturation of these transport mechanisms at a

certain luminal zinc level is reflected by an absorption plateau with a half saturation constant

(Km) of cellular zinc uptake of 29–55 µM zinc in vivo [92,131,136]. However, at higher luminal

zinc concentrations, zinc uptake was also reported to be nonsaturable, indicating passive

diffusion [3,119,125]. Notably, the so called ‘high zinc concentrations’ applied in these

studies varied from >200–1000 µM [119,125,135]. However, this might not be relevant in

vivo under normal zinc administration, as physiologically relevant concentrations in the

intestinal lumen after consummation of a standard meal were reported to vary around 100

µM [4,5,137], a concentration range where a saturable and carrier-mediated transport

kinetic was observed both in in vitro and in vivo studies.

Human fractional absorption of dietary zinc was estimated to be around 16–50% [15,88,138-

141] increasing inversely related to the oral zinc intake [11]. Accordingly, human zinc

resorption is more efficient from low zinc diets and was even shown to adapt to low dietary

zinc intake [15]. Moreover, net absorption is regulated by body zinc homeostasis and thus

dependent on the individual zinc status adapting to prolonged low zinc diets. Consequently,

Literature Review

16

zinc deficient humans and animals showed an increased fractional zinc resorption [89,142-

144], absorbing almost 92% of dietary zinc [89,144]. It has to be noted that zinc resorption is

also affected by its orally administered form. Herein, the net absorption was reported to be

higher from orally administered aqueous zinc solutions than the resorption of the same

amount of zinc included in a meal [119]. This is mainly because absorption of the mineral is

dependent on its bioavailability in the intestinal lumen and highly affected by food

components, which will be discussed in detail in section 2.2.3.

2.2.2.1 Intestinal Zinc Transporters

Intestinal zinc absorption is mainly mediated by ZIP-4 (SLC39A4), which imports ionic zinc

from the lumen into enterocytes [145,146], and ZnT-1 (SLC30A1), a basolateral membrane

protein exporting zinc on the basolateral site of enterocytes into the portal blood [147]. The

basolaterally localized transporter ZIP-5 (SLC39A5) and ZIP-14 (SLC39A14) complement these

two transporters by importing zinc from circulation into enterocytes [148,149]. Moreover,

the protein ZnT-5 variant B (SLC305B) was shown to be localized at the apical membrane of

enterocytes [132,137] and is discussed to function in a bidirectional manner, transporting

both luminal zinc into the enterocytes and cellular ions back into the lumen [132,134].

Earlier findings indicated involvement of the divalent metal transporter (DMT)-1, a broad

specific cation transporter, in intestinal zinc uptake [150]. The raise of ZIP-4 as the major

transporter for zinc uptake and contradictory results in several in vitro studies [151-155],

however, leads to questioning the role of DMT-1 in physiological zinc transport.

Even though the exact transport mechanisms of ZIP and ZnT transporters are not yet fully

investigated, it is known that these proteins transport ionic zinc [74,156]. Dietary zinc in the

intestinal lumen, however, is mainly complexed by food components influencing the actual

available and absorbable zinc concentration (in detail discussed in section 2.2.3, p. 20 ff.). In

addition to the uptake of the ionic form, zinc was also suggested to be absorbed as complex

with certain amino acids possibly utilizing another transport pathway than the ionic zinc

[157].

Literature Review

17

2.2.2.2 (Cellular) Regulation of Intestinal Zinc Absorption

The discovery of intestinal zinc transporters and elucidation of the role of the zinc binding

MT proteins in maintaining enterocytes’ zinc homeostasis contributed to the increased

knowledge of regulatory parameters of intestinal zinc resorption.

Figure 2.6 Regulation of intestinal zinc resorption

Shown is a potential regulatory mechanism of zinc resorption in enterocytes during (A) zinc excess (ZE), (B) adequate supply and (C) zinc deficiency (ZD) based on experimental data on the zinc-dependent expression pattern of zinc transporters and metallothionein (MT) in enterocytes and what we know about general zinc homeostasis in cells (refer to Figure 2.4). Enterocytes’ zinc homeostasis is discussed to be controlled by these proteins regulating the amount of intestinal absorbed and basolateral exported zinc [158]. Both protein and messenger ribonucleic acid (mRNA) of MTs are increasing with absorbed zinc concentration (A) and decreasing during ZD (C) [132,159-161]. Thus, together with increased sequestering of the cation into vesicles, due to upregulation of znt-2 [162] and znt-4 [162,163] in response to zinc intake, MTs might tightly regulate cytoplasmic free zinc levels in enterocytes (A,B). During zinc excess, the zinc importer Zrt-, Irt-like protein (ZIP)-4 at the apical membrane of enterocytes is endocytosed and degraded [164-166] (A), while during ZD its mRNA is stabilized leading to accumulation of the protein at the apical membrane [166,167] (C). Basolateral ZIP-14 is unaffected by dietary zinc [163] (A-C), while the ZIP-5 protein is decreased during zinc deficiency (C). Data for the basolateral zinc transporter (ZnT)-1 during ZE and ZD are contradictory and scarce regarding its differential expression in humans. The mRNA and protein of ZnT-1 are decreased during ZD [168] (C) and both down- und up-regulated [117,132,147,162] after zinc supply (A, C). The bidirectional transporter ZnT-5B is not affected during ZD, but was reported to localize to the plasma membrane as response to dietary zinc [134,137].

In the intestine, mainly MT-1 and MT-2 are expressed [115], these MT isoforms are

ubiquitously found in all tissues with particularly high expression in liver, pancreas, kidney

and intestine [169]. In the following, the singular form of MT refers to both MT-isoforms for

Literature Review

18

the sake of convenience and readability. Similar to its role in cellular zinc homeostasis in

general, MT plays an important role in regulating enterocytes’ zinc homeostasis binding zinc

that is absorbed into the cells [131]. Thus, the protein controls free levels of the cation and is

discussed to mediate zinc trafficking through the cell as well as its transfer to other proteins

such as zinc transporters (Figure 2.4 and Figure 2.6) [106,170]. Hence, MT’s zinc buffering

and muffling properties might consequently regulate the amount of zinc that is finally

exported into the portal blood and distributed into the body.

Expression of MT is related to changes in enterocytes’ zinc levels as elevated cellular free

zinc itself induces mt expression via the metal regulatory transcription factor 1 (MTF-1) [171]

(Figure 2.5). Protein and messenger ribonucleic acid (mRNA) levels of intestinal MT was

reported to increase in response to elevated dietary zinc in animals and humans in vivo

[132,159-161], acting as an initial defense mechanism against high luminal zinc

concentrations [161]. Furthermore, it was suggested that its upregulation has an impact on

zinc transport kinetics and decreases the luminal zinc absorption [125,172-174]. In this

regard, serum and body zinc levels were observed to decrease with elevated intestinal MT

[117,173,175]. Interestingly, in an earlier study, MT was also suggested to be involved in zinc

export from enterocytes back into the intestinal lumen [174]. In fact, luminal secretion of MT

after treatment with physiological zinc concentrations was observed in a three-dimensional

in vitro intestinal model, indicating that MT might also mediate enterocytes’ zinc

homeostasis by apically sequestering excess zinc [176]. As already mentioned in section

2.2.1 there is evidence that MT is not the only protein involved in cellular zinc trafficking

[116-118,161]. In this context, Cousins and coworkers proposed the involvement of the

cysteine rich intestinal protein (CRIP) as an additional mediator of enterocytes’ zinc

trafficking possibly competing with MT [177-179]. However, CRIP was later shown to be

expressed in nearly all organs and suggested to play a role in immune response [180]. It is

most likely, that there is another moiety involved in zinc muffling and transfer through the

cell, possibly similar to chaperones involved in enterocytes’ iron and copper homeostasis

[181].

Similar to MT, intestinal zinc transporters are not only part in the maintenance of

enterocytes’ zinc homeostasis but are also decisive for zinc resorption, thereby regulating

body zinc homeostasis. The main intestinal zinc importer ZIP-4 is regulated by dietary zinc in

a transcriptional, translational and post-translational manner [182]. ZIP-4 is essential for zinc

resorption. This is particularly demonstrated in the zinc malabsorption disease

acrodermatitis enteropathica which originates from different mutations in the gene

encoding the human ZIP-4 protein [145,146,183,184]. Moreover, surface localization of ZIP-4

of enterocytes is regulated by zinc concentrations in the cytoplasm [166]. Under zinc

deficiency, zip-4 mRNA was shown to be stabilized [184-186] and the protein accumulated at

the apical plasma membrane of enterocytes resulting in higher zinc uptake [166,167]. Zinc

repletion results in endocytosis of the protein [166] and ubiquitin-mediated degradation at

even higher zinc concentrations [164,165], but not downregulation on the mRNA level [132].

In contrast to ZIP-4, zip-5 mRNA abundance is not changed by dietary zinc but its translation

was reported to be zinc-dependent [185]. During zinc insufficiency, the basolateral plasma

Literature Review

19

membrane protein is internalized while its mRNA is associated with polysomes, minimizing

the secretion of body zinc from the blood into the intestinal tract [185,186]. After zinc

repletion the protein is again rapidly accumulated at the membrane [185]. Consequently,

ZIP-5 is important for the control of systemic zinc homeostasis and was suggested to even be

involved in sensing zinc body status [187].

Regulation of ZIP-14 mRNA on the other hand was not altered during dietary zinc deficiency

or excess in mice [163].

ZnT-1 mRNA expression is zinc-dependent and, similar to MT, was shown to be regulated by

MTF-1 [147,162]. MTF-1 directly senses cytoplasmic zinc concentration in enterocytes and

regulates ZnT-1 expression guaranteeing a reasonable export of the cation into the portal

blood and controlling intracellular free zinc levels [106]. To this end, in animal studies, high

oral zinc doses increased protein [147] and mRNA expression [117,147,162]. Conversely, znt-

1 mRNA and the corresponding protein were downregulated after zinc supplementation in

humans in vivo [132]. Zinc restriction on the other hand resulted in downregulation of mRNA

and protein in weanling rats [168], whereas no effect was reported in mature rats [162].

Interestingly, in contrast to the aforementioned MT knockout mice, ZnT-1 knockout mice

already died in an early embryonic state [188].

The apically localized bidirectional zinc transporter ZnT-5B was reported to be

downregulated [132] and upregulated [134,137] with elevated cellular zinc availability in in

vitro and in vivo studies. This converse regulation indicates a rather complex role in zinc

homeostasis and was suggested to be based on both transcriptional repression and

stabilization of its mRNA [133]. Aside of its apically located variant B, ZnT5 is also distributed

in cytoplasm of enterocytes and goblet cells and was shown to be essential for zinc

homeostasis, as ZnT-5 knockout mice displayed impaired growth and bone development

[189].

All in all, this zinc sensitive regulation of intestinal zinc transporters and metallothionein

expression is pivotal for the control of enterocytes’ zinc absorption and release into

circulation. However, the distinct molecular parameters by which the transporters are

regulated and zinc is absorbed remain to be fully understood. To this end, the involvement

of a systemic moiety in regulating zinc transporters, such as the humoral factor hepcidin,

was already suggested and is topic of ongoing research [130,190].

Literature Review

20

2.2.3 Zinc in Nutrition and Intestinal Bioavailability

2.2.3.1 Human Zinc Requirements

To maintain human body zinc homeostasis, the essential trace element has to be supplied

with food on a daily basis. Herein, daily intestinal and non-intestinal losses of endogenous

zinc have to be counterbalanced with the nutrition [76,191]. Based on several human studies

these losses include fecal zinc excretions and excretions with urine, sweat, menstrual flow

and semen (for adults) as well as zinc losses with hair, nails and desquamated skin [138] and

were estimated to be around 2–3 mg per day for healthy adults [13,76,192,193]. Moreover,

additional zinc is needed for particular physiological states such as pregnancies, lactation or

early infancy [191]. To this end, nowadays human requirements are mostly estimated using

a factorial approach which considers the overall zinc losses including the additional

physiological requirements as well as bioavailability of the mineral from the diet [138,191].

The latter is very important as the actual amount of absorbed zinc is highly dependent on its

availability in the intestinal lumen from the diet. Therefore, a sufficient zinc supply is only

guaranteed when accessibility and availability of the mineral from food is taken into account.

Table 2.3 depicts daily recommendations for dietary zinc intake from different governmental

agencies and non-governmental organizations. Estimations for the physiological

recommendations from the World Health Organization (WHO) differ between high (50%),

moderate (30%) and low (15%) zinc bioavailability which represents molar phytate : zinc

ratios <5, 5-15 and >15 and other factors impacting zinc absorption in diets [13]. Likewise,

the European Food Safety Authority (EFSA) includes different phytate levels for their

recommendations for adults using a trivariate model from Miller et al. to assess the

relationship between total absorbed zinc, dietary phytate and dietary zinc [194]. In contrast,

the German Society for Nutrition (ger. Deutsche Gesellschaft für Ernährung, DGE) does not

classify their recommended daily zinc intake by its bioavailability from food [195].

Literature Review

21

Table 2.3 Recommended daily allowance for dietary zinc intake for selected life-stages

WHO [13] EFSA [192] DGE [195]

Age, Sex RW [kg] RNI [mg/d]

Age, Sex RW [kg] PRI [mg/d] Age, Sex RDI [mg/d] High

a Mod

b Low

c

7-12 mos 9 0.8d; 2.5

e 4.1 8.4 7-11 mos 2.9 < 4 mos 1.0

1-3 yr 12 2.4 4.1 8.3 1-3 yr 11.9 4.3 4-12 mos 2.0

4-6 yr 17 2.9 4.8 9.6 4-6 19.0 5.5 1- 4 yr 3.0

7-9 yr 25 3.3 5.6 11.2 7-10 28.7 7.4 4-7 yr 5.0

10-18 yr, m 49 5.1 8.6 17.1 m f m f 7-10 yr 7.0

10-18 yr, f 47 4.3 7.2 14.4 11-14 yr 44.0 45.1 9.4 9.4 m f

19-65 yr, m 65 4.2 7.0 14.0 15-17 yr 64.1 56.4 12.5 10.4 10-13 yr 9.0 7.0

19-65 yr, f 55 3.0 4.9 9.8 Age Phytate [mg/d] 13-15 yr 9.5 7.0

> 65 yr, m 65 4.2 7.0 14.0 ≥ 18 yr 300 68.1 58.5 9.4 7.5 15-19 yr 10.0 7.0

> 65 yr, f 55 3.0 4.9 9.8 ≥ 18 yr 600 68.1 58.5 11.7 9.3 19-25 yr 10.0 7.0

Pregnancy ≥ 18 yr 900 68.1 58.5 14.0 11.0 25-51 yr 10.0 7.0

1st

trimester 3.4 5.5 11.0 ≥ 18 yr 1200 68.1 58.5 16.3 12.7 51-65 yr 10.0 7.0

2nd

trimester 4.2 7.0 14.0 ≥ 65 yr 10.0 7.0

3rd

trimester 6.0 10.0 20.0 Pregnancy +1.6 Pregnancy 10.0

Lactation Lactation +2.9 Lactation 11.0

0-3 mo 5.8 9.5 19.0

3-6 mo 5.3 8.8 17.5

6-12 mo 4.3 7.2 14.4

BV, bioavailability; EFSA, European Food Safety Authority; DGE, German Society for Nutrition; ger.: Deutsche Gesellschaft für Ernährung; f, female; m, male, mos, months; PRI, population

reference intake; RDI, recommended daily intake; RNI, recommended nutrient intake; RW, reference weight; WHO, World Health Organization; yr, years; aHigh bioavailability (50%);

bModerate

bioavailability (30%); c

Low bioavailability (15%); dexclusively breastfed infants (bioavailability 80%);

enot exclusively breastfed.

Literature Review

22

2.2.3.2 Zinc in Nutrition

The amount of food that has to be consumed to meet the daily recommended levels highly

depends on its zinc content and as already mentioned, on the bioavailability of this

micronutrient. In Table 2.4 the general zinc content for selected foods is depicted,

illustrating the general variance of the trace element in esculents. Highest zinc

concentrations are found in animal products from pork, beef, poultry, fish and shellfish,

particularly in their flesh and organs. Regarding plant-based food, zinc content is high in

seeds and nuts, followed by legumes (beans and lentils) and whole-grain cereals while lesser

amounts can be found in vegetables, fruit and refined cereal grain [138].

Table 2.4 Zinc and phytate content, as well as phytate : zinc-molar ratios of foods adapted from [138]

Food group Zinc content

[mg/100g]

Phytate content

[mg/100g]

Phytate : zinc

molar ratio

Liver, kidney (beef, poultry) 4.2-6.1 0 0

Meat (beef, pork) 2.9-4.7 0 0

Poultry (chicken, duck, etc.) 1.8-3.0 0 0

Seafood (fish, etc.) 0.5-5.2 0 0

Eggs (chicken, duck) 1.1-1.4 0 0

Dairy (milk, cheese) 0.4-3.1 0 0

Seeds, nuts 2.9-7.8 1,760-4,710 22-88

Beans, lentils 1.0-2.0 110-617 19-56

Whole-grain cereal 0.5-3.2 211-618 22-53

Refined cereal grain 0.4-0.8 30-439 16-54

Bread 0.9 30 3

Fermented cassava root 0.7 70 10

Tubers 0.3-0.5 93-131 26-31

Vegetables 0.1-0.8 0-116 0-42

Fruits 0-0.2 0-63 0-31

Phytate : zinc-molar ratio was estimated based on (mg phytate/660) / (mg zinc/65.4) [138].

An insufficient zinc absorption results in zinc deficiency with severe health consequences,

such as poor growth, retardation in development, decreased brain function and impairment

of the immune defense [79,196-198]. Severe and prolonged deficiency increases the risk of

infection, often connected with diarrhea and impaired wound healing, all causing high

morbidity rates [199,200]. According to the WHO, one third of the world’s population are at

risk for zinc deficiency [13], placing this micronutrient deficiency among the ten highest risks

for human health in developing countries with high morbidity rates [10].

The main obstacle in this situation, however, is the lack of a suitable biomarker for

physiological zinc status and thus a low possibility to recognize insufficient zinc absorption,

particularly in the early stages of a mild zinc deficiency [76,201]. Inadequacy of zinc status is

often connected to an insufficient food supply but mostly dependent on poor bioavailability

Literature Review

23

of the mineral from the consumed diet [11]. In this context, particularly phytate content of

consumed food is severely impairing zinc bioavailability in the intestine, the underlying

physico-chemical interactions will be discussed in detail in section 2.2.3.3 on p. 23 ff. Herein,

the molar ratio of food’s phytate and zinc content was shown to be more important than the

phytate content of the product itself [202,203]. As illustrated in Table 2.4, plant-based diets

contain higher phytate levels than mixed diets, providing the person with less intestinally

available zinc than from meat-based diets [204,205]. Therefore, people suffering from zinc

deficiency are more likely to live in developing or poor countries [206,207]. Vegans [204],

vegetarians [204], elderly [138,207] and people with disorders connected with a diminished

zinc absorption, such as acrodermatitis enteropathica or celiac disease [208] as well as

diseases that cause increased zinc loss, such as Crohn’s disease [209] and inflammatory

bowel disease [210], are also at risk.

Symptoms of zinc deficiency are reversible when the essential trace element is administered

again [76,196,197,199]. In most cases, (pharmaceutical) zinc supplementation in addition to

dietary zinc provides a convenient option to compensate inadequate zinc intake,

malabsorption or increased zinc loss due to intestinal diseases [211-213]. For this a variety of

zinc compounds are available: zinc complexes with aspartate, acetate, ascorbate, citrate,

gluconate, histidine, methionine, oxide, chloride or sulfate [138,214].

2.2.3.3 Intestinal Zinc Bioavailability

According to the International Zinc Nutrition Consultative Group (IZiNCG), zinc bioavailability

from a mixed or vegetarian diet, based on refined cereal grains, was estimated to be 26–

34%, whereas 18–26% was resorbed from an unrefined cereal based diet [138]. As already

mentioned above, the actual absorbed amount of zinc not only depends on the zinc content

of the consumed diet but is highly affected by its intestinal zinc bioaccessibility and -

availability. Bioavailability describes the amount of zinc absorbed by the cells that is

subsequently released into the blood and therefore available for the systemic circulation and

body homeostasis [215]. The term bioaccessibility in this context includes the potentially

free and absorbable zinc concentration in the intestinal lumen and represents a preselection

in addition to bioavailability of the nutrient [215,216].

Due to the digestion process, a wide range of different zinc species is available in the

intestinal lumen, mainly complexed by food-derived macromolecules or even low molecular

weight ligands [3]. Hence the accessibility and availability of the essential micronutrient is

certainly dependent on its solubility and stability of the respective complexes in the

intestinal lumen. This is affected by the particular diet as well as by physiological factors such

as the mucus layer and the intestinal fluid. Together these luminal factors alter the

speciation of the metal as well as its luminal free and available concentration, consequently

affecting its absorption by the intestinal epithelium. In the following, the beneficial or

inhibitory impact of these diet-derived and physiological luminal factors on the intestinal

zinc bioavailability will be briefly summarized.

Literature Review

24

2.2.3.4 Dietary Factors Recognized to Influence Zinc Absorption

The fractional zinc resorption is mainly dependent on the zinc intake itself, as the efficiency

of this process is declining with increased zinc consumption [15,16,87] (refer to section 2.2.2,

p. 15 ff.). Additionally, the zinc species itself influences its intestinal absorption, which is

particularly of great interest regarding their use as pharmaceutical zinc supplement. Many

studies aimed to characterize different zinc complexes regarding their bioavailability, which

is mainly dependent on their solubility in intestinal environment. Herein, zinc oxide has the

lowest availability, whereas results for other zinc complexes, with citrate, chloride,

gluconate, sulfate, amino acids or even ethylene-diamine-tetra-acetic acid (EDTA) are

contradictory suggesting that their bioavailability is quite comparable [121,138,217-220].

Particularly phytate, a natural component of plants, severely decreases the intestinal zinc

bioavailability and is regarded as the main inhibitor of zinc resorption. Notably, the term

phytate includes magnesium, calcium or potassium salts from phytic acid and comprises a

mixture of myo-inositol hexa-, penta-, tetra- and triphosphates [221]. Actually, tetra- and

triphosphates were described to have little impact on zinc absorption, whereas hexa- and

pentaphosphates of inositol severely impaired the intestinal zinc availability in in vivo studies

[221-223]. Nevertheless, phytate can be hydrolyzed by phytase, an enzyme that degrades

the molecule to tetra- and triphosphates, consequently increasing the zinc availability

[224,225]. In contrast to sheep and pigs, which are able to degrade phytate with their own

intestinal phytase, levels of this enzyme in human small intestine are very low and thus

phytate degradation is highly dependent on phytogenic and microbiotic phytase

[221,224,226,227]. Phytogenic phytase, particularly in grains, can be activated during

fermentation and food processing [224,225,228], subsequently enhancing zinc absorption

[228].

Zinc is bound by phosphates of the molecule yielding a 2:1 stoichiometry of the zinc-

phytate-complex [229] with strong binding affinities: 1. 8·106 L mol-1 (site 1) and 8·104 L mol-1

(site 2) (myo-inositol hexaphosphates at 37°C) [230]. Moreover, stability of the zinc-phytate-

complex is pH dependent, illustrating moderate solubility at low pH and poor solubility at pH

7 [230]. Hence, zinc must not be bound to phytate when consumed with the meal [231], at

intestinal pH (luminal pH 6–7.4), however, phytate binds the cation effectively, forming

stable complexes with low solubility and bioaccessibility [232,233]. Consequentially,

complexed zinc is not available for absorption and is excreted with the feces [234]. Phytate is

also discussed to severely impact body zinc homeostasis by binding endogenous zinc that is

excreted into the lumen by inhibiting its reabsorption [3]. Thus the total phytate content of

the diet can affect the overall zinc bioavailability in one meal. More precisely, the inhibitory

effect of phytate on zinc absorption is concentration-dependent and the molar phytate :

zinc-ratio of the diet (Table 2.4) is applied to estimate zinc bioavailability [13] (for details

refer to section 2.2.3.2., p. 22 ff.). Significant changes in zinc absorption in humans were

observed beginning with a molar ratio of 5, reducing fractional zinc absorption from 21%

without phytate to moderate levels of 11-16% at a molar phytate : zinc-ratio of 5–15 and

lower 4-11% bioavailability at molar ratios >15 [235]. Additionally, these complexes were

Literature Review

25

observed to be stronger when calcium is present, suggesting calcium as an inhibitor of zinc

absorption in the presence of phytate [235]. Contradictory, the calcium content did not

increase the inhibition of zinc absorption by phytate in several human dietary studies

[141,231,236]. Other than phytate, fiber like cellulose seems to have no significant impact on

zinc absorption [225,234].

Dietary protein levels were shown to positively correlate with zinc uptake [12,141]. In

general, human zinc absorption is higher in the presence of protein from animal-sources

than plant-based protein, mainly because of the phytate content of the latter [237]. In fact,

addition of animal protein to vegetable-based food significantly improved its zinc

bioavailability in vivo. [238]. This beneficial impact, however, was discussed to be rather

based on the fact that protein itself counteracts the impairing effect of phytate and not

because of its animal origin [239]. It has to be noted that to investigate the role of plant-

derived protein on intestinal zinc uptake, phytate has to be removed in advance, as it would

strongly affect the zinc availability [11]. In this manner, zinc resorption from soy-protein

increased after phytate-removal [240]. Casein is the main zinc binding protein in cow’s milk

and was suggested to be the main reason why zinc from cow’s milk is less bioavailable than

from human milk [241]. This, however, was not confirmed when adding isolated casein to

test meals [239]. In this in vivo human study only the addition of bovine serum albumin and

soy protein decreased zinc absorption [239].

Protein is digested in the gastrointestinal tract and degraded into peptides or amino acids

[30]. These low molecular weight (LMW) compounds form complexes with zinc increasing its

bioavailability by enhancing the solubility of the cation in the intestinal lumen [11] and

possibly by being resorbed via amino acid transporters [157]. The latter increase their

relevance for zinc supplementation in zinc malabsorption diseases, such as acrodermatitis

enteropathica [157]. Several studies investigated the impact of amino acids on zinc

absorption, yielding contradictory results. For example, histidine was shown to increase zinc

bioavailability in humans [140,242] and was even reported to elevate its absorption from

zinc-phytate-complexes together with methionine [243]. Moreover, tryptophan, histidine,

imidazole, proline, and pyroglutamate were shown to increase zinc absorption in perfused

rat intestine [244]. In contrast, cysteine and histidine had no beneficial effect on zinc uptake

of isolated rat enterocytes in vitro [245] and methionine even reduced zinc resorption in rats

in vivo [246].

The interrelation of micronutrients on their resorption is still topic of ongoing research. The

possible inhibitory impact of calcium on the intestinal zinc bioavailability was already

discussed above. Further, negative effects of both heme-iron and inorganic iron on zinc

absorption were reported by several in vivo studies [242,247-250], whereas the effect was

greater when iron was administered as aqueous solutions than together with the meal

[242,251]. Copper on the other hand seems to have no decreasing impact on zinc absorption

[252]. High zinc doses though were described to crucially affect intestinal copper resorption,

thus a balanced zinc and copper nutrition is by all means very important [253]. Lastly,

cadmium [254] and tin were reported to inhibit zinc absorption [255], although it has to be

Literature Review

26

noted that the latter study applied unrealistically high amounts of tin. Nonetheless, naturally

occurring tin concentrations were observed to severely affect zinc homeostasis by increasing

urinary excretion of zinc [256].

In contrast to its beneficial role in iron resorption [251], ascorbic acid has no effect on

intestinal zinc bioavailability [237,257,258]. Citrate on the other hand positively influences

zinc availability and thus zinc-citrate-complexes are already used as zinc supplements [219].

Citrate is the main low-molecular weight ligand binding zinc in milk possibly influencing

zinc’s bioavailability from milk and milk products [259]. Concentrations of zinc-citrate-

complexes are higher in human milk compared to cow’s milk [260], which might explain the

higher human zinc absorption from human milk than from cow’s milk [241].

Lastly, chemical and physical food processing were also shown to affect the bioaccessibility

and -availability of the essential cation [261]. In this context, particularly the formation of

heat-derived zinc binding ligands such as Maillard browning products [262,263] decreased its

availability, whereas fermentation or germination elevated its accessibility due to phytate

reduction [224,264].

2.2.3.5 Physiological Factors Discussed to Regulate Zinc Absorption

Aside of dietary components various physiological factors in the intestinal lumen influences

the solubility of the mineral and its subsequent availability for the intestinal epithelium.

Most notably, at intestinal pH of 6–7.4 [31] the divalent zinc cation tends to form insoluble

hydroxypolymers (Zn(OH)2) which would affect its luminal availability [265]. About three

decades ago, the intestinal mucus layer was suggested to have an impact on luminal zinc

uptake, possibly binding luminal zinc and enhancing its availability for the intestinal

epithelium [130,266]. Subsequent animal studies confirmed this hypothesis and even

indicated zinc buffering properties of this physical barrier [23,24,267]. In fact, mucins were

already shown to bind several other metals, such as iron, lead, calcium and aluminum

[46,268,269], with increased affinity from M+ < M2+ < M3+ [24]. Consequentially, this

competitive binding was suggested to influence the luminal availability of these metals for

the underlying epithelium which might also explain the mutual interference observed in

intestinal trace element resorption. Hence, there is evidence that mucins represent an

important luminal factor for zinc resorption, influencing luminal accessibility of the cation

and consequently its bioavailability. Nevertheless, there is still a lot we do not know about

this physiological factor, including its zinc binding properties as well as its distinct role in

luminal zinc uptake into enterocytes.

Most recently, systemic factors were discussed to play a role in intestinal zinc resorption by

regulating the uptake and transport into the systemic circulation. In this context, Hennigar et

al. studied the impact of the liver-derived humoral factor hepcidin, which plays an important

role in iron resorption [270], on enterocyte’s zinc transport [190]. Herein, basolaterally

added hepcidin was shown to reduce serosal zinc transport into the blood by post-

translationally downregulating ZnT-1 in Caco-2 cells. Furthermore, zinc concentration in

Literature Review

27

enterocytes increased and mt-1a was upregulated, possibly controlling subcellular zinc pools

[190].

In this manner, zinc resorption seems to not only be influenced by luminal factors affecting

its bioavailability, but might also be managed by basolateral factors possibly providing a

direct regulating mechanism of systemic zinc supply by the intestine.

2.3 In vitro Studies on Intestinal Zinc Resorption

In the past 50 years, several analytical approaches have been applied to investigate

intestinal zinc resorption and its underlying mechanisms. The latter was mainly elucidated

with ex vivo animal studies, such as everted rat gut sacs [121,122], isolated vascularly

perfused rat intestine [92,131,143,172,271] and intestinal brush-border membrane vesicles

from rat [135] and pig [272,273] small intestine. Moreover, some human studies using

perfused intestine were performed as well [4,5]. Conversely, zinc transport kinetics,

fractional absorption, efficiency of transport, and the impact of dietary components on zinc

bioavailability were studied with in vivo human and animal studies using mostly (stable)

isotope techniques [15,87,234,259]. Distinct processes on the cellular level, like the role of

zinc transporters and metallothionein, however, were mostly investigated with in vitro cell

models [132,137,148,166,190], as they provide a standardized and easy platform to study

various cellular processes.

Furthermore, in times were the three R paradigm of animal testing [274] is getting more

important, calling for refined and reduced animal studies, the application of suitable in vitro

cellular models has to be increased to achieve the “third R” of replacing animal experiments

[275]. The aim of this thesis was to establish a three-dimensional in vitro intestinal model to

investigate intestinal zinc transport and to further elucidate regulatory parameters of its

resorption. Hence, the following should only focus on the application of in vitro cell models

in the investigation of intestinal zinc resorption to summarize the hitherto obtained results

using in vitro intestinal models.

2.3.1 Investigation of Zinc Transport using in vitro Intestinal Models

Until now, predominantly the Caco-2 cell model was used to elucidate human intestinal zinc

resorption and transport with in vitro studies. This model is widely employed to determine

the absorption of various drug compounds as well as the uptake and transport kinetics of

(micro-) nutrients [276-279]. When cultured for 21 d, the colon carcinoma cell line Caco-2

differentiates resembling human enterocytes functionally and morphologically [280,281].

They form an intact monolayer with important characteristics of the intestinal epithelium,

including microvilli as well as tight junction proteins, and express several important proteins

of intestinal transport [282,283]. Furthermore, this model is a validated intestinal model to

study drug absorption, recognized by the FDA, giving promising correlations for fractional

resorption of several drug components [284].

Literature Review

28

Figure 2.7 Schematic representation of the three-dimensional in vitro intestinal model Caco-2

(A) The intestinal epithelium in vivo is mostly composed of enterocytes and goblet cells [28], which represent about 90% of intestinal cells of the brush border membrane [27,285], and are covered by a viscoelastic gel: the mucus layer. This physical barrier is synthesized and secreted by goblet cells and serves as a protective layer for the underlying epithelium. (B) Three-dimensional Caco-2 monoculture in the so called “Transwell® system”. The intestinal cell line Caco-2 is cultured in transwell inserts on a permeable membrane, mostly composed of polycarbonate. This forms three compartments: an apical compartment representing the intestinal lumen, a basolateral side illustrating the serosal blood side of enterocytes, and the intestinal barrier which is formed by differentiated Caco-2 cells.

In three-dimensional cultures, the cellular monolayer, seeded on transwell inserts, forms an

intact barrier mimicking the intestinal epithelium, whereas the apical transport chamber

depicts the intestinal lumen and the basolateral side illustrates the serosal blood side [18].

By these means, uptake and transport of a substance of interest can be tracked in all three

compartments. In detail, transport of apically added zinc inside the cells and via the

intestinal epithelium into the blood side can thus be determined.

Whilst zinc resorption and transport kinetics (Table 2.5) were characterized using three-

dimensional Caco-2 models, two-dimensional culture of these cells was additionally applied

to investigate zinc uptake parameters. Furthermore, this model was widely used to study the

effect of various dietary food components on intestinal zinc bioavailability [157,279,286-301]

and was also applied to elucidate the regulatory role of intestinal zinc transporters and

metallothionein in zinc resorption [132,134,137,302-310]. Notably, the impact of dietary zinc

on zinc transporters and metallothionein expression in Caco-2 cells is very well comparable

to the homeostatic regulation of these proteins in human small intestine [132]. For detailed

results of these studies, refer to Table S5 and S6 in Appendix F, which summarizes their main

study design and outcome.

Literature Review

29

Table 2.5 summarizes studies of zinc resorption using three-dimensional Caco-2

monocultures, depicting parameters of cell models including buffer composition and the

main outcome of the study. Regardless of the detailed experimental setting almost all

transport studies obtained with Caco-2 models observed a saturable apical zinc uptake and

transport kinetic. Interestingly, Km values for zinc uptake of 41 µM [127], 11.7 µM [125]

obtained with Caco-2 are in the same order of magnitude as those observed with in vitro rat

intestine (Km = 10-12 µM, [125]), rat perfused intestine (Km = 32 µM, [92]; Km = 29 µM, [136];

Km = 55 µM, [131]), or brush-border membrane vesicles from pig (Km = 67 µM, [254] or rat

(Km = 24 µM, [311]). Accordingly, Caco-2 cells are very well suitable to study intestinal zinc

uptake. Of note, two different studies observed no saturable zinc uptake from the apical

side, both using regular cell culture medium with 10% FCS for their apical zinc treatment

[128,129] and very high and not physiologic zinc concentrations [129]. Therefore it is very

important that the amount of applied zinc corresponds to physiological concentrations in

lumen in vivo, particularly when analyzing transport and uptake kinetics to prevent the

analysis of artefacts. Additionally, medium or buffer constituents have to be carefully

considered when investigating metal uptake and transport with in vitro models, particularly

with regard to metal binding components [19]. In fact, speciation of zinc in cell culture

medium or buffer severely affects its availability and cellular uptake [19,20].

It has to be noted that the in vitro Caco-2 model lacks one very important factor of the

intestinal epithelium: the mucus layer. The intestinal epithelium is not only composed of

enterocytes but also includes goblet cells, producing and secreting mucins that cover the

whole gastrointestinal tract (in detail discussed in section 2.1.2, p. 7 ff.). The co-culture of

Caco-2 cells together with the goblet cell line HT-29-MTX improves this disadvantage,

forming a mucus layer, which was shown to cover the whole cell layer [312]. This Caco-2/HT-

29-MTX model is well characterized [313-316] and was already used to investigate the

resorption of different metal species [312,317-319], the effect of nanoparticles on nutrient

absorption [320], and bacterial adhesion [321]. However, this model has never been used to

study intestinal zinc resorption, even though mucins are discussed to play an important role

in this process.

Literature Review

30

Table 2.5 Zinc transport studies using in vitro intestinal models

Cell model Incubation parameter Zinc Quantification

Main Outcome Reference

Caco-2 cells Cultivation time: 14 d 3D Transwell (PC membrane) 14 d

ZnCl2

20 µM (Kinetic 0-50min) 0-100 µM (10 min) (in salt buffer on apical and basolateral side) Inhibitor: oubain, vanadate, dinitrophenol, sodium cyanide, ammonium vanadate Potential zinc ligands: histidine, cysteine, proline, glutathione

65Zn

- cellular zinc uptake is saturable process - Km = 41 µM

Vmax= 0.3 nmol/cm²/10min - basolateral zinc uptake was partially inhibited

(30%) by oubain and vanadate, suggesting an involvement of the (Na-K)-ATPase in serosal uptake

- apical Zn uptake was not affected by metabolic inhibitors and ligands

- basolateral zinc uptake (50 min) ~0.47 nmol/cm² - zinc transport ~0.8 nmol/cm² (20 µM; after 50

min) - transport from basolateral to apical is higher than

from the apical to the basolateral compartment

Raffaniello et al. 1992 [127]

Caco-2 cells Cultivation time: 18-21 d 3D Transwell

ZnSO4 10-1000 µM (for 90 min) 10 nM 1α,25-dihydroxyvitamin D3 (preincubation for 72 h) + 100 µM ZnSO4 (for 90 min) Apical: MES-buffer with NaCl, KCl, MgSO4, CaCl2, glutamine, glucose, Basolateral: 2.5 mg/mL BSA in Hepes with NaCl, KCl, MgSO4, CaCl2, glutamine, glucose,

65Zn

- saturable zinc uptake kinetic up to 1000 µM - Km = 226 µM - zinc transport rate (after 90min):

~ 10 µM: ~ 0.12 nmol/cm² ~ 50 µM: ~ 0.25 nmol/cm²

- zinc transport increased in vitamin D3 incubated cells

- mt-2a mRNA and protein was increased with increased zinc concentrations

- Crip mRNA (30% less expressed in Caco-2 cells than in rat mucosa) was decreased by vitamin D3 treatment

Fleet et al. 1993 [322]

Literature Review

31

Caco-2 cells Cultivation time: 21 d 2D,3D Transwell (PE membrane)

n.a. 1-200 µM (in DMEM + 10% FCS on apical and basolateral side) for 0 - 30 h

65Zn

- saturable zinc uptake at the basolateral membrane

- apical zinc uptake and zinc transport, both from apical to basolateral and vice versa, were not saturable

- higher transport from apical to basolateral - transport rate

50 µM: 6 pmol/h/cm² - transport from apical to basolateral was

independent from basolateral zinc concentration - study indicates that zinc uptake and transcellular

movement are different at the apical and basolateral side

Finley et al. 1995 [128]

Caco-2 cells Cultivation time: 14 - 16 d 3D Transwell (Polyethylene terephthalate membrane)

ZnSO4

0-1000 µM (in DMEM + 10% FCS on apical) and 0-450µM (in DMEM + 10% FCS on basolateral side) for 24 h

FAAS

- applied 0-1000 µM zinc on apical or 7 – 450 µM zinc basolateral side

- transport occurs from both sides to the other compartment

- accumulation in the cells was low, particularly when zinc was added on the apical side

- zinc toxicity on cell viability and integrity of the intestinal barrier (TEER) 0-2000 µM zinc:

- observed higher toxicity when adding high zinc concentrations to the basolateral side

Rossi et al. 1996 [129]

Caco-2 cells Cultivation time: 18-21 d 3D Transwell (PC)

ZnCl2 50-200 µM (in serum free medium on apical and basolateral side) for 6 h, 12 h, 24 h

65Zn

- zinc transport an MT secretion (HPLC analysis) - this study suggest that MT is secreted into the

gastrointestinal lumen and plays a role in intestinal zinc uptake

- zinc transport (after 6 h) - 100 µM:~ 2.0 nmol/cm²

Moltedo et al. 2000 [176]

Literature Review

32

Caco-2 cells Cultivation time: 21 d 3D (PES-HD membranes)

ZnSO4

5 µM or 25 µM (in DMEM + 10% FCS on apical and basolateral) (preincubation for 7d)

65Zn

- zinc uptake and transport were measured in both apical (AP) and basolateral (BL) directions

- rate of apical zinc uptake and transport rate to basolateral was lower in cells pretreated 25 µM zinc

- basolateral zinc release was higher in cells treated with 25 µM

- cellular zinc uptake 2-3 nmol mg-1

protein - induction of MT (analyzed using radiolabeled

cadmium) was zinc-dependent, increasing with zinc concentration

Reeves et al. 2001 [323]

Caco-2 cells Cultivation time: 21 d 3D Transwell (PC)

ZnSO4

15.6 - 500 µM ( apical: KHB buffer, basolateral: KHB-buffer + 5% BSA)

ICP-MS

- comparison with zinc transport across isolated rat small intestine

- rat: Km = 10-12.1 µM - Caco-2 Km = 11.7 µM

Vmax = 31.8 pmol min-1

cm-2

- transport across Caco-2 monolayers is carrier-

mediated and energy-dependent - zinc transport into basolateral chamber followed

a saturated process

- transport rate: 50 µM: 39 pmol min

-1 cm

-²

- mRNA expression of zip-4, zip-5, znt-1, mt1, mt2 in duodenum, jejunum and ileum of isolated rat small intestine

Yasuno et al. 2012 [125]

Caco-2 cells Cultivation time: 17 d 3D Transwell (Polytetrafluoroethylene)

ZnSO4 100 µM (serum free medium on apical and basolateral side)

for 3-24 h 1 µM hepcidin

67Zn

- hepcidin reduces basolateral zinc export by post-translationally downregulation of ZnT-1

- cells incubated with hepcidin showed less zinc export while cellular zinc and mt-1a mRNA increased; cell surface ZnT-1 as well as ZnT-1 protein decreased

- hepcidin might play a role in controlling zinc resorption and enterocytes’ subcellular zinc pool

Hennigar et al. 2016 [190]

3D, three-dimensional; BSA, bovine serum albumin; DMEM, Dulbecco’s Modified Eagles Medium; FAAS, flame atomic absorption spectrometry; FCS, fetal calf serum; HBSS, Hank's Balanced Salt Solution; HD, high density; ICP-MS, inductively-coupled plasma mass spectrometry; KHB, Krebs-Henseleit buffer; n.a., not available; PC, polycarbonate; PE, polyethylene; PES, polyester; Zn, zinc.

Literature Review

33

2.3.2 Analytical Approaches to Study in vitro Zinc Resorption and Zinc

Bioavailability

Zinc resorption and bioavailability in humans’ in vivo is mostly analyzed with (stable) isotope

tracer techniques, primarily measuring fractional zinc resorption [324]. In earlier studies, the

zinc radioisotope 65Zn was also used to investigate zinc homeostasis [325] and bioavailability

in humans [326], but is nowadays mostly replaced by non-radioactive and stable isotopes

[327] and solely employed in vitro [125,127,143].

In three-dimensional in vitro intestinal models, the quantity of the metal in the apical and

basolateral compartment as well as the cellular zinc content is analyzed to determine the

amount of absorbed and actually transported zinc to the blood side (Figure 2.8A). Thus,

transport kinetics and bioavailability of luminally added zinc species is investigated. Aside of

(stable) isotope techniques, zinc is generally quantified with inductively coupled mass

spectrometry (ICP-MS), inductively coupled plasma optical emission spectrometry (ICP-OES)

or atomic absorption spectrometry (AAS) [328].

Aside of determining enterocytes’ zinc uptake or transport, in vitro intestinal models offer

the great opportunity to scrutinize (sub-) cellular concentration of the metal, providing

additional information about its disposition and cellular availability after its absorption into

enterocytes (Figure 2.8B). For this, fluorescent zinc sensors are employed, offering a

versatile tool to analyze small subcellular changes of free zinc [100]. These sensors bind free

or mobile zinc, which represents a particular small part of the cellular zinc content. In fact,

this cellular zinc pool includes zinc that is in transit through the cell or serves as a cellular

signal [97,329] (for details refer to 2.2.1.1, p. 12 ff.).

Literature Review

34

Figure 2.8 Application of in vitro intestinal models to study intestinal zinc transport

In vitro intestinal cell models provide a standardized and versatilely applicable microenvironment to study enterocytes’ zinc uptake and transport via the intestinal epithelium. Zinc resorption and transport kinetics can be analyzed using three-dimensional cell models, such as the Caco-2/HT-29-MTX model shown in A. For this, zinc is quantified in all three compartments (apical, cellular, basolateral) with conventional analytical approaches, such as inductively coupled mass spectrometry (ICP-MS) or flame atomic absorption spectrometry (FAAS). Furthermore, the application of chemical- or protein-based fluorescent zinc sensors in enterocytes provides additional information about the (sub-) cellular distribution of the micronutrient upon its uptake into the cell. More precisely, these sensors bind intracellular free zinc and track already small changes of intracellular free zinc. Depending on the (sub)-cellular accumulation of the sensor, the cytoplasmic free zinc pool or free zinc in organelles, like vesicles and the endoplasmic reticulum etc. (as circled in red), can be investigated (B).

Generally, fluorescent zinc sensors are mainly classified into low molecular weight sensors

(or chemical sensors) and genetically encoded biosensors (Figure 2.9) [100]. In the following

their function and application in vitro, as well as advantages and disadvantages of the two

classes of sensors, are briefly summarized.

The principle of most LMW sensors is based on photo-induced electron transfer (PET)

between the fluorophore and a chelating unit, which in case of a non-ratiometric sensor

quenches fluorescence when no metal is present. Metal binding leads to disruption of PET

and increase of fluorescence (in detail reviewed in [330]). After entering the cells by passive

diffusion, changes in their fluorescence upon binding of intracellular free zinc can then be

analyzed with fluorescence spectrometric methods to quantify free zinc concentration or

using fluorescence microscopy to image spatial distribution of the cation [330,331].

Literature Review

35

Figure 2.9 Chemical- and protein-based fluorescent sensors

(A) Shown are selected low molecular weight (LMW) or chemical sensors and genetically encoded biosensors or protein sensors (B). LMW sensors Zinypr-1, Fluozin-3 and Zinquin are all non-ratiometric zinc probes that lead to increase of fluorescence upon zinc binding [330], in which Zinpyr-1 provides two zinc binding sides and Fluozin-3 and Zinquin bind only one zinc molecule. The sensors’ fluorescent domain is indicated in color. (B) Schematic representation of Förster resonance energy transfer (FRET) -sensor (eCFP-Atox1-linker-WD4-YFP) (eCalwy) [332] and bioluminescence resonance energy transfer (BRET)-sensor Zinch-3 [333] from Merkx and co-workers. These genetically encoded zinc probes on the other hand are ratiometric sensors based on FRET or BRET. Changes of FRET or BRET-ratio decreases (in the case of eCalwy) or increases (Zinch-3) due to conformational change upon zinc binding by these proteins can be measured.

In addition to low molecular weight probes, genetically encoded sensors are applied to study

intracellular zinc concentration in a less invasive and more sensitive way than chemical

probes [100]. Main principal of these sensors is comparable to LMW probes, resulting in

measurable changes upon zinc binding. Various ratiometric biosensors have been developed

based on Förster resonance energy transfer (FRET) or bioluminescence resonance energy

transfer (BRET), respectively, between two fluorescent molecules [332-336]. These fusion

proteins are composed of two fluorescent domains and a metal binding site connected by a

flexible linker (Figure 2.9B). In detail, emission wavelength of donor fluorescent domain

overlaps with excitation wavelength of acceptor domain, resulting in a FRET or BRET signal

when these fluorescent molecules are in proximity. Conformational changes upon zinc

binding consequentially lead to a shift of FRET or BRET signals [330,333]. Most recently, the

group from Palmer et al. developed a biosensor based on a single fluorescent protein [337].

In contrast to low molecular weight sensors these probes are genetically encoded and thus

transfected as plasmids into the cells [332]. Consequently, the cell produces the sensor

controlling its subcellular concentrations and distribution, which makes them particularly

convenient for long-term measurements and not as invasive as chemical probes [97,100].

Literature Review

36

Introduction of BRET-based biosensors improved some disadvantages of FRET-sensors

including autofluorescence and photobleaching of fluorophores due to the illumination of

the sample, which is necessary for the excitation of the donor domain [333]. Furthermore,

FRET analysis requires an elaborate technical approach mostly based on laser scanning

microscopy determining FRET or fluorescence life time imaging (FLIM)-FRET, almost entirely

analyzing single cells [338]. BRET-based biosensors instead can be employed in high

throughput screening assays using bioluminescence plate readers [333,339,340].

Of particularly interest are zinc biosensors with organelle-specific targeting, accumulating in

distinct organelles within the cell, such as mitochondria, Golgi apparatus and endoplasmic

reticulum and cell membrane [334,341-343]. Although subcellular distribution of LMW

sensors is generally not easy to control, chemical probes with specific cellular targeting were

already successfully developed [344-346].

In terms of application of these sensors in human intestinal cell lines to either measure zinc

uptake or analyze its subcellular distribution, low molecular weight sensors Zinpyr-1 [157]

and Fluozin-3 [190,347,348] were already used in Caco-2 and HT-29 cells. Until now,

however, genetically encoded biosensors have never been applied in intestinal cells to study

intestinal zinc resorption.

Objectives and Structure of Thesis

37

Chapter 3. Objectives and Structure of Thesis

The aim of this thesis was to investigate the intestinal zinc resorption using in vitro intestinal

models. For this a three-dimensional in vitro model had to be established that resembles the

in vivo situation of the intestinal epithelium as close as possible, optimizing the conventional

Caco-2 model with regard to its cellular composition as well as luminal and basolateral

factors. Herein, the impact of these factors on in vitro cellular zinc uptake and transport to

the basolateral side should be scrutinized in detail. Furthermore, low molecular weight and

genetically encoded zinc sensors should be applied in the intestinal cell line Caco-2 to

provide an additional intestinal model system to determine intestinal zinc absorption aside

from conventional analytic approaches (ICP-MS, FAAS) and to elucidate its subsequent

intracellular distribution in enterocytes.

This thesis is based on three accepted peer-reviewed publications, which are structured in

three Chapters (Chapter 4 – Chapter 6) and include the following subjects:

In Chapter 4, the application of zinc biosensors in enterocytes is addressed. Therefore, the in

vitro intestinal cell line Caco-2 was stably transfected with the zinc biosensor (eCFP-Atox1-

linker-WD4-YF (eCalwy). Before its use, the Caco-2-eCalwy clone had to be characterized

regarding characteristic features of the intestinal cell line Caco-2 and changes in its zinc

homoeostasis compared to Caco-2-wild type (WT) cells (detailed parameters of FAAS

measurements in Appendix D). Furthermore, functionality and application of Caco-2-eCalwy

cells to measure enterocytes’ zinc uptake was analyzed. Supplemental material of this

manuscript can be found in Appendix A.

Chapter 5 deals with the importance of serum albumin as a basolateral zinc acceptor for

intestinal zinc resorption and examines the critical aspects of medium composition,

particularly the use of FCS and albumin, with regard to zinc speciation and availability when

investigating zinc uptake using in vitro cell models. In this context, the impact of apical

protein on short and long-term zinc uptake in Caco-2 cells as well as the influence of in vitro

digestion on the zinc availability from a protein matrix was examined. Supplemental material

of this manuscript can be found in Appendix B. Moreover, detailed parameters of ICP-MS

and FAAS measurements are depicted in Appendix D as well as additional results of zinc

transport using mono- and co-cultures with and without basolateral added albumin in

Appendix E.

Lastly, the role of intestinal mucins in the intestinal zinc resorption is discussed in Chapter 6.

For this, the binding capacity and affinity of mucins as well as the effect of mucins on short-

and long-term zinc absorption into enterocytes and goblet cells in two-dimensional

experiments was analyzed. Moreover the impact of the intestinal mucus layer on zinc

resorption was investigated with zinc transport-studies using three-dimensional intestinal

models: mucus-lacking Caco-2 monocultures and mucin-producing Caco-2/HT-29-MTX co-

cultures (detailed parameters of FFAS and ICP-MS measurements are listed in Appendix D).

Supplemental material of this manuscript can be found in Appendix C as well as additional

Objectives and Structure of Thesis

38

results of zinc transport using mono- and co-cultures with and without basolateral added

albumin in Appendix E.

The main findings from Chapter 4-6 will be related and discussed in the light of current

knowledge in Chapter 7. References of Chapter 1, 2 and 7 can be found at the end of this

manuscript, whereas references of Chapters 4-6 are at the end of the respective chapter.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

39

Chapter 4. Characterization of Caco-2 cells Stably Expressing

the Protein-based Zinc Probe eCalwy-5 as a Model System for

Investigating Intestinal Zinc Transport2

Abstract

Intestinal zinc resorption, in particular its regulation and mechanisms, are not yet fully

understood. Suitable intestinal cell models are needed to investigate zinc uptake kinetics and

the role of labile zinc in enterocytes in vitro. Therefore, a Caco-2 cell clone was produced,

stably expressing the genetically encoded zinc biosensor eCalwy-5. The aim of the present

study was to reassure the presence of characteristic enterocyte-specific properties in the

Caco-2-eCalwy clone. Comparison of Caco-2-WT and Caco-2-eCalwy cells revealed only slight

differences regarding subcellular localization of the tight junction protein occludin and

alkaline phosphatase activity, which did not affect basic integrity of the intestinal barrier or

the characteristic brush border membrane morphology. Furthermore, introduction of the

additional zinc binding protein in Caco-2 cells did not alter mRNA expression of the major

intestinal zinc transporters (zip4, zip5, znt-1 and znt-5), but increased metallothionein 1a

expression and cellular resistance to higher zinc concentrations. Moreover, this study

examines the effect of sensor expression level on its saturation with zinc. Fluorescence cell

imaging indicated considerable intercellular heterogeneity in biosensor-expression. However,

FRET-measurements confirmed that these differences in expression levels have no effect on

fractional zinc-saturation of the probe.

2 The following article is the accepted version and appears as journal version in:

Maria Maares, Claudia Keil, Susanne Thomsen, Dorothee Günzel, Burkhard Wiesner, Hajo Haase. "Characterization of Caco-2 cells stably expressing the protein-based zinc probe eCalwy-5 as a model system for investigating intestinal zinc transport." Journal of Trace Elements in Medicine and Biology 2018, 49: 296-304, DOI: 10.1016/j.jtemb.2018.01.004, https://doi.org/10.1016/j.jtemb.2018.01.004 https://www.sciencedirect.com/science/article/pii/S0946672X17309033?via%3Dihub

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

40

4.1 Introduction

The essential trace element zinc plays an important role for a variety of biological processes

in the human body [1]. Zinc homoeostasis is mainly dependent on dietary intake and

bioaccessibility from food in the intestine, where zinc is resorbed [2]. Intestinal zinc

homeostasis is primarily mediated by four zinc transporters: zinc uptake from the

gastrointestinal lumen is controlled by ZIP4, which is expressed on the apical membrane of

the intestinal epithelium, whereas ZIP5 at the basolateral membrane transports zinc from

the blood into enterocytes. The zinc exporters ZnT-1 and ZnT-5 are essential for zinc

transport from the enterocytes into the blood or back into the intestinal lumen, respectively

[3]. Despite ongoing research [4], knowledge of the amount of labile zinc and zinc uptake

kinetics in intestinal cells is scarce. In particular, the mechanism and regulation of human

zinc resorption is not fully understood [5].

There are several ways to investigate cellular zinc resorption in vitro. Alongside the

quantification of whole cellular zinc using analytical methods such as ICP-MS (inductively-

coupled plasma mass spectrometry) and FAAS (flame atomic absorption spectrometry) [6],

fluorescence- based sensors are increasingly used to study intracellular free zinc [7,8].

Recently, several genetically encoded fluorescent sensorproteins have been developed to

analyze intracellular labile zinc. Among them are the eCalwy sensors from the Merkx group,

which are based on Förster resonance energy transfer (FRET) between two fluorescent

domains bound to the metal binding domains WD4 and ATOX1, connected by a flexible

linker [9]. Upon zinc binding, FRET is decreasing because of conformational change and

growing distance between the donor-domain (cerulean) and the acceptor mCitrine. Zinc

biosensors offer several advances compared to synthetic sensors. They are produced by the

cell itself, providing control over their subcellular distribution and concentration and are

well-suited for long term-measurements [10,11]. In this sense, applying these sensors in

enterocytes would be suitable to monitor free zinc in these cells and illuminate sensitive

parameters of intestinal zinc resorption. Therefore, a Caco-2 cell clone was produced stably

expressing eCalwy-5.

Future research aims to include Caco-2-eCalwy-5 cells in in vitro models for the intestinal

epithelium. Therefore, this study investigates if characteristic properties of the wildtype cells

are conserved in the stably transfected cell clone and examines the effect of sensor-

expression level on its zinc-saturation.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

41

4.2 Experimental

4.2.1 Materials

ApaLI (NEB, Ipswich, USA), Cell Counting Kit-8/WST-8 (Sigma Aldrich, Munich, Germany), Cy-

2-labeled goat anti mouse and Cy-5- labeled goat anti rabbit secondary antibodies (Jackson

ImmunoResearch, Dianova, Hamburg, Germany), Claudin 2 polyclonal antibody, Occludin

monoclonal antibody and ZO-1 polyclonal antibody (Invitrogen, ThermoFisher Scientific,

USA), Cloning cylinders (Sigma Aldrich, Munich, Germany), DAPI (Invitrogen, ThermoFisher

Scientific, USA), DMEM (PAN-Biotech, Aidenbach, Germany), FCS (CCPro, Oberdorla,

Germany), G 418 disulfate (Geneticin) (Santa Cruz Biotechnology, Dallas, Germany), Invisorb

Spin Tissue Mini Kit (Stratec Molecular GmbH, Berlin, Germany), iScript cDNA Synthesis Kit

(Quantabio, Beverly, USA), Lipofectamin 2000 (Invitrogen, ThermoFisher Scientific, USA),

MTT (Carl Roth, Karlsruhe, Germany), NucleoSpin II (Macherey-Nagel GmbH & Co. KG, Berlin,

Germany), Opti-Mem (Sigma Aldrich, Munich, Germany), PCR Clean-Up System (Promega,

Madison, USA), peCalwy-plasmid (addgene, Cambridge, USA), pNPP (PanReac AppliChem,

Glenview, USA), pNP (Sigma Aldrich, Munich, Germany), ProTaqs Mount Fluor (Biocyc,

Luckenwalde, Germany), Purified Mouse Anti-E-Cadherin (BD Transduction Laboratories™,

BD Biosience, New Jersey, USA), SYBR™- Green (Quantabio, Beverly, USA), Transwell inserts

(Corning, New York, USA), TPEN (Sigma Aldrich, Munich, Germany), Qiagen Plasmid Maxi Kit

(Qiagen, Venlo, Netherlands), ZnSO4·7H2O (Sigma Aldrich, Munich, Germany). All other

chemicals were purchased from standard sources.

4.2.2 Cell Culture

Cells were cultured at 37°C, 5% CO2 and humidified atmosphere in Dulbecco’s Modified

Eagles Medium (DMEM), containing 10% fetal calf serum (FCS), 100 U mL-1 penicillin and 100

µg mL-1 streptomycin. Media were changed every other day. A final concentration of 0.6 mg

mL-1 G418 was added to Caco-2-eCalwy clones.

4.2.3 Stable Transfection

Plasmid-DNA of peCalwy-5 was introduced to E. coli, isolated using Qiagen Plasmid Maxi Kit

and purified using Invisorb Spin Tissue Mini Kit, according to the manufacturers’ protocols.

Purified peCalwy-5-DNA was linearized using ApaLI and successful linearization was

confirmed by gel-electrophoresis. Finally, 4·104 Caco-2-WT cells were seeded into 24 well-

plates, cultivated for 24 h and transfected by adding a total of 0.5 μg plasmid DNA and 2.5 μg

Lipofectamin 2000 directly to the medium. After additional 24 h, medium was changed and

transfection was checked by fluorescence microscopy. 48 h after transfection 1.2 mg mL-1 G

418 disulfate was added to cells and cultured for additional 6 days. Finally cells were

transferred to a 100 cm2 dish and isolated cell clones were selected, picked by using cloning

cylinders and transferred into 24 well plates.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

42

4.2.4 Alkaline Phosphatase Activity

Alkaline phosphatase (ALP) catalyzes the hydrolyzation of pNPP to yellow p-nitrophenol,

which can be measured photometrically at 405 nm. Extracellular ALP-activity was analyzed

on live cultured cells according to the method described by Ferruzza et al. [12]. Caco-2

clones (5000 cells/well) were cultured for 21 days in 96 well plates, washed with PBS, and 10

µM pNPP in reaction buffer (10 mM Tris-HCl, 150 mM NaCl, pH 8.0) was added to the cells.

Samples were collected at different time points and p-nitrophenol was quantified using an

external calibration. Protein was quantified using the BCA-assay as described [13] and ALP-

activity is depicted in mU/mg protein (Hydrolyzation of 0.346 nmol pNPP/min=1 mU ALP).

4.2.5 Immunofluorescent Staining

Immunofluorescence analysis was performed with cells (8·104 cells/well) grown on

polycarbonate transwell membranes (pore size 0.4 μm) for 21 d. Cell culture media were

changed every other day and integrity of the cell layer was monitored by measuring TEER

(transepithelial electrical resistance) with the epithelial volt-ohm meter Millicell® ERS-2

(Millipore, USA). Differentiated cells were washed with PBS, fixed with 2 %

paraformaldehyde for 20 min at room temperature and permeabilized with 0.5 % Triton-X-

100 in PBS for 10 min. Cells were incubated overnight at 4°C with primary antibodies against

claudin-2, occludin, zonula occludens-1 (ZO-1) protein and E-cadherin diluted 1:200 in PBS.

After three washes, cells were incubated with Cy-2-labeled goat anti mouse and Cy-5-labeled

goat anti rabbit secondary antibodies and 4′,6-diamidino-2-phenylindole (DAPI) (final

concentration 1 μg mL-1) for 45 min at room temperature. Subsequently cells were washed,

dehydrated with 95% ethanol and mounted in ProTaqs Mount Fluor. The mounting medium

was allowed to solidify for 1 h in the dark, before fluorescence measurements were

performed using confocal laser scanning microscopy (Zeiss LSM780) at excitation

wavelengths of 488 nm (Cy-2), 633 nm (Cy-5) and 405 nm (DAPI).

4.2.6 Transmission Electron Microscopy (TEM) and Scanning Electron

Microscopy (SEM)

Cells grown on cover slips (6·104 cells/well; 12 well plates), were fixed with 2.5%

glutaraldehyde in 0.1 M sodium-cacodylat buffer for 30 min at room temperature and

dehydrated trough a graded series of ethanol. TEM imaging was performed using an EM906

(Zeiss, Germany). SEM measurements were conducted with a DSM982Gemini (Zeiss,

Germany).

4.2.7 Viability Assays

Cells were seeded with an initial density of 5000 cells/well in 96 well plates and cultured for

21 d. For acute zinc toxicity, differentiated cells were incubated with 0–1000 μM ZnSO4·7H2O

for 24 h in DMEM without phenol red and fetal calf serum. Subsequently, cells were washed

with PBS and cell viability was analyzed incubating either with the water soluble tetrazolium

salt (WST)-8 (diluted 1:10) or 1 mg mL-1 3- (4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

43

bromide (MTT) in DMEM without phenol red for 30-60 min. In case of the MTT-assay cells

were lysed in isopropanol. The absorption was determined on a well plate reader (M200,

Tecan, Swiss) at 490 nm (WST) or 570 nm (MTT), respectively. A total of 0.01% Triton-X-100

was used as a positive control. Data were analyzed with GraphPad Prism software version

5.01 (GraphPad Software Inc., CA, USA) and a non-linear regression using a sigmoidal dose–

response curve with variable slope as a function of the logarithm of concentration was

applied. Chronic zinc toxicity was measured by incubating the cells with 0, 5 and 10 μM

ZnSO4·7H2O in standard cell culture medium during differentiation. Cell viability of

differentiated cells was measured as the amount of cellular protein using the

sulforhodamine B (SRB)-assay as described [14].

4.2.8 Quantitative Real Time PCR (qPCR)

1.2·105 cells were seeded in 6 well plates and cultured for 21 d. Cells were harvested in PBS

on ice using a cell scraper, and cell pellets were stored at -80°C. RNA was isolated with

Nucleo Spin II and cDNA synthesized using iScript cDNA Synthesis Kit according to the

manufacturers’ protocols. Finally, mRNA-levels were quantified by qPCR with SYBR™Green

Super Mix, using the primer listed in Table 4.1, and normalized to β-actin.

Table 4.1: Oligonucleotide sequences used for qPCR

Primer NCBI Reference

Sequence

Sequence fwd 5'-3' Sequence rev 5'-3' Ref

zip-4 NM_017767 AGACTGAGCCCAGAGTTGAGGCTA TGTCGCAGAGTGCTACGTAGAGGA [15]

zip-5 NM_173596 GAGCAGGAGCAGAACCATTACCTG CAATGAGTGGTCCAGCAACAGAAG [15]

znt-1 NM_021194 GGCCAATACCAGCAACTCCAA TGCAGAAAAACTCCACGCATGT [16]

znt-5 NM_024055 AAGGACATCATGACAGTGCTCTAACTC CCAACTTTACAACACAAAGCCAGTAC [16]

mt1a NM_005946.2 CTCCTGCAAGAAGAGCTGCTG CAGCCCTGGGCACACTT

alp NM_001631.4 CCGCTTTAACCAGTGCAACA CCCATGAGATGGGTCACAGA

β-actin NG_007992.1 CGCCCCAGGCACCAGGGC GCTGGGGTGTTGAAGGT [17]

4.2.9 Atomic Absorption Spectrometry

1.2·105 cells were seeded in 6 well plates and cultured for 21d. Cell layers were harvested on

ice with a cell scraper, and an aliquot was collected for protein quantification as described

[13]. Subsequently cells were dissolved in a mixture of 67% ultrapure HNO3 and 30% H2O2

(50/50; v/v) and dried at 92°C overnight using a thermoshaker. Residues were dissolved in

0.67% HNO3 and samples were analyzed by FAAS using a Perkin Elmer AAnalyst800 (Perkin

Elmer, Germany).

4.2.10 Live Cell Imaging and FRET-Measurements

Confluent Caco-2-eCalwy cells on cover slips were used for live cell imaging with a confocal

laser scanning microscope (Zeiss LSM710, Germany). Cells were washed twice with buffer

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

44

(120 mM NaCl, 5.4 mM KCl; 5 mM Glucose; 1 mM CaCl2; 1 mM MgCl2; 1 mM NaH2PO4; 10

mM HEPES; pH 7.35) and a fluorescence emission scan was conducted after excitation at 458

nm (40×/1.3 oil objective; λem=462 nm–590 nm). Additionally, an emission scan of the

acceptor-domain mCitrine was measured after direct excitation at λex=514 nm (λem =526nm–

590 nm). FRET is expressed as the fluorescence intensity ratio of mCitrine (λem =527 nm) and

cerulean (λem =478 nm) after excitation at 458 nm. Ratios were measured in buffer alone

(Rapo), or 10 min after the addition of 20 μM TPEN (Rmax) or 400 μM zinc (Rmin) (final

concentrations). The measurements using a multiphoton laser scanning microscope (Zeiss

LSM510-META-NLO, Germany) were conducted as described above, but after multiphoton

excitation (810 nm) using a tunable IR-laser (λem (cerulean)=478 nm, λem(mCitrine)=532 nm).

The concentration of free zinc was determined using the following equation [11,18] and a

dissociation constant for the zinc-eCalwy-5-complex of 1.85 nM [9]: [Zinc]=Kd ×

[(Rapo−Rmax)/(Rmin−Rapo) × (Sf2/Sb2)]. The proportionality coefficients Sf2 and Sb2 are

corresponding to the emission intensities of the free and bound forms of the sensor at 532

nm.

4.2.11 Fluorescence Lifetime Imaging Microscopy (FLIM)-FRET

FLIM-measurements were performed with Caco-2-eCalwy cells using a multiphoton LSM510-

META microscope equipped with a time resolved LSM upgrade setup (Becker&Hickl,

Germany). FLIM of the donor-domain cerulean was measured in buffer and in the presence

of 20 μM TPEN or 400 μM zinc (40×/1.3 oil objective, λex =810 nm, 450–490 nm band pass

filter). To analyze FLIM of cerulean in the absence of the acceptor domain mCitrine, Caco-2-

WT clones were transfected with a cerulean-construct. FLIM data were analyzed using the

SPC Image software from Becker&Hickl and FRET-efficiency was calculated using the

following equation: E(%)=(1−(τDA/τD)) × 100, where τDA is the fluorescence lifetime of the

donor-domain cerulean in intact eCalwy-protein and τD in the absence of mCitrine.

4.2.12 Statistical Analysis

Statistical significance was analyzed by Student’s t-test (for paired samples), or one- or two-

way analysis of variance (ANOVA) (for multiple comparisons), followed by Bonferroni or

Dunnett’s multiple comparison post hoc tests, as indicated in the respective figure legends,

using GraphPad Prism software version 5.01 (GraphPad Software Inc., CA, USA). Error bars

represent standard deviation of three independent biological replicates.

4.3 Results

4.3.1 Characteristic Cellular Features

To ensure the presence of characteristic features of the intestinal cell line Caco-2 after

introduction of eCalwy-plasmids, cell differentiation, morphology, as well as barrier integrity

were investigated. First, activity and expression of the differentiation marker alkaline

phosphatase were compared. Both cells expressed large amounts of functional ALP;

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

45

however, ALP-activity and -expression by Caco-2-eCalwy were significantly lower (each p <

0.001, Figure 4.1) compared to wild type cells.

Figure 4.1: Alkaline phosphatase (ALP) in Caco-2-WT and Caco-2-eCalwy cells.

(A) Enzyme activity in differentiated cells was measured using an ALP-assay and is displayed relative to cellular protein. (B) Relative alp expression in differentiated cells was analyzed using qPCR. Data are shown as means + SD of three independent experiments. Means of each cell clone are significantly different as indicated (***p < 0.001, Student’s t-test).

Ultrastructure comparison by TEM detected no remarkable differences in brush border

formation. Figure 4.2 shows the distribution of mitochondria, lysosomes, desmosomes,

nexus and tight junctions, and microvilli of about 500–1000 nm length in differentiated cells.

Notably, TEM images revealed the presence of more lysosomal structures in Caco-2-eCalwy.

SEM images show the characteristic columnar shape of differentiated Caco-2 cells and

evenly distributed microvilli on the surface of both cell clones (Figure 4.3).

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

46

Figure 4.2: Transmission electron micrographs.

Differentiated Caco-2-WT clones (A) and Caco-2-eCalwy (B) were analyzed for typical features of the brush border (BB) of intestinal cells, showing microvilli (MV) of 500–1000 nm length. Tight junctions (TJ), desmosomes (D), nexus (Nx), mitochondria (M), lysosomes (L) and nuclei (N) are indicated. Scale bar 1000 nm.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

47

Figure 4.3: Scanning electron microscope images.

Shown are the apical surface of Caco-2-WT (A, B) and Caco-2-eCalwy clones (C, D). Both cells display characteristic columnar shape (A, C) and microvilli formation (B, D). Scale bars 2 μm (B, D) and 10 μm (A, C).

Barrier integrity, measured as TEER, did not differ between the two cell clones

(Supplemental Figure S4.1). Tight junction formation was analyzed using immunofluorescent

staining of selected proteins: claudin-2, occludin, ZO-1 and E-cadherin (Figure 4.4 and

Supplemental Figure S4.2) to investigate whether stable transfection with eCalwy-5 had

affected their distribution.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

48

Figure 4.4: Localization of tight junction proteins.

Fluorescence imaging was conducted using a confocal laser scanning microscope, showing results for differentiated Caco-2-WT (upper panel) and Caco-2-eCalwy (lower panel) cells. Shown are selected images of Caco-2-WT and -eCalwy clones after immunofluorescent staining of claudin-2 (A, E), occludin (B, F), zonola occludens-1 (ZO-1) (C, G) and E-cadherin (D, H) of two independent experiments. Scale bar 5 μm.

Immunofluorescent images in Figure 4.4 show the intracellular distribution of the tight

junction proteins as x-y-scans and Supplemental Figure S4.2 additionally depicts tight

junction localization as z-scans together with stained nuclei. Localization and generation of

claudin-2 and ZO-1 were not altered. Staining of claudin-2 revealed a membranous and

cytoplasmic localization as well as a nuclear enrichment in Caco-2-WT and -eCalwy cells. In

both cases, the peripheral protein ZO-1 displayed a homogenous localization in the

cytoplasmic membrane. In contrast, immunofluorescent images of occludin and E-cadherin

revealed slight differences. Occludin is located in the plasma membrane of Caco-2-WT,

whereas it shows tendencies of a lateral staining, with decreasing sensitivity from apical to

basolateral membrane, in Caco-2-eCalwy (Figure 4.4B,F and Supplemental Figure S4.2B,F).

Imaging of E-cadherin revealed membranous distribution and vesicular accumulation of this

protein in Caco-2-WT, whilst in eCalwy transfected cells E-cadherin was mostly found in the

apical cell membrane with only a slight vesicular localization (Figure 4.4D,H and

Supplemental Figure S4.2D,H).

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

49

4.3.2 Zinc Homeostasis

The introduction of an additional zinc binding protein into Caco-2 cells might change zinc

homeostasis in the stably transfected cell clone, potentially enhancing the resistance against

elevated zinc concentrations, or increasing the requirement for zinc during growth and

differentiation. Thus, viabilities of differentiated Caco-2-WT and -eCalwy after incubation

with 0–1000 μM zinc for 24 h were compared by measuring mitochondrial activity (Figure

4.5A,B). In contrast to Caco-2-WT, the Caco-2-eCalwy cells had higher LC50 values in MTT

(865 μM vs. 394 μM for wildtype) and WST assays (803 μM vs. 427 μM for wildtype).

Figure 4.5: Zinc-toxicity on differentiated Caco-2-WT and -eCalwy.

Cells were treated with different zinc concentrations for 24 h and mitochondrial activity was analyzed using two different assays based on the tetrazolium salts MTT (A) and WST-8 (B). A total of 0.01% Triton X-100 was used as positive control. Data are shown as means ± SD of three independent experiments. Sigmoidal dose-response curves were fitted by nonlinear regression and means significantly different from the untreated controls are indicated (**p < 0.01; ***p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test).

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

50

Furthermore, cellular protein content after differentiation in the presence of different zinc

concentrations was analyzed (Figure 4.6). The amounts did not change in Caco-2-WT,

whereas addition of 5 and 10 μM zinc to the culture media led to a slight but significant

increase in Caco-2-eCalwy.

Figure 4.6: Effect of zinc on cellular protein levels.

Caco-2-WT and -eCalwy cells were cultivated for 21 d with different concentrations of zinc added to

the culture medium, and protein content was analyzed using the SRB assay. Data are shown as

means + SD of three independent experiments. Significant differences are indicated (*p < 0.05; **p <

0.01; two-way ANOVA with Bonferroni post hoc test).

The expression of proteins important for zinc homeostasis in enterocytes, such as

metallothionein 1a (mt1a) and the zinc transporters zip4, zip5, znt-1 and znt-5, were

examined (Figure 4.7). eCalwy-5 did not affect mRNA expression of zinc transporters, but

significantly elevated mt1a-expression (p < 0.05) by 20%. In contrast, there were no

significant differences in basal cellular zinc levels between Caco-2-WT and -eCalwy cells

(Figure 4.7B).

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

51

Figure 4.7: Zinc homeostasis in Caco-2-WT and -eCalwy.

(A) Gene expression of proteins involved in zinc-homeostasis in differentiated Caco-2-WT and Caco-2-eCalwy clones was analyzed by qPCR. (B) Basal zinc content of differentiated Caco-2-WT and Caco-2-eCalwy clones was analyzed using FAAS. Data are shown as means + SD of three independent experiments. Significant differences between the two cell clones are indicated (*p < 0.05; Student’s t-test).

4.3.3 Functionality and Application

Cellular distribution of the sensor-protein in Caco-2-eCalwy cells was analyzed by live cell

fluorescent imaging, using a confocal laser scanning microscope. Fluorescence images show

the cytoplasmic localization of the two fluorescent domains cerulean (Figure 4.8A) and

mCitrine (Figure 4.8B), as well as the occurrence of FRET (Figure 4.8C) in resting Caco-2-

eCalwy cells. Notably, there is considerable intercellular heterogeneity in biosensor-

expression. This was also confirmed by frequency distribution analysis of the mCitrine-

concentration in different Caco-2-eCalwy cells. Because fluorescence of the acceptor-domain

mCitrine after direct excitation at 514 nm is independent of zinc binding [19], its

fluorescence signal is directly proportional to the sensor-concentration. mCitrine

measurements revealed a wide range of sensor-expression within different cells of the same

clone (Figure 4.8D).

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

52

Figure 4.8: Life cell imaging.

(A–C) Cellular distribution of sensor-protein in Caco-2-eCalwy cells was analyzed by fluorescence live cell imaging using a confocal laser scanning microscope. Shown are fluorescence images of (A) cerulean emission (λex=458 nm; λem=462 nm–496 nm), (B) mCitrine emission (λex=514 nm; λem=526 nm–590 nm) and (C) FRET, not corrected for spectral bleed-trough (λex=458 nm; λem=526 nm–590 nm). Scale bar 50 μm. (D) Frequency distribution of eCalwy-sensor concentration in Caco-2-eCalwy cells, analyzed by direct excitation of mCitrine (λex=458 nm; λem=527 nm).

Due to high background fluorescence and light scattering, the sensitivity of confocal

microscopy was insufficient for quantification of free zinc in the stably transfected Caco-2-

eCalwy clone. As this was not the case with multiphoton laser scanning microscopy, further

FRET-measurements were performed using two photon excitation.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

53

Figure 4.9: Two photon microscopy.

(A) Emission spectra of Caco-2-eCalwy in the presence of TPEN (20 μM) or zinc (400 μM). (B) Fluorescence lifetime of FRET-donor domain cerulean of eCalwy-biosensor in Caco-2-eCalwy. Fluorescence halftime is shown for the donor in the absence of the acceptor-domain (cerulean), in resting states (buffer) and after addition of zinc. Data are shown as means + SD of three independent experiments. Significant differences between the fluorophore-lifetimes under the different conditions are indicated (***p < 0.001; ANOVA followed by Bonferroni’s multiple comparison test). (C) FRET-efficiency of eCalwy-5 in Caco-2-eCalwy cells based on FLIM-FRET-data compared to theoretical calculations using fluorescence intensities. (D) Impact of cellular sensor-expression level on zinc-saturation of the sensor. Shown is the correlation of fluorescence emission ratio (mCitrine/cerulean) and eCalwy-sensor expression, depicted as mCitrine fluorescence when directly excited at 514 nm (λem=532 nm) using a multiphoton laser scanning microscope.

Figure 4.9A compares normalized FRET-emission spectra of Caco-2-eCalwy cells after zinc-

depletion by TPEN or addition of excess zinc, showing a twofold decrease in fluorescence of

mCitrine in response to saturation with zinc. Next, sensor-efficiency after stable transfection

and performance of the eCalwy sensor after two photon excitation were investigated using

two different approaches: FLIM-FRET based on lifetimes of the donor fluorophore, and

theoretical photobleaching using fluorescence intensities. FLIM-FRET analysis confirmed

proper function of the eCalwy probe after stable transfection into Caco-2-eCalwy (Figure

4.9B). Fluorescence lifetime of the donor-domain (τDA) cerulean increased significantly after

zinc addition (one-way ANOVA, p < 0.001). Besides, a significant difference between τD

(fluorescence lifetime of the donor in the absence of the acceptor-domain mCitrine) and in

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

54

τDA was observed. FRET-efficiency is defined as the amount of energy that is transferred from

donor to acceptor-domain after excitation [20]. Calculated FRET-efficiencies based on

fluorescence lifetime and the actual fluorescence intensities are in good agreement (Figure

4.9C), with an energy transfer of about 60% in the resting state (buffer) and 40% after

addition of zinc. Finally, the correlation of sensor-concentration and FRET-ratio was

examined with multiphoton microscopy. Despite differences in sensor concentrations of

more than one order of magnitude (measured by directly exciting the mCitrine-domain),

there was no correlation with the FRET-emission ratio (Figure 4.9D). Linear regression

analysis of these data provided additional confirmation that the slope is not significantly

different from zero. This demonstrates that the expression level of the eCalwy-biosensor had

no effect on fractional probe-saturation with zinc.

4.4 Discussion and Conclusion

Several zinc-FRET-biosensors, such as eCalwy, eZinch, and ZapCy2, have been created and

were used in various different cell types, but until now only after transient transfection

[19,21–24]. This study describes a Caco-2 cell clone stably expressing the zinc-biosensor

eCalwy- 5. Several approaches were applied to reassure enterocyte specific properties,

including morphological characterization as well as investigating the maintenance of the

intestinal barrier integrity, especially tight junction formation and epithelial resistance,

which need to be considered when using intestinal cell models [25]. ALP-activity is a marker

for intestinal cell differentiation and its increasing activity during differentiation of Caco-2

cells was demonstrated before [12]. Moreover, the dependence of intestinal ALP-activity on

zinc availability is well known [26] and might be reflected in Caco-2-eCalwy. Differences in

ALP-expression and -activity were observed in the present study, with the ALP-activity of

Caco-2-eCalwy being only 50% of that measured in Caco-2-WT. Yet, the activity in Caco-2-

eCalwy is still one order of magnitude higher than levels observed in other differentiated

intestinal cell lines [27], indicating successful differentiation.

A morphology characteristic for the intestinal brush border membrane was still present after

stable transfection of the FRET-biosensor into Caco-2 cells, and is comparable to previous

studies with Caco-2-WT cells [27–29]. Comparison of Caco-2-WT and Caco-2-eCalwy clones

revealed only slight differences regarding subcellular localization of the tight junction protein

occludin, which did not affect basic integrity of the intestinal barrier. Tight junction proteins

play a key role in creating a selective transport barrier and maintain cell polarity, while

guaranteeing a certain permeability [30]. Claudin-2 is important for paracellular ion

permeability and was reported to decrease tightness of the epithelial barrier [31] and to be

localized in plasma membrane and cytoplasm of Caco-2-WT cells [32]. The peripheral tight

junction protein ZO-1 is crucial for assembly of tight junctions and binds to the

transmembrane protein occludin [33]. While both cell clones show typical localization of ZO-

1 [33], distribution of occludin in Caco-2-eCalwy differs from Caco-2-WT and observations in

previous studies [32,33]. An increase in occludin-expression might elevate TEER and thus

could affect the tightness of epithelium [34]. However, no differences in TEER between the

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

55

two cell clones were observed. The vesicular distribution of E-cadherin in Caco-2-WT clones

is no artifact caused by the experimental conditions, because it was not observed in Caco-2-

eCalwy cells incubated in parallel. Possibly, excess E-cadherin is internalized in Caco-2-WT

cells and E-cadherin-expression in Caco-2-eCalwy is lower as a result of the transfection.

Nevertheless, localization and expression of E-cadherin and occludin were shown to be zinc-

dependent [33], thus the differences between the two cell clones could also be explained by

the introduction of the additional zinc binding eCalwy-protein. However, E-cadherin is still

evenly distributed at the cell-cell junction membrane of both cell clones, where it plays an

important role in cell-cell adhesion during Caco-2 cell differentiation [35]. Taken together,

important features of the intestinal cell line Caco-2 were largely preserved during stable

transfection.

Intracellular free zinc is regulated by zinc binding proteins, such as MT, maintaining free zinc

concentrations in the low or even sub-nanomolar range [36]. Thus, the expression of an

additional zinc binding protein, such as eCalwy-5, might interfere with cellular zinc

homeostasis, possibly by buffering zinc. Accordingly, investigations of acute zinc toxicity

revealed that Caco-2-eCalwy cells were more resistant to high zinc concentrations. We also

analyzed the effect of long term administration of sub-toxic concentrations of zinc in

addition to the basal content of cell culture medium of 3 μM. Caco-2-WT cells were

unaffected, whereas Caco-2-eCalwy seem to require more zinc during differentiation, since

the protein level was higher after cultivation in the presence of additional zinc. This indicates

that the development of Caco-2-eCalwy cells is already significantly enhanced when

cultivated with a 2.7-fold higher zinc concentration. However, the transfection of eCalwy-

protein showed only slight effects on intracellular zinc homeostasis, as expression of major

intestinal zinc transporters was not impaired, the basal cellular zinc content did not differ,

and only mt1a-expression was elevated in Caco-2-eCalwy. More precisely, the zinc

transporters ZIP4, ZIP5, ZnT-5 and ZnT-1 play a key role in intestinal zinc resorption and

together with proteins of the MT-family these transporters are important for maintaining

intracellular zinc homeostasis [37]. The slightly elevated mt1a-expression in Caco-2-eCalwy

clones might therefore also be a reason for the increased resistance against acute zinc

toxicity.

Finally, performance of the zinc-biosensor in Caco-2-eCalwy clones and cellular distribution

as well as the effect of its expression-level was investigated. Beforehand, sensor-activity had

to be scrutinized, since it is well known that the use of multi photon microscopy for sensors

created for one photon approaches is not always working properly [38]. FLIM-FRET-

measurements confirmed a proper and efficient function of the stably transfected eCalwy-

protein. Due to the off-FRET, τDA increases after zinc-addition, because of the decreased

energy transfer to the acceptor-domain. Notably, τDA in the presence of zinc did not reach

the full amount of τD. This is not surprising, as some remaining energy transfer to the

acceptor-domain is always present in the FRET-sensor, even after conformational change

due to zinc binding [9]. In this manner, correct functionality of the eCalwy-sensor after stable

transfection into Caco-2 and by two-photon excitation was confirmed as well.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

56

Fluorescence cell imaging showed no organelle-specific distribution of the eCalwy-protein

but indicated considerable intercellular heterogeneity in expression of the FRET-biosensor.

Quantification of free zinc in the resting state yielded a concentration of ∼2 nM, which is in

line with previous studies analyzing free zinc with low molecular weight sensors in the

intestinal cell line HT29 [39] and other cell lines [40]. It is well known that the introduction of

a zinc binding molecule into cells can perturb cellular zinc homeostasis, which might lead to

a disturbance of the cellular equilibria of free and bound zinc [39,41]. Thus the intracellular

sensor-concentration must be as low as possible, particularly when analyzing cellular free

zinc using synthetic probes, where sensor-concentration was shown to be critical for zinc-

quantification [39,42]. However, it remains to be fully understood inasmuch the expression

of genetically encoded zinc-sensors impacts zinc-quantification. The expression-level of the

eCalwy-biosensor in stably transfected Caco-2-eCalwy cells had no effect on the FRET-ratio.

A comparable observation was also reported by Qin et al., where different concentrations of

the transiently transfected FRET-biosensor ZapCY2 had no influence on percental saturation

of the sensor molecule in Hela cells [24]. Furthermore, in that study, the ZapCY2-sensor was

expressed in nucleus and cytoplasm of Hela cells, while Caco-2-eCalwy revealed an even

cytoplasmic distribution of the biosensor. In fact, intracellular biosensor-concentrations

were estimated to be 1–10 μM [24], which is considerably lower compared to the reported

millimolar concentrations of low molecular weight-sensors in cells [41]. Certainly, final low

molecular weight sensor-concentration is dependent on its subcellular distribution and

volume of the cellular compartment where it accumulates [40], whereas biosensor-

concentrations are controlled by their expression and the effect of different expression-

levels on their subcellular concentration is yet unknown. Thus, it might be that because both

biosensors are localized in the cytoplasm of the cells and not in a smaller organelle, the

sensor-concentration was not sufficiently high to influence cellular labile zinc. However,

there are several zinc biosensors with organelle-specific targeting [19,21–23]. Therefore it is

certainly necessary to further investigate the effect of cellular expression of those organelle-

specific biosensors on the quantification of free zinc.

This study describes a well characterized stable Caco-2 cell clone expressing the eCalwy-5

FRET-sensor, which can be used to investigate intestinal zinc resorption. It also highlights the

importance of reassuring characteristic features of wildtype cells in the transfected cell

clones, as well as the investigation of possible changes in zinc-homeostasis when stably

transfecting zinc-biosensors. Furthermore, it is not only crucial to characterize subcellular

distribution of the biosensor but also to check the impact of sensor-expression on its zinc-

saturation.

4.5 Conflict of Interest

The authors declare they have no conflict of interest.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

57

4.6 Funding

The work of HH is supported by the Deutsche Forschungsgemeinschaft (TraceAge – DFG

Research Unit on Interactions of essential trace elements in healthy and diseased elderly,

Potsdam-Berlin-Jena, FOR 2558/1, HA 4318/4-1).

4.7 References

1. Maares, M.; Haase, H. Zinc and immunity: An essential interrelation. Arch. Biochem.

Biophys. 2016, 611, 58-65.

2. Ford, D. Intestinal and placental zinc transport pathways. Proc. Nutr. Soc. 2004, 63,

21-29.

3. Cousins, R.J.; Liuzzi, J.P.; Lichten, L.A. Mammalian zinc transport, trafficking, and

signals. J. Biol. Chem. 2006, 281, 24085-24089.

4. Maret, W.; Sandstead, H.H. Zinc requirements and the risks and benefits of zinc

supplementation. J. Trace Elem. Med. Biol. 2006, 20, 3-18.

5. Wang, X.; Zhou, B. Dietary zinc absorption: A play of ZIPs and ZnTs in the gut. IUBMB

life 2010, 62, 176-182.

6. Cerchiaro, G.; Manieri, T.M.; Bertuchi, F.R. Analytical methods for copper, zinc and

iron quantification in mammalian cells. Metallomics 2013, 5, 1336-1345.

7. Carter, K.P.; Young, A.M.; Palmer, A.E. Fluorescent sensors for measuring metal ions

in living systems. Chem. Rev. 2014, 114, 4564-4601.

8. Huang, Z.; Lippard, S.J. Illuminating mobile zinc with fluorescence: From cuvettes to

live cells and tissues. Methods Enzymol. 2012, 505, 445-468.

9. Vinkenborg, J.L.; Nicolson, T.J.; Bellomo, E.A.; Koay, M.S.; Rutter, G.A.; Merkx, M.

Genetically encoded FRET sensors to monitor intracellular Zn2+ homeostasis. Nat.

Methods 2009, 6, 737-740.

10. Hessels, A.M.; Merkx, M. Genetically-encoded FRET-based sensors for monitoring

Zn2+ in living cells. Metallomics 2014, 7, 258-266.

11. Carpenter, M.C.; Lo, M.N.; Palmer, A.E. Techniques for measuring cellular zinc. Arch.

Biochem. Biophys. 2016, 611, 20-29.

12. Ferruzza, S.; Rossi, C.; Scarino, M.L.; Sambuy, Y. A protocol for in situ enzyme assays

to assess the differentiation of human intestinal Caco-2 cells. Toxicol. In Vitro 2012,

26, 1247-1251.

13. Smith, P.K.; Krohn, R.I.; Hermanson, G.T.; Mallia, A.K.; Gartner, F.H.; Provenzano,

M.D.; Fujimoto, E.K.; Goeke, N.M.; Olson, B.J.; Klenk, D.C. Measurement of protein

using bicinchoninic acid. Anal. Biochem. 1985, 150, 76-85.

14. Vichai, V.; Kirtikara, K. Sulforhodamine b colorimetric assay for cytotoxicity screening.

Nat. Protocols 2006, 1, 1112-1116.

15. Leung, K.W.; Liu, M.; Xu, X.; Seiler, M.J.; Barnstable, C.J.; Tombran-Tink, J. Expression

of ZnT and ZIP zinc transporters in the human RPE and their regulation by

neurotrophic factors. Invest. Ophthalmol. Vis. Sci.2008, 49, 1221-1231.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

58

16. Overbeck, S.; Uciechowski, P.; Ackland, M.L.; Ford, D.; Rink, L. Intracellular zinc

homeostasis in leukocyte subsets is regulated by different expression of zinc

exporters ZnT-1 to ZnT-9. J. Leukoc. Biol. 2008, 83, 368-380.

17. Wolf, K.; Schulz, C.; Riegger, G.A.J.; Pfeifer, M. Tumour necrosis factor‐α induced

CD70 and interleukin‐7R mRNA expression in beas‐2b cells. Eur. Respir. J.2002, 20,

369-375.

18. Grynkiewicz, G.; Poenie, M.; Tsien, R.Y. A new generation of Ca2+ indicators with

greatly improved fluorescence properties. J. Biol. Chem. 1985, 260, 3440-3450.

19. Dittmer, P.J.; Miranda, J.G.; Gorski, J.A.; Palmer, A.E. Genetically encoded sensors to

elucidate spatial distribution of cellular zinc. J. Biol. Chem. 2009, 284, 16289-16297.

20. Wallrabe, H.; Periasamy, A. Imaging protein molecules using FRET and FLIM

microscopy. Curr. Opin. Chem. Biol. 2005, 16, 19-27.

21. Chabosseau, P.; Tuncay, E.; Meur, G.; Bellomo, E.A.; Hessels, A.; Hughes, S.; Johnson,

P.R.; Bugliani, M.; Marchetti, P.; Turan, B.; Lyon, A.R.; Merkx, M.; Rutter, G.A.

Mitochondrial and ER-targeted eCalwy probes reveal high levels of free Zn2+. ACS

Chem. Biol. 2014, 9, 2111-2120.

22. Hessels, A.M.; Chabosseau, P.; Bakker, M.H.; Engelen, W.; Rutter, G.A.; Taylor, K.M.;

Merkx, M. eZinch-2: A versatile, genetically encoded FRET sensor for cytosolic and

intraorganelle Zn2+ imaging. ACS Chem. Biol. 2015, 10, 2126-2134.

23. Qin, Y.; Dittmer, P.J.; Park, J.G.; Jansen, K.B.; Palmer, A.E. Measuring steady-state and

dynamic endoplasmic reticulum and Golgi Zn2+ with genetically encoded sensors.

Proc. Natl. Acad. Sci. USA 2011, 108, 7351-7356.

24. Qin, Y.; Miranda, J.G.; Stoddard, C.I.; Dean, K.M.; Galati, D.F.; Palmer, A.E. Direct

comparison of a genetically encoded sensor and small molecule indicator:

Implications for quantification of cytosolic Zn2+. ACS Chem. Biol. 2013, 8, 2366-2371.

25. Le Ferrec, E.; Chesne, C.; Artusson, P.; Brayden, D.; Fabre, G.; Gires, P.; Guillou, F.;

Rousset, M.; Rubas, W.; Scarino, M.L. In vitro models of the intestinal barrier. The

report and recommendations of ecvam workshop 46. European centre for the

validation of alternative methods. Altern. Lab. Anim.2001, 29, 649-668.

26. Williams, R.B. Intestinal alkaline phosphatase and inorganic pyrophosphatase

activities in the zinc-deficient rat. Br. J. Nutr. 1972, 27, 121-130.

27. Pinto, M.; Robineleon, S.; Appay, M.D.; Kedinger, M.; Triadou, N.; Dussaulx, E.;

Lacroix, B.; Simonassmann, P.; Haffen, K.; Fogh, J.; Zweigbaum, A. Enterocyte-like

differentiation and polarization of the human-colon carcinoma cell-line Caco-2 in

culture. Biol.Cell 1983, 47, 323-330.

28. Schimpel, C.; Werzer, O.; Frohlich, E.; Leitinger, G.; Absenger-Novak, M.; Teubl, B.;

Zimmer, A.; Roblegg, E. Atomic force microscopy as analytical tool to study physico-

mechanical properties of intestinal cells. Beilstein J. Nanotechnol. 2015, 6, 1457-1466.

29. Biazik, J.M.; Jahn, K.A.; Su, Y.; Wu, Y.-N.; Braet, F. Unlocking the ultrastructure of

colorectal cancer cells in vitro using selective staining. World J. Gastroenterol. 2010,

16, 2743-2753.

Characterization of Caco-2 cells Stably Expressing the Protein-based Zinc Probe eCalwy-5 as a Model System for Investigating Intestinal Zinc Transport

59

30. Günzel, D.; Yu, A.S.L. Claudins and the modulation of tight junction permeability.

Physiol. Rev. 2013, 93, 525-569.

31. Amasheh, S.; Fromm, M.; Günzel, D. Claudins of intestine and nephron – a correlation

of molecular tight junction structure and barrier function. Acta Physiol. 2011, 201,

133-140.

32. Fischer, A.; Gluth, M.; Weege, F.; Pape, U.-F.; Wiedenmann, B.; Baumgart, D.C.;

Theuring, F. Glucocorticoids regulate barrier function and claudin expression in

intestinal epithelial cells via mkp-1. Am. J. Physiol. Gastrointest. Liver Physiol. 2014,

306, G218-G228.

33. Finamore, A.; Massimi, M.; Conti Devirgiliis, L.; Mengheri, E. Zinc deficiency induces

membrane barrier damage and increases neutrophil transmigration in Caco-2 cells. J.

Nutr. 2008, 138, 1664-1670

34. McCarthy, K.M.; Skare, I.B.; Stankewich, M.C.; Furuse, M.; Tsukita, S.; Rogers, R.A.;

Lynch, R.D.; Schneeberger, E.E. Occludin is a functional component of the tight

junction. J. Cell Sci. 1996, 109, 2287-2298.

35. Schreider, C.; Peignon, G.; Thenet, S.; Chambaz, J.; Pinçon-Raymond, M. Integrin-

mediated functional polarization of caco-2 cells through e-cadherin-actin

complexes. ‎J. Cell Sci. 2002, 115, 543-552.

36. Maret, W. Molecular aspects of human cellular zinc homeostasis: Redox control of

zinc potentials and zinc signals. BioMetals 2009, 22, 149 - 157.

37. Murakami, M.; Hirano, T. Intracellular zinc homeostasis and zinc signaling. Cancer Sci.

2008, 99, 1515-1522.

38. Day, R.N.; Tao, W.; Dunn, K.W. A simple approach for measuring FRET in fluorescent

biosensors using two-photon microscopy. Nat. Protocols 2016, 11, 2066-2080.

39. Krężel, A.; Maret, W. Zinc-buffering capacity of a eukaryotic cell at physiological pZn.

J. Biol. Inorg. Chem. 2006, 11, 1049-1062.

40. Maret, W. Analyzing free zinc(II) ion concentrations in cell biology with fluorescent

chelating molecules. Metallomics 2015, 7, 202-211.

41. Dineley, K.E.; Malaiyandi, L.M.; Reynolds, I.J. A reevaluation of neuronal zinc

measurements: Artifacts associated with high intracellular dye concentration. Mol.

Pharmacol. 2002, 62, 618-627.

42. Haase, H.; Hebel, S.; Engelhardt, G.; Rink, L. Flow cytometric measurement of labile

zinc in peripheral blood mononuclear cells. Anal. Biochem. 2006, 352, 222-230.

60

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

61

Chapter 5. The Impact of Apical and Basolateral Albumin on

Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture

Model3,4

Abstract

The molecular mechanisms of intestinal zinc resorption and its regulation are still topic of

ongoing research. To this end, the application of suitable in vitro intestinal models, optimized

with regard to their cellular composition and medium constituents, is of crucial importance.

As one vital aspect, the impact of cell culture media or buffer compounds, respectively, on the

speciation and cellular availability of zinc has to be considered when investigating zinc

resorption. Thus, the present study aims to investigate the impact of serum, and in particular

its main constituent serum albumin, on zinc uptake and toxicity in the intestinal cell line

Caco-2. Furthermore, the impact of serum albumin on zinc resorption is analyzed using a co-

culture of Caco-2 cells and the mucin-producing goblet cell line HT-29-MTX. Apically added

albumin significantly impaired zinc uptake into enterocytes and buffered its cytotoxicity. Yet,

undigested albumin does not occur in the intestinal lumen in vivo and impairment of zinc

uptake was abrogated by digestion of albumin. Interestingly, zinc uptake, as well as gene

expression studies of mt1a and selected intestinal zinc transporters after zinc incubation for

24 h, did not show significant differences between 0 and 10% serum. Importantly, the

basolateral application of serum in a transport study significantly enhanced fractional apical

zinc resorption, suggesting that the occurrence of a zinc acceptor in the plasma considerably

affects intestinal zinc resorption. This study demonstrates that the apical and basolateral

medium composition is crucial when investigating zinc, particularly its intestinal resorption,

using in vitro cell culture.

3 The following article is the accepted version and appears as journal version in:

Maria Maares, Ayşe Duman, Claudia Keil, Tanja Schwerdtle, Hajo Haase. "The impact of apical and basolateral albumin on intestinal zinc resorption in the Caco-2/HT-29-MTX co-culture model." Metallomics 2018. 10(7): 979-991, DOI: 10.1039/C8MT00064F. 4 Reproduced by permission of The Royal Society of Chemistry;

https://doi.org/10.1039/C8MT00064F https://pubs.rsc.org/en/Content/ArticleLanding/2018/MT/C8MT00064F#!divAbstract

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

62

5.1 Introduction

Human zinc homeostasis is tightly regulated, mainly by the resorption and excretion of zinc

[1]. Zinc is resorbed from food in the small intestine, where it is transported into enterocytes

mainly by the Zrt-, Irt-like protein (ZIP)-4 (SLC39A4) and exported into the blood via zinc

transporter (ZnT)-1 (SLC30A1) at the basolateral enterocyte membrane. Further important

intestinal zinc transporters are ZIP-5 (SLC39A5), transporting zinc from the blood back into

the enterocytes, and ZnT-5 (SLC30A5), which is localized at the apical membrane and

mediates zinc efflux back into the intestinal lumen [2]. In spite of recent advances in

investigating regulatory parameters of intestinal zinc resorption, particularly the role of the

intestinal zinc transporters and the zinc binding protein metallothionein in maintaining

enterocyte zinc homoeostasis, the distinct molecular mechanisms that regulate intestinal

zinc transport remain to be fully understood [3].

Thus, suitable in vitro intestinal models are needed to further scrutinize distinct parameters

of zinc resorption. More precisely, these in vitro models need to be as close to the in vivo

situation as possible, with respect to their cellular composition as well as the apical and

basolateral matrix [4,5]. The human intestinal epithelium is composed of several different

cell types, among which enterocytes and goblet cells are the most abundant [6]. Caco-2 is

widely used as an in vitro cell model to mimic the intestinal epithelium [5]. The cells

differentiate after some time in confluent culture into a phenotype that, morphologically

and functionally, resembles that of human enterocytes [7,8]. Monocultures of Caco-2 cells

were used previously to investigate zinc resorption kinetics in the intestines [9,10], the

impact of food components on fractional zinc resorption [11,12] and enterocyte zinc

homeostasis [13–15]. In contrast, the co-culture of Caco-2 and the mucin-producing cell line

HT-29-MTX has never been used to study zinc resorption before, but has been shown to

offer various benefits compared to conventional Caco-2 monocultures [16,17]. Co-cultures

with HT-29-MTX were reported to particularly modify the permeability of the cell

monolayers [17,18], especially with respect to the Pgp-protein, which is highly expressed in

conventional Caco-2 monocultures [17]. More importantly, Caco-2 monocultures lack the

presence of an intestinal mucus layer [19], which covers the intestinal epithelium in vivo and

is constituted of about 1-5% mucins [20]. In fact, these mucins are secreted by the goblet cell

line HT-29-MTX in the co-cultures and were shown to play an important role in the

resorption of other trace elements such as iron [19].

Although some studies used special assay buffers for the apical application of zinc, the

basolateral composition in transport studies was rarely acknowledged. With respect to the

media composition, it is particularly relevant to consider the occurrence of macromolecules

with high zinc binding affinities, which tremendously decrease the actual free zinc

concentration. In this context, it has to be noted that there is no such thing as actual free

zinc in biological systems, but the terms ‘‘free’’, ‘‘mobile’’ or ‘‘loosely bound’’ zinc are

frequently employed to describe the zinc pool of non-protein bound zinc that is complexed

by small molecule ligands [21,22]. Hence, in the following the term ‘‘free’’ zinc will be used

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

63

for this exchangeable pool, which was defined as being in ‘‘transit during re-distribution’’

[21]. Notably, free zinc was suggested to be the biologically active form [23] and the

speciation of zinc was shown to be crucial when analyzing zinc uptake in several cell lines in

the presence of albumin [24]. Albumin is the main zinc binding protein in human plasma,

buffering the actual free zinc concentration to a narrow nanomolar range [24]. In detail, the

zinc binding ability of albumin is mediated by two binding sites (yielding a 2:1 stoichiometry

of the zinc–albumin-complex), of which site A was reported to bind zinc with an approximate

dissociation constant of 100 nM [25]. This 64 kDa protein is also the main component of

serum, which is a commonly used additive in cell culture media [26]. While this might be

suitable for some cell types, such as immune or epithelial cells [24,27], the application of

high albumin concentrations might constitute a significant distortion when investigating

intestinal zinc resorption.

Thus, this study aims to investigate the role of apical and basolateral addition of albumin in

zinc uptake of in vitro intestinal cells and resorption studies, using a Caco-2/HT-29-MTX co-

culture to further elucidate its role in zinc buffering in in vitro cell culture.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

64

5.2 Methods

5.2.1 Materials

Bovine Serum Albumin (BSA) (Sigma Aldrich, Munich, Germany), Cell Counting Kit-8/water

soluble tetrazolium (WST)-8 (Sigma Aldrich, Munich, Germany), Dulbecco’s Modified Eagles

Medium (DMEM) (PAN-Biotech, Aidenbach, Germany), Element mix (SPETEC, Erding,

Germany), fetal calf serum (FCS) (CCPro, Oberdorla, Germany), 4-(2-hydroxyethyl)-1-

piperazine-ethanesulfonic acid (HEPES) (Carl Roth, Karlsruhe, Germany), iScript cDNA

Synthesis Kits (Quantabio, Beverly, USA), 3-(4,5-dimethylthiazol-2-yl)- 2,5-

diphenyltetrazolium bromide (MTT) (Carl Roth, Karlsruhe, Germany), a Nanosep centrifugal

device (Pall Life Sciences, Michigan, USA), Neutral red (Sigma Aldrich, Munich, Germany),

non-essential amino acids (NEAA) (Sigma Aldrich, Munich, Germany), NucleoSpin II

(Macherey-Nagel GmbH & Co. KG, Berlin, Germany), 4-(2-pyridylazo)resorcinol (PAR) (Sigma

Aldrich, Munich, Germany), Rotiphorese Gel 40 (Carl Roth, Karlsruhe, Germany),

sulforhodamine B (SRB) (Sigma Aldrich, Munich, Germany), SYBR™-Green (Quantabio,

Beverly, USA), Transwell inserts (Corning, New York, USA), N,N,N’,N’- tetrakis(2-

pyridylmethyl)ethylenediamine (TPEN) (Sigma Aldrich, Munich, Germany), powdered trypsin

1:250 (AppliChem, Darmstadt, Germany), Zinpyr-1 (Santa Cruz Biotechnology, Dallas, USA),

and ZnSO4·7H2O (Sigma Aldrich, Munich, Germany) were used. All other chemicals were

purchased from standard sources.

5.2.2 Cell Culture

The cell lines Caco-2 and HT-29-MTX-E12 were obtained from the European Collection of

Authenticated Cell Cultures (ECACC, Porton Down, UK). Cells were cultured at 37°C, in a 5%

CO2 and humidified atmosphere in DMEM, containing 10% FCS (zinc content of 100% FCS: 30

µM as determined by ICP-MS), 100 U ml-1 penicillin and 100 µg ml-1 streptomycin. The

medium for HT-29-MTX cells additionally contained 1% NEAA. The medium was changed

every other day. Caco-2 cells (initially seeding in 96-well plates: 5000 cells per well; 6-well

plates: 120,000 cells per well) were cultured for 21 d to fully differentiate into an enterocyte-

like phenotype and to form a monolayer, as shown before [7,8]. Analysis of the proper

differentiation, morphology and barrier integrity of the Caco-2 cells cultured in our lab was

reported previously [28].

5.2.3 Tryptic Digestion of BSA

300 mg mL-1 BSA (containing traces of 1.3 ng zinc per mg BSA) was incubated with 150 mg

mL-1 trypsin in tris(hydroxylmethyl)-aminomethan (tris)-buffered saline (TBS-buffer) (50 mM

Tris, 110 mM NaCl, pH 7.5) at 37°C for 72 h on a thermoshaker. Successful in vitro digestion

of BSA was checked by electrophoresis, which was performed using polyacrylamide gels (4%

stacking and 20% resolving gel) and denaturing conditions [29]. Prior to zinc-uptake assays

using Zinpyr-1, trypsin was removed from the samples by ultracentrifugation using 10 kDa

membrane filters.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

65

5.2.4 Zinc Uptake Assay using Zinpyr-1

Cellular zinc uptake was analyzed using the fluorescent zinc probe Zinpyr-1. To this end, zinc

resorption was measured as the increase of the cellular free zinc concentration [nM] using

the equation by Grynkiewicz et al. [Zinc] = Kd × [(F - Fmin) / (Fmax - F)] [30] and a dissociation

constant for the zinc-Zinpyr-1-complex of 0.7 nM.[31] Caco-2 cells grown in 96-well plates

were loaded with 2.5 µM Zinpyr-1 in incubation buffer (120 mM NaCl, 5.4 mM KCl, 5 mM

glucose, 1 mM CaCl2, 1 mM MgCl2, 1 mM NaH2PO4, 10 mM HEPES, 0.3% BSA, pH 7.35) for 30

min. Afterwards, the supernatants were removed and the adherent cell monolayers were

washed twice with assay buffer (incubation buffer w/o BSA) to remove extracellular Zinpyr-

1. To determine the minimal (Fmin) or maximal (Fmax) fluorescence signal, 100 µL of assay

buffer containing either 20 µM TPEN, a zinc chelator, or 600 µM ZnSO4·7H2O, respectively,

were added to a subset of wells used for calibration. All other wells were filled with 100 µL

of assay buffer. After incubation for 20 min at 37°C, baseline Zinpyr-1 fluorescence (λex = 508

nm and λem = 527 nm) was measured using a fluorescence plate reader (SPARK, Tecan

Switzerland). Subsequently, 25 µL of 5-fold concentrated zinc-solutions containing either BSA

(digested or undigested) or FCS, respectively, were added to the cells in order to measure

zinc uptake (equal volumes of 20 µM TPEN (Fmin) and 600 µM ZnSO4·7H2O (Fmax) were added

to the respective wells used for calibration in order to yield comparable buffer volumes in all

wells). Fluorescence was measured again after incubation for an additional 20 min at 37°C.

5.2.5 Zinc Binding Properties of Digested BSA

The zinc binding properties of BSA after tryptic digestion were measured using the chelating

chromophore PAR. To this end, 20 µM PAR (stock solution 25 mM in H2O) and 15 mg mL-1 of

BSA or digested BSA, respectively, were titrated with 0–1000 µM ZnSO4·7H2O in 50 mM

HEPES-buffer at pH 7.4 and the absorption of the zinc-(PAR)2-complex was measured at 492

nm using a well plate reader (M200, Tecan, Switzerland) [32]. Data were analyzed with

GraphPad Prism software version 5.01 (GraphPad Software Inc., CA, USA) and a non-linear

regression assuming a one site specific binding with Hill slope as a function of the zinc-

concentration was applied.

5.2.6 Viability Assays

Caco-2 cells grown in 96-well plates were incubated with 0–1000 µM ZnSO4·7H2O in DMEM

without phenol red with 0 or 10% FCS, respectively, for 24 h. A total of 0.01% Triton-X-100

was used as a positive control. Cell viability was analyzed using three different endpoints:

dehydrogenase activity with MTT and WST, respectively, lysosomal cytotoxicity by neutral

red uptake (NRU-assay), and the amount of cellular protein using the SRB-assay. MTT, WST

and SRB were performed as previously described [28,33]. A final concentration of 40 mg L-1

neutral red was used and the cells were incubated for 3 h followed by washing with PBS and

lysis with 50% ethanol in H2O. The absorbance was measured at 540 nm [34]. Data were

analyzed with GraphPad Prism software version 5.01 (GraphPad Software Inc., CA, USA) and

a non-linear regression using a sigmoidal dose–response curve with variable slope as a

function of the logarithm of concentration was applied.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

66

5.2.7 Zinc Uptake Assay using Atomic Absorption Spectrometry

Cells were cultured in 6-well plates and incubated with 0-100 µM ZnSO4·7H2O in DMEM

without phenol red for 24 h. Finally, the cells were washed with ice-cold PBS and harvested

with a cell scraper on ice. The cell pellets were dissolved in a mixture of equal volumes of

67% ultrapure HNO3 and 30% H2O2 and dried at 92°C overnight using a thermoshaker. The

residues were dissolved in 0.67% HNO3 and the samples were analyzed by flame atomic

absorption spectrometry (FAAS) using a Perkin Elmer AAnalyst800 (Perkin Elmer, Germany).

5.2.8 Quantitative Real Time PCR (qPCR)

Cells were cultured in 6-well plates. After harvesting the differentiated cells on ice with PBS,

RNA was isolated using the Nucleo Spin II Kit and cDNA was synthesized using the iScript

cDNA Synthesis Kit according to the manufacturers’ protocols. Finally, mRNA-levels were

quantified by qPCR with SYBR™Green Super Mix, using the primers listed in Table 5.1.

Relative quantification of mRNA was realized using the 2δδCt-method [35] with Ct-values

normalized to β-actin and referred to non-treated control cells.

Table 5.1. Oligonucleotide sequences used for qPCR

Primer NCBI Reference

Sequence

Sequence fwd 5'-3' Sequence rev 5'-3' Ref.

zip-4 NM_017767 AGACTGAGCCCAGAGTTGAGGCTA TGTCGCAGAGTGCTACGTAGAGGA [36]

zip-5 NM_173596 GAGCAGGAGCAGAACCATTACCTG CAATGAGTGGTCCAGCAACAGAAG [36]

znt-1 NM_021194 GGCCAATACCAGCAACTCCAA TGCAGAAAAACTCCACGCATGT [37]

znt-5 NM_024055 AAGGACATCATGACAGTGCTCTAACTC CCAACTTTACAACACAAAGCCAGTAC [37]

mt1a NM_005946.2 CTCCTGCAAGAAGAGCTGCTG CAGCCCTGGGCACACTT

alp NM_001631.4 CCGCTTTAACCAGTGCAACA CCCATGAGATGGGTCACAGA [38]

β-actin NG_007992.1 CGCCCCAGGCACCAGGGC GCTGGGGTGTTGAAGGT [36]

5.2.9 Zinc Transport Studies

Zinc transport studies were performed using a Caco-2/HT-29-MTX co-culture. The cell ratio

of Caco-2 and HT-29-MTX cells was slightly modified after Nollevaux et al. [39], with an initial

cell ratio of 75% Caco-2 and 25% HT-29-MTX cells and alternated cell seeding. For this

purpose, 60,000 Caco-2 cells were transferred onto polycarbonate transwell membranes

(pore size 0.4 µm, culture area 1.12 cm2). 2 d after seeding of Caco-2, 20,000 HT-29-MTX

cells were added and co-cultured for an additional 18 d in DMEM with 10% FCS, 100 U ml-1

penicillin and 100 µg ml-1 streptomycin and 1% NEAA. After 21 d, the cells were incubated

with 0 µM, 25 µM, 50 µM or 100 µM ZnSO4·7H2O in transport buffer (Hanks’ Balanced Salt

Solution (HBSS)-buffer (130 mM NaCl, 10 mM KCl, 1 mM MgSO4, 50 mM HEPES, pH 7.5, 5

mM Glucose, mix of amino acids (398 µM L-arginine, 322 µM L-cysteine, 3.9 µM L-glutamine,

399 µM glycine, 115 µM L-histidine, 798 µM L-isoleucine, 798 µM L-leucine, 800 µM L-lysine;

201 µM L-methionine, 399 µM L-phenylalanine, 399 µM L-serine, 799 µM L-threonine, 78

µM L-tryptophan, 508 µM L-tyrosine, 798 µM L-valine))) on the apical side of the transport

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

67

chamber for 8 h. The basolateral compartment consisted of a cell culture medium with 0 or

30 mg mL-1 BSA. Membrane integrity was monitored by measuring TEER (transepithelial

electrical resistance) prior to and after the experiment with the epithelial volt-ohm meter

Millicell®ERS-2 (Millipore, USA). Additionally, the permeability of the cell monolayer was

determined by examining the basolateral concentrations of Fluorescein isothiocyanate

(FITC)-Dextran-20 (FD-20; 20,000 MWCO), which was used before to characterize Caco-2/HT-

29-MTX co-cultures [39] during the experiment. To this end, a final concentration of 0.25 mg

mL-1 FD-20 was added with the incubation buffer in the apical transwell chamber at the start

of the experiment and the amount of FD-20 in the basolateral and apical compartments at

the end of the experiment was analyzed by fluorescence measurements (λex = 485 nm, λem =

520 nm). Permeability for FD-20 was then calculated as the apparent permeability using the

following equation: Papp = (dQ/dt) × (1/A x c0), with dQ/dt FD-20 transport in mg s-1; A is the

area of the transwell (1.12 cm2) and c0 is the initial FD-20-concentration [40].

At the end of the experiment, the media of the apical and basolateral compartments were

collected and cells were harvested on ice in PBS, homogenized and centrifuged (800g). An

aliquot of cell homogenates was collected for protein quantification using the bicinchoninic

acid (BCA)-assay as described [41]. Subsequently, the cells were dried at 92°C overnight as

described above and dissolved in 0.67% HNO3. Zinc quantification in the apical, basolateral

and cellular compartments was conducted by inductively-coupled plasma mass

spectrometry (ICP-MS), after dilution (1:10 and 1:200) in 2% HNO3 containing 5 µg L-1

rhodium, using an Agilent 8800 ICP-QQQ (Agilent Technologies Deutschland GmbH,

Böblingen, Germany) in the single quadmode.

5.2.10 Statistical Analysis

Statistical significance was analyzed by one- or two-way analysis of variance (ANOVA),

followed by Bonferroni or Dunnett’s multiple comparison post hoc tests, as indicated in the

respective figure legends, using GraphPad Prism software version 5.01 (GraphPad Software

Inc., CA, USA). Error bars represent standard deviation or standard error of the mean, as

indicated, of three independent biological replicates.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

68

5.3 Results

5.3.1 Influence of Apical Serum on Zinc Uptake

Zinc resorption was analyzed as the increase of free zinc, measured with the zinc probe

Zinpyr-1. The effect of the apical serum concentration on the initial zinc uptake into Caco-2

cells was analyzed in the presence of 0 or 10% FCS. Notably, no effect of FCS on intracellular

free zinc was detected in control cells (0 µM zinc), but the presence of 10% FCS led to a

significantly lower increase of free zinc after incubation with 25 and 50 µM zinc (Figure 5.1).

In contrast, short-term zinc uptake measured with FAAS showed no statistically significant

change in the total cellular zinc content after addition of 50 µM extracellular zinc,

irrespective of the presence or absence of FCS (Figure 5.1B). Furthermore, to examine a

possible enhancement of zinc absorption due to the intracellular zinc sensor Zinpyr-1,

cellular zinc uptake was compared between controls and cells loaded with 2.5 µM Zinpyr-1,

showing that the probe had no significant impact on cellular zinc uptake (Figure 5.1C).

Figure 5.1: Impact of serum and Zinpyr-1 on short-term zinc uptake.

Zinc uptake into differentiated Caco-2 cells in the presence of 0 or 10% FCS, respectively, was investigated after 20 min by measuring the increase of intracellular free zinc with the low molecular weight probe Zinpyr-1 (A) or total cellular zinc with FAAS (B). (C) The impact of Zinpyr-1 on cellular zinc uptake was analyzed in differentiated Caco-2 cells loaded with 0 µM or 2.5 µM of the probe for 30 min, followed by incubation with zinc for 20 min and measurement of total cellular zinc by FAAS. Data are shown as means + SD of three independent experiments. Significant differences between 0 and 10% FCS are indicated (**p < 0.01; ***p < 0.001; two-way ANOVA with Bonferroni post hoc test).

Because albumin is the major protein in FCS [26], zinc resorption was also studied with

purified albumin, yielding similar results. In Figure 5.2A, zinc uptake into Caco-2 cells in the

presence of 15 mg ml-1 BSA is shown. Without BSA, free zinc increased proportional to the

added zinc concentration, but no noteworthy zinc uptake was observed in the presence of

albumin. Zinc uptake was partly restored by tryptic digestion of albumin. Moreover, cell free

measurements with the zinc chelating chromophore PAR confirmed the altered zinc binding

properties of BSA after tryptic digestion (Figure 5.2B). Here, 15 mg ml-1 BSA, corresponding

to 234.4 µM albumin, were saturated by addition of approximately 500 µM zinc,

corresponding well to the reported 1:2 stoichiometry of the albumin-zinc-complex [25]. The

digestion of albumin, however, diminished its zinc binding affinity, leading to a partial

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

69

recovery of zinc uptake. Gel electrophoresis of BSA before and after in vitro digestion shows

the typical band at 64 kDa, which vanishes after tryptic digestion, indicating a successful

degradation of the protein (Figure 5.2C). Besides, the band with an apparent molecular mass

of 20 kDa can be correlated with bovine trypsin, which has a molecular mass of 24 kDa [42].

Figure 5.2: Effect of albumin digestion on intestinal zinc resorption.

(A) Zinc resorption is shown as increase of cellular free zinc measured with Zinpyr-1 in differentiated Caco-2 cells. Cells were incubated for 10 min with 0–50 µM ZnSO4·7H2O alone and in the presence of either 15 mg mL-1 undigested BSA or after tryptic digestion. Additionally, blank digestion without BSA was analyzed as a control (CTR). Data are displayed as means ± SEM of three independent experiments. Significant differences between the incubated zinc concentrations are indicated (*p < 0.05; ***p < 0.001; two-way ANOVA with Bonferroni post hoc test). (B) Comparison of the zinc binding properties of 15 mg mL-1 BSA before and after tryptic digestion analyzed with the colorimetric zinc chelator 4-(2-pyridylazo)resorcinol (PAR). Shown is the absorption of the zinc-(PAR)2-complex relative to the maximal absorption at 492 nm. Data are displayed as means ± SD of three independent experiments. (C) Verification of successful tryptic digestion of albumin with either 0 or 150 mg mL-1 trypsin by gel electrophoresis.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

70

5.3.2 Impact of Serum on Zinc Cytotoxicity

The impact of apically added FCS was investigated with four different cell viability assays. All

showed a trend of reduced cytotoxicity in the presence of serum (Figure 5.3 and

Supplemental Table S5.2). This was statistically significant for the cellular retention of

neutral red, based on the 95% confidence intervals of the lethal concentration (LC50) values

(Supplemental Table S5.2), indicating that zinc cytotoxicity is buffered by high serum

concentrations.

Figure 5.3: Impact of serum on zinc cytotoxicity.

Differentiated Caco-2 cells were incubated with 0–1000 µM ZnSO4·7H2O for 24 h and cell viability was analyzed using several assays with different endpoints. Mitochondrial activity was assayed using WST (A) and MTT (B) and lysosomal activity using NRU (C). Moreover, the effect of zinc on cellular protein content was analyzed with SRB (D). Data are shown as means ± SD of three independent experiments. Significant differences are indicated (*,# p < 0.05; **,## p < 0.01; ***,### p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test).

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

71

5.3.3 Effect of Serum on Zinc Uptake after Long-term Incubation

In addition to short-term zinc uptake in the presence of FCS, zinc absorption after long-term

incubation was investigated using FAAS. Remarkably, no statistically significant difference in

zinc uptake after 24 h incubation of 0–100 µM zinc was observed (Figure 5.4). Regardless of

the FCS concentration, the relative zinc uptake decreased with increased added zinc

concentration, showing a regulated zinc absorption into Caco-2 cells (Figure 5.4A). The

fractional zinc absorption shown in this figure depends on the apically added volume, as it

affects the total amount of zinc that is added. Yet, the values are comparable within the

experiment, as the same volume was added to all samples. When the data were normalized

to the cellular protein content, they showed an increase of cellular zinc after treatment with

zinc both in the presence and absence of FCS, and displayed a general trend to higher values

in the absence of FCS (Figure 5.4B).

Figure 5.4: Influence of serum on zinc uptake in Caco-2 cells after incubation for 24 h.

Differentiated Caco-2 cells were incubated with 0–100 µM zinc with 0% or 10% FCS, respectively, and cellular zinc uptake was quantified using FAAS. The cellular zinc uptake is shown relative to the initially added zinc concentration (A) as well as relative to the protein content of the cells (B). Data are displayed as means + SD of three independent experiments.

Moreover, gene expression of important proteins for cellular zinc homeostasis was analyzed,

yielding only a slight but not statistically significant difference for mt1a, and no effect for the

zinc transporters zip4, zip5, znt-1 and znt-5 after incubation with 50 µM or 100 µM zinc in

the presence of 0 or 10% FCS (Figure 5.5). However, within the same serum concentrations

mt1a and znt-1 were slightly upregulated by 50 µM and 100 µM zinc, while the other zinc

transporters showed no zinc-dependent changes of expression at all.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

72

Figure 5.5: Impact of zinc concentration and serum on gene expression of proteins involved in cellular zinc homeostasis.

Differentiated Caco-2 cells were cultivated for 24 h with 0–100 µM ZnSO4·7H2O with 0% and 10% FCS, respectively. Subsequently, gene expression of selected proteins of intestinal zinc homeostasis was analyzed using qPCR and is depicted relative to cells incubated with 0 µM zinc. Data are shown as means + SD of three independent experiments.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

73

5.3.4 Effect of the Basolateral Albumin Concentration on Zinc Resorption

Zinc transport studies were performed using the transwell system and a co-culture of the

intestinal cell line Caco-2 and mucin producing goblet cells HT-29-MTX. Before performing

the transport studies, the co-culture of Caco-2 and HT-29-MTX was characterized concerning

the suitable cellular ratio by investigating the mucin secretion using alcianblue- and periodic

acid-Schiff (PAS)-staining (Supplemental Figure S5.1) and adequate differentiation of

enterocytes by monitoring the activity of the differentiation marker alkaline phosphatase

(Supplemental Figure S5.2) and the expression of alp and muc5ac (Supplemental Figure S5.3

and Supplemental Figure Table S5.1) as shown previously [39].

Next, the impact of serum albumin on zinc resorption was analyzed by comparing zinc

transport in the presence or absence of 30 mg mL-1 albumin (BSA) in the basolateral

compartment. The integrity of the cell monolayer was monitored at the beginning and end

of the experiments by measuring TEER and during the experiments by determining the

paracellular permeability for FD-20, revealing no impairment of both parameters during the

transport assay (Supplemental Figure S5.4). In Figure 5.6, the zinc uptake from the apical

compartment and the fractional resorption after 8 h of zinc incubation, relative to the

amount of added zinc, and the zinc transport rates in nmol per cm2 resorption area, are

shown. Of note, the apical and basolateral volumes added to the cells are important factors

affecting relative zinc uptake, as they determine the total amount of added zinc.

Consequently, this study reports the resorption in the presence or absence of BSA using

identical volumes for all samples, in order to obtain comparable results. Nevertheless,

absolute numbers for fractional resorption will vary if experiments are conducted under

different conditions. Figure 5.7 depicts detailed quantitative data of zinc uptake into the

cells, the cellular zinc content and the amount of zinc transported into the basolateral

compartment, each in ng cm-2. After zinc supplementation, the decrease of apical zinc

concentration was more pronounced in the transport system with albumin, while cellular

zinc levels were even slightly lower than those in the absence of albumin, resulting from a

higher zinc export to the basolateral side in the presence of albumin (Figure 5.7C).

Zinc transport is increasing with added zinc, resulting in higher zinc transport rates in the

presence of albumin on the basolateral side of the transwell chamber (1.8–3.6 nmol zinc cm-

2, Figure 5.6C and D). The fractional zinc resorption, which represents the amount of finally

resorbed (basolateral) zinc relative to the luminal zinc concentration, shows an inversely

proportional relation to the initially added zinc (Figure 5.6E and F). While zinc uptake from

the apical site is slightly higher in the transport system with albumin (Figure 5.6A and B),

there is increased zinc transport into the basolateral chamber in the presence of BSA,

leading to a zinc resorption of 2.5–6% added zinc. According to a two-way ANOVA followed

by a Bonferroni posttest, there are significant differences in fractional zinc resorption

between the two assays after adding 25 µM zinc (p < 0.001) and 50 µM zinc (p < 0.05) to the

apical chamber.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

74

Figure 5.6: Effect of serum albumin as a basolateral zinc acceptor on intestinal zinc resorption

in a Caco-2/HT-29-MTX co-culture.

Shown are the amounts of zinc that disappeared from the apical side (apical zinc uptake) (A and B) and the fractional zinc resorption (E and F) relative to the added zinc concentration of the transport-assay without (A and E) and with 30 mg mL-1 albumin (B and F) in the basolateral compartment. Moreover, zinc transport rates in nmol zinc per cm2 resorption area in the absence (C) and presence of albumin (D) are displayed. Data are presented as means + SD of three independent experiments. Significant differences to control cells (0 µM zinc) are indicated (*p < 0.05; ***p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test). According to a two-way ANOVA with a Bonferroni post hoc test comparing the results within one added zinc concentration, there are significant differences between the apical zinc concentration (25 µM: p < 0.001) and the fractional zinc-resorption (25 µM: p < 0.001; 50 µM: p < 0.05) with or without albumin in the basolateral compartment.

Additionally, cellular zinc uptake normalized to basal zinc levels (0 µM zinc) increases

significantly with the extracellular added zinc concentration. Interestingly, the amount of

apically absorbed zinc that stays in the cell seems to be higher when there is no BSA present

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

75

in the basolateral compartment (Figure 5.7). These findings support the abovementioned

observation of an increased zinc resorption in the presence of higher basolateral albumin

concentrations.

Figure 5.7: Impact of basolateral albumin concentration on cellular zinc uptake and transport.

(A and B) The cellular zinc uptake of the Caco-2/HT-29-MTX co-culture is shown relative to cellular protein content after subtracting basal cellular zinc content. Data are displayed as means + SD of three independent experiments. Significant differences to control cells (0 µM zinc) are indicated (*p < 0.05; **p < 0.01; ***p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test). (C) Shown are the amounts of zinc that disappeared from the apical compartment (zinc uptake), the cellular zinc content, and the amounts resorbed into the basolateral compartment, all in ng zinc per resorption area (in cm2). Data are presented as means + SD of three independent experiments.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

76

5.4 Discussion

In vitro cell culture models in general, and especially when investigating intestinal resorption

of food components, notably trace elements, always need to reflect the in vivo situation as

close as possible. The speciation of zinc in a cell culture medium has a strong impact on its

availability and cellular uptake [4,24]. Thus, the composition of the incubation medium has

to be considered when analyzing zinc resorption with intestinal cell models. FCS is commonly

used in the cell culture medium and contains essential components for cell growth and

proliferation [43]. Furthermore, albumin accounts for about 60% of the total FCS protein

concentration [26], is well known for its high zinc binding affinity [25], and is the major zinc

transport protein in blood serum in vivo [44]. Previous studies of our group already

demonstrated that FCS buffers the free zinc concentration in cell culture medium, resulting

in a considerably smaller available zinc concentration for immune cells [24]. The present

study investigates the impact of FCS on zinc uptake of enterocytes. First, short-term zinc

uptake of enterocytes in the presence of 10% FCS was analyzed, resulting in a decrease of

zinc absorption (Figure 5.1). A concentration of 10% FCS results in an albumin content of

1.55 mg mL-1 albumin [45] (corresponding to 24.2 µM). Based on studies demonstrating that

albumin has, at least under in vitro conditions, two zinc binding sites [25], this should be

sufficient for complexing a twofold molar excess of zinc. Accordingly, 10% FCS was sufficient

to reduce the short-term uptake of 50 µM zinc into enterocytes. Similar observations were

made when analyzing the effect of 15 mg mL-1 purified albumin, representing physiological

but low serum concentrations, on zinc uptake in the intestinal cell line Caco-2 (Figure 5.2). In

accordance with the present results, our previous studies showed that even smaller

concentrations bind zinc, reducing the cellular zinc availability for an intracellular zinc probe

[24]. However, it has to be considered that in some cases, like for immune cells, the

application of 100% FCS is more realistic compared to their environment in vivo [24]. In

contrast, for intestinal cells the apical occurrence of serum or albumin, respectively, is not

realistic considering the digestion process throughout the whole gastrointestinal tract. For

this reason, the observed negative effect of albumin on zinc bioavailability cannot be

transferred to the in vivo situation in the intestinal lumen. Consequently, the effect of

digested albumin on intestinal zinc uptake was investigated resulting in a significant increase

of intracellular free zinc compared to the zinc uptake in the presence of undigested protein

(Figure 5.2A). However, digested albumin still binds zinc, impairing its uptake into

enterocytes (Figure 5.2A and B), which is in agreement with the effect of albumin on zinc

resorption in vivo [46].

Next, the impact of apically added high serum concentrations on zinc cytotoxicity after long-

term incubation was analyzed, indicating that albumin buffers zinc cytotoxicity in

enterocytes, starting at 250 µM extracellular added zinc (Figure 5.3). It was shown that

already narrow changes in free zinc can determine its toxicity [23] and that cells could be

protected against zinc cytotoxicity by albumin [24], suggesting that zinc is toxic for Caco-2

cells as soon as the buffer capacity of albumin is exceeded. The investigation of zinc uptake

after long-term incubation (Figure 5.4) revealed a regulated and concentration-dependent

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

77

cellular zinc uptake that is comparable to previous studies obtained with Caco-2 cells after

24 h [47,48] and consistent with the kinetics of zinc resorption [9,10,49]. In contrast to the

short-term zinc uptake in Caco-2 cells, no impact of the extracellular albumin concentration

on long-term zinc absorption was observed. However, these results are not contradictory to

those of the cytotoxicity studies, because the uptake assays were performed for non-toxic

zinc concentrations (0-100 µM) and the serum-dependent impairment of cell viability is only

observed starting at 250 µM zinc. Moreover, studies with endothelial cells reported a

facilitated zinc resorption in the presence of albumin [27], thus the Caco-2 cells might have

resorbed albumin-bound zinc after 24 h incubation. Besides, the phenomenon of zinc uptake

together with albumin was already observed 30 min after incubation of peripheral blood

mononuclear cells (PBMC) with zinc and BSA [24]. In contrast, no significant change in

cellular zinc was determined using FAAS (Figure 5.1B and C) after 20 min as well as in the

first 6 h [48] of incubating zinc together with albumin in Caco-2 cells. These observations

indicate that zinc uptake in Caco-2 cells is not detectable using FAAS after short-term

incubation, probably because the resulting changes are too small compared to the total

amount of cellular zinc, which is measured by FAAS. This underlines the importance of low

molecular weight sensors to conduct initial zinc uptake measurements and detect small

changes in the intracellular free zinc pool. Yet again, these observations cannot be

transferred to the situation in the intestinal lumen in vivo and are only valid for in vitro

experiments with Caco-2 cells. Additionally, gene expression after 24 h incubation showed

no significant differences in the expression of mt1a and selected zinc transporters with or

without serum (Figure 5.5). Consistent with previous findings in Caco-2 cells [14,15,48] there

is a concentration-dependent trend for upregulation of mt1a and znt-1 after treatment with

zinc for 24 h (Figure 5.5A and D). According to earlier studies, the reason for the increase of

metallothionein and znt-1, both regulated via zinc-dependent transcription factor MTF-1

[50], after zinc incubation, is mainly to maintain intracellular zinc homeostasis [15,51].

Furthermore, no zinc-dependent expression of the other zinc transporters was observed,

which is comparable to previous studies with Caco-2 cells [14,15].

Finally, the effect of the basolateral albumin concentration on zinc resorption was examined

with zinc transport studies using a co-culture of Caco-2 cells and the goblet cell line HT-29-

MTX. Intestinal zinc resorption is mainly controlled by internalization or expression of

enterocyte zinc transporters [2], but the definite molecular mechanisms for regulating zinc

resorption must be further investigated [52]. Notably the basolateral zinc export into the

blood stream is mediated by the expression of the basolateral exporter ZnT-1, which is

upregulated at high zinc concentrations to enhance zinc transport and to maintain

intracellular homeostasis in enterocytes [2]. According to the present study, a basolateral

zinc acceptor, such as serum albumin, might also play an important role in the regulation of

zinc resorption. The fractional zinc resorption is significantly augmented when adding

albumin on the basolateral side (Figure 5.6), indicating that basolateral serum albumin

enhances zinc resorption and acts as a basolateral zinc acceptor. Furthermore, a basolateral

acceptor seems to increase the cellular release to the basolateral compartment, as the

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

78

cellular zinc concentration tends to be higher in the absence of albumin (Figure 5.7). Hence,

apically absorbed zinc does not remain inside the cells in the presence of albumin, but is

transported to the basolateral compartment (Figure 5.7C). This was observed before,

suggesting that plasma albumin is not only involved in zinc transport but also acts as a zinc-

carrier by removing zinc from intestinal-mucosal cells [53]. Besides, the albumin

concentration in this study (30 mg mL-1) is in the vicinity of the physiological human plasma

albumin (HSA) concentration (typically 35–50 mg mL-1) [24]. Yet, it has to be noted that this

study aims to clarify the impact of albumin in in vitro models of intestinal absorption, which

typically involve BSA, while the function of albumin as an acceptor in humans would have to

be confirmed with HSA instead of BSA. In contrast to the human plasma zinc concentrations

of 12–16 µM [54], the basolateral medium contains only 3 µM zinc, resulting in a 110-fold

molar excess of albumin compared to a molar albumin : zinc-ratio of 30 in vivo [24].

However, the fractional resorption of zinc in vivo was reported to be around 20–30%, but

highly dependent on the amount of consumed zinc [55,56]. The fractional absorption in this

study shows the same concentration-dependency as observed in vivo and is comparable to

studies conducted with Caco-2 monocultures [57] regarding the results without BSA.

Additionally, results obtained for the zinc resorption in the presence of albumin in this study

are even higher than those observed before [58,59]. In fact, the applied zinc concentrations

in this experimental setting represent luminal zinc concentrations in vivo after consumption

of a meal containing normal amounts of zinc [13]. As the digested food components offer

various possibilities for zinc buffering, it has to be considered that these estimated

concentrations differ from free and available zinc concentrations. This factor though could

be easily incorporated into the intestinal model by combining it with an in vitro digestion

model [60]. Furthermore, while zinc transport was conducted using a co-culture of

enterocytes and mucin producing goblet cells, there are still several factors missing

compared to the in vivo situation, particularly concerning the basolateral side of the

intestinal epithelium. Among them, the transport and distribution of zinc into the other

organs and most importantly humoral factors, which are still topics of current investigations

[61]. Nevertheless, these outcomes underline the crucial character of distinct apical and

basolateral media composition for enterocytes in zinc resorption studies: the apical buffer

has to be as close to the luminal fluid in vivo as possible and the basal media must resemble

the serum in vivo.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

79

5.5 Conclusion

In summary, our work clearly demonstrates the impact of the zinc binding protein albumin

on zinc resorption using in vitro cell models. Albumin reduces the amount of cellular

available zinc and has a severe impact on initial zinc uptake in enterocytes when applied

apically. Yet, it plays an important role as a basolateral zinc acceptor enhancing fractional

zinc resorption and augmenting the cellular release of zinc into the blood stream. Thus, it is

not only important to reduce the amount of FCS to a minimum in the apical medium when

using in vitro intestinal models to investigate zinc uptake, but it is also essential to adapt the

basolateral matrix to the in vivo situation when analyzing zinc resorption in the future.

5.6 Conflict of Interest

There are no conflicts to declare.

5.7 Acknowledgements

The work of HH and TS was supported by the Deutsche Forschungsgemeinschaft (TraceAge –

DFG Research Unit on Interactions of essential trace elements in healthy and diseased

elderly, Potsdam-Berlin-Jena, FOR 2558/1, HA 4318/4-1, SCHW903/16-1).

5.8 References

1. Hambidge, K.M.; Miller, L.V.; Westcott, J.E.; Sheng, X.; Krebs, N.F. Zinc bioavailability

and homeostasis. Am. J. Clin. Nutr. 2010, 91, 1478S-1483S.

2. Wang, X.; Zhou, B. Dietary zinc absorption: A play of ZIPs and ZnTs in the gut. IUBMB

life 2010, 62, 176-182.

3. King, J.C. Zinc: An essential but elusive nutrient. Am. J. Clin. Nutr. 2011, 94, 6795-

6845.

4. Ollig, J.; Kloubert, V.; Weßels, I.; Haase, H.; Rink, L. Parameters influencing zinc in

experimental systems in vivo and in vitro. Metals 2016, 6, 71.

5. Langerholc, T.; Maragkoudakis, P.A.; Wollgast, J.; Gradisnik, L.; Cencic, A. Novel and

established intestinal cell line models – an indispensable tool in food science and

nutrition. Trends Food Sci. Tech. 2011, 22, S11-S20.

6. Madara, J. L. Functional Morphology of Epithelium of the Small Intestine. In

Comprehensive Physiology; Terjung, R., Ed.; JohnWiley & Sons, Inc.: Hoboken, NJ, USA,

2011.

7. Pinto, M.; Robineleon, S.; Appay, M.D.; Kedinger, M.; Triadou, N.; Dussaulx, E.;

Lacroix, B.; Simonassmann, P.; Haffen, K.; Fogh, J., et al. Enterocyte-like differentiation

and polarization of the human-colon carcinoma cell-line Caco-2 in culture. Biol. Cel.

1983, 47, 323-330.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

80

8. Sambuy, Y.; de Angelis, M.; Ranaldi, G.; Scarino, M.L.; Stammati, A.; Zucco, F. The

Caco-2 cell line as a model for the intestinal barrier: Influence of cell and culture-

related factors on Caco-2 cell functional characteristics. Cell Biol. Toxicol. 2005, 21, 1-

26.

9. Yasuno, T.; Okamoto, H.; Nagai, M.; Kimura, S.; Yamamoto, T.; Nagano, K.;

Furubayashi, T.; Yoshikawa, Y.; Yasui, H.; Katsumi, H.; Sakane, T.; Yamamoto, A. In

vitro study on the transport of zinc across intestinal epithelial cells using Caco-2

monolayers and isolated rat intestinal membranes. Biol. Pharm. Bull. 2012, 35, 588-

593.

10. Raffaniello, R.D.; Shih-Yu Lee; Teichberc, S.; Wapnir, R.A. Distinct mechanisms of zinc

uptake at the apical and basolateral membranes of Caco-2 cells. J. Cell. Physiol. 1992,

152, 356-361.

11. Jovanı, M.; Barbera, R.; Farre, R.; de Aguilera, E.M. Calcium, iron, and zinc uptake from

digests of infant formulas by Caco-2 cells. J. Agric. Food Chem.2001, 49, 3480−3485.

12. Jou, M.-Y.; Du, X.; Hotz, C.; Lönnerdal, B. Biofortification of rice with zinc: Assessment

of the relative bioavailability of zinc in a Caco-2 cell model and suckling rat pups. J.

Agric. Food Chem. 2012, 60, 3650-3657.

13. Cragg, R.A.; Phillips, S.R.; Piper, J.M.; Varma, J.S.; Campbell, F.C.; Mathers, J.C.; Ford,

D. Homeostatic regulation of zinc transporters in the human small intestine by dietary

zinc supplementation. Gut 2005, 54, 469-478.

14. Zemann, N.; Zemann, A.; Klein, P.; Elmadfa, I.; Huettinger, M. Differentiation- and

polarization-dependent zinc tolerance in Caco-2 cells. Eur. J. Nutr. 2011, 50, 379-386.

15. Shen, H.; Qin, H.; Guo, J. Cooperation of metallothionein and zinc transporters for

regulating zinc homeostasis in human intestinal Caco-2 cells. Nutr. Res. 2008, 28, 406-

413.

16. Lozoya-Agullo, I.; Araújo, F.; González-Álvarez, I.; Merino-Sanjuán, M.; González-

Álvarez, M.; Bermejo, M.; Sarmento, B. Usefulness of Caco-2/HT29-MTX and Caco-

2/HT29-MTX/Raji b coculture models to predict intestinal and colonic permeability

compared to Caco-2 monoculture. Mol. Pharm. 2017, 14, 1264-1270.

17. Hilgendorf, C.; Spahn-Langguth, H.; Regardh, C.G.; Lipka, E.; Amidon, G.L.; Langguth, P.

Caco-2 versus Caco-2/HT29-MTX co-cultured cell lines: Permeabilities via diffusion,

inside- and outside-directed carrier-mediated transport. J. Pharm. Sci. 2000, 89, 63-

75.

18. Beduneau, A.; Tempesta, C.; Fimbel, S.; Pellequer, Y.; Jannin, V.; Demarne, F.;

Lamprecht, A. A tunable Caco-2/Ht29-MTX co-culture model mimicking variable

permeabilities of the human intestine obtained by an original seeding procedure. Eur.

J. Pharm. Biopharm. 2014, 87, 290-298.

19. Mahler, G.J.; Shuler, M.L.; Glahn, R.P. Characterization of Caco-2 and HT29-MTX

cocultures in an in vitro digestion/cell culture model used to predict iron

bioavailability. J. Nutr. Biochem. 2009, 20, 494-502.

20. Bansil, R.; Turner, B.S. The biology of mucus: Composition, synthesis and organization.

Adv. Drug Deliv. Rev. 2018, 124, 3-15.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

81

21. Maret, W. Analyzing free zinc(II) ion concentrations in cell biology with fluorescent

chelating molecules. Metallomics 2015, 7, 202-211.

22. Carpenter, M.C.; Lo, M.N.; Palmer, A.E. Techniques for measuring cellular zinc. Arch.

Biochem. Biophys. 2016, 611, 20-29.

23. Bozym, R.A.; Chimienti, F.; Giblin, L.J.; Gross, G.W.; Korichneva, I.; Li, Y.; Libert, S.;

Maret, W.; Parviz, M.; Frederickson, C.J., et al. Free zinc ions outside a narrow

concentration range are toxic to a variety of cells in vitro. Exp. Biol. Med. (Maywood)

2010, 235, 741-750.

24. Haase, H.; Hebel, S.; Engelhardt, G.; Rink, L. The biochemical effects of extracellular

Zn2+ and other metal ions are severely affected by their speciation in cell culture

media. Metallomics 2015, 7, 102-111.

25. Bal, W.; Sokołowska, M.; Kurowskaa, E.; Faller, P. Binding of transition metal ions to

albumin: Sites, affinities and rates Biochim. Biophys. Acta 2013, 1830, 5444–5455.

26. Zheng, X.; Baker, H.; Hancock, W.S.; Fawaz, F.; McCaman, M.; Pungor, E. Proteomic

analysis for the assessment of different lots of fetal bovine serum as a raw material

for cell culture. Part IV. Application of proteomics to the manufacture of biological

drugs. Biotechnol. Prog. 2006, 22, 1294-1300.

27. Tibaduiza, E.C.; Bobilya, D.J. Zinc transport across an endothelium includes vesicular

cotransport with albumin. J. Cell. Physiol. 1996, 167, 539-547.

28. Maares, M.; Keil, C.; Thomsen, S.; Gunzel, D.; Wiesner, B.; Haase, H. Characterization

of Caco-2 cells stably expressing the protein-based zinc probe eCalwy-5 as a model

system for investigating intestinal zinc transport. J. Trace Elem. Med. Biol. 2018, 49,

296-304.

29. Laemmli, U.K. Cleavage of structural proteins during the assembly of the head of

bacteriophage T4. Nature 1970, 227, 680-685.

30. Grynkiewicz, G.; Poenie, M.; Tsien, R.Y. A new generation of Ca2+ indicators with

greatly improved fluorescence properties. J. Biol. Chem. 1985, 260, 3440-3450.

31. Burdette, S.C.; Walkup, G.K.; Spingler, B.; Tsien, R.Y.; Lippard, S.J. Fluorescent sensors

for Zn2+ based on a fluorescein platform: Synthesis, properties and intracellular

distribution. J. Am. Chem. Soc. 2001, 123, 7831-7841.

32. Kocyła, A.; Pomorski, A.; Krężel, A. Molar absorption coefficients and stability

constants of metal complexes of 4-(2-pyridylazo)resorcinol (par): Revisiting common

chelating probe for the study of metalloproteins. J. Inorg. Biochem. 2015, 152, 82-92.

33. Vichai, V.; Kirtikara, K. Sulforhodamine b colorimetric assay for cytotoxicity screening.

Nat. Protoc. 2006, 1, 1112-1116.

34. Repetto, G.; del Peso, A.; Zurita, J.L. Neutral red uptake assay for the estimation of

cell viability/cytotoxicity. Nat. Protoc. 2008, 3, 1125-1131.

35. Livak, K.J.; Schmittgen, T.D. Analysis of relative gene expression data using real-time

quantitative pcr and the 2(-delta delta C(t)) method. Methods 2001, 25, 402-408.

36. Leung, K.W.; Liu, M.; Xu, X.; Seiler, M.J.; Barnstable, C.J.; Tombran-Tink, J. Expression

of ZnT and ZIP zinc transporters in the human RPE and their regulation by

neurotrophic factors. Invest. Ophthalmol. Vis. Sci. 2008, 49, 1221-1231.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

82

37. Overbeck, S.; Uciechowski, P.; Ackland, M.L.; Ford, D.; Rink, L. Intracellular zinc

homeostasis in leukocyte subsets is regulated by different expression of zinc

exporters ZnT-1 to ZnT-9. J. Leukoc. Biol. 2008, 83, 368-380.

38. Wolf, K.; Schulz, C.; Riegger, G.A.J.; Pfeifer, M. Tumour necrosis factor‐α induced

CD70 and interleukin‐7R mRNA expression in BEAS‐2B cells. Eur. Respir. J. 2002, 20,

369-375.

39. Nollevaux, G.; Deville, C.; El Moualij, B.; Zorzi, W.; Deloyer, P.; Schneider, Y.J.; Peulen,

O.; Dandrifosse, G. Development of a serum-free co-culture of human intestinal

epithelium cell-lines (Caco-2/HT29-5M21). BMC Cell Biol. 2006, 7, 20.

40. Artursson, P.; Magnusson, C. Epithelial transport of drugs in cell culture. Ii: Effect of

extracellular calcium concentration on the paracellular transport of drugs of different

lipophilicities across monolayers of intestinal epithelial (Caco-2) cells. J. Pharm. Sci.

1990, 79, 595-600.

41. Smith, P.K.; Krohn, R.I.; Hermanson, G.T.; Mallia, A.K.; Gartner, F.H.; Provenzano,

M.D.; Fujimoto, E.K.; Goeke, N.M.; Olson, B.J.; Klenk, D.C. Measurement of protein

using bicinchoninic acid. Anal. Biochem. 1985, 150, 76-85.

42. Walsh, K.A. Trypsinogens and trypsins of various species. Meth. Enzymol. 1970; Vol.

19, pp 41-63.

43. Gstraunthaler, G. Alternatives to the use of fetal bovine serum: Serum-free cell

culture. Altex 2003, 20, 275-281.

44. Scott, B.J.; Bradwell, A.R. Identification of the serum binding proteins for iron, zinc,

cadmium, nickel, and calcium. Clin. Chem. 1983, 29, 629-633.

45. Kenneth, V.H.; John, A.S.; Walter, C. Fetal bovine serum: A multivariate standard.

Proc. Soc. Exp. Biol. Med. 1975, 149, 344-347.

46. Davidsson, L.; Almgren, A.; SandströM, B.; Juillerat, M.E.-A.; Hurrell, R.F. Zinc

absorption in adult humans: The effect of protein sources added to liquid test meals.

Br. J. Nutr. 2007, 75, 607-613.

47. Cheng, Z.; Tako, E.; Yeung, A.; Welch, R.M.; Glahn, R.P. Evaluation of metallothionein

formation as a proxy for zinc absorption in an in vitro digestion/Caco-2 cell culture

model. Food Funct. 2012, 3, 732-736.

48. Gefeller, E.M.; Bondzio, A.; Aschenbach, J.R.; Martens, H.; Einspanier, R.; Scharfen, F.;

Zentek, J.; Pieper, R.; Lodemann, U. Regulation of intracellular Zn homeostasis in two

intestinal epithelial cell models at various maturation time points. J. Physiol. Sci.

2015, 65, 317-328.

49. Reeves, P.G.; Briske-Anderson, M.; Johnson, L. Pre-treatment of Caco-2 cells with zinc

during the differentiation phase alters the kinetics of zinc uptake and transport. J.

Nutr. Biochem. 2001, 12, 674–684.

50. Langmade, S.J.; Ravindra, R.; Daniels, P.J.; Andrews, G.K. The transcription factor

MTF-1 mediates metal regulation of the mouse ZnT1 gene. J. Biol. Chem.2000, 275,

34803- 34809.

The Impact of Apical and Basolateral Albumin on Intestinal Zinc Resorption in the Caco-2/HT-29-MTX Co-culture Model,

83

51. McMahon, R.J.; Cousins, R.J. Regulation of the zinc transporter ZnT-1 by dietary zinc.

Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 4841-4846.

52. Krebs, N.F.; Hambidge, K.M. Zinc metabolism and homeostasis: The application of

tracer techniques. BioMetals 2001, 14, 397-412.

53. Smith, K.T.; Failla, M.L.; Cousins, R.J. Identification of albumin as the plasma carrier

for zinc absorption by perfused rat intestine. Biochem. J. 1979, 184, 627-633.

54. Maares, M.; Haase, H. Zinc and immunity: An essential interrelation. Arch. Biochem.

Biophys. 2016, 611, 58-65.

55. Kristensen, M.B.; Hels, O.; Morberg, C.M.; Marving, J.; Bugel, S.; Tetens, I. Total zinc

absorption in young women, but not fractional zinc absorption, differs between

vegetarian and meat-based diets with equal phytic acid content. Br. J. Nutr. 2006, 95,

963-967.

56. King, J.C.; Shames, D.M.; Woodhouse, L.R. Zinc homeostasis in humans. J. Nutr. 2000,

130, 1360S-1366S.

57. Jou, M.Y.; Du, X.; Hotz, C.; Lönnerdal, B. Biofortification of rice with zinc: Assessment

of the relative bioavailability of zinc in a Caco-2 cell model and suckling rat pups. J.

Agric. Food Chem. 2012, 60, 3650-3657.

58. Moltedo, O.; Verde, C.; Capasso, A.; Parisi, E.; Remondelli, P.; Bonatti, S.; Alvarez-

Hernandez, X.; Glass, J.; Alvino, C.G.; Leone, A. Zinc transport and metallothionein

secretion in the intestinal human cell line Caco-2. J. Biol. Chem. 2000, 275, 31819-

31825.

59. Finley, J.W.; Briske-Anderson, M.; Reeves, P.G.; Johnson, L.K. Zinc uptake and

transcellular movement by Caco-2 cells: Studies with media containing fetal bovine

serum. J. Nutr. Biochem. 1995, 6, 137-144.

60. Lichtenstein, D.; Ebmeyer, J.; Knappe, P.; Juling, S.; Bohmert, L.; Selve, S.; Niemann,

B.; Braeuning, A.; Thunemann, A.F.; Lampen, A. Impact of food components during in

vitro digestion of silver nanoparticles on cellular uptake and cytotoxicity in intestinal

cells. Biol. Chem. 2015, 396, 1255-1264.

61. Hennigar, S.R.; McClung, J.P. Hepcidin attenuates zinc efflux in Caco-2 cells. FASEB J.

2016, 30, 292.293.

84

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

85

Chapter 6. In vitro Studies on Zinc Binding and Buffering by

Intestinal Mucins5

Abstract

The investigation of luminal factors influencing zinc availability and accessibility in the

intestine is of great interest when analyzing parameters regulating intestinal zinc resorption.

Of note, intestinal mucins were suggested to play a beneficial role in the luminal availability

of zinc. Their exact zinc binding properties, however, remain unknown and the impact of

these glycoproteins on human intestinal zinc resorption has not been investigated in detail.

Thus, the aim of this study is to elucidate the impact of intestinal mucins on luminal uptake of

zinc into enterocytes and its transfer into the blood. In the present study, in vitro zinc binding

properties of mucins were analyzed using commercially available porcine mucins and

secreted mucins of the goblet cell line HT-29-MTX. The molecular zinc binding capacity and

average zinc binding affinity of these glycoproteins demonstrates that mucins contain

multiple zinc binding sites with biologically relevant affinity within one mucin molecule. Zinc

uptake into the enterocyte cell line Caco-2 was impaired by zinc-depleted mucins. Yet this

does not represent their form in the intestinal lumen in vivo under zinc adequate conditions.

In fact, zinc uptake studies into enterocytes in the presence of mucins with differing degree of

zinc saturation revealed zinc buffering by these glycoproteins, indicating that mucin-bound

zinc is still available for the cells. Finally, the impact of mucins on zinc resorption using three-

dimensional cultures was studied comparing the zinc transfer of a Caco-2/HT-29-MTX co-

culture and conventional Caco-2 monoculture. Here, the mucin secreting co-cultures yielded

higher fractional zinc resorption and elevated zinc transport rates, suggesting that intestinal

mucins facilitate the zinc uptake into enterocytes and act as a zinc delivery system for the

intestinal epithelium.

5 The following article is the accepted version and appears as journal version in:

Maria Maares, Claudia Keil, Jenny Koza, Sophia Straubing, Tanja Schwerdtle, Hajo Haase. "In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins." International Journal of Molecular Science 2018. 19(9): 2662, DOI: 10.3390/ijms19092662, https://doi.org/10.3390/ijms19092662 https://www.mdpi.com/1422-0067/19/9/2662

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

86

6.1 Introduction

The essential trace element zinc is predominantly resorbed in the small intestine, where it is

absorbed by enterocytes and transported into the blood stream, primarily mediated by the

apically located Zrt-, Irt-like transporter (ZIP)-4 and the basolateral zinc exporter (ZnT)-1 [1].

These two transporters are complemented by the basolateral transporter ZIP-5, importing

zinc from the blood into the enterocytes, and the apical transporter ZnT-5, which exports

zinc back into the intestinal lumen [1]. However, despite ongoing research, the molecular

mechanisms regulating zinc absorption are not yet fully elucidated. Human intestinal zinc

resorption was shown to be a regulated process, as the fractional zinc resorption varies

between 20–60%, generally decreasing with elevated zinc intake [2]. The amount of

absorbed zinc is not only influenced by oral zinc intake, but particularly depends on its

accessibility in the intestine, which is strongly influenced by food components such as

phytate, impairing the intestinal zinc availability. Additionally, amino acids and several trace

elements were reported to impact enterocytes’ zinc uptake [3]. Notably, the intestinal

availability of trace elements in general does not exclusively depend on food components,

but is also influenced by the intestinal mucus layer. Specifically, the mucus was shown to

bind ions such as iron, lead, and zinc preventing their hydroxypolymerisation at intestinal pH

and increasing their solubility and availability for the intestinal epithelium [4–6]. It has been

suggested that the affinity of mucins for metals increases from M+ < M2+ < M3+ leading to a

competitive binding to the glycoproteins and consequentially influencing their bioavailability

[5,7,8]. In fact, while the impact of mucins for iron resorption was investigated in detail,

there is evidence that the mucus layer might also be important for zinc uptake by the human

intestinal mucosa, as zinc binding by mucins was observed in animal studies [4,5,9,10].

Nevertheless, the detailed role of the intestinal mucus on human zinc absorption has not yet

been investigated.

Mucus, synthesized and secreted by goblet cells, covers the entire gastrointestinal tract

protecting the underlying epithelium against the luminal content, and plays an essential role

in nutrition and health [11]. While a single loosely bound mucus layer supports the

resorption of nutrients in the small intestine, this physical barrier is extended by an

additional adherent mucus layer in the stomach and colon [11]. The gastrointestinal mucus is

mainly constituted of water, ions, lipids and 5–10% highly glycosylated proteins: the mucins.

These proteins maintain their macromolecular network-like structure by being largely

composed of Serine, Threonine and Proline tandem-repeats and O-linked oligosaccharides as

well as of fewer O-glycosylated cysteine-rich regions (recently reviewed in [12,13]). Thus, the

intestinal epithelium does not only consist of enterocytes, absorbing the nutrients from the

luminal content, but contains a variety of other cell types of which goblet cells are the most

abundant, constantly secreting mucins into the lumen [14].

In vitro intestinal models provide a standardized and easy platform to analyze the

bioavailability of nutrients, such as trace elements, as well as transport kinetics [15], offering

a promising tool to illuminate distinct molecular aspects of intestinal zinc resorption. Not

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

87

only are changes in cellular zinc tracked by using inductively-coupled plasma mass

spectrometry (ICP-MS) and flame atomic absorption spectrometry (FAAS), but the

application of low molecular weight sensors as an approach to measure zinc uptake [16–18]

gained importance to determine small changes in the intracellular zinc pool [19]. These

models always need to resemble the in vivo situation, not only concerning buffer and

medium constituents, but also cellular composition. Until now, in vitro studies on intestinal

zinc uptake were mainly conducted using the Caco-2 model, which was already shown to

express the main intestinal zinc transporters [17]. This cell line is very well characterized and

differentiates into a cell monolayer morphologically and functionally representing the

enterocytes in vivo [20]. Some disadvantages of this in vitro cell model concerning

overexpression of the Pgp-protein and lack of a mucus layer were improved by introducing

the Caco-2/HT-29-MTX co-culture [21,22]. This well characterized co-culture of Caco-2 cells

and the goblet cell line HT-29-MTX [22,23] was shown to be covered by mucus with a

thickness of at least 2–10 µm after fixation [21] and has already been used to investigate the

role of mucins on bacterial adhesion [24] as well as on the resorption of nutrients [21,25,26].

Furthermore, this co-culture was recently applied by our group to study the impact of a

basolateral zinc acceptor on zinc resorption [17].

The aim of this study is to examine the role of intestinal mucins for zinc resorption. Herein,

zinc binding properties of these glycoproteins and their zinc affinity are investigated in cell-

free measurements as well as in the presence of intestinal cells to clarify the role of the

mucus layer on zinc uptake. Finally, zinc transport was measured comparing a conventional

Caco-2 monoculture and the Caco-2/HT-29-MTX co-culture to investigate the impact of

mucins on the actual zinc transfer.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

88

6.2 Results

6.2.1 Zinc Binding by Intestinal Mucins

First, the property of mucins to bind and release zinc was investigated using the zinc-

chelating chromophore 4-(2-pyridylazo)resorcinol (PAR) (additional spectrophotometric

titration of the zinc-(PAR)2-complex in Supplementary Figure S6.1). Figure 6.1 shows a

significant decrease of the free zinc concentration after applying 2.5–10 mg mL-1 zinc-

depleted mucins (one-way analysis of variance (ANOVA) with Dunnett’s multiple comparison

test; p < 0.001), indicating zinc binding by the gastrointestinal glycoproteins.

Figure 6.1: Effect of mucins on zinc availability for 4-(2-pyridylazo)resorcinol (PAR).

Effect of mucins on zinc availability for PAR is shown as free zinc relative to the initially added zinc concentration. Different concentrations of zinc-depleted porcine mucin were incubated with 8 µM zinc and free zinc was analyzed using the colorimetric zinc chelator PAR. Data are presented as means + standard deviation (SD) of at least three independent experiments. Significant differences to the control are indicated (*** p < 0.001; one-way analysis of variance (ANOVA) with Dunnett’s multiple comparison test).

Subsequently, the binding capacity of these mucins was further investigated after dialysis of

porcine mucins with different zinc concentrations for 12 h against Tris(hydroxymethyl)-

aminomethane (Tris)-buffered saline (TBS) (selection of the appropriate dialysis time in

Supplementary Figure S6.2). Here, the zinc content of zinc-loaded mucins increased

significantly compared to mucins without added zinc (Figure 6.2).

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

89

Figure 6.2: Zinc binding properties of gastrointestinal mucins.

Different zinc concentrations were added to 25 mg mL-1 zinc-depleted porcine mucins and dialyzed against (Tris)-buffered saline (TBS) for 12 h. Subsequently the amount of zinc retained by binding to mucins was measured using flame atomic absorption spectrometry (FAAS) (A). Moreover, the zinc content of mucins after zinc loading is shown relative to the glycoprotein content measured by quantitative periodic acid Schiff (PAS)-assay (B), and relative to protein content of mucins measured by bicinchoninic acid (BCA)-assay (C). Data are shown as means + SD of at least three independent experiments. Significant differences to the control are indicated (** p < 0.01; *** p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test).

The addition of 10,000 µM zinc before dialysis resulted in a beginning zinc-saturation of the

glycoproteins (Figure 6.2A). These samples were defined as high zinc-loaded mucins. In total,

three different degrees of zinc-loaded mucins were selected for further analysis: in addition

to the high zinc-loaded mucins, also medium and low zinc-loaded mucins were dialyzed in

the presence of 5000 µM or 2500 µM zinc, respectively. Next, zinc values were normalized to

the total glycoprotein and protein content of the mucin samples after dialysis. These also

depended on the amount of zinc present during dialysis, and maximum amounts of 5.7 mg

zinc per g glycoprotein (Figure 6.2B) and of 94.2 mg per g protein (Figure 6.2C) were

determined for the high zinc-loaded mucins.

Finally, the zinc binding affinity of zinc-depleted porcine mucin was analyzed using the

colorimetric reagent 2-carboxy-20-hydroxy-50-sulfoformazylbenzene monosodium salt

(zincon) (additional spectrophotometric titration in Supplementary Figure S6.3) and

compared to the affinity of zinc-depleted mucins harvested from the goblet cell line HT-29-

MTX (Figure 6.3A,B). This analysis yielded similar dissociation constants for the zinc-mucin-

complexes with 6.8 µM for porcine and 5 µM for HT-29-MTX mucin.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

90

Figure 6.3: Zinc binding affinity of gastrointestinal mucins.

Shown are zinc binding affinities of two different zinc-depleted mucins analyzed with the chromophore Zincon. For this, 1 mg mL-1 porcine mucins (A) and mucins harvested from HT-29-MTX (B) were used. Data were analyzed with GraphPad Prism software version 5.01 (GraphPad Software Inc., San Diego, CA, USA) and a non-linear regression assuming a one site-specific binding with Hill slope as a function of the zinc concentration was applied to calculate the dissociation constants of the mucin-zinc-complex as indicated. Data are presented as means ± SD of three independent experiments.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

91

6.2.2 Role of Zinc Buffering by Mucins on Zinc Uptake into Goblet Cells

Next, the impact of mucins on zinc uptake by goblet cells was investigated. Zinc absorption

of mucin-producing HT-29-MTX cells with or without mucin removal as well as the

enterocyte cell line HT-29 was analyzed with the fluorescent low molecular weight zinc

probe Zinpyr-1 and is depicted as the increase of intracellular free zinc (Figure 6.4A–C). Of

note, the term “free” zinc is frequently used to describe the zinc pool that is complexed by

small molecule ligands [27], which was already employed to investigate short-term zinc

uptake in intestinal cells [16,17]. In detail, HT-29 showed a concentration-dependent zinc

absorption (Figure 6.4A), while zinc uptake of the mucin-producing HT-29-MTX cells was very

slight and concentration-independent (Figure 6.4B). When extracellular mucins were

depleted with N-acetylcysteine (NAC), zinc uptake of HT-29-MTX cells increased (Figure

6.4C). Notably, analysis of the intracellular distribution of the fluorescent zinc sensor

revealed a vesicular accumulation in both cell lines (Figure 6.4D). Extracellular mucins were

visualized with the high molecular fluorescein isothiocyanate (FITC)-dextran 20 kDa (FD-20).

It intercalates in the mucus layer due to its high molecular weight [28], and was analyzed

using confocal laser scanning microscopy (CLSM) together with staining of the cell

membrane (Figure 6.4E). Z-scans showed differences in the mucin thickness of HT-29 and

HT-29-MTX, with HT-29 showing only slight FD-20-staining, whereas the HT-29-MTX goblet

cells produced a thick extracellular mucin layer, which decreased visibly after mucin

depletion. Moreover, mucin secretion of HT-29-MTX cells and its successful depletion with

NAC was further investigated by immunofluorescent staining of the MUC5AC-apoprotein

together with nuclear staining using Hoechst (Figure 6.4F). It shows a diffuse distribution of

the MUC5AC-apoprotein over the cell layer of HT-29-MTX, whereas no MUC5AC-staining

was observed for mucin-depleted HT-29-MTX and HT-29 cells.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

92

Figure 6.4: Effect of mucin depletion on zinc-resorption in HT-29-MTX.

(A–C) Zinc uptake of the enterocytes HT-29 was analyzed with the fluorescent probe Zinpyr-1 (A) and compared to the zinc-uptake of the goblet cell line HT-29-MTX (B). Additionally, the effect of mucin depletion on zinc absorption of HT-29-MTX was analyzed after removing extracellular mucins using 10 mM N-acetylcysteine (C). Data are presented as means ± standard error of the mean (SEM) of three independent experiments. (D) Cellular distribution of Zinpyr-1 in HT-29 and HT-29-MTX cells together with nuclear staining using Hoechst was analyzed by fluorescence microscopy. Scale bar 20 µm. (E) Visualization of extracellular mucins with fluorescein isothiocyanate (FITC)-dextran was conducted using confocal laser scanning microscopy. Shown are z-stacks of HT-29, HT-29-MTX and HT-29-MTX without mucins after incubation with FITC dextran (green). The cell layer is stained with a cell membrane-tracker (red). Scale bar 50 µm. (F) Immunochemical detection of MUC5AC and nuclear staining using Hoechst using fluorescence microscopy in HT-29, regular HT-29-MTX and HT-29-MTX after removal of mucins. Scale bar 20 µm.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

93

6.2.3 Impact of Extracellular Mucins on Zinc Uptake into Enterocytes

The impact of different mucin concentrations on short-term zinc uptake into the intestinal

cell line Caco-2 was investigated with Zinpyr-1 (Figure 6.5A). Cellular free zinc is increasing

after adding 25 and 50 µM zinc, respectively, but decreases significantly with added zinc-

depleted porcine mucin (one-way ANOVA with the Bonferroni post hoc test: 1.25 mg mL-1

(25 µM): p < 0.05; 2.5 mg mL-1 (25 µM and 50 µM): p < 0.05; 5 mg mL-1 (25 µM and 50 µM): p

< 0.05). Moreover, Figure 6.5B presents imaging of the cellular localization of Zinpyr-1

fluorescence in Caco-2 cells unveiling a predominantly vesicular distribution of the

fluorescence.

Figure 6.5: Impact of zinc-depleted mucins on zinc uptake by enterocytes.

(A) Zinc uptake in Caco-2 cells after zinc incubation for 40 min in the presence of different concentrations of zinc-depleted porcine mucin is shown as the increase of free zinc using the fluorescent zinc probe Zinpyr-1. Data are shown as means + SD of at least three independent experiments. Significant differences from 0 mg mL-1 zinc-depleted porcine gastric mucin within one

zinc concentration are indicated (*, #, •p < 0.05; **p < 0.01; ***p < 0.001, one-way ANOVA with

Dunnett’s multiple comparison test). (B) Fluorescence microscopy showing the intracellular distribution of the zinc-dependent signal of the fluorescent probe Zinpyr-1 in Caco-2 cells together with nuclear staining using Hoechst. Scale bar 20 µm.

Short-term zinc uptake in the presence of 5 mg mL-1 zinc-depleted porcine mucins was also

investigated by FAAS, yielding no significant change of the cellular zinc content either with or

without zinc-depleted mucins (Figure 6.6A). In comparison, long-term zinc absorption in the

presence of 0 and 5 mg mL-1 zinc-depleted mucins, also conducted with FAAS, resulted in a

significant increase of the cellular zinc content of Caco-2 cells after applying zinc without

mucins (Figure 6.6B; one-way ANOVA with Dunnett’s multiple comparison test; 25 µM: p <

0.001; 50 µM: p < 0.01). The incubation of zinc together with 5 mg mL-1 mucins did not

significantly change the cellular zinc content.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

94

Figure 6.6: Impact of extracellular mucins on zinc uptake in Caco-2 cells measured with FAAS.

The influence of extracellular addition of zinc-depleted mucins on the uptake of different zinc concentrations in enterocytes was investigated after incubation for 30 min (A) and 24 h (B). Cellular zinc content was analyzed using FAAS and is shown relative to cellular protein. Data are shown as means + SD of three independent experiments. Means significantly different from the untreated controls are indicated (* p < 0.05; ** p < 0.01; one-way ANOVA with Dunnett’s multiple comparison test).

Next, the zinc buffering capacity of mucins and their zinc release into enterocytes was

investigated in an experimental setting closer to the in vivo situation by using the

aforementioned low, medium and high zinc-loaded porcine mucins. First, cellular zinc uptake

from zinc-loaded mucins diluted to a final zinc concentration of 25 µM was compared to the

absorption of 25 µM zinc without mucins (Figure 6.7A). While low and medium zinc-loaded

mucins resulted in a comparable rise of intracellular free zinc, high zinc-loaded mucins

caused a significant increase. However, they did not reach the levels of cellular zinc that

were observed without mucins (Figure 6.7A).

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

95

Figure 6.7: Effect of mucin zinc saturation on zinc uptake by enterocytes.

The impact of the degree of zinc saturation of extracellular added mucins on zinc uptake in Caco-2 cells is shown by measuring the increase of free zinc using the fluorescent zinc probe Zinpyr-1. Porcine mucins were incubated with 2500 µM, 5000 µM and 10,000 µM zinc, resulting in mucins with differing zinc content (low, medium, high) and degree of zinc saturation. (A) Zinc absorption of Caco-2 cells after 40 min of treatment with 25 µM zinc alone or with zinc-loaded mucins, diluted to a final zinc concentration of 25 µM zinc (final mucin concentration: 1 mg mL-1 low zinc mucin, 0.46 mg mL-1 medium zinc mucin, 0.31 mg mL-1 high zinc mucin). (B) Zinc uptake after 40 min incubation with 0 µM or 25 µM zinc in the presence of 0 mg mL-1 or 0.25 mg mL-1 zinc-loaded or zinc-depleted mucins, respectively. Data are presented as means + SEM of at least three independent experiments. Significant differences were analyzed by repeated-measures ANOVA with the Bonferroni post hoc test. Bars sharing a letter (a, b, c, d) are not significantly different.

To elaborate the influence of the zinc buffering capacity of mucins on cellular zinc uptake,

the absorption of 25 µM zinc in the presence of 0.25 mg mL-1 zinc-loaded or zinc-depleted

mucins was analyzed (Figure 6.7B). This was then compared to the uptake of 25 µM zinc

without mucins. Here, intracellular free zinc increased significantly, whereas the presence of

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

96

zinc-depleted mucins diminished the zinc uptake comparable to the results shown in Figure

6.5. Incubating the cells with 0.25 mg mL-1 zinc-loaded mucins (w/o additional zinc), a similar

slight increase of free zinc regardless of the mucins’ remaining zinc binding capacity and zinc

content was detected. Adding zinc-loaded mucins and 25 µM zinc simultaneously to the cells

yielded a significant rise of free zinc in the presence of high zinc-loaded mucins (repeated-

measures ANOVA; high zinc-loaded mucins with 25 µM zinc compared to high zinc mucins

with 0 µM: p < 0.05). This increase was similar to the absorption of 25 µM zinc in the

absence of mucins. In contrast, low and medium zinc mucins, still not completely saturated

with zinc, showed no additional zinc absorption in the presence of 25 µM zinc.

6.2.4 Comparison of Zinc Resorption in Different Intestinal Cell Culture Models:

The Role of Mucins

Finally, the role of intestinal mucins on zinc resorption was analyzed comparing zinc

transport by a Caco-2/HT-29-MTX co-culture with a Caco-2 monoculture. The integrity of the

cell monolayers was monitored during the experiments by measuring the paracellular

permeability for FD-20 and by detecting transepithelial electrical resistance (TEER) at the

beginning and end of the experiments, revealing no impairment of both parameters during

the resorption study (Supplementary Figure S6.4). The presence of goblet cells in the co-

culture did not influence the permeability of the cell monolayer, as the transepithelial

resistance of the co-cultures and Caco-2 monocultures did not differ significantly (co-

cultures: 1146.7 ± 25.5 Ω cm2; monocultures: 1288.9 ± 239.2 Ω cm2) and the paracellular

permeability was comparable to those measured in the absence of goblet cells

(Supplementary Figure S6.4).

Apical zinc uptake by the monoculture, relative to the initially applied zinc concentrations,

was declining with increasing amounts of zinc. In contrast, the Caco-2/HT-29-MTX co-culture

absorbed comparable amounts between 12.8% and 14.2% of all added zinc concentrations

(Figure 6.8A,B). Regardless of the differences in the apical zinc uptake, the fractional zinc

resorption into the basolateral compartment of both intestinal models declined inversely

related to the initially added zinc. Yet, the fractional resorption was significantly higher in

Caco-2/HT-29-MTX co-cultures (Figure 6.8C,D; two-way ANOVA with the Bonferroni post hoc

test comparing the mono- and co-cultures: 25 µM: p < 0.001; 50 µM: p < 0.05). More

precisely, fractional resorption by monocultures dropped from 1.6% to 0.9%, showing only

slight concentration dependence, whereas resorption by co-cultures declined from 4.2% to

1.9% of the initially added zinc.

Additionally, in both intestinal models the cellular zinc uptake increased with added zinc,

yielding a significantly higher uptake after addition of 50 µM zinc by the co-culture (Figure

6.8E,F; two-way ANOVA with the Bonferroni post hoc test, p < 0.05).

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

97

Figure 6.8: Comparison of zinc resorption in Caco-2 monocultures and Caco-2/HT-29-MTX co- cultures.

Zinc transport studies were conducted using the enterocytes Caco-2 (A,C,E) and a co-culture of Caco-2 with the mucus-producing goblet cell line HT-29-MTX (B,D,F). Shown are the decrease of apical zinc (apical zinc uptake) (A,B) and fractional zinc resorption after 4 h incubation relative to the initially added amount of zinc (C,D). Moreover, cellular zinc uptake is shown relative to cellular protein content (E,F). Data are shown as means + SD of three independent experiments and means significantly different from the untreated controls are indicated (* p < 0.05; ** p < 0.01; *** p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test). According to a two-way ANOVA with the Bonferroni post hoc test comparing the results within one added zinc concentration of the mono- and co-cultures there are significant differences regarding the fractional zinc resorption (25 µM: p < 0.001; 50 µM: p < 0.05), and zinc uptake (50 µM: p < 0.05).

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

98

Supplementary Table S6.1 summarizes detailed quantitative data of zinc uptake into the

cells, cellular zinc content, and the amount of zinc transported to the basolateral

compartment. Over all, the zinc transport study resulted in higher zinc transport rates for

the mucin-producing co-culture (Figure 6.9A,B; monoculture: 0.3–1.29 nmol zinc cm-2; co-

culture: 1.1–2.3 nmol zinc cm-2). In detail, according to a two-way ANOVA with the

Bonferroni post hoc test comparing the results of the two intestinal models within one

added zinc concentration, the co-culture resulted in a significantly higher zinc transport rate

of initially added 100 µM (p < 0.05).

Figure 6.9: Zinc transport rates in Caco-2 monocultures and Caco-2/HT-29-MTX co-cultures.

Zinc transport rates in nmol zinc per cm2 resorption area in mono- and co-cultures are displayed. Data are presented as means + SD of three independent experiments. Significant differences to control cells (0 µM zinc) are indicated (** p < 0.01; one-way ANOVA with Dunnett’s multiple comparison test). According to a two-way ANOVA with the Bonferroni post hoc test comparing the results within one added zinc concentration of the mono- and co-cultures, there is a significant difference between the zinc transport rate at 100 µM (p < 0.001).

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

99

6.3 Discussion

Zinc binding by the mucus layer was already observed in animal studies [4,5,9,10],

suggesting that mucins could play a role in human intestinal zinc absorption. However, little

was known about the distinct zinc binding properties of these glycoproteins. The present

study demonstrated that zinc binding is not only dependent on the amount of the available,

or rather, free zinc concentration, but also on the zinc : mucin ratio. Assuming an

approximate molecular mass of intestinal mucins of 2.5 MDa [29,30], the employed mucin

concentrations between 2.5–10 mg mL-1 correspond to 1 µM–4 µM mucin, resulting in a 2–

8-fold molar excess of zinc in the assay. When applying an 8-fold molar zinc excess, 50% of

zinc was retained by the glycoproteins, declining to almost no free zinc at a molar zinc :

mucin ratio of 2 (Figure 6.1). This suggests multiple zinc-binding sites within one mucin

molecule with biologically relevant affinity.

Notably, gastrointestinal mucins were able to decrease the amount of available zinc for the

high affinity colorimetric reagent PAR (dissociation constant of the zinc-(PAR)2-complex

7.08·10-13 M2 [31]), indicating that at least some of the binding sites have high affinity for

zinc. The coordination of metals by proteins strongly depends on the metal ion preference

for particular electron donors [32]. Zinc tends to form stronger covalent binding to nitrogen

and sulfur and weak complexes with oxygen [32]. The latter are highly present in mucins by

carboxylate groups of the O-glycans [13]. Nitrogen and Sulfur are part of both human and

porcine intestinal mucins containing N-acetyl groups (N-acetyl-galactosamine (GalNac), N-

acetylneuraminic acid (NeuAc)) [33,34] and on average about 8–14% free thiols [35,36],

possibly playing an important role in the zinc binding affinity of mucins.

The zinc binding capacity of mucins was further investigated by dialysis of pig gastric mucins

against different zinc concentrations, resulting in low, medium and high zinc-loaded mucins

(Figure 6.2). According to the present study, mucins have an average molar zinc binding

capacity of about 200, indicating a multiplicity of zinc-binding sites within one mucin

molecule, possibly providing a broad spectrum of different binding affinities. Mucins are

highly glycosylated proteins, consisting of approximately 80% carbohydrates [11,13]. Thus,

the total amount of zinc bound per g protein is one order of magnitude higher than per g

glycoprotein (Figure 6.2B,C). These results are comparable to those obtained by Quarterman

et al. analyzing zinc binding of porcine mucins at different pH levels, which led to a zinc

content of 10 mg zinc per g mucin at pH 7.5 [4]. Furthermore, the binding constants of

porcine gastric mucin and mucins obtained from the cell line HT-29-MTX are in good

agreement with each other (Figure 6.3), and additionally are of the same magnitude as

luminal zinc levels [37,38]. Given the high number of binding sites, they probably represent

an average dissociation constant of the zinc-mucin-complex, constituting a mixture of

several binding sites with varying affinities.

Diet-derived luminal factors influence intestinal zinc bioavailability [39]. Together with

physiological factors such as luminal fluid and mucus layer, these components represent the

luminal matrix, which influences zinc speciation, consequently affecting its availability for

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

100

enterocytes [4,18]. To illuminate the impact of apically present zinc binding proteins on zinc

uptake into enterocytes, a recent study of our group investigated the effect of albumin on

zinc absorption. Albumin significantly reduced short-term zinc uptake measured with the

low molecular weight sensor Zinpyr-1 [17]. By the same technique, the role of mucins in

short-term zinc absorption was analyzed in the present study. The goblet cell line HT-29-MTX

produces and secretes mucins covering the cell surface, as shown by qualitative analysis

using immunochemical staining of the MUC5AC-apoprotein (Figure 6.4E), which is

comparable to previous MUC5AC-stainings of HT-29-MTX mucins [40]. Yet, the impact of the

mucus layer on their zinc uptake has not been investigated before. Zinc uptake of HT-29-

MTX before and after mucin depletion was analyzed and compared to the intestinal

absorptive cells HT-29. Overall, mucin depletion caused a FD-20- and MUC5AC staining

almost similar to that of HT-29 cells, confirming a successful removal of extracellular mucins

by N-acetylcysteine. Notably, short-term zinc uptake of HT-29-MTX was impaired by

extracellular mucins, which indicates zinc binding and buffering by the glycoproteins

produced by these cells. In Caco-2 cells, cellular zinc absorption also decreased significantly

with elevated concentrations of zinc-depleted porcine mucins (Figure 6.5A), comparable to

the impairment by albumin [17]. In addition to the analysis with the low molecular weight

sensor Zinpyr-1, short-term zinc uptake by Caco-2 cells was also measured with FAAS

resulting in no significant change of cellular zinc, either with or without mucins (Figure 6.6A).

Only long-term zinc incubation of Caco-2 cells resulted in a significant increase of cellular

zinc in the absence of mucins (Figure 6.6B). Consistent with previous findings, changes in

intracellular zinc after short-term incubation of Caco-2 cells are probably too small

compared to the cellular zinc content to be detected by FAAS [17].

From the uptake studies shown in Figure 6.5A and Figure 6.6 it could be concluded that

mucins impair intestinal zinc availability. However, these porcine mucins were zinc-depleted,

which is compulsory for investigating zinc binding capacity and affinity, but does not

represent the in vivo situation in the intestinal lumen under zinc adequate conditions

[38,41]. This is corroborated by the fact that the commercially available porcine mucins

contained considerable amounts of zinc before zinc-depletion with Chelex® 100 Resin.

Accordingly, the impact of zinc-loaded mucins on zinc uptake into enterocytes was examined

using zinc-containing mucins (low, medium and high zinc-loaded mucins; Figure 6.7). Here,

all three types of mucins seemed to release part of their zinc, resulting in an increase of

cellular free zinc compared to control cells (Figure 6.7A). Interestingly, this release appeared

to be largely independent of the mucins’ zinc content and their remaining zinc binding

capacity, adding the same concentration of differentially loaded mucins to the cells resulted

in similar zinc uptake (Figure 6.7B). These observations suggest that mucins buffer the

cellular available zinc concentration in the lumen, still keeping it available for enterocytes. In

animal studies, intestinal mucins were discussed to absorb zinc from the luminal content

transferring it to the mucosa [9,10]. Zinc transfer to the intestinal cells occurred slower than

the initial zinc binding by mucins [9]. In this manner, the intestinal mucus layer might lead to

retention of luminal available zinc, possibly providing intestinal cells with zinc for extended

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

101

periods of time after food intake. This implies that mucins act as a zinc delivery system from

the lumen to the intestinal epithelium, which was already postulated for iron [5], and led to

our hypothesis that zinc-saturated mucins might facilitate zinc delivery to enterocytes. In the

present study, however, cellular zinc uptake in the presence of zinc-saturated mucins was

similar, but not augmented, compared to the absence of mucins (Figure 6.7B). This

discrepancy might be due to insufficient equivalence to the in vivo situation: first, the

basolateral compartment, where the absorbed zinc can be exported, is lacking. Second, the

mucins applied in this study were commercially available isolated porcine mucins, which

might not be entirely comparable to the native mucins produced by goblet cells [42].

Although these mucins are often used as a standard model for the mucus layer to investigate

characteristics of gastrointestinal mucins [43] and their role in intestinal metal uptake [28],

they are also known to have weaker gel-forming abilities [44], possibly due to protease

treatment during the isolation and purification process [42].

To overcome both issues, the impact of chemically unprocessed mucins on intestinal zinc

resorption in a three-dimensional culture was investigated by transport studies in the mucin-

producing Caco-2/HT-29-MTX co-culture, which were then compared to zinc transport in the

absence of a mucus layer using conventional Caco-2 monocultures. Herein, the mucin-

producing co-culture clearly yielded higher zinc absorption than the mucus-lacking

monoculture (Figure 6.8). The Caco-2/HT-29-MTX model resulted in a stronger decrease of

the apical zinc concentration, varying around 12.7–14.1% independent of the initially added

zinc, as well as an elevated cellular zinc uptake (Figure 6.8E,F), indicating that the cellular

zinc uptake is facilitated by the mucus layer. Indeed, this supports the aforementioned

hypothesis that the mucus layer might act as a zinc delivery system for the intestinal

epithelium. Moreover, the glycoproteins seemed to assist the zinc transfer across the

intestinal epithelium as the fractional zinc resorption was significantly higher (1.8–2.4-fold

higher) in the presence of mucin-producing goblet cells. Likewise, the zinc transport rate was

elevated using co-cultures. Here, the basal zinc transport rate without the addition of

exogenous zinc was already 3.8-fold higher (1.13 ng zinc cm-2 resorption area) than that of

monocultures (0.3 ng zinc cm-2) (Figure 6.9), possibly due to resorption of the basal zinc

levels of mucins. These basal zinc levels might originate from the cell culture medium

(containing 3 µM zinc), and were incorporated by the mucins during the 21 days of

cultivation of the cell model. Thus, the mucus-secreting co-culture not only absorbed more

zinc from the apical compartment, but showed augmented zinc export to the basolateral

side, supporting previous observations that intestinal mucins represent an important factor

for intestinal zinc resorption [4,9,10].

The beneficial role of mucins for the resorption of other trace elements, such as iron, is

already well documented [21,28,45]. Concerning the essential nutrient iron, the mucus layer

not only mediates its availability for the intestinal epithelium by maintaining its solubility [5],

but was also proposed to provide its delivery to the absorptive cells by the mucin-integrin

mobilferrin pathway [45,46]. A similar mechanism might also be effective for zinc, as the

present study indicates that mucins not only influence zinc uptake by increasing its luminal

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

102

solubility, as discussed before [4], but rather promote zinc absorption by additionally acting

as a zinc delivery system for the mucosa, which would be in good agreement with

observations from animal studies [9,10].

Metal ion binding by mucins might have further implications for metal ion homeostasis. The

impact of other trace elements on zinc resorption was previously investigated and is still a

topic of ongoing research [3,47]. Mucins are not only involved in the mucosal uptake of

single trace elements, but were also suggested to influence their bioavailability by

competitively binding different metals [5]. Thus, in addition to competing for transport

proteins, competition for binding sites in mucins could be another factor for the mutual

interferences observed in intestinal trace element absorption. Furthermore, mucins not only

support intestinal zinc absorption, but might also be involved in fecal zinc loss. A

considerable amount of endogenous zinc is excreted with feces [2,48]. Even during extreme

zinc deficiency, the fecal excretion of zinc can be observed, which is defined as the

“obligatory fecal loss” [2]. The intestinal epithelium and the overlying mucus layer undergo

cycles of renewal [49,50] resulting in a complete turnover of the mucins, which is said to

occur much faster than that of the underlying mucosa [51]. Considering the amount of zinc

bound to mucins, the fast renewal of the intestinal mucus layer might play an important role

in the fecal zinc loss as the mucins are possibly excreted together with a remainder of tightly

bound zinc.

On the one hand, gastrointestinal mucins are important for zinc absorption in the intestinal

tract. On the other hand, zinc was also reported to be important for mucin synthesis, as the

gene expression profile of different mucin proteins was shown to depend on zinc supply

[52], and abdominal production as well as secretion of mucins were also decreased in zinc-

deficient animals [53,54]. Thus, further investigations of the interplay of zinc with intestinal

mucins and distinct molecular mechanisms of the mucosal delivery system are needed for a

better understanding of their role in intestinal zinc uptake as well as gastrointestinal

disorders.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

103

6.4 Materials and Methods

6.4.1 Materials

CellMask™DeepRed (ThermoFisher Scientific,Waltham, MA, USA); Cy3® goat anti rabbit IgG

(Jackson ImmunoResearch, Dianova, Hamburg, Germany); Chelex® 100 Resin (Bio-Rad,

Hercules, CA, USA); Dulbecco’s modified Eagles medium (DMEM) (PAN-Biotech, Aidenbach,

Germany); Hoechst 33258 (Sigma Aldrich, Munich, Germany); fluorescein isothiocyanate

(FITC)-dextran 20 (TdB, Uppsala, Sweden); fetal calf serum (FCS) (CCPro, Oberdorla,

Germany); mucin from porcine stomach Type II (Sigma Aldrich, Munich, Germany); PAR

(Sigma Aldrich, Munich, Germany); Transwell inserts (Corning, New York, NY, USA);

N,N,N’,N’-Tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) (Sigma Aldrich, Munich,

Germany); WillCo-dish® glass bottom dish (WillCo, Amsterdam, The Netherlands); Zincon

(Sigma Aldrich, Munich, Germany); ZnSO4·7H2O (Sigma Aldrich, Munich, Germany). All other

chemicals were purchased from standard sources.

6.4.2 Cell Culture

Caco-2, HT-29-MTX and HT-29 cells were cultured at 37°C, 5% CO2 and in a humidified

atmosphere in DMEM, containing 10% FCS, 100 U mL-1 penicillin and 100 µg mL-1

streptomycin. Additionally, medium for HT-29-MTX contained 1% non-essential amino acids

(NEAA). Media were changed every other day. Analysis of proper differentiation,

morphology and barrier integrity of Caco-2 cells cultured in our lab into an enterocyte-like

phenotype after 21 days, as shown before [20,55], was reported previously [56]. Caco-2, HT-

29-MTX and HT-29 cells were obtained from European Collection of Cell Cultures (ECACC,

Porton Down, UK).

6.4.3 4-(2-Pyridylazo)Resorcinol (PAR) Assay

Zinc binding by mucins was investigated with the colorimetric reagent PAR. The assay was

performed in TBS, containing 50 mM Tris(hydroxylmethyl)aminomethan and 150 mM NaCl,

at pH 7.5. Stock solutions of 25 mg mL-1 porcine mucins were prepared in TBS and zinc-

depleted using Chelex® 100 Resin according to the manufacturer’s protocol (zinc content

before depletion: 436 µg g-1 mucin; zinc content after depletion: 16.2 µg g-1 mucin). Different

mucin concentrations were incubated with 8 µM ZnSO4·7H2O and 20 µM PAR (stock solution

25 mM in H2O) was added. Absorption of the zinc-(PAR)2-complex was measured at 485 nm

using a well plate reader (M200, Tecan, Switzerland) and the effect of mucins on the

availability of zinc for PAR was analyzed using an external calibration of 0–10 µM zinc

obtained by spectrophotometric titrations followed by linear regression analysis.

6.4.4 Zinc Binding Capacity

The zinc binding capacity of gastrointestinal mucins was investigated by dialysis. 25 mg mL

mL-1 mucins were incubated with 0–100 mM ZnSO4·7H2O overnight and dialyzed against 1.5

L TBS for 6, 12 and 24 h. Finally, zinc concentrations were determined with FAAS using a

Perkin Elmer AAnalyst800 (Perkin Elmer, Rodgau, Germany) and zinc binding capacity was

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

104

calculated relative to the zinc concentrations before dialysis. Furthermore, the amount of

protein in the mucin samples after dialysis was analyzed using BCA-assay as described [57]

and the glycoprotein content in the mucin samples was detected using the quantitative PAS-

assay from Schömig et al. [44].

6.4.5 Zinc Binding Affinity of Mucins

The binding affinity of commercially available porcine gastric mucins and mucins produced

by the cell line HT-29-MTX were investigated with the chelating chromophore zincon [31].

Secreted mucins from HT-29-MTX were collected after culturing the cells for 13 days,

washing with phosphate buffered saline (PBS) and additional incubation in DMEM without

phenol red (with 100 U mL-1 penicillin and 100 µg mL-1 streptomycin, w/o FCS) for 24 h.

Subsequently, the cell supernatant was collected and secreted mucins were concentrated to

an 8-fold increase after desalting and washing with TBS using ultrafiltration (molecular

weight cut-off: 50 kDA) followed by zinc depletion using Chelex® 100 Resin (zinc content

before depletion: 59.2 µg g-1 protein; zinc content after depletion: 4.1 µg g-1 protein). Protein

content was analyzed using BCA assay as described [57]. Finally, zincon in TBS pH 7.5 (final

concentration 50 µM) was added to either 1 mg mL-1 commercially available porcine mucin

or the equivalent protein amount of secreted mucins from HT-29-MTX, respectively, and

titrated with 0–60 µM ZnSO4·7H2O. Subsequently, the absorption was determined on a well

plate reader (M200, Tecan, Switzerland) at 620 nm. The amount of mucin-bound zinc was

calculated using an external calibration. Data were analyzed with GraphPad Prism software

version 5.01 (GraphPad Software Inc., San Diego, CA, USA) and a non-linear regression

assuming a one site-specific binding with Hill slope as a function of the zinc concentration

was applied.

6.4.6 Cellular Zinc Uptake Measured by Zinpyr-1

Short-term zinc uptake in Caco-2, HT-29 and HT-29-MTX was quantified as the increase of

free zinc [nM] using the low molecular zinc probe Zinpyr-1. The concentration of free zinc

was determined using the following equation of Grynkiewicz et al. [Zinc] = Kd × [(F - Fmin) /

(Fmax - F)] [58] and a dissociation constant for the zinc-Zinpyr-1-complex of 0.7 nM [59]. Cells

were transferred to 96-well plate and cultured for 21 days (Caco-2; initial cell number per

well: 5000) or 7 days (HT-29 and HT-29-MTX; initial cell number per well: 10,000),

respectively. On the day of the experiment, cells were incubated with 2.5 µM Zinpyr-1 and

the uptake of zinc, in the presence or absence of mucins as indicated in the respective figure

legends, was measured as the increase of free zinc on a fluorescence well plate reader

(Spark, Tecan, Switzerland; Zinpyr-1 fluorescence: λex = 508 nm and λem = 527 nm) as

described [17].

6.4.7 Removal of Extracellular Mucins by N-Acetylcysteine

Extracellular mucins of the goblet cell line HT-29-MTX were removed by reducing the intra-

and intermolecular disulfide bounds in the glycoproteins using NAC [21,60]. Before analyzing

zinc uptake of HT-29-MTX, cells were treated twice with 10 mM NAC in PBS for 10 min.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

105

Between treatments, cells were incubated in DMEM for 1 h. Successful removal of

extracellular mucins was examined using immunochemical staining of the main glycoprotein

of HT-29-MTX, MUC5AC, and fluorescence microscopic visualization with FD-20.

6.4.8 Immunochemical Staining of the MUC5AC Glycoprotein

Immunochemical staining of MUC5AC was performed using the mucin-specific antiserum

MAN-5ACI for the polypeptides of MUC5AC [61,62]. Therefore, HT-29-MTX and HT-29 cells

were seeded on glass slides and cultivated for 7 days. Depletion of extracellular mucins of

HT-29-MTX prior to immunochemical staining was performed as described above. Cells were

fixed on ice with a final concentration of 3.7% formaldehyde directly added to the cell

medium, washed with cold PBS and permeabilized using 0.5% Triton-X-100 in PBS for 20 min

on ice. After washing with PBS, cells were blocked with 10% FCS in TBS for 1 h, incubated

with MAN-5ACI antiserum (1:500 in TBS with 20% Tween (TBST)) overnight, followed by

additional washing and blocking. Subsequently, Cy3® goat anti rabbit IgG (indocarbocyanin

goat-anti rabbit immunoglobulin G) (1:500 in TBST) was incubated for 1 h at 37°C.

Additionally cellular nuclei were stained with Hoechst 33258. Finally, cells were washed with

TBST and evaluated by fluorescence microscopy (Axio Imager M1, Zeiss, Germany) at

excitation wavelengths of 546 nm (Cy-3) and 358 nm (Hoechst).

6.4.9 Visualizing Extracellular Mucins with Fluorescein Isothiocyanate (FITC)-

Dextran

Extracellular mucins produced by HT-29-MTX were visualized using high molecular dextran

labeled with fluorescein isothiocyanate (FD-20). Due to its size, FD-20 is trapped in the high

molecular glycoproteins, possibly due to steric hindrance [63]. To this end, HT-29 and HT-29-

MTX cells were cultured in glass bottom dishes for 14 d. Prior to the experiment, medium

was carefully removed and 100 µM FD-20 together with cell membrane tracker

CellMask™DeepRed (3.3 ng mL-1) in assay buffer (120 mM NaCl, 5.4 mM KCl, 5 mM Glucose,

1 mM CaCl2, 1 mM MgCl2, 1 mM NaH2PO4, 10 mM HEPES, pH 7.35) was incubated for 15 min

at 37°C. Subsequently, cells were washed with assay buffer and live cell imaging of

fluorescently labeled extracellular mucins was performed with a CLSM (Leica TCS SP8; λex

(FITC) = 488 nm, λem (FITC) = 510 nm; λex (CellMask™DeepRed) = 552 nm, λem

(CellMask™DeepRed) = 695 nm).

6.4.10 Total Cellular Zinc Content Measured by Flame Atomic Absorption

Spectrometry (FAAS)

For the determination of long-term zinc uptake with FAAS, 1.2·105 Caco-2 cells were seeded

in 6 well plates and cultured for 21 days. Fully differentiated cells were incubated with

different zinc concentrations with 0 or 5 mg mL-1 porcine mucin in DMEM w/o phenol red

and incubated for 24 h. Short-term uptake was conducted after 30 min incubation using 0 or

5 mg mL-1 mucin in assay buffer. Finally, cells were harvested on ice with a cell scraper, and

an aliquot was collected for protein quantification as described [57]. Subsequently, cells

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

106

were dissolved in a mixture of 67% ultrapure HNO3 and 30% H2O2 (50/50; v/v) and dried at

92°C overnight using a thermoshaker. Residues were dissolved in 0.67% HNO3 and samples

were analyzed by FAAS.

6.4.11 Zinc Transport Assay

Zinc transport studies were performed using monocultures of Caco-2 and Caco-2/HT-29-MTX

co-cultures. Co-cultures were realized with an initial cell ratio of 75% Caco-2 and 25% HT-29-

MTX cells and alternated cell seeding, modified after Nollevaux et al. [64]. Herein, the co-

culture of Caco-2 and HT-29-MTX cells in our lab was characterized concerning the proper

cellular ratio by investigating the mucin secretion and adequate differentiation of the

enterocytes as reported before [17].

Eighty thousand cells were transferred onto polycarbonate transwell membranes (pore size

0.4 µm, culture area 1.12 cm2) and cultured for 21 days in DMEM with 10% FCS, 100 U mL-1

penicillin, 100 µg mL-1 streptomycin and 1% NEAA. For the co-cultures, 25% 20,000 HT-29-

MTX cells were added 2 days after seeding of Caco-2. After 21 days, the cells were incubated

with 0 µM, 25 µM, 50 µM and 100 µM ZnSO4·7H2O in 0.5 mL transport buffer [17] on the

apical side of the transport chamber for 4 h. The basolateral compartment constituted 1.5

mL cell culture medium with 30 mg mL-1 BSA. Prior and after the experiment, barrier

integrity was monitored by measuring TEER with the epithelial volt-ohm meter Millicell®

ERS-2 (Millipore, Burlington, MA, USA). Additionally, permeability of cell monolayers during

the experiment was determined using FD-20 [64] as reported before [17]. At the end of the

experiment, the media of the apical and basolateral compartments were collected, and cells

were harvested on ice in PBS, homogenized and centrifuged (800g). An aliquot of cell

homogenates was collected for protein quantification using BCA [57]. Subsequently, cells

were dried at 92°C overnight as described above and dissolved in 0.67% HNO3. Zinc

quantification in apical, basolateral and cellular compartment was conducted by ICP-MS

after dilution (1:10 and 1:200) in 2% HNO3 containing 5 µg L-1 rhodium, using an Agilent 8800

ICP-QQQ (Agilent Technologies Deutschland GmbH, Böblingen, Germany) in the single quad-

mode [17].

6.4.12 Statistical Analysis

Statistical significance was analyzed by one- or two-way ANOVA (for multiple comparisons),

followed by Bonferroni or Dunnett’s multiple comparison post hoc tests, as indicated in the

respective figure legends, using GraphPad Prism software version 5.01 (GraphPad Software

Inc., San Diego, CA, USA). Error bars represent standard deviation or standard error of the

mean, as indicated, of at least three independent biological replicates.

6.5 Conclusion

This study provides the first comprehensive assessment of the zinc binding properties of

mucins and their impact on in vitro intestinal zinc resorption. By clarifying the molecular zinc

binding capacity of these glycoproteins and their average affinity for the bivalent cation, we

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

107

could demonstrate that mucins bind multiple zinc ions with physiologically relevant affinity.

Hereby, zinc-free mucins impair zinc uptake, but this is not the form in which they are

present in the gastrointestinal tract. 2D-experiments with isolated porcine mucins show that

mucin-bound zinc is still available for cellular uptake, but not superior to free zinc. In

contrast, the 3D co-culture of enterocytes and mucin-secreting goblet cells suggests that

mucins even facilitate zinc uptake by enterocytes, making them an integral part of intestinal

zinc resorption.

6.6 Author Contributions

Conceptualization, M.M., C.K. and H.H.; Data curation, M.M.; Formal analysis, M.M. and

H.H.; Funding acquisition, T.S. and H.H.; Investigation, M.M., J.K. and S.S.; Methodology, T.S.;

Project administration, H.H.; Resources, T.S. and H.H.; Supervision, M.M., C.K. and H.H.;

Writing-original draft, M.M.; Writing-review and editing, C.K., T.S. and H.H.

6.7 Funding

The work of H.H. and T.S. is funded by the Deutsche Forschungsgemeinschaft (TraceAge–

DFG Research Unit on Interactions of essential trace elements in healthy and diseased

elderly, Potsdam-Berlin-Jena, FOR 2558/1, HA 4318/4-1, SCHW903/16-1).

6.8 Acknowledgements

The authors would like to thank Vera Meyer from the Department of Applied and Molecular

Microbiology (Berlin Institute of Technology) for their kind support with the CLSM

measurements, David J. Thornton (University of Manchester) for generously providing the

mucin-specific antiserum MAN-5ACI, and Ayşe Duman for her excellent technical work.

6.9 Conflicts of Interest

The authors declare no conflict of interest.

6.10 References

1. Lichten, L.A.; Cousins, R.J. Cousins. Mammalian zinc transporters: Nutritional and

physiologic regulation. Annu Rev. Nutr. 2009, 29, 153–176.

2. King, J.C.; Shames, D.M.; Woodhouse, L.R. Zinc homeostasis in humans. J. Nutr. 2000,

130, 1360S–1366S.

3. Lönnerdal, B. Dietary factors influencing zinc absorption. J. Nutr. 2000, 130, 1378S–

1383S.

4. Powell, J.J.; Jugdaohsingh, R.; Thompson, R.P. The regulation of mineral absorption in

the gastrointestinal tract. Proc. Nutr. Soc. 1999, 58, 147–153.

5. Conrad, M.E.; Umbreit, J.N.; Moore, E.G. A role for mucin in the absorption of

inorganic iron and other metal cations. A study in rats. Gastroenterology 1991, 100,

129–136.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

108

6. Powell, J.J.; Whitehead, M.W.; Ainley, C.C.; Kendall, M.D.; Nicholson, J.K.; Thompson,

R.P.H. Dietary minerals in the gastrointestinal tract: Hydroxypolymerisation of

aluminium is regulated by luminal mucins. J. Inorg. Biochem. 1999, 75, 167–180.

7. Whitehead, M.W.; Thompson, R.P.; Powell, J.J. Regulation of metal absorption in the

gastrointestinal tract. Gut 1996, 39, 625–628.

8. Powell, J.J.; Whitehead, M.W.; Lee, S.; Thompson, R.P.H. Mechanisms of

gastrointestinal absorption—Dietary minerals and the influence of beverage

ingestion. Food Chem. 1994, 51, 381–388.

9. Quarterman, J. Metal absorption and the intestinal mucus layer. Digestion 1987, 37,

1–9.

10. Glover, C.N.; Hogstrand, C. In vivo characterization of intestinal zinc uptake in

freshwater rainbow trout. J. Exp. Biol. 2002, 205, 141–150.

11. Johansson, M.E.; Sjovall, H.; Hansson, G.C. The gastrointestinal mucus system in

health and disease. Nat. Rev. Gastroenterol. 2013, 10, 352–361.

12. Birchenough, G.M.; Johansson, M.E.; Gustafsson, J.K.; Bergstrom, J.H.; Hansson, G.C.

New developments in goblet cell mucus secretion and function. Mucosal Immunol.

2015, 8, 712–719.

13. Bansil, R.; Turner, B.S. The biology of mucus: Composition, synthesis and

organization. Adv. Drug Deliv. Rev. 2018, 124, 3–15.

14. Madara, J. L. Functional Morphology of Epithelium of the Small Intestine. In

Comprehensive Physiology; Terjung, R., Ed.; JohnWiley & Sons, Inc.: Hoboken, NJ,

USA, 2011.

15. Langerholc, T.; Maragkoudakis, P.A.; Wollgast, J.; Gradisnik, L.; Cencic, A. Novel and

established intestinal cell line models - An indispensable tool in food science and

nutrition. Trends Food Sci. Tech. 2011, 22, S11–S20.

16. Sauer, A.K.; Pfaender, S.; Hagmeyer, S.; Tarana, L.; Mattes, A.K.; Briel, F.; Kury, S.;

Boeckers, T.M.; Grabrucker, A.M. Characterization of zinc amino acid complexes for

zinc delivery in vitro using Caco-2 cells and enterocytes from hiPSC. Biometals 2017,

30, 643–661.

17. Maares, M.; Duman, A.; Keil, C.; Schwerdtle, T.; Haase, H. The impact of apical and

basolateral albumin on intestinal zinc resorption in the Caco-2/HT-29-MTX co-culture

model. Metallomics 2018, 10, 979–991.

18. Haase, H.; Hebel, S.; Engelhardt, G.; Rink, L. The biochemical effects of extracellular

Zn2+ and other metal ions are severely affected by their speciation in cell culture

media. Metallomics 2015, 7, 102–111.

19. Carter, K.P.; Young, A.M.; Palmer, A.E. Fluorescent sensors for measuring metal ions

in living systems. Chem. Rev. 2014, 114, 4564–4601.

20. Pinto, M.; Robineleon, S.; Appay, M.D.; Kedinger, M.; Triadou, N.; Dussaulx, E.;

Lacroix, B.; Simonassmann, P.; Haffen, K.; Fogh, J.; et al. Enterocyte-like

differentiation and polarization of the human-colon carcinoma cell-line Caco-2 in

culture. Biol. Cell 1983, 47, 323–330.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

109

21. Mahler, G.J.; Shuler, M.L.; Glahn, R.P. Characterization of Caco-2 and HT29-MTX

cocultures in an in vitro digestion/cell culture model used to predict iron

bioavailability. J. Nutr. Biochem. 2009, 20, 494–502.

22. Hilgendorf, C.; Spahn-Langguth, H.; Regardh, C.G.; Lipka, E.; Amidon, G.L.; Langguth,

P. Caco-2 versus Caco-2/HT29-MTX co-cultured cell lines: Permeabilities via diffusion,

inside- and outside-directed carrier-mediated transport. J. Pharm. Sci. 2000, 89, 63–

75.

23. Walter, E.; Janich, S.; Roessler, B.J.; Hilfiger, J.M.; Amidon, G.L. HT29-MTX/Caco-2

cocultures as an in vitro model for the intestinal epithelium: In vitro−in vivo

correlation with permeability data from rats and humans. J. Pharm. Sci. Exp.

Pharmacol. 1996, 85, 1070–1076.

24. Laparra, J.M.; Sanz, Y. Comparison of in vitro models to study bacterial adhesion to

the intestinal epithelium. Lett. Appl. Microbiol. 2009, 49, 695–701.

25. Calatayud, M.; Vazquez, M.; Devesa, V.; Velez, D. In vitro study of intestinal transport

of inorganic and methylated arsenic species by Caco-2/HT29-MTX cocultures. Chem.

Res. Toxicol. 2012, 25, 2654–2662.

26. Vazquez, M.; Calatayud, M.; Velez, D.; Devesa, V. Intestinal transport of

methylmercury and inorganic mercury in various models of Caco-2 and HT29-MTX

cells. Toxicology 2013, 311, 147–153.

27. Maret, W. Analyzing free zinc(II) ion concentrations in cell biology with fluorescent

chelating molecules. Metallomics 2015, 7, 202–211.

28. Jin, F.; Welch, R.; Glahn, R. Moving toward a more physiological model: Application of

mucin to refine the in vitro digestion/Caco-2 cell culture system. J. Agric. Food Chem.

2006, 54, 8962–8967.

29. Johansson, M.E.; Hansson, G.C. Mucus and the goblet cell. Dig. Dis. 2013, 31, 305–

309.

30. Holmen Larsson, J.M.; Thomsson, K.A.; Rodriguez-Pineiro, A.M.; Karlsson, H.;

Hansson, G.C. Studies of mucus in mouse stomach, small intestine, and colon. III.

Gastrointestinal MUC5AC and MUC2 mucin o-glycan patterns reveal a regiospecific

distribution. Am. J. Physiol. Gastrointest. Liver Physiol. 2013, 305, G357–G363.

31. Kocyła, A.; Pomorski, A.; Krężel, A. Molar absorption coefficients and stability

constants of metal complexes of 4-(2-pyridylazo)resorcinol (par): Revisiting common

chelating probe for the study of metalloproteins. J. Inorg. Biochem. 2015, 152, 82–92.

32. Krężel, A.; Maret, W. The biological inorganic chemistry of zinc ions. Arch. Biochem.

Biophys. 2016, 611, 3–19.

33. Song, X.; Ju, H.; Lasanajak, Y.; Kudelka, M.R.; Smith, D.F.; Cummings, R.D. Oxidative

release of natural glycans for functional glycomics. Nat. Methods 2016, 13, 528–534.

34. Hennebicq-Reig, S.; Lesuffleur, T.; Capon, C.; De Bolos, C.; Kim, I.; Moreau, O.; Richet,

C.; Hemon, B.; Recchi, M.A.; Maes, E.; et al. Permanent exposure of mucin-secreting

HT-29 cells to benzyl-n-acetyl-alpha-d-galactosaminide induces abnormal O-

glycosylation of mucins and inhibits constitutive and stimulated MUC5AC secretion.

Biochem. J. 1998, 334, 283–295.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

110

35. Mantle, M.; Stewart, G.; Zayas, G.; King, M. The disulphide-bond content and

rheological properties of intestinal mucins from normal subjects and patients with

cystic fibrosis. Biochem. J. 1990, 266, 597–604.

36. Fogg, F.J.; Hutton, D.A.; Jumel, K.; Pearson, J.P.; Harding, S.E.; Allen, A.

Characterization of pig colonic mucins. Biochem. J. 1996, 316, 937–942.

37. Cragg, R.A.; Christie, G.R.; Phillips, S.R.; Russi, R.M.; Kury, S.; Mathers, J.C.; Taylor,

P.M.; Ford, D. A novel zinc-regulated human zinc transporter, hZTL1, is localized to

the enterocyte apical membrane. J. Biol. Chem. 2002, 277, 22789–22797.

38. Casey, C.E.; Walravens, P.A.; Hambidge, K.M.; Silverman, A. The zinc content of

human duodenal secretions. Nutr. Res. 1986, 6, 725–728.

39. Cousins, R.J. Gastrointestinal factors influencing zinc absorption and homeostasis. Int.

J. Vitam. Nutr. Res. 2010, 80, 243–248.

40. Navabi, N.; McGuckin, M.A.; Linden, S.K. Gastrointestinal cell lines form polarized

epithelia with an adherent mucus layer when cultured in semi-wet interfaces with

mechanical stimulation. PLoS ONE 2013, 8, e68761.

41. Matseshe, J.W.; Phillips, S.F.; Malagelada, J.R.; McCall, J.T. Recovery of dietary iron

and zinc from the proximal intestine of healthy man: Studies of different meals and

supplements. Am. J. Clin. Nutr. 1980, 33, 1946–1953.

42. Svensson, O.; Arnebrant, T. Mucin layers and multilayers - Physicochemical properties

and applications. Curr. Opin. Colloid Interface Sci. 2010, 15, 395–405.

43. Sumarokova, M.; Iturri, J.; Weber, A.; Maares, M.; Keil, C.; Luis Toca-Herrera, J.

Influencing the adhesion properties and wettability of mucin protein films by

variation of the environmental pH. Sci. Rep. 2018, 8, 9660.

44. Schömig, V.J.; Kasdorf, B.T.; Scholz, C.; Bidmon, K.; Lieleg, O.; Berensmeier, S. An

optimized purification process for porcine gastric mucin with preservation of its

native functional properties. RSC Adv. 2016, 6, 44932–44943.

45. Conrad, M.E.; Umbreit, J.N.; Moore, E.G. Regulation of iron absorption: Proteins

involved in duodenal mucosal uptake and transport. J. Am. Coll. Nutr. 1993, 12, 720–

728.

46. Simovich, M.; Hainsworth, L.N.; Fields, P.A.; Umbreit, J.N.; Conrad, M.E. Localization

of the iron transport proteins mobilferrin and dmt-1 in the duodenum: The surprising

role of mucin. Am. J. Hematol. 2003, 74, 32–45.

47. Iyengar, V.; Pullakhandam, R.; Nair, K.M. Coordinate expression and localization of

iron and zinc transporters explain iron-zinc interactions during uptake in Caco-2 cells:

Implications for iron uptake at the enterocyte. J. Nutr. Biochem. 2012, 23, 1146–

1154.

48. McCance, R.A.; Widdowson, E.M. The absorption and excretion of zinc. Biochem. J.

1942, 36, 692–696.

49. Van der Flier, L.G.; Clevers, H. Stem cells, self-renewal, and differentiation in the

intestinal epithelium. Annu. Rev. Physiol. 2009, 71, 241–260.

50. Schneider, H.; Pelaseyed, T.; Svensson, F.; Johansson, M.E.V. Study of mucin turnover

in the small intestine by in vivo labeling. Sci. Rep. 2018, 8, 5760.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

111

51. Johansson, M.E. Fast renewal of the distal colonic mucus layers by the surface goblet

cells as measured by in vivo labeling of mucin glycoproteins. PLoS ONE 2012, 7,

e41009.

52. Liu, P.; Pieper, R.; Rieger, J.; Vahjen, W.; Davin, R.; Plendl, J.; Meyer, W.; Zentek, J.

Effect of dietary zinc oxide on morphological characteristics, mucin composition and

gene expression in the colon of weaned piglets. PLoS ONE 2014, 9, e91091.

53. Quarterman, J.; Humphries, W.R.; Morrison, J.; Jackson, F.A. The effect of zinc

deficiency on intestinal and salivary mucins. Biochem. Soc. Trans. 1973, 1, 101.

54. Quarterman, J.; Jackson, F.A.; Morrison, J.N. The effect of zinc deficiency on sheep

intestinal mucin. Life Sci. J. 1976, 19, 979–986.

55. Sambuy, Y.; de Angelis, M.; Ranaldi, G.; Scarino, M.L.; Stammati, A.; Zucco, F. The

Caco-2 cell line as a model for the intestinal barrier: Influence of cell and culture-

related factors on Caco-2 cell functional characteristics. Cell Biol. Toxicol. 2005, 21, 1–

26.

56. Maares, M.; Keil, C.; Thomsen, S.; Gunzel, D.; Wiesner, B.; Haase, H. Characterization

of Caco-2 cells stably expressing the protein-based zinc probe eCalwy-5 as a model

system for investigating intestinal zinc transport. J. Trace Elem. Med. Biol. 2018, 49,

296–304.

57. Smith, P.K.; Krohn, R.I.; Hermanson, G.T.; Mallia, A.K.; Gartner, F.H.; Provenzano,

M.D.; Fujimoto, E.K.; Goeke, N.M.; Olson, B.J.; Klenk, D.C. Measurement of protein

using bicinchoninic acid. Anal. Biochem. 1985, 150, 76–85.

58. Grynkiewicz, G.; Poenie, M.; Tsien, R.Y. A new generation of Ca2+ indicators with

greatly improved fluorescence properties. J. Biol. Chem. 1985, 260, 3440–3450.

59. Burdette, S.C.; Walkup, G.K.; Spingler, B.; Tsien, R.Y.; Lippard, S.J. Fluorescent sensors

for Zn2+ based on a fluorescein platform: Synthesis, properties and intracellular

distribution. J. Am. Chem. Soc. 2001, 123, 7831–7841.

60. Behrens, I.; Pena, A.I.; Alonso, M.J.; Kissel, T. Comparative uptake studies of

bioadhesive and non-bioadhesive nanoparticles in human intestinal cell lines and

rats: The effect of mucus on particle adsorption and transport. Pharm. Res. 2002, 19,

1185–1193.

61. Kirkham, S.; Sheehan, J.K.; Knight, D.; Richardson, P.S.; Thornton, D.J. Heterogeneity

of airways mucus: Variations in the amounts and glycoforms of the major oligomeric

mucins MUC5aC and MUC5B. Biochem. J. 2002, 361, 537–546.

62. Thornton, D.J.; Carlstedt, I.; Howard, M.; Devine, P.L.; Price, M.R.; Sheehan, J.K.

Respiratory mucins: Identification of core proteins and glycoforms. Biochem. J. 1996,

316, 967–975.

63. Leal, J.; Smyth, H.D.C.; Ghosh, D. Physicochemical properties of mucus and their

impact on transmucosal drug delivery. Int. J. Pharm. 2017, 532, 555–572.

64. Nollevaux, G.; Deville, C.; El Moualij, B.; Zorzi, W.; Deloyer, P.; Schneider, Y.J.; Peulen,

O.; Dandrifosse, G. Development of a serum-free co-culture of human intestinal

epithelium cell-lines (Caco-2/HT29-5M21). BMC Cell Biol. 2006, 7, 20.

In vitro Studies on Zinc Binding and Buffering by Intestinal Mucins

112

General Discussion

113

Chapter 7. General Discussion

Zinc resorption mainly occurs in the small intestine. Here, zinc is transported into

enterocytes and excreted into the blood stream. In the past four decades, the zinc

resorption site and kinetic parameters of zinc uptake into the intestinal epithelium as well as

the fractional resorption and bioavailability of this micronutrient were investigated mainly

using in vivo or in vitro animal models. The investigation of metallothionein as an important

zinc binding protein whose expression is dependent on dietary zinc intake as well as the

discovery of the main intestinal zinc transporters ZIP-4, ZIP-5, ZnT-1, and ZnT-5 profoundly

contributed to the understanding of regulatory processes of intestinal zinc resorption.

However, there are still many questions to be answered, particularly considering molecular

parameters that regulate the uptake of zinc into enterocytes and its release into blood

circulation as well as dietary factors affecting its luminal availability for the intestinal

epithelium.

For this purpose, the application of in vitro intestinal models provides a standardized and

versatile platform to analyze zinc resorption across the intestinal epithelium as well as to

elucidate its cellular disposition and transfer after its absorption into enterocytes. In this

way, sensitive regulatory parameters of zinc resorption as well as bioavailability of zinc

species can be scrutinized. One of the biggest challenges of in vitro cell models, however, is

the difficulty to resemble the in vivo situation as close as possible.

Therefore, the scope of this thesis was to investigate zinc resorption using in vitro intestinal

models. What is more, the aim was to apply an in vitro model closer to the in vivo situation

to analyze zinc transport, as well as fluorescent sensors to study enterocytes’ zinc uptake in

more detail. First, an improved three-dimensional co-culture of Caco-2 and the mucin-

producing goblet cell line HT-29-MTX was established, critically addressing luminal factors

(Chapter 5 and 6) as well as the basolateral medium composition simulating the intestinal

blood side in vivo (Chapter 5). Moreover, the LMW sensor Zinpyr-1 was used to analyze the

bioavailability of different zinc species and short-time zinc uptake (Chapter 5-6) and,

moreover, a stable Caco-2 clone expressing the genetically encoded zinc sensor eCalwy-5

was produced and characterized providing a promising intestinal model system to study

enterocytes’ zinc resorption (Chapter 4).

On the following pages the main results of this thesis, composed of the findings presented in

Chapters 4-6, are related and consequently discussed in the light of current knowledge.

General Discussion

114

7.1 Application of fluorescent sensors to study intestinal zinc uptake

Aside of determining changes of total cellular zinc with conventional analytical approaches,

using inductively-coupled plasma mass spectrometry (ICP-MS) or flame atomic absorption

spectrometry (FAAS), respectively, the investigation of intracellular free zinc upon zinc

absorption provides a fruitful approach to elucidate already considerably small fluctuations

of intracellular zinc and its spatial distribution. For this, a wide range of low molecular weight

(LMW) or genetically encoded fluorescent zinc sensors can be used to quantify intracellular

free zinc and depict its intracellular localization (for details refer to Chapter 2, p. 33 ff.).

In this thesis, LMW zinc sensor Zinpyr-1 was used to measure zinc uptake into enterocytes

and goblet cells. The cellular free zinc pool is strictly regulated by zinc binding proteins like

metallothionein and zinc transporters, maintaining the intracellular free zinc concentration

in the nanomolar range in both cell lines (Chapter 5 and Chapter 6). More specifically, basal

free zinc concentration in Caco-2 cells in these studies was determined to be ~0.2 nM,

whereas free levels in the goblet cell line HT-29-MTX varied around~0.5 nM. Incubation of

both cell lines with zinc immediately increased this zinc pool depending on the luminal

available zinc concentration. Hence due to its high bioavailability and mobility, free zinc

served as an interesting zinc species to analyze intestinal zinc transport. Consequently,

already small changes of free zinc upon uptake of apical zinc into enterocytes and goblet

cells could be measured. This is particularly of interest when analyzing samples with very low

zinc content or short-time zinc uptake into enterocytes Caco-2. The latter was observed in

studies of Chapter 5 and Chapter 6 of this thesis, where changes of the intracellular free zinc

pool after short-term incubation of enterocytes seem to be too small compared to the total

cellular zinc content to be detected with conventional analytical methods, such as FAAS,

whereas the uptake was determined using low molecular weight sensor Zinpyr-1 (Chapter 5

and Chapter 6).

The application of genetically encoded biosensors offers various advantages compared to

low molecular weight sensors, as already discussed in Chapter 2 on p. 33 ff. Nevertheless,

these protein zinc sensors have never been applied in intestinal cells before. To generate the

first Caco-2 cell clone expressing a zinc biosensor, Caco-2 cells were stably transfected with

the protein sensor eCalwy-5 (Kd (eCalwy-5) = 1.85 nM [332]) (Chapter 4). Since future

research aims to use Caco-2-eCalwy clones as a cell model representing the intestinal

epithelium, the development of enterocyte specific properties in Caco-2-eCalwy was

reassured (Chapter 4). Hence, Caco-2-eCalwy clones provide a well characterized intestinal

model to analyze intestinal zinc uptake in addition to LMW sensors. The functionality of the

sensor and its zinc-dependent FRET signal in these stable Caco-2 clones was measured using

different techniques: laser scanning and two photon microscopy as well as detection of

FLIM-FRET (Chapter 4). These analytical approaches are all quite elaborate and in contrast to

BRET-based biosensors, change of FRET-ratio upon zinc uptake cannot be measured with

plate reader assays [333,339].

Conversely to genetically encoded sensors, spatial distribution and accumulation of LMW

sensors inside cells is not easy to control [100] and is often different between cell lines. This

General Discussion

115

has to be considered, particularly when comparing free zinc concentrations between cell

lines or sensors. Moreover, subcellular zinc concentrations were reported to be different

between organelles [74,342,343], hence prior characterization of intracellular sensor

distribution is quite important [343]. Zinpyr-1, for example, was described to co-localize with

staining of the endoplasmic reticulum, Golgi and mitochondria in Hela cells [349]. In the

neuronal cell line SH-SY5Y Zinpyr-1 showed cytoplasmic distribution [350], whereas in COS-7

cells the sensor was suggested to stain Golgi-associated vesicles [351,352]. In this thesis, the

low molecular weight sensor accumulated in cytoplasmic vesicles in Caco-2, HT-29-MTX as

well as HT-29 cells (Chapter 5 and Chapter 6). The eCalwy-5-sensor, on the other hand, is

distributed in the cytoplasm of Caco-2-eCalwy clones (Chapter 4) similar to the distribution

eCalwy-4 in transient transfected INS-1(832/13)-cells [332].

When comparing the basal free zinc levels of HT-29 cells determined with Zinpyr-1 in this

thesis (~0.8 nM) to a study where the less affine sensor Fluozin-3 (Kd (Fluozin-3) = 8.9 nM

[348], Kd (Zinpyr-1) = 0.7 nM [352]) was used to measure free zinc in HT-29 (0.4-0.7 nM)

[348], the correlation is actually quite good. In fact, LMW sensor concentration was shown

to be more critical for quantification of intracellular free zinc [348,353,354] than its actual

binding affinity, as the intracellular quantity of LMW sensors probably varies in several

hundred micromolar and deranges the steady state of the intracellular free zinc pool more

severely [354]. In contrast, intracellular concentration of biosensors depends on their

expression levels and was reported to be at least one order of magnitude smaller than that

of chemical probes [335]. Likewise, this applies also for cytoplasmic eCalwy sensor

concentration, as the heterogenic sensor expression between Caco-2-eCalwy cells did not

affect zinc-saturation of the biosensor.

The examination of free zinc (pools) in Caco-2 cells with eCalwy biosensor (Chapter 4) and

LMW sensor Zinpyr-1 (Chapter 5 and Chapter 6) documents that enterocytes comprise at

least two different zinc pools: cytoplasmic free zinc and vesicular stored zinc, both

containing nanomolar traces of the metal in basal states. Zinc uptake by enterocytes

increased the vesicular stored zinc pool as already mentioned above (Chapter 5 and Chapter

6) yet zinc treatment of Caco-2-eCalwy also augmented the cytoplasmic free zinc level

(Chapter 4). In the light of the suggested regulatory mechanism of enterocytes’ zinc

homeostasis during the resorption process (Chapter 2, p. 17 ff.), these findings indicate that

free zinc in cytoplasm increases to a certain level after entering the cell but is subsequently

sequestered into cellular zinc storages to maintain the intracellular zinc pool. These

processes have to be further scrutinized; in particular the chronology of the zinc transfer

through the enterocytes upon its absorption and its subsequent basolateral release into the

blood circulation needs to be unraveled in more detail.

General Discussion

116

7.2 Role of Luminal and Basolateral Factors in in vitro Zinc Resorption

As already discussed in Chapter 2, only free zinc that is not complexed by macromolecules is

available for the intestinal epithelium. Dependent on their affinity for zinc, these complexes

can critically impact zinc bioavailability. This is also discussed in Chapter 5 and Chapter 6,

where two different proteins with high zinc binding capacity, albumin and zinc-depleted

gastrointestinal mucin, decreased apical availability of the cation for enterocytes.

Consequently, medium and buffer composition has to be considered when analyzing zinc

with in vitro models, as they could affect zinc speciation and its actual available

concentration for cells [20]. Since the aim of this thesis was to analyze zinc resorption in an

experimental setting close to the in vivo situation in the intestine, these constituents have to

both simulate the physiological environment in vivo, while guaranteeing suitable in vitro

vicinity for the cells as well.

A particular problem in this context is fetal calf serum (FCS), which proves to be an

unpredictable factor due to its variability [355] and contains about 60% albumin [356].

Notably, FCS is commonly used in cell culture [357], just as 10% FCS is used to culture cells in

the present thesis, resulting in a final albumin concentration of 1.55 mg mL-1 [355]

(corresponding to 24.2 µM) in the medium. Albumin is the main zinc transporting protein in

blood serum in vivo [84] and binds the metal with high affinity [358]. Consequently, zinc

binding to apically added FCS or bovine serum albumin (BSA) severely impacts its

bioavailability for enterocyte Caco-2 cells as shown in decreased cytotoxicity and short-time

zinc uptake in presence of these proteins in Chapter 5.

However adding albumin to the luminal side of enterocytes certainly does not represent the

in vivo situation in the intestinal lumen and is consequently not included in apical medium

composition of the in vitro intestinal models in this thesis. Nevertheless, this protein was

used as an apical component in previous in vitro zinc transport studies using three-

dimensional Caco-2 models (as shown in Chapter 2, Table 2.5, p. 30 ff.). Notably,

enterocytes’ zinc transport in three-dimensional models is strongly altered by the apical

medium composition. This becomes obvious in comparison to a previous zinc resorption

study where zinc together with 10% FCS was apically applied in a Caco-2 monoculture [128]

in contrast to the Caco-2 monocultures from this thesis, where no FCS was added to the

apical side and no BSA was added basolaterally (Appendix E). The presence of 10% FCS alters

cellular available zinc, as the zinc transport rate was certainly smaller with apical FCS than in

the absence of apical albumin (Caco-2 monocultures + apical 10% FCS: transport rate of 50

µM after 8h: 0.05 nmol cm-2 [128]; Caco-2 monoculture without basolateral albumin

addition: transport rate of 50 µM zinc after 8 h: 1.16 nmol cm-2, Appendix E, Figure S1, Table

S4).

Actually, the authors of this previous study [128] purposely applied 10% FCS on the apical

side of their intestinal model to mimic luminal protein matrix during transport studies. The

occurrence of such protein concentrations, however, does not reflect the luminal

environment in vivo, as the protein would have already been degraded into smaller

molecules (peptides or amino acids) due to the digestion process in the gastrointestinal

General Discussion

117

tract. In fact in the present thesis, in vitro digestion led to degradation of BSA and increased

the zinc bioavailability for Caco-2 cells compared to undigested protein (Chapter 5).

Interestingly, comparable to in vivo studies [239], digested BSA still binds luminal zinc

diminishing its bioavailability compared to protein free controls. Nevertheless, the protein

content itself has a positive effect on zinc resorption in the intestinal lumen mainly because

of its release of amino acids and peptides upon its degradation, possibly increasing luminal

solubility of the metal and consequently enhancing its availability into enterocytes [12,245].

Figure 7.1 The role of the intestinal mucus layer as a luminal factor of intestinal zinc resorption

The intestinal mucus layer prevents hydroxypolymerization of zinc in intestinal lumen by binding the metal. Subsequently mucins buffer zinc that is available for enterocytes for the intestinal epithelium making them an integral part of intestinal zinc resorption. Zinc that is transported to the basolateral side of enterocytes is bound to albumin, which is the main zinc transporter protein in the blood. Basolateral albumin levels were shown to influence zinc resorption in three-dimensional intestinal model in this thesis by acting as a basolateral zinc acceptor.

At intestinal pH zinc naturally forms zinc hydroxide, which is highly insoluble [265] and thus

lowers its availability for enterocytes (Figure 7.1). This precipitation, however, is prevented

by the intestinal mucus layer, which binds the trace element and subsequently plays a

beneficial role for its resorption [24]. In the present study, discussed in Chapter 6, the

binding properties of gastrointestinal mucins are investigated in detail, showing that these

glycoproteins contain multiple zinc binding sites with physiologically relevant affinity.

Keeping in mind that the mucus layer is not static but represents a dynamic and viscoelastic

gel [43,64], these glycoproteins might assist zinc transport via this physical barrier to the

General Discussion

118

underlying epithelium. In fact, their ability to bind the cation and buffer its free levels that

subsequently would be available for intestinal cells was also observed in short-term zinc

uptake studies of goblet cells and enterocytes in this thesis (Chapter 6). This indicates that

this process might even lead to retention of luminal available zinc and was already discussed

in animal studies [23,266] but was neither investigated in detail nor included in previous in

vitro intestinal models to investigate intestinal zinc transport. Our findings in Chapter 6

comparing the mucin-producing Caco-2/HT-29-MTX model with mucus-lacking Caco-2

monocultures in the presence of basolateral albumin, supports the hypothesis of a beneficial

role of mucins for intestinal zinc absorption. In this study enhanced apical zinc uptake and

higher fractional resorption was observed when a mucus layer was present (for detailed

results of this study refer to Table 7.1, p. 121). Consequently, the application of a mucus

layer or mucin-producing cells in in vitro models to study intestinal zinc resorption must not

be neglected.

Simulation of the mucus layer by adding isolated (porcine) mucins on top of three-

dimensional Caco-2 monocultures was already critically discussed in connection with iron

transport studies [44,312]. These mucins do not display similar viscoelastic and gel forming

properties of gastrointestinal mucus layer in vivo, possibly because of their isolation and

purification process [359,360] as already discussed in detail in Chapter 6. Moreover, these

isolated mucins do not simulate transmembrane mucins which, however, represent an

important fraction of the mucus layer in vivo (for details refer to Chapter 2, p.7). In this

context it is also worth noticing that the Caco-2/HT-29-MTX model is discussed to provide a

more physiological in vitro model than Caco-2 monocultures [361]. In fact, introducing HT-

29-MTX to Caco-2 monocultures does not only improve this intestinal model regarding the

presence of a mucus layer, but was also reported to optimize cellular permeability of

conventional Caco-2 cultures [313,362].

In contrast to the apical side, where FCS addition has to be excluded to resemble the in vivo

situation in lumen, the blood serum in vivo contains about 30–50 mg mL-1 human serum

albumin (HSA) [20], the main zinc binding protein in serum [84] buffering the total zinc level

of 12–16 µM [81-83] on the serosal side of enterocytes to the nanomolar range [363-365];

hence 30 mg mL-1 BSA was added to the basolateral medium of the intestinal model

(Chapter 5 and Chapter 6). In fact, transport studies with and without basolateral albumin in

this thesis (Chapter 5) clearly demonstrate that albumin acts as a basolateral zinc acceptor

and enhances serosal zinc export in vitro, while its role as a zinc acceptor in human

resorption has to be confirmed with HSA. Regardless, in vivo basolateral applied BSA was

shown to enhance fractional zinc resorption in vascular perfusion experiments of rat small

intestine [366]. These findings emphasize the relevance of basolateral constituents,

representing the blood side of the intestinal epithelium, when investigating zinc resorption.

It has to be noted, that the basolateral medium in our study contained only 3 µM zinc at the

beginning of experiments leading to a 110-fold molar excess of albumin which does not

reflect the molar albumin : zinc-ratio of 30 in vivo [20]. Due to its high zinc binding affinity

[358], the elevated basolateral albumin level provides indeed an additional thermodynamic

General Discussion

119

sink for the metal. However, only basolateral zinc excretion of the cells is enhanced by this

zinc acceptor, whereas cellular zinc uptake seems to be unaffected by the albumin level

(Chapter 5). Hence, higher zinc transport in the presence of albumin is not based on a simple

diffusion process, following a zinc concentration gradient from apical to basolateral side as

apical zinc uptake into the cell is not perturbed by basolateral albumin. This also reiterates

previous knowledge on intestinal zinc uptake and transport kinetics (in detail described in

Chapter 2, p. 15 ff.). In vivo apical to basolateral zinc transport is a saturable and carrier-

mediated process [125], where apical zinc uptake is suggested to be the rate limiting step

[131]. This transport process is mainly mediated by the apical zinc importer ZIP-4 and

basolateral zinc exporter ZnT-1 [367], which are both regulated by dietary zinc (for details

refer to Chapter 2, p. 17 ff.). Recent findings demonstrate that basolateral zinc export by

ZnT-1 in Caco-2 cells into the basolateral compartment are attenuated by the humoral factor

hepcidin [190] (for details refer to Chapter 2, p. 26 ff.). The underlying regulatory

parameters, however, that enhance the basolateral release of zinc in presence of albumin

need to be further investigated.

Some of the previous in vitro zinc transport studies using Caco-2 models (Chapter 2, Table

2.5, p. 30) used cell culture medium with 10% FCS for the basolateral compartment

[125,128,129,323]. Since FCS contains about 15.5 mg mL-1 albumin [355], this FCS

concentration corresponds to only 3–5% of the serum albumin concentration in vivo (10%

FCS: 1.55 mg mL-1 albumin; blood serum in vivo: 30-50 mg mL-1 HSA [20]). Although these in

vitro studies applied zinc to the basolateral compartment mainly to investigate the serosal

zinc uptake into the intestinal epithelium [127-129,323], the apical zinc transport in the

presence of physiologic zinc and albumin concentrations on the basolateral side has not yet

been investigated. Until now, zinc content on the basolateral side of three-dimensional

intestinal models only originated from FCS [125,128,129,323], which generally contains

higher amounts of zinc than cell culture medium [368]. The exact basolateral zinc

concentration in these studies though are unknown, as FCS’ zinc content highly varies

leading to final zinc concentrations between 3 [137] and 14 µM [369]. The addition of

albumin apart from the amount already present in FCS, to the basolateral side of the in vitro

model as done in this thesis, however, was only performed in two in vitro transport studies

with Caco-2 monocultures before, applying very low (2.5 mg mL-1) [322] and physiologic

albumin concentrations (5% BSA; corresponding to 50 mg mL-1 albumin) [125]. Regardless,

the effect of serum albumin on zinc resorption by in vitro intestinal models has not been

discussed in these studies.

General Discussion

120

7.3 Comparison of Zinc Transport in the Three-Dimensional in vitro Intestinal Model Caco-

2/HT-29-MTX with Specific in vitro Caco-2 Models and in vivo Zinc Resorption

In this thesis the conventional Caco-2 model was expanded regarding its cellular composition

as well as luminal and basolateral factors including mucin-producing HT-29-MTX cells and

physiological serum concentrations. Yet, it has to be noted that the impact of basolateral

albumin on zinc transport, discussed in Chapter 5, was studied already using the mucin-

producing Caco-2/HT-29-MTX co-cultures, whereas the effect of mucins on zinc resorption

from Chapter 6 was obtained using mono- and co-cultures in presence of basolateral

albumin (for details refer to Table 7.1, p. 121). Hence, the influence of basolateral albumin

(Chapter 5) and apical mucin (Chapter 6) on zinc transport by three-dimensional cell models

was not investigated separately. In the following, the role of mucins and basolateral albumin

on zinc transport, after apical zinc treatment for 8 h, is discussed separately and compared

to findings from Chapter 5 and Chapter 6. For this, Table S3 and S4 as well as Figure S1 in

Appendix E summarize the results of the zinc transport studies from Chapter 5 and Chapter

6 together with additional data of zinc transport in Caco-2 mono- and co-cultures with and

without basolaterally added albumin after incubation for 4 h and 8 h.

Comparison of our findings with previous in vitro transport studies is challenging, as

incubation time and unit of reported results differ. Regardless, Moltedo et al. reported zinc

transport rates of 2.0 nmol zinc per cm² after treatment of Caco-2 monocultures with 100

µM zinc for 6 h. They used serum free medium on apical and basolateral side. Likewise,

apical addition of 100 µM zinc to Caco-2 monocultures without basolateral albumin from this

thesis for 8 h, yielded zinc transport rates of 1.6 nmol zinc per cm² resorption area (Appendix

E, Figure S1). Basolateral addition of albumin increased zinc transport of this model to 1.6-

fold, leading to transport rates of 2.6 nmol zinc per cm² (Appendix E, Figure S1). This is

additionally augmented in the presence of mucins as Caco-2/HT-29-MTX reached zinc

transport of 3.6 nmol zinc per cm2 resorption area (Chapter 5, Appendix E, Figure S1).

In general, the introduction of basolateral albumin to Caco-2 monocultures and to Caco-

2/HT-29-MTX co-cultures increased serosal zinc export of cell monolayers after 8 h, as

cellular zinc content seems to be lower (for details refer to Chapter 5 and Appendix E, Figure

S1A-B, Table S3) and the resorbed amount of zinc was elevated in both in vitro models in the

presence of this protein, whereas apical zinc uptake was not different with or without

albumin (for details refer to Appendix E, Table S3, Table S4).

When implementing the mucus layer in Caco-2 monocultures by way of the co-culture with

mucin-producing HT-29-MTX cells, the beneficial impact of mucins on cellular zinc

absorption, described in Chapter 6, occurs also without basolateral albumin. More precisely,

zinc uptake into Caco-2/HT-29-MTX in absence of basolateral albumin was elevated (Chapter

5, Figure 7A and Appendix E, Figure S1A, two-way ANOVA with Bonferroni post hoc test

comparing mono-and co-culture without basolateral BSA: 100 µM: p < 0.01), whereas zinc

transport rates and fractional zinc resorption of conventional Caco-2 monocultures and co-

cultures were comparable (Appendix E, Figure S1C). After adding albumin to the basolateral

side of these models, the fractional zinc resorption as well as zinc transport rate were

General Discussion

121

increasing (for detailed results refer to Chapter 5 and Appendix E, Figure S1B, D, F, Table S4).

Interestingly, the basolateral zinc acceptor shows a stronger effect on the fractional zinc

resorption of mucin-producing co-cultures than on the mucus-lacking Caco-2 model, as net

absorption of Caco-2/HT-29-MTX is significantly elevated in the presence of albumin

(Appendix E, Table S4; Figure S1; two-way ANOVA with Bonferroni post hoc test comparing

co-cultures without and with basolateral BSA: 25 µM: p < 0.001; 50 µM: p < 0.05).

Comparable to observations from Chapter 6, investigating zinc transport in the presence of

albumin after 4 h, net resorption of co-cultures with basolateral BSA after 8 h is 1.4–1.8-fold

higher than that of Caco-2 models without apical mucus layer (Appendix E, Figure S1F, Table

S4).

All in all, these findings indicate that mucins assist apical zinc uptake and basolateral albumin

increases enterocytes’ zinc excretion to the blood side; this also holds true when examined

separately. Consequently, fractional zinc resorption of apically applied zinc (25–100 µM)

increased from 0.9–1.6% in the absence of mucins, to 1.9–4.2% in the presence of mucins

(Chapter 6) and from 2% in the absence of basolateral albumin to 2.9–5.8% with albumin

(Chapter 5). Table 7.1 summarizes the main results of zinc transport studies of the present

work which are discussed in detail in Chapter 5 and 6.

Table 7.1 Main Results of in vitro zinc transport studies from this thesis

Cell model Incubation parameter

Zinc Quantification

Main Outcome

Chapter 5 Caco-2/HT-29-MTX co-culture Differentiation time: 21 d 3D Transwell (PC)

ZnSO4 0 –100 µM (apical: serum free transport buffer, basolateral: DMEM +10% FCS + 0 or 30 mg mL

-1 BSA)

for 8 h

ICP-MS

- albumin has a role in in vitro zinc resorption as a basolateral zinc acceptor

- cellular uptake is not significantly different with or w/o basolateral added albumin

- basolateral serum albumin enhances cellular zinc export to the basolateral side

- fractional resorption (25 -100 µM): w/o BSA: ~ 2% with BSA: 5.8 - 2.9%

- zinc transport rates (0-100 µM): w/o BSA: 0.1 - 2.2 nmol cm

-2

with BSA: 1.1 - 3.6 nmol cm-2

Chapter 6 Caco-2/HT-29-MTX co-culture and Caco-2 monoculture Differentiation time: 21 d 3D Transwell (PC)

ZnSO4 0 –100 µM (apical: serum free transport buffer, basolateral: DMEM +10% FCS + 30 mg mL

-1 BSA)

for 4 h

ICP-MS

- intestinal mucins influence cellular zinc uptake and zinc transport

- results suggest that mucins facilitate zinc uptake into enterocytes and act as a zinc delivery system

- mucins are an integral part of intestinal zinc resorption

- fractional resorption (25 -100 µM): monoculture: 1.6 - 0.9% co-culture: 4.2 - 1.9%

- zinc transport rates (0-100 µM): monoculture: 0.3 - 1.3 nmol cm

-2

co-culture: 1.1 - 2.3 nmol cm-2

BSA = bovine serum albumin; DMEM = Dulbecco’s Modified Eagles Medium; FCS = fetal calf serum; ICP-MS = inductively-

coupled plasma mass spectrometry; PC = Polycarbonate membrane.

General Discussion

122

The in vitro intestinal model Caco-2/HT-29-MTX established in the present thesis provides a

suitable platform to investigate zinc resorption in an experimental setting closer to the in

vivo situation than conventional Caco-2 monocultures. Nevertheless, comparing the

determined fractional resorption of the in vitro model (Table 7.1) with estimated net

resorption of 16–50% for humans in vivo [15,88,138-141], these results appear really low.

The mentioned human fractional resorption was observed in several in vivo resorption

studies investigating the net resorption and bioavailability of the micronutrient (for details

refer to Chapter 2, p. 15 ff.), generally reporting an inverse relation to the luminal available

zinc level. Likewise zinc resorption in in vitro intestinal model Caco-2/HT-29-MTX illustrated

the same concentration-dependency. Notably, most of these in vivo studies investigated

fractional zinc absorption from meals containing dietary ligands that additionally affect its

bioavailability in the intestinal lumen [11], whereas in the present in vitro studies, zinc was

added as liquid solutions and without food matrix. However, in vivo studies estimating the

fractional zinc resorption from liquid solutions with comparable zinc concentrations to those

applied in the in vitro studies from this thesis are scarce [5,119]. Hence in the following,

findings of studies from Chapter 5 and 6 are compared to data from a human in vivo study

were comparable zinc concentrations were applied with a meal.

In an in vivo study by Hunt et al. intake of a meal containing 17 mg zinc yielded fractional

zinc resorption of 24% [15], whereas 49% were absorbed when 4.3 mg zinc was

administered [15]. Assuming an average intestinal resorption area of 30.000 m² [33] and

intestinal liquid of 3 L [28], oral zinc intake of 17 mg zinc would correspond to a luminal

concentration of maximum 86 µM, while 4.3 mg zinc would result in 22 µM luminal zinc. Net

absorption of the intestinal model Caco-2/HT-29-MTX in Chapter 5 yielded 5.8 ± 0.9% or 2.9

± 1.1% after treatment with 25 or 100 µM in aqueous solutions for 8 h. This is certainly

lower than the fractional absorption of comparable luminal zinc concentrations in vivo from

the abovementioned study by Hunt et al. and yet it represents 1/10 of fractional zinc

resorption in vivo.

Applying in vitro models to mimic processes in vivo always needs a critical discussion of their

limitations. Even though the in vitro intestinal model Caco-2/HT-29-MTX in this thesis

represents an improved and more physiological model than conventional monocultures

[285] that were used before to study zinc transport, there are certain differences to the in

vivo intestinal epithelium that have to be critically examined.

Two important physical factors in the intestine in vivo are lacking in this in vitro model:

intestinal and blood fluid flow as well as peristaltic motions. Intestinal peristalsis enables

movement of chyme along the intestine and increases mechanical degradation of food

components [28], which is important for the digestion process and availability of nutrients

for absorption. Furthermore, the present in vitro model did not include the intestinal

digestion process, which might also influence the luminal bioaccessibility of the metal. Even

though a combination of the in vitro intestinal model with in vitro digestion is principally

possible [312], this was not applied on purpose, because at first the resorption of liquid zinc

samples had to be investigated solely.

General Discussion

123

In vivo, resorbed zinc is bound to albumin and continuously transported within the blood

circulation distributing the cation in the whole body [84]. This sink is missing in vitro and

resorbed zinc accumulates in the basolateral compartment of the three-dimensional cell

models. It seems that the addition of physiological albumin concentrations, although not yet

completely saturated with zinc, is not enough to fulfill this task. This insinuates the

involvement of another parameter in enterocytes’ zinc release into the blood, probably a yet

unidentified humoral factor, similar to hepcidin, but with converse impact on zinc

resorption.

Moreover, the ratio of intestinal liquid per resorption area has to be taken into account

when comparing in vitro and in vivo fractional resorption. The resorption area in vitro (1.12

cm2) is certainly smaller compared to the size of the intestinal epithelium in vivo (~30.000

cm2) [33]. It has to be noted that this estimation does not include the actual amount of

absorptive enterocytes to the resorption area of the intestinal epithelium in vivo and

disregards the factor that microvilli of Caco-2 cells would incorporate into the actual

resorption area in vitro. The volume of intestinal liquid in lumen in vivo amounts to around 3

L [28] (corresponding to 0.1 mL cm-2), whereas in the in vitro model the volume to area ratio

is 0.45 ml per cm2, leading to a 4.5-fold higher apically applied liquid volume per cm2

resorption area in vitro. Hence the total amount of zinc that has to be transported per

resorption area in vitro is greater than in vivo, which consequently influences the estimation

of fractional resorption. It is well known, that fractional zinc resorption declines with

increased zinc concentration and that zinc uptake into enterocytes is a saturable process.

Accordingly, assuming that expression and activity of the main zinc transporters in Caco-2

cells in vitro correspond to the expression in vivo which we certainly not know yet, the

higher ratio of zinc per cm2 resorption area in the in vitro model could be one explanation

for the smaller fractional resorption. Nevertheless, adjusting the liquid volume that is

applied into apical chambers of in vitro intestinal models to the volume per cm2 ratio in vivo

(0.1 mL cm-2) would attenuate cellular viability and consequently negatively affect the zinc

resorption process.

In addition to net absorption, it would be interesting to compare the final amounts of zinc

that are resorbed into the blood circulation in vivo with zinc levels that were transported to

the basolateral side of the three-dimensional in vitro model. For this, Table 7.2 depicts

estimated amounts of actual transported zinc per cm2 resorption area in vitro and in vivo,

using data of a study from Hunt et al. [15].

General Discussion

124

Table 7.2 Total amounts of resorbed zinc in vivo and in the Caco-2/HT-29-MTX model of this thesis

In vitro (A = 1.12 cm2, Volume: 500 µL)

Apical zinc Fractional resorption

[%]

Resorbed zinc [µg/total resorption

area] Resorbed zinc [µg cm

-2]

100 µM = 3.23 µg/1.12 cm2

2.9 0.09 0.08

25 µM = 0.82 µg/1.12 cm2

5.8 0.05 0.04

In vivo [15] (A = 30-40 m2~30 m2, Volume: 3-6 L ~3 L)

17 mg/30 m2 = 86 µM

24 4080 0.14

4.3 mg/30 m2 = 21 µM

49 2100 0.07

According to the above calculation, the amounts of actually transported zinc with the

optimized in vitro model of this thesis are by all means quite similar to the estimated

amounts transported in vivo. Compared to conventional Caco-2 monocultures, not only is

the fractional resorption enhanced when using the mucin-producing Caco-2/HT-29-MTX

model from this thesis, but the amount of transported zinc to the basolateral side of the

model is also higher than reported before after 8 h [128] (detailed data in Chapter 2, Table

2.5, p.30). Hence, this model provides a suitable platform for future investigation of human

zinc transport kinetics, bioavailability of different zinc species as well as molecular regulatory

parameters of intestinal zinc resorption.

7.4 Conclusion

All in all, the work of this thesis demonstrates that the luminal and basolateral medium

composition is crucial when studying zinc resorption and that the intestinal mucus layer

plays a beneficial role in zinc resorption. These factors are all combined in the three-

dimensional in vitro model Caco-2/HT-29-MTX which provides a standardized

microenvironment to investigate intestinal zinc transport and resorption in an experimental

setting closer to the in vivo situation. Even though fractional zinc resorption obtained with

this model represents only ~10% of that reported in in vivo studies, when applying zinc

levels comparable to the ones typically found in the intestinal lumen after a meal, the total

amount of transported zinc in this in vitro model resembles those estimated in vivo after zinc

resorption. With this model the zinc resorption process can be further elucidated measuring

zinc uptake into the intestinal epithelium as well as its transport to the basolateral chamber,

consequently providing information about zinc transport kinetics and net resorption of the

apically applied zinc species.

By applying low molecular weight sensor Zinpyr-1 in Caco-2 cells and Caco-2-eCalwy clones

the cellular distribution of the essential metal upon its absorption into enterocytes and

General Discussion

125

throughout the resorption process can be additionally examined. What is more, using these

two sensors already small changes of the intracellular zinc pool were investigated, which is

of particular interest for short-term zinc uptake and samples with low zinc content. To this

end, involvement of two different cellular free zinc pools in the maintenance of enterocytes’

zinc homeostasis during zinc resorption could be illuminated. Moreover, Caco-2-eCalwy

clones offer a well characterized intestinal model system to investigate enterocyte free zinc

levels in addition to LMW sensors, which might be worth to be investigated in combination

with goblet cells to study enterocytes’ zinc homeostasis in co-culture.

Finally, by applying these three-dimensional and two-dimensional intestinal models the

impact of albumin as a basolateral zinc acceptor and the role of the mucus layer as an apical

zinc transfer system to the underlying brush border membrane were elucidated. These

findings contribute to the closer understanding of the in vitro and in vivo zinc uptake and

transport process on the apical mucosal membrane as well as on the basolateral side of

enterocytes.

7.5 Future Perspectives

As far as the in vitro intestinal model of this thesis is concerned, further work needs to be

done to investigate whether basolateral addition of zinc closer to the serum concentrations

in vivo will impact zinc transport kinetics of the three-dimensional Caco-2/HT-29-MTX model

and if the model has to be optimized accordingly. Besides, to validate the role of albumin as

a basolateral zinc acceptor in human zinc resorption, zinc transport has to be investigated in

the presence of human serum albumin. Moreover, future implementation of the in vitro

intestinal model of this thesis should consider a combination of the in vitro model with an in

vitro digestion model to study zinc bioavailability from complex food samples.

In the course of this thesis the involvement of humoral factors in regulating zinc resorption,

particularly the basolateral zinc export into the blood circulation were suggested. It was

already shown that hepcidin, a humoral factor produced by hepatocytes, affects basolateral

zinc transport [190]. Hence, similar to its involvement in iron resorption [270], the liver

might play an important role in secreting humoral factors to regulate zinc resorption.

Therefore, a triple culture of Caco-2 mono- or co-cultures with basolateral seeded

hepatocytes would provide a suitable platform to tackle this question and to elucidate the

role of the liver in this process.

The findings in this study demonstrate that mucins bind zinc with biologically relevant

affinity. It was already suggested that intestinal mucins influence bioavailability of trace

elements by competitively binding them. Hence, it would be valuable to scrutinize zinc

binding to mucins in the presence of other metals in in vitro cell free analysis. Together with

the determination of zinc resorption and transport kinetics in the presence of other trace

elements with the mucin-producing Caco-2/HT-29-MTX model, these measurements could

provide further insights into the mutual interference of their absorption in the intestine.

Furthermore, stably transfected cells such as Caco-2-eCalwy offer the great opportunity to

be co-cultured with other cell lines, like HT-29-MTX for example. With this intestinal model

zinc can be analyzed specifically in this cell type, tracking the micronutrient after its

General Discussion

126

absorption into the enterocytes, in the presence of goblet cells and a mucus layer, while

LMW probes would always stain the entire model. In addition to the above discussed impact

of mucins on zinc resorption, this could contribute to further understanding about the

regulatory role of mucins in enterocytes’ zinc absorption and help characterize zinc

homoeostasis of enterocytes in presence of goblet cells.

With respect to the complex analytical techniques to measure FRET in transfected cell

clones, the application of BRET-sensors is quite promising. Herein, Caco-2 clones stably

expressing BRET-sensor Zinch-3 from the Merkx group [333] was already produced in our

group. With these means, free zinc of Caco-2-Zinch clones in monoculture as well as in co-

culture with HT-29-MTX cells, can be measured using plate readers. Moreover, future

research should aim to produce stably transfected Caco-2 clones with organelle-specific

biosensors, as presented in Chapter 2, to contribute to the knowledge of subcellular

distribution of zinc after its absorption into enterocytes.

All in all, these future perspectives show that there are several points that await to be

answered. Their investigation would not only enhance the current knowledge on zinc

resorption and molecular regulatory parameters of its luminal absorption and transport into

the blood circulation, but also contribute to the overall understanding of enterocytes’ zinc

homeostasis in addition to the results of this thesis.

References

127

Chapter 8. References

1. Maret, W. Zinc in cellular regulation: The nature and significance of "zinc signals". International Journal of Molecular Sciences 2017, 18, 2285-2297.

2. Rink, L.; Gabriel, P. Zinc and the immune system. Proceedings of the Nutrition Society 2000, 59, 541-552.

3. Krebs, N.F. Overview of zinc absorption and excretion in the human gastrointestinal tract. The Journal of Nutrition 2000, 130, 1374S-1377S.

4. Steinhardt, H.J.; Adibi, S.A. Interaction between transport of zinc and other solutes in human intestine. The American Journal of Physiology 1984, 247, G176-182.

5. Lee, H.H.; Prasad, A.S.; Brewer, G.J.; Owyang, C. Zinc absorption in human small intestine. The American Journal of Physiology 1989, 256, G87-91.

6. Ford, D. Intestinal and placental zinc transport pathways. The Proceedings of the Nutrition Society 2004, 63, 21-29.

7. King, J.C. Zinc: An essential but elusive nutrient. The American Journal of Clinical Nutrition 2011, 94, 6795-6845.

8. Maywald, M.; Wessels, I.; Rink, L. Zinc signals and immunity. International Journal of Molecular Sciences 2017, 18, 2222-2256.

9. Alker, W.; Haase, H. Zinc and sepsis. Nutrients 2018, 10, 976. 10. World Health Organization. The World Health Report: 2002: Reducing Risks,

Promoting Healthy Life. World Health Organization: Geneva, Switzerland, 2002. 11. Lönnerdal, B. Dietary factors influencing zinc absorption. The Journal of Nutrition

2000, 130, 1378S-1383S. 12. Wapnir, R.A. Zinc deficiency, malnutrition and the gastrointestinal tract. The Journal

of Nutrition 2000, 130, 13885-13925. 13. World Health Organization / Food and Agricultural Organization. Vitamin and Mineral

Requirements in Human Nutrition. 2 ed.; World Health Organization: Geneva, Switzerland, 2004.

14. Trame, S.; Wessels, I.; Haase, H.; Rink, L. A short 18 items food frequency questionnaire biochemically validated to estimate zinc status in humans. Journal of Trace Elements in Medicine and Biology 2018, 49, 285-295.

15. Hunt, J.R.; Beiseigel, J.M.; Johnson, L.K. Adaptation in human zinc absorption as influenced by dietary zinc and bioavailability. The American Journal of Clinical Nutrition 2008, 87, 1336-1345.

16. Hambidge, K.M.; Miller, L.V.; Westcott, J.E.; Sheng, X.; Krebs, N.F. Zinc bioavailability and homeostasis. The American Journal of Clinical Nutrition 2010, 91, 1478S-1483S.

17. Dosh, R.H.; Essa, A.; Jordan-Mahy, N.; Sammon, C.; Le Maitre, C.L. Use of hydrogel scaffolds to develop an in vitro 3D culture model of human intestinal epithelium. Acta Biomaterialia 2017, 62, 128-143.

18. Langerholc, T.; Maragkoudakis, P.A.; Wollgast, J.; Gradisnik, L.; Cencic, A. Novel and established intestinal cell line models – an indispensable tool in food science and nutrition. Trends in Food Science & Technology 2011, 22, S11-S20.

19. Ollig, J.; Kloubert, V.; Weßels, I.; Haase, H.; Rink, L. Parameters influencing zinc in experimental systems in vivo and in vitro. Metals 2016, 6, 71.

20. Haase, H.; Hebel, S.; Engelhardt, G.; Rink, L. The biochemical effects of extracellular Zn2+ and other metal ions are severely affected by their speciation in cell culture media. Metallomics 2015, 7, 102-111.

References

128

21. Bansil, R.; Turner, B.S. The biology of mucus: Composition, synthesis and organization. Advanced Drug Delivery Reviews 2018, 124, 3-15.

22. Johansson, M.E.; Sjovall, H.; Hansson, G.C. The gastrointestinal mucus system in health and disease. Nature Reviews. Gastroenterology & Hepatology 2013, 10, 352-361.

23. Glover, C.N.; Hogstrand, C. In vivo characterisation of intestinal zinc uptake in freshwater rainbow trout. Journal of Experimental Biology 2002, 205, 141-150.

24. Powell, J.J.; Jugdaohsingh, R.; Thompson, R.P. The regulation of mineral absorption in the gastrointestinal tract. The Proceedings of the Nutrition Society 1999, 58, 147-153.

25. DeSesso, J.M.; Jacobson, C.F. Anatomical and physiological parameters affecting gastrointestinal absorption in humans and rats. Food and Chemical Toxicology 2001, 39, 209-228.

26. Madara, J.L. Functional morphology of epithelium of the small intestine. In Comprehensive Physiology; Terjung, R., Ed.; JohnWiley & Sons, Inc.: Hoboken, NJ, USA, 2011.

27. Van der Flier, L.G.; Clevers, H. Stem cells, self-renewal, and differentiation in the intestinal epithelium. Annual Review of Physiology 2009, 71, 241-260.

28. Welcome, M.O. Gastrointestinal Physiology - Development, Principles and Mechanisms of Regulation. Springer International Publishing: Switzerland, 2018.

29. Shen, L. Functional morphology of the gastrointestinal tract. Current Topics in Microbiology and Immunology 2009, 337, 1-35.

30. Kiela, P.R.; Ghishan, F.K. Physiology of intestinal absorption and secretion. Best Practice & Research Clinical Gastroenterology 2016, 30, 145-159.

31. Fallingborg, J. Intraluminal ph of the human gastrointestinal tract. Danish Medical Bulletin 1999, 46, 183-196.

32. Schaal, S.; Kunsch, K.; Kunsch, S. Der Mensch in Zahlen. Springer Spektrum: Berlin, Heidelberg, 2016.

33. Helander, H.F.; Fandriks, L. Surface area of the digestive tract - revisited. Scandinavian Journal of Gastroenterology 2014, 49, 681-689.

34. Günzel, D.; Fromm, M. Claudins and other tight junction proteins. Comprehensive Physiology 2012, 2, 1819-1852.

35. Lee, S.H. Intestinal permeability regulation by tight junction: Implication on inflammatory bowel diseases. Intestinal Research 2015, 13, 11-18.

36. Stewart, A.S.; Pratt-Phillips, S.; Gonzalez, L.M. Alterations in intestinal permeability: The role of the “leaky gut” in health and disease. Journal of Equine Veterinary Science 2017, 52, 10-22.

37. Van Roy, F.; Berx, G. The cell-cell adhesion molecule E-cadherin. Cellular and Molecular Life Sciences 2008, 65, 3756-3788.

38. Sjoberg, A.; Lutz, M.; Tannergren, C.; Wingolf, C.; Borde, A.; Ungell, A.L. Comprehensive study on regional human intestinal permeability and prediction of fraction absorbed of drugs using the ussing chamber technique. European Journal of Pharmaceutical Sciences 2013, 48, 166-180.

39. Peterson, L.W.; Artis, D. Intestinal epithelial cells: Regulators of barrier function and immune homeostasis. Nature Reviews. Immunology 2014, 14, 141-153.

40. Gassler, N.; Newrzella, D.; Böhm, C.; Lyer, S.; Li, L.; Sorgenfrei, O.; van Laer, L.; Sido, B.; Mollenhauer, J.; Poustka, A., et al. Molecular characterisation of non-absorptive and absorptive enterocytes in human small intestine. Gut 2006, 55, 1084-1089.

References

129

41. Allaire, J.M.; Crowley, S.M.; Law, H.T.; Chang, S.Y.; Ko, H.J.; Vallance, B.A. The intestinal epithelium: Central coordinator of mucosal immunity. Trends in Immunology 2018, 39, 677-696.

42. Kucharzik, T.; Lugering, N.; Rautenberg, K.; Lugering, A.; Schmidt, M.A.; Stoll, R.; Domschke, W. Role of M cells in intestinal barrier function. Annals of the New York Academy of Sciences 2000, 915, 171-183.

43. Leal, J.; Smyth, H.D.C.; Ghosh, D. Physicochemical properties of mucus and their impact on transmucosal drug delivery. International Journal of Pharmaceutics 2017, 532, 555-572.

44. Jin, F.; Welch, R.; Glahn, R. Moving toward a more physiological model: Application of mucin to refine the in vitro digestion/Caco-2 cell culture system. Journal of Agriculture and Food Chemistry 2006, 54, 8962-8967.

45. Johansson, M.E.; Ambort, D.; Pelaseyed, T.; Schutte, A.; Gustafsson, J.K.; Ermund, A.; Subramani, D.B.; Holmen-Larsson, J.M.; Thomsson, K.A.; Bergstrom, J.H., et al. Composition and functional role of the mucus layers in the intestine. Cellular and Molecular Life Sciences 2011, 68, 3635-3641.

46. Conrad, M.E.; Umbreit, J.N.; Moore, E.G. A role for mucin in the absorption of inorganic iron and other metal cations. A study in rats. Gastroenterology 1991, 100, 129-136.

47. Powell, J.J.; Whitehead, M.W.; Ainley, C.C.; Kendall, M.D.; Nicholson, J.K.; Thompson, R.P.H. Dietary minerals in the gastrointestinal tract: Hydroxypolymerisation of aluminium is regulated by luminal mucins. Journal of Inorganic Biochemistry 1999, 75, 167-180.

48. Johansson, M.E.; Larsson, J.M.; Hansson, G.C. The two mucus layers of colon are organized by the MUC2 mucin, whereas the outer layer is a legislator of host-microbial interactions. Proceedings of the National Academy of Sciences of the United States of America 2010, 108 Suppl 1, 4659-4665.

49. Allen, A.; Flemström, G. Gastroduodenal mucus bicarbonate barrier: Protection against acid and pepsin. American Journal of Physiology-Cell Physiology 2005, 288, C1-C19.

50. Atuma, C.; Strugala, V.; Allen, A.; Holm, L. The adherent gastrointestinal mucus gel layer: Thickness and physical state in vivo. American Journal of Physiology- Gastrointestinal and Liver Physiology 2001, 280, G922-G929.

51. Johansson, M.E.; Phillipson, M.; Petersson, J.; Velcich, A.; Holm, L.; Hansson, G.C. The inner of the two MUC2 mucin-dependent mucus layers in colon is devoid of bacteria. Proceedings of the National Academy of Sciences of the United States of America 2008, 105, 15064-15069.

52. Cone, R.A. Barrier properties of mucus. Advanced Drug Delivery Reviews 2009, 61, 75-85.

53. Johansson, M.E. Fast renewal of the distal colonic mucus layers by the surface goblet cells as measured by in vivo labeling of mucin glycoproteins. PLoS One 2012, 7, e41009.

54. Schneider, H.; Pelaseyed, T.; Svensson, F.; Johansson, M.E.V. Study of mucin turnover in the small intestine by in vivo labeling. Scientific Reports 2018, 8, 5760-5771.

55. Tailford, L.; Crost, E.; Kavanaugh, D.; Juge, N. Mucin glycan foraging in the human gut microbiome. Frontiers in Genetics 2015, 6, 81-99.

56. Johansson, M.E.; Hansson, G.C. Mucus and the goblet cell. Digestive Diseases 2013, 31, 305-309.

References

130

57. Holmen Larsson, J.M.; Thomsson, K.A.; Rodriguez-Pineiro, A.M.; Karlsson, H.; Hansson, G.C. Studies of mucus in mouse stomach, small intestine, and colon. III. Gastrointestinal MUC5AC and MUC2 mucin O-glycan patterns reveal a regiospecific distribution. American Journal of Physiology-Gastrointestinal and Liver Physiology 2013, 305, G357-363.

58. Rama Bansil, B.S.T. Mucin structure, aggregation, physiological functions and biomedical applications. Current Opinion in Colloid & Interface Science 2006, 11, 164–170.

59. Bell, S.L.; Xu, G.; Forstner, J.F. Role of the cystine-knot motif at the c-terminus of rat mucin protein MUC2 in dimer formation and secretion. Biochemical Journal 2001, 357, 203-209.

60. Podolsky, D.K. Oligosaccharide structures of human colonic mucin. The Journal of Biological Chemistry 1985, 260, 8262-8271.

61. Bergstrom, K.S.; Xia, L. Mucin-type o-glycans and their roles in intestinal homeostasis. Glycobiology 2013, 23, 1026-1037.

62. Jin, C.; Kenny, D.T.; Skoog, E.C.; Padra, M.; Adamczyk, B.; Vitizeva, V.; Thorell, A.; Venkatakrishnan, V.; Lindén, S.K.; Karlsson, N.G. Structural diversity of human gastric mucin glycans. Molecular & Cellular Proteomics 2017, 16, 743-758.

63. Sheng, Y.H.; Hasnain, S.Z.; Florin, T.H.J.; McGuckin, M.A. Mucins in inflammatory bowel diseases and colorectal cancer. Journal of Gastroenterology and Hepatology 2012, 27, 28-38.

64. McGuckin, M.A.; Linden, S.K.; Sutton, P.; Florin, T.H. Mucin dynamics and enteric pathogens. Nature Reviews. Microbiology 2011, 9, 265-278.

65. Singh, P.K.; Hollingsworth, M.A. Cell surface-associated mucins in signal transduction. Trends in Cell Biology 2006, 16, 467-476.

66. Corfield, A.P. Mucins: A biologically relevant glycan barrier in mucosal protection. Biochimica et Biophysica Acta 2015, 1850, 236-252.

67. Robbe, C.; Capon, C.; Coddeville, B.; Michalski, J.-C. Structural diversity and specific distribution of O-glycans in normal human mucins along the intestinal tract. The Biochemical journal 2004, 384, 307-316.

68. Turner, J.R. Intestinal mucosal barrier function in health and disease. Nature Reviews. Immunology 2009, 9, 799-809.

69. Asker, N.; Axelsson, M.A.; Olofsson, S.O.; Hansson, G.C. Dimerization of the humanMUC2 mucin in the endoplasmic reticulum is followed by a N-glycosylation-dependent transfer of the mono- and dimers to the golgi apparatus. The Journal of Biological Chemistry 1998, 273, 18857-18863.

70. Theodoropoulos, G.; Carraway, K.L. Molecular signaling in the regulation of mucins. Journal of Cellular Biochemistry 2007, 102, 1103-1116.

71. Kim, J.J.; Khan, W.I. Goblet cells and mucins: Role in innate defense in enteric infections. Pathogens 2013, 2, 55-70.

72. Maret, W. The metals in the biological periodic system of the elements: Concepts and conjectures. International Journal of Molecular Sciences 2016, 17.

73. Lim, K.H.; Riddell, L.J.; Nowson, C.A.; Booth, A.O.; Szymlek-Gay, E.A. Iron and zinc nutrition in the economically-developed world: A review. Nutrients 2013, 5, 3184-3211.

74. Hara, T.; Takeda, T.A.; Takagishi, T.; Fukue, K.; Kambe, T.; Fukada, T. Physiological roles of zinc transporters: Molecular and genetic importance in zinc homeostasis. The Journal of Physiological Sciences 2017, 67, 283-301.

References

131

75. Matthews, J.M.; Bhati, M.; Lehtomaki, E.; Mansfield, R.E.; Cubeddu, L.; Mackay, J.P. It takes two to tango: The structure and function of LIM, RING, PHD and MYND domains. Current Pharmaceutical Design 2009, 15, 3681-3696.

76. Maret, W.; Sandstead, H.H. Zinc requirements and the risks and benefits of zinc supplementation. Journal of Trace Elements in Medicine and Biology 2006, 20, 3-18.

77. Fukada, T.; Yamasaki, S.; Nishida, K.; Murakami, M.; Hirano, T. Zinc homeostasis and signaling in health and diseases: Zinc signaling. Journal of Biological Inorganic Chemistry 2011, 16, 1123-1134.

78. Frederickson, C.J.; Koh, J.Y.; Bush, A.I. The neurobiology of zinc in health and disease. Nature Reviews. Neuroscience 2005, 6, 449-462.

79. Sandstead, H.H.; Frederickson, C.J.; Penland, J.G. History of zinc as related to brain function. The Journal of Nutrition 2000, 130, 496S-502S.

80. Jackson, M.J. Physiology of zinc: General aspects. In Zinc in Human Biology, Mills, C.F., Ed. Springer London: London, 1989; pp 1-14.

81. Driessen, C.; Hirv, K.; Kirchner, H.; Rink, L. Zinc regulates cytokine induction by superantigens and lipopolysaccharide. Immunology 1995, 84, 272-277.

82. Wessells, K.R.; Jorgensen, J.M.; Hess, S.Y.; Woodhouse, L.R.; Peerson, J.M.; Brown, K.H. Plasma zinc concentration responds rapidly to the initiation and discontinuation of short-term zinc supplementation in healthy men. The Journal of Nutrition 2010, 140, 2128-2133.

83. Hess, S.Y.; Peerson, J.M.; King, J.C.; Brown, K.H. Use of serum zinc concentration as an indicator of population zinc status. Food and Nutrition Bulletin 2007, 28, S403-429.

84. Scott, B.J.; Bradwell, A.R. Identification of the serum binding proteins for iron, zinc, cadmium, nickel, and calcium. Clinical Chemistry 1983, 29, 629-633.

85. Brown, K.H.; Wuehler, S.E.; Peerson, J.M. The importance of zinc in human nutrition and estimation of the global prevalence of zinc deficiency. Food and Nutrition Bulletin 2001, 22, 113-125.

86. King, J.C.; Shames, D.M.; Lowe, N.M.; Woodhouse, L.R.; Sutherland, B.; Abrams, S.A.; Turnlund, J.R.; Jackson, M.J. Effect of acute zinc depletion on zinc homeostasis and plasma zinc kinetics in men. The American Journal of Clinical Nutrition 2001, 74, 116-124.

87. Lowe, N.M.; Woodhouse, L.R.; Sutherland, B.; Shames, D.M.; Burri, B.J.; Abrams, S.A.; Turnlund, J.R.; Jackson, M.J.; King, J.C. Kinetic parameters and plasma zinc concentration correlate well with net loss and gain of zinc from men. The Journal of Nutrition 2004, 134, 2178-2181.

88. Hunt, J.R.; Matthys, L.A.; Johnson, L.K. Zinc absorption, mineral balance, and blood lipids in women consuming controlled lactoovovegetarian and omnivorous diets for 8 wk. The American Journal of Clinical Nutrition 1998, 67, 421-430.

89. Taylor, C.M.; Bacon, J.R.; Aggett, P.J.; Bremner, I. Homeostatic regulation of zinc absorption and endogenous losses in zinc-deprived men. The American Journal of Clinical Nutrition 1991, 53, 755-763.

90. Johnson, P.E.; Hunt, C.D.; Milne, D.B.; Mullen, L.K. Homeostatic control of zinc metabolism in men: Zinc excretion and balance in men fed diets low in zinc. The American Journal of Clinical Nutrition 1993, 57, 557-565.

91. Jackson, M.J.; Jones, D.A.; Edwards, R.H.; Swainbank, I.G.; Coleman, M.L. Zinc homeostasis in man: Studies using a new stable isotope-dilution technique. The British Journal of Nutrition 1984, 51, 199-208.

References

132

92. Hoadley, J.E.; Leinart, A.S.; Cousins, R.J. Kinetic analysis of zinc uptake and serosal transfer by vascularly perfused rat intestine. The American Journal of Physiology 1987, 252, G825-831.

93. Ziegler, E.E.; Serfass, R.E.; Nelson, S.E.; Figueroa-Colon, R.; Edwards, B.B.; Houk, R.S.; Thompson, J.J. Effect of low zinc intake on absorption and excretion of zinc by infants studied with 70Zn as extrinsic tag. The Journal of Nutrition 1989, 119, 1647-1653.

94. Baer, M.T.; King, J.C. Tissue zinc levels and zinc excretion during experimental zinc depletion in young men. The American Journal of Clinical Nutrition 1984, 39, 556-570.

95. Jackson, M.J.; Jones, D.A.; Edwards, R.H. Tissue zinc levels as an index of body zinc status. Clinical Physiology 1982, 2, 333-343.

96. Wellenreuther, G.; Cianci, M.; Tucoulou, R.; Meyer-Klaucke, W.; Haase, H. The ligand environment of zinc stored in vesicles. Biochemical and Biophysical Research Communications 2009, 380, 198-203.

97. Maret, W. Analyzing free zinc(II) ion concentrations in cell biology with fluorescent chelating molecules. Metallomics 2015, 7, 202-211.

98. Bozym, R.A.; Chimienti, F.; Giblin, L.J.; Gross, G.W.; Korichneva, I.; Li, Y.; Libert, S.; Maret, W.; Parviz, M.; Frederickson, C.J., et al. Free zinc ions outside a narrow concentration range are toxic to a variety of cells in vitro. Experimental Biology and Medicine 2010, 235, 741-750.

99. Maret, W. Molecular aspects of human cellular zinc homeostasis: Redox control of zinc potentials and zinc signals. BioMetals 2009, 22, 149 - 157.

100. Chabosseau, P.; Woodier, J.; Cheung, R.; Rutter, G.A. Sensors for measuring subcellular zinc pools. Metallomics 2018.

101. Colvin, R.A.; Holmes, W.R.; Fontaine, C.P.; Maret, W. Cytosolic zinc buffering and muffling: Their role in intracellular zinc homeostasis. Metallomics 2010, 2, 306-317.

102. Huang, L.; Tepaamorndech, S. The SLC30 family of zinc transporters - a review of current understanding of their biological and pathophysiological roles. Molecular Aspects of Medicine 2013, 34, 548-560.

103. Jeong, J.; Eide, D.J. The SLC39 family of zinc transporters. Molecular Aspects of Medicine 2013, 34, 612-619.

104. Palmiter, R.D. Protection against zinc toxicity by metallothionein and zinc transporter 1. Proceedings of the National Academy of Sciences of the United States of America 2004, 101, 4918-4923.

105. Maret, W. Redox biochemistry of mammalian metallothioneins. Journal of Biological Inorganic Chemistry 2011, 16, 1079-1086.

106. Kimura, T.; Kambe, T. The functions of metallothionein and ZIP and ZnT transporters: An overview and perspective. International Journal of Molecular Sciences 2016, 17, 336.

107. Krȩżel, A.; Maret, W. Dual nanomolar and picomolar Zn(II) binding properties of metallothionein. Journal of the American Chemical Society 2007, 129, 10911-10921.

108. Bell, S.G.; Vallee, B.L. The metallothionein/thionein system: An oxidoreductive metabolic zinc link. Chembiochem 2008, 10, 55-62.

109. Coyle, P.; Philcox, J.C.; Carey, L.C.; Rofe, A.M. Metallothionein: The multipurpose protein. Cellular and Molecular Life Sciences 2002, 59, 627-647.

110. Maret, W. The function of zinc metallothionein: A link between cellular zinc and redox state. The Journal of Nutrition 2000, 130, 1455S-1458S.

References

133

111. Krȩżel, A.; Hao, Q.; Maret, W. The zinc/thiolate redox biochemistry of metallothionein and the control of zinc ion fluctuations in cell signaling. Archives of Biochemistry and Biophysics 2007, 463, 188-200.

112. Maret, W.; Li, Y. Coordination dynamics of zinc in proteins. Chemical Reviews 2009, 109, 4682-4707.

113. Maret, W. Zinc coordination environments in proteins as redox sensors and signal transducers. Antioxidants & Redox Signaling 2006, 8, 1419-1441.

114. Heuchel, R.; Radtke, F.; Georgiev, O.; Stark, G.; Aguet, M.; Schaffner, W. The transcription factor MTF-1 is essential for basal and heavy metal-induced metallothionein gene expression. The EMBO Journal 1994, 13, 2870-2875.

115. Tapiero, H.; Tew, K.D. Trace elements in human physiology and pathology: Zinc and metallothioneins. Biomedicine & Pharmacotherapy 2003, 57, 399-411.

116. Kelly, E.J.; Quaife, C.J.; Froelick, G.J.; Palmiter, R.D. Metallothionein I and II protect against zinc deficiency and zinc toxicity in mice. The Journal of Nutrition 1996, 126, 1782-1790.

117. Davis, S.R.; McMahon, R.J.; Cousins, R.J. Metallothionein knockout and transgenic mice exhibit altered intestinal processing of zinc with uniform zinc-dependent zinc transporter-1 expression. The Journal of Nutrition 1998, 128, 825-831.

118. Colvin, R.A.; Bush, A.I.; Volitakis, I.; Fontaine, C.P.; Thomas, D.; Kikuchi, K.; Holmes, W.R. Insights into Zn2+ homeostasis in neurons from experimental and modeling studies. American Journal of Physiology-Cell Physiology 2008, 294, C726-742.

119. Sandström, B. Dose dependence of zinc and manganese absorption in man. Proceedings of the Nutrition Society 1992, 51, 211-218.

120. Davies, N.T. Studies on the absorption of zinc by rat intestine. The British Journal of Nutrition 1980, 43, 189-203.

121. Seal, C.J.; Heaton, F.W. Chemical factors affecting the intestinal absorption of zinc in vitro and in vivo. The British Journal of Nutrition 1983, 50, 317-324.

122. Seal, C.J.; Mathers, J.C. Intestinal zinc transfer by everted gut sacs from rats given diets containing different amounts and types of dietary fibre. The British Journal of Nutrition 1989, 62, 151-163.

123. Antonson, D.L.; Barak, A.J.; Vanderhoof, J.A. Determination of the site of zinc absorption in rat small intestine. The Journal of Nutrition 1979, 109, 142-147.

124. Wang, S.C.; Chen, Y.S.; Chen, S.M.; Young, T.K. Possible site of decreased intestinal zinc absorption in chronic uremic rats. Nephron 2001, 89, 208-214.

125. Yasuno, T.; Okamoto, H.; Nagai, M.; Kimura, S.; Yamamoto, T.; Nagano, K.; Furubayashi, T.; Yoshikawa, Y.; Yasui, H.; Katsumi, H., et al. In vitro study on the transport of zinc across intestinal epithelial cells using Caco-2 monolayers and isolated rat intestinal membranes. Biological & Pharmaceutical Bulletin 2012, 35, 588-593.

126. Reyes, J.G. Zinc transport in mammalian cells. The American Journal of Physiology 1996, 270, C401-410.

127. Raffaniello, R.D.; Shih-Yu Lee; Teichberc, S.; Wapnir, R.A. Distinct mechanisms of zinc uptake at the apical and basolateral membranes of Caco-2 cells. Journal of Cellular Physiology 1992, 152, 356-361.

128. Finley, J.W.; Briske-Anderson, M.; Reeves, P.G.; Johnson, L.K. Zinc uptake and transcellular movement by Caco-2 cells: Studies with media containing fetal bovine serum. The Journal of Nutritional Biochemistry 1995, 6, 137-144.

References

134

129. Rossi, A.; PoveriniI, R.; Di Lullo, G.; Modesti, A.; Modica, A.; Scarinos, M.L. Heavy metal toxicity following apical and basolateral exposure in the human intestinal cell line Caco-2. Toxicology in Vitro 1996, 10, 27-36.

130. Lönnerdal, B. Intestinal absorption of zinc. In Zinc in Human Biology, Mills, C.F., Ed. Springer London: London, 1989; pp 33-55.

131. Steel, L.; Cousins, R.J. Kinetics of zinc absorption by luminally and vascularly perfused rat intestine. The American Journal of Physiology 1985, 248, G46-53.

132. Cragg, R.A.; Phillips, S.R.; Piper, J.M.; Varma, J.S.; Campbell, F.C.; Mathers, J.C.; Ford, D. Homeostatic regulation of zinc transporters in the human small intestine by dietary zinc supplementation. Gut 2005, 54, 469-478.

133. Jackson, K.A.; Helston, R.M.; McKay, J.A.; O'Neill, E.D.; Mathers, J.C.; Ford, D. Splice variants of the human zinc transporter ZnT5 (SLC30A5) are differentially localized and regulated by zinc through transcription and mrna stability. The Journal of Biological Chemistry 2007, 282, 10423-10431.

134. Valentine, R.A.; Jackson, K.A.; Christie, G.R.; Mathers, J.C.; Taylor, P.M.; Ford, D. Znt5 variant b is a bidirectional zinc transporter and mediates zinc uptake in human intestinal Caco-2 cells. The Journal of Biological Chemistry 2007, 282, 14389-14393.

135. Menard, M.P.; Cousins, R.J. Zinc transport by brush border membrane vesicles from rat intestine. The Journal of Nutrition 1983, 113, 1434-1442.

136. Ghishan, F.K.; Sobo, G. Intestinal maturation: In vivo zinc transport. Pediatric Research 1983, 17, 148-151.

137. Cragg, R.A.; Christie, G.R.; Phillips, S.R.; Russi, R.M.; Kury, S.; Mathers, J.C.; Taylor, P.M.; Ford, D. A novel zinc-regulated human zinc transporter, hZTL1, is localized to the enterocyte apical membrane. The Journal of Biological Chemistry 2002, 277, 22789-22797.

138. International Zinc Nutrition Consultative Group; Brown, K.H.; Rivera, J.A.; Bhutta, Z.; Gibson, R.S.; King, J.C.; Lönnerdal, B.; Ruel, M.T.; Sandtrom, B.; Wasantwisut, E., et al. International zinc nutrition consultative group (IZiNCG) technical document #1. Assessment of the risk of zinc deficiency in populations and options for its control. Food and Nutrition Bulletin 2004, 25, S99-203.

139. Sandström, B. Dietary pattern and zinc supply. In Zinc in Human Biology, Mills, C.F., Ed. Springer, London: London, 1989; pp 351-363.

140. Scholmerich, J.; Freudemann, A.; Kottgen, E.; Wietholtz, H.; Steiert, B.; Lohle, E.; Haussinger, D.; Gerok, W. Bioavailability of zinc from zinc-histidine complexes. I. Comparison with zinc sulfate in healthy men. The American Journal of Clinical Nutrition 1987, 45, 1480-1486.

141. Sandström, B.; Arvidsson, B.; Cederblad, A.; Björn-Rasmussen, E. Zinc absorption from composite meals. I. The significance of whest extraction rate, zinc, calcium, and protein content in meals based on bread. The American Journal of Clinical Nutrition 1980, 33, 739-745.

142. Wada, L.; Turnlund, J.R.; King, J.C. Zinc utilization in young men fed adequate and low zinc intakes. The Journal of Nutrition 1985, 115, 1345-1354.

143. Yasuno, T.; Okamoto, H.; Nagai, M.; Kimura, S.; Yamamoto, T.; Nagano, K.; Furubayashi, T.; Yoshikawa, Y.; Yasui, H.; Katsumi, H., et al. The disposition and intestinal absorption of zinc in rats. European Journal of Pharmaceutical Sciences 2011, 44, 410-415.

References

135

144. Istfan, N.W.; Janghorbani, M.; Young, V.R. Absorption of stable 70Zn in healthy young men in relation to zinc intake. The American Journal of Clinical Nutrition 1983, 38, 187-194.

145. Kury, S.; Dreno, B.; Bezieau, S.; Giraudet, S.; Kharfi, M.; Kamoun, R.; Moisan, J.P. Identification of SLC39A4, a gene involved in acrodermatitis enteropathica. Nature genetics 2002, 31, 239-240.

146. Wang, K.; Zhou, B.; Kuo, Y.M.; Zemansky, J.; Gitschier, J. A novel member of a zinc transporter family is defective in acrodermatitis enteropathica. American Journal of Human Genetics 2002, 71, 66-73.

147. McMahon, R.J.; Cousins, R.J. Regulation of the zinc transporter Znt-1 by dietary zinc. Proceedings of the National Academy of Sciences of the United States of America 1998, 95, 4841-4846.

148. Wang, F.; Kim, B.E.; Petris, M.J.; Eide, D.J. The mammalian zip5 protein is a zinc transporter that localizes to the basolateral surface of polarized cells. The Journal of Biological Chemistry 2004, 279, 51433-51441.

149. Guthrie, G.J.; Aydemir, T.B.; Troche, C.; Martin, A.B.; Chang, S.M.; Cousins, R.J. Influence of ZIP14 (SLC39A14) on intestinal zinc processing and barrier function. American Journal of Physiology-Gastrointestinal and Liver Physiology 2015, 308, G171-178.

150. Gunshin, H.; Mackenzie, B.; Berger, U.V.; Gunshin, Y.; Romero, M.F.; Boron, W.F.; Nussberger, S.; Gollan, J.L.; Hediger, M.A. Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature 1997, 388, 482-488.

151. Tandy, S.; Williams, M.; Leggett, A.; Lopez-Jimenez, M.; Dedes, M.; Ramesh, B.; Srai, S.K.; Sharp, P. Nramp2 expression is associated with ph-dependent iron uptake across the apical membrane of human intestinal Caco-2 cells. The Journal of Biological Chemistry 2000, 275, 1023-1029.

152. Kordas, K.; Stoltzfus, R.J. New evidence of iron and zinc interplay at the enterocyte and neural tissues. The Journal of Nutrition 2004, 134, 1295-1298.

153. Ferguson, C.J.; Wareing, M.; Ward, D.T.; Green, R.; Smith, C.P.; Riccardi, D. Cellular localization of divalent metal transporter DMT-1 in rat kidney. American Journal of Physiology-Renal Physiology 2001, 280, F803-814.

154. Garrick, M.D.; Dolan, K.G.; Horbinski, C.; Ghio, A.J.; Higgins, D.; Porubcin, M.; Moore, E.G.; Hainsworth, L.N.; Umbreit, J.N.; Conrad, M.E., et al. DMT1: A mammalian transporter for multiple metals. BioMetals 2003, 16, 41-54.

155. Yamaji, S.; Tennant, J.; Tandy, S.; Williams, M.; Singh Srai, S.K.; Sharp, P. Zinc regulates the function and expression of the iron transporters DMT1 and IREG1 in human intestinal Caco-2 cells. FEBS Letters 2001, 507, 137-141.

156. Fukada, T.; Kambe, T. Molecular and genetic features of zinc transporters in physiology and pathogenesis. Metallomics 2011, 3, 662-674.

157. Sauer, A.K.; Pfaender, S.; Hagmeyer, S.; Tarana, L.; Mattes, A.K.; Briel, F.; Kury, S.; Boeckers, T.M.; Grabrucker, A.M. Characterization of zinc amino acid complexes for zinc delivery in vitro using Caco-2 cells and enterocytes from hiPSC. BioMetals 2017, 30, 643-661.

158. Hennigar, S.R.; McClung, J.P. Zinc transport in the mammalian intestine. Comprehensive Physiology 2018, 9, 59-74.

159. Richards, M.P.; Cousins, R.J. Mammalian zinc homeostasis: Requirement for RNA and metallothionein synthesis. Biochemical and Biophysical Research Communications 1975, 64, 1215-1223.

References

136

160. Menard, M.P.; McCormick, C.C.; Cousins, R.J. Regulation of intestinal metallothionein biosynthesis in rats by dietary zinc. The Journal of Nutrition 1981, 111, 1353-1361.

161. Reeves, P.G. Adaptation responses in rats to long-term feeding of high-zinc diets: Emphasis on intestinal metallothionein. The Journal of Nutritional Biochemistry 1995, 6, 48-54.

162. Liuzzi, J.P.; Blanchard, R.K.; Cousins, R.J. Differential regulation of zinc transporter 1, 2, and 4 mRNA expression by dietary zinc in rats. The Journal of Nutrition 2001, 131, 46-52.

163. Liuzzi, J.P.; Bobo, J.A.; Lichten, L.A.; Samuelson, D.A.; Cousins, R.J. Responsive transporter genes within the murine intestinal-pancreatic axis form a basis of zinc homeostasis. Proceedings of the National Academy of Sciences of the United States of America 2004, 101, 14355-14360.

164. Mao, X.; Kim, B.E.; Wang, F.; Eide, D.J.; Petris, M.J. A histidine-rich cluster mediates the ubiquitination and degradation of the human zinc transporter, hZIP4, and protects against zinc cytotoxicity. The Journal of Biological Chemistry 2007, 282, 6992-7000.

165. Huang, Z.L.; Dufner-Beattie, J.; Andrews, G.K. Expression and regulation of SLC39A family zinc transporters in the developing mouse intestine. Developmental Biology 2006, 295, 571-579.

166. Kim, B.E.; Wang, F.; Dufner-Beattie, J.; Andrews, G.K.; Eide, D.J.; Petris, M.J. Zn2+-stimulated endocytosis of the mZIP4 zinc transporter regulates its location at the plasma membrane. The Journal of Biological Chemistry 2004, 279, 4523-4530.

167. Hashimoto, A.; Nakagawa, M.; Tsujimura, N.; Miyazaki, S.; Kizu, K.; Goto, T.; Komatsu, Y.; Matsunaga, A.; Shirakawa, H.; Narita, H., et al. Properties of ZIP4 accumulation during zinc deficiency and its usefulness to evaluate zinc status: A study of the effects of zinc deficiency during lactation. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 2016, 310, R459-468.

168. Jou, M.Y.; Hall, A.G.; Philipps, A.F.; Kelleher, S.L.; Lönnerdal, B. Tissue-specific alterations in zinc transporter expression in intestine and liver reflect a threshold for homeostatic compensation during dietary zinc deficiency in weanling rats. The Journal of Nutrition 2009, 139, 835-841.

169. Davis, S.R.; Cousins, R.J. Metallothionein expression in animals: A physiological perspective on function. The Journal of Nutrition 2000, 130, 1085-1088.

170. Cousins, R.J.; Liuzzi, J.P.; Lichten, L.A. Mammalian zinc transport, trafficking, and signals. Journal of Biological Chemistry 2006, 281, 24085-24089.

171. Andrews, G.K. Regulation of metallothionein gene expression by oxidative stress and metal ions. Biochemical Pharmacology 1999, 59, 95-104.

172. Smith, K.T.; Cousins, R.J. Quantitative aspects of zinc absorption by isolated, vascularly perfused rat intestine. The Journal of Nutrition 1980, 110, 316-323.

173. Hinskens, B.; Philcox, J.C.; Coyle, P.; Rofe, A.M. Increased zinc absorption but not secretion in the small intestine of metallothionein-null mice. Biological Trace Element Research 2000, 78, 231-240.

174. Hoadley, J.E.; Leinart, A.S.; Cousins, R.J. Relationship of 65Zn absorption kinetics to intestinal metallothionein in rats: Effects of zinc depletion and fasting. The Journal of Nutrition 1988, 118, 497-502.

175. Coppen, D.E.; Davies, N.T. Studies on the effects of dietary zinc dose on 65Zn absorption in vivo and on the effects of Zn status on 65Zn absorption and body loss in young rats. The British Journal of Nutrition 1987, 57, 35-44.

References

137

176. Moltedo, O.; Verde, C.; Capasso, A.; Parisi, E.; Remondelli, P.; Bonatti, S.; Alvarez-Hernandez, X.; Glass, J.; Alvino, C.G.; Leone, A. Zinc transport and metallothionein secretion in the intestinal human cell line Caco-2. The Journal of Biological Chemistry 2000, 275, 31819-31825.

177. Hempe, J.M.; Cousins, R.J. Cysteine-rich intestinal protein and intestinal metallothionein: An inverse relationship as a conceptual model for zinc absorption in rats. The Journal of Nutrition 1992, 122, 89-95.

178. Chai, F.; Truong-Tran, A.Q.; Ho, L.H.; Zalewski, P.D. Regulation of caspase activation and apoptosis by cellular zinc fluxes and zinc deprivation: A review. Immunology and Cell Biology 1999, 77, 272-278.

179. O'Dell, B.L. Cysteine-rich intestinal protein (CRIP): A new intestinal zinc transport protein. Nutrition Reviews 1992, 50, 232-233.

180. Lanningham-Foster, L.; Green, C.L.; Langkamp-Henken, B.; Davis, B.A.; Nguyen, K.T.; Bender, B.S.; Cousins, R.J. Overexpression of crip in transgenic mice alters cytokine patterns and the immune response. American Journal of Physiology-Endocrinology and Metabolism 2002, 282, E1197-E1203.

181. Nishito, Y.; Kambe, T. Absorption mechanisms of iron, copper, and zinc: An overview. Journal of Nutritional Science and Vitaminology 2018, 64, 1-7.

182. Kasana, S.; Din, J.; Maret, W. Genetic causes and gene-nutrient interactions in mammalian zinc deficiencies: Acrodermatitis enteropathica and transient neonatal zinc deficiency as examples. Journal of Trace Elements in Medicine and Biology 2015, 29, 47-62.

183. Andrews, G.K. Regulation and function of ZIP4, the acrodermatitis enteropathica gene. Biochemical Society Transactions 2008, 36, 1242-1246.

184. Dufner-Beattie, J.; Wang, F.; Kuo, Y.-M.; Gitschier, J.; Eide, D.; Andrews, G.K. The acrodermatitis enteropathica gene ZIP4 encodes a tissue-specific, zinc-regulated zinc transporter in mice. Journal of Biological Chemistry 2003, 278, 33474-33481.

185. Weaver, B.P.; Dufner-Beattie, J.; Kambe, T.; Andrews, G.K. Novel zinc-responsive post-transcriptional mechanisms reciprocally regulate expression of the mouse SLC39A4 and SLC39A5 zinc transporters (ZIP4 and ZIP5). Biological Chemistry 2007, 388, 1301-1312.

186. Dufner-Beattie, J.; Kuo, Y.M.; Gitschier, J.; Andrews, G.K. The adaptive response to dietary zinc in mice involves the differential cellular localization and zinc regulation of the zinc transporters ZIP4 and ZIP5. The Journal of Biological Chemistry 2004, 279, 49082-49090.

187. Lichten, L.A.; Cousins, R.J. Mammalian zinc transporters: Nutritional and physiologic regulation. Annual Review of Nutrition 2009, 29, 153-176.

188. Andrews, G.K.; Wang, H.; Dey, S.K.; Palmiter, R.D. Mouse zinc transporter 1 gene provides an essential function during early embryonic development. Genesis 2004, 40, 74-81.

189. Inoue, K.; Matsuda, K.; Itoh, M.; Kawaguchi, H.; Tomoike, H.; Aoyagi, T.; Nagai, R.; Hori, M.; Nakamura, Y.; Tanaka, T. Osteopenia and male-specific sudden cardiac death in mice lacking a zinc transporter gene, ZnT5. Human Molecular Genetics 2002, 11, 1775-1784.

190. Hennigar, S.R.; McClung, J.P. Hepcidin attenuates zinc efflux in Caco-2 cells. The FASEB Journal 2016, 30, 292.293.

191. Gibson, R.S.; King, J.C.; Lowe, N. A review of dietary zinc recommendations. Food and Nutrition Bulletin 2016, 37, 443-460.

References

138

192. EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific opinion on dietary reference values for zinc. EFSA Journal 2014 2014, 12.

193. King, J.C.; Shames, D.M.; Woodhouse, L.R. Zinc homeostasis in humans. The Journal of Nutrition 2000, 130, 1360S-1366S.

194. Miller, L.V.; Krebs, N.F.; Hambidge, K.M. A mathematical model of zinc absorption in humans as a function of dietary zinc and phytate. The Journal of Nutrition 2007, 137, 135-141.

195. Deutsche Gesellschaft für Ernährung; Österreichische Gesellschaft für Ernährung; Schweizerische Gesellschaft für Ernährung. Referenzwerte für die Nährstoffzufuhr. 2 ed.; Neuer Umschau Buchverlag: Bonn, Germany, 2015; Vol. 1.

196. Carter, J.P.; Grivetti, L.E.; Davis, J.T.; Nasiff, S.; Mansour, A.; Mousa, W.A.; Atta, A.E.; Patwardhan, V.N.; Abdel Moneim, M.; Abdou, I.A., et al. Growth and sexual development of adolescent egyptian village boys. Effects of zinc, iron, and placebo supplementation. The American Journal of Clinical Nutrition 1969, 22, 59-78.

197. Prasad, A.S.; Miale, A., Jr.; Farid, Z.; Sandstead, H.H.; Schulert, A.R. Zinc metabolism in patients with the syndrome of iron deficiency anemia, hepatosplenomegaly, dwarfism, and hypognadism. The Journal of Laboratory and Clinical Medicine 1963, 61, 537-549.

198. Plum, L.M.; Rink, L.; Haase, H. The essential toxin: Impact of zinc on human health. International Journal of Environmental Research and Public Health 2010, 7, 1342-1365.

199. Walker, C.L.; Black, R.E. Zinc for the treatment of diarrhoea: Effect on diarrhoea morbidity, mortality and incidence of future episodes. International Journal of Epidemiology 2010, 39 Suppl 1, i63-i69.

200. Walker, C.F.; Black, R.E. Zinc and the risk for infectious disease. Annual Review of Nutrition 2004, 24, 255-275.

201. King, J.C.; Brown, K.H.; Gibson, R.S.; Krebs, N.F.; Lowe, N.M.; Siekmann, J.H.; Raiten, D.J. Biomarkers of nutrition for development (BOND)-zinc review. The Journal of Nutrition 2016.

202. Oberleas, D.; Harland, B.F. Phytate content of foods: Effect on dietary zinc bioavailability. Journal of the American Dietetic Association 1981, 79, 433-436.

203. Davies, N.T.; Olpin, S.E. Studies on the phytate: Zinc molar contents in diets as a determinant of Zn availability to young rats. The British Journal of Nutrition 1979, 41, 590-603.

204. Foster, M.; Samman, S. Vegetarian diets across the lifecycle: Impact on zinc intake and status. Advances in Food and Nutrition Research 2015, 74, 93-131.

205. Kristensen, M.B.; Hels, O.; Morberg, C.M.; Marving, J.; Bugel, S.; Tetens, I. Total zinc absorption in young women, but not fractional zinc absorption, differs between vegetarian and meat-based diets with equal phytic acid content. The British Journal of Nutrition 2006, 95, 963-967.

206. Broadley, D.B.K.; Edward, J.M.J.; Ander, E.L.; Michael, J.W.; Scott, D.Y.; Sue, W.; Martin, R. Dietary calcium and zinc deficiency risks are decreasing but remain prevalent. Scientific Reports 2015, 5.

207. Turnlund, J.R.; Durkin, N.; Costa, F.; Margen, S. Stable isotope studies of zinc absorption and retention in young and elderly men. The Journal of Nutrition 1986, 116, 1239-1247.

208. McClain, C.J. Zinc metabolism in malabsorption syndromes. Journal of the American College of Nutrition 1985, 4, 49-64.

References

139

209. Solomons, N.W.; Rosenberg, I.H.; Sandstead, H.H.; Vo-Khactu, K.P. Zinc deficiency in Crohn's disease. Digestion 1977, 16, 87-95.

210. Valberg, L.S.; Flanagan, P.R.; Kertesz, A.; Bondy, D.C. Zinc absorption in inflammatory bowel disease. Digestive Diseases and Sciences 1986, 31, 724-731.

211. Prasad, A.S.; Beck, F.W.; Bao, B.; Fitzgerald, J.T.; Snell, D.C.; Steinberg, J.D.; Cardozo, L.J. Zinc supplementation decreases incidence of infections in the elderly: Effect of zinc on generation of cytokines and oxidative stress. The American Journal of Clinical Nutrition 2007, 85, 837-844.

212. Haase, H.; Overbeck, S.; Rink, L. Zinc supplementation for the treatment or prevention of disease: Current status and future perspectives. Experimental Gerontology 2008, 43, 394-408.

213. Liberatoa, S.C.; Singha, G.; Mulhollanda, K. Zinc supplementation in young children: A review of the literature focusing on diarrhoea prevention and treatment. Clinical Nutrition 2015, 34, 181–188.

214. Scientific Panel on Food Additives and Nutrient Sources added to food (ANS). Magnesium aspartate, potassium aspartate, magnesium potassium aspartate, calcium aspartate, zinc aspartate, and copper aspartate as sources for magnesium, potassium, calcium, zinc, and copper added for nutritional purposes to food supplements. EFSA Journal 2008, 6, 883.

215. Holst, B.; Williamson, G. Nutrients and phytochemicals: From bioavailability to bioefficacy beyond antioxidants. Current Opinion in Biotechnology 2008, 19, 73-82.

216. Carbonell-Capella, J.M.; Buniowska, M.; Barba, F.J.; Esteve, M.J.; Frígola, A. Analytical methods for determining bioavailability and bioaccessibility of bioactive compounds from fruits and vegetables: A review. Comprehensive Reviews in Food Science and Food Safety 2014, 13, 155-171.

217. Hotz, C.; DeHaene, J.; Woodhouse, L.R.; Villalpando, S.; Rivera, J.A.; King, J.C. Zinc absorption from zinc oxide, zinc sulfate, zinc oxide + EDTA, or sodium-zinc EDTA does not differ when added as fortificants to maize tortillas. The Journal of Nutrition 2005, 135, 1102-1105.

218. Lopez de Romana, D.; Lönnerdal, B.; Brown, K.H. Absorption of zinc from wheat products fortified with iron and either zinc sulfate or zinc oxide. The American Journal of Clinical Nutrition 2003, 78, 279-283.

219. Wegmuller, R.; Tay, F.; Zeder, C.; Brnic, M.; Hurrell, R.F. Zinc absorption by young adults from supplemental zinc citrate is comparable with that from zinc gluconate and higher than from zinc oxide. The Journal of Nutrition 2014, 144, 132-136.

220. Zhang, S.Q.; Yu, X.F.; Zhang, H.B.; Peng, N.; Chen, Z.X.; Cheng, Q.; Zhang, X.L.; Cheng, S.H.; Zhang, Y. Comparison of the oral absorption, distribution, excretion, and bioavailability of zinc sulfate, zinc gluconate, and zinc-enriched yeast in rats. Molecular Nutrition & Food Research 2018, 62, e1700981.

221. Lönnerdal, B. Phytic acid–trace element (Zn, Cu, Mn) interactions. International Journal of Food Science & Technology 2002, 37, 749-758.

222. Sandström, B.; Sandberg, A.S. Inhibitory effects of isolated inositol phosphates on zinc absorption in humans. Journal of Trace Elements and Electrolytes in Health and Disease 1992, 6, 99-103.

223. Lönnerdal, B.; Sandberg, A.-S.; Sandström, B.; Kunz, C. Inhibitory effects of phytic acid and other inositol phosphates on zinc and calcium absorption in suckling rats. The Journal of Nutrition 1989, 119, 211-214.

References

140

224. Sandberg, A.; Andlid, T. Phytogenic and microbial phytases in human nutrition. International Journal of Food Science and Technology 2002, 37, 823-833.

225. Nävert, B.; Sandström, B.; Cederblad, A. Reduction of the phytate content of bran by leavening in bread and its effect on zinc absorption in man. The British Journal of Nutrition 1985, 53, 47-53.

226. Kemme, P.A.; Schlemmer, U.; Mroz, Z.; Jongbloed, A.W. Monitoring the stepwise phytate degradation in the upper gastrointestinal tract of pigs. Journal of the Science of Food and Agriculture 2006, 86, 612-622.

227. Iqbal, T.H.; Lewis, K.O.; Cooper, B.T. Phytase activity in the human and rat small intestine. Gut 1994, 35, 1233-1236.

228. Turk, M.; Sandberg, A.S. Phytate degradation during breadmaking: Effect of phytase addition. Journal of Cereal Science 1992, 15, 281-294.

229. Vasca, E.; Materazzi, S.; Caruso, T.; Milano, O.; Fontanella, C.; Manfredi, C. Complex formation between phytic acid and divalent metal ions: A solution equilibria and solid state investigation. Analytical and Bioanalytical Chemistry 2002, 374, 173-178.

230. Tang, N.; Skibsted, L.H. Zinc bioavailability from phytate-rich foods and zinc supplements. Modeling the effects of food components with oxygen, nitrogen, and sulfur donor ligands. Journal of Agricultural and Food Chemistry 2017, 65, 8727-8743.

231. Lönnerdal, B.; Cederblad, A.; Davidson, L.; Sandström, B. The effect of individual components of soy formula and cows’ milk formula on zinc bioavailability. The American Journal of Clinical Nutrition 1984, 40, 1064-1070.

232. Khouzam, R.B.; Pohl, P.; Lobinski, R. Bioaccessibility of essential elements from white cheese, bread, fruit and vegetables. Talanta 2011, 86, 425-428.

233. Simpson, C.J.; Wise, A. Binding of zinc and calcium to inositol phosphates (phytate) in vitro. British Journal of Nutrition 1990, 64, 225-232.

234. Turnlund, J.R.; King, J.C.; Keyes, W.R.; Gong, B.; Michel, M.C. A stable isotope study of zinc absorption in young men: Effects of phytate and alpha-cellulose. The American Journal of Clinical Nutrition 1984, 40, 1071-1077.

235. Fredlund, K.; Isaksson, M.; Rossander-Hulthén, L.; Almgren, A.; Sandberg, A.-S. Absorption of zinc and retention of calcium: Dose-dependent inhibition by phytate. Journal of Trace Elements in Medicine and Biology 2006, 20, 49-57.

236. Hunt, J.R.; Beiseigel, J.M. Dietary calcium does not exacerbate phytate inhibition of zinc absorption by women from conventional diets. The American Journal of Clinical Nutrition 2009, 89, 839-843.

237. Sandström, B.; Cederblad, A. Zinc absorption from composite meals. Ii. Influence of the main protein source. The American Journal of Clinical Nutrition 1980, 33, 1778-1783.

238. Sandström, B.; Almgren, A.; Kivisto, B.; Cederblad, A. Effect of protein level and protein source on zinc absorption in humans. The Journal of Nutrition 1989, 119, 48-53.

239. Davidsson, L.; Almgren, A.; SandströM, B.; Juillerat, M.E.-A.; Hurrell, R.F. Zinc absorption in adult humans: The effect of protein sources added to liquid test meals. British Journal of Nutrition 1996, 75, 607-613.

240. Lönnerdal, B.; Bell, J.G.; Hendrickx, A.G.; Burns, R.A.; Keen, C.L. Effect of phytate removal on zinc absorption from soy formula. American Journal of Clinical Nutrition 1988, 48, 1301-1306.

References

141

241. Sandström, B.; Cederblad, A.; Lönnerdal, B. Zinc absorption from human milk, cow's milk, and infant formulas. American Journal of Diseases of Children (1960) 1983, 137, 726-729.

242. Sandström, B.; Davidsson, L.; Cederblad, A.; Lönnerdal, B. Oral iron, dietary ligands and zinc absorption. The Journal of Nutrition 1985, 115, 411-414.

243. Sandstead, H.H.; Smith, J.C., Jr. Deliberations and evaluations of approaches, endpoints and paradigms for determining zinc dietary recommendations. The Journal of Nutrition 1996, 126, 2410s-2418s.

244. Wapnir, R.A.; Stiel, L. Zinc intestinal absorption in rats: Specificity of amino acids as ligands. The Journal of Nutrition 1986, 116, 2171-2179.

245. Raffaniello, R.D.; Wapnir, R.A. Zinc uptake by isolated rat enterocytes: Effect of low molecular weight ligands. Proceedings of the Society for Experimental Biology and Medicine 1989, 192, 219-224.

246. Hempe, J.M.; Cousins, R.J. Effect of EDTA and zinc-methionine complex on zinc absorption by rat intestine. The Journal of Nutrition 1989, 119, 1179-1187.

247. Davidsson, L.; Kastenmayer, P.; Hurrell, R.F. Sodium iron EDTA [NaFe (III)EDTA] as a food fortificant: The effect on the absorption and retention of zinc and calcium in women. The American Journal of Clinical Nutrition 1994, 60, 231-237.

248. Davidsson, L.; Almgren, A.; Sandström, B.; Hurrell, R.F. Zinc absorption in adult humans: The effect of iron fortification. The British Journal of Nutrition 1995, 74, 417-425.

249. Turnlund, J.R.; Keyes, W.R.; Hudson, C.A.; Betschart, A.A.; Kretsch, M.J.; Sauberlich, H.E. A stable-isotope study of zinc, copper, and iron absorption and retention by young women fed vitamin B-6-deficient diets. The American Journal of Clinical Nutrition 1991, 54, 1059-1064.

250. Solomons, N.W.; Jacob, R.A. Studies on the bioavailability of zinc in humans: Effects of heme and nonheme iron on the absorption of zinc. The American Journal of Clinical Nutrition 1981, 34, 475-482.

251. Sandström, B. Micronutrient interactions: Effects on absorption and bioavailability. The British Journal of Nutrition 2001, 85 Suppl 2, S181-185.

252. August, D.; Janghorbani, M.; Young, V.R. Determination of zinc and copper absorption at three dietary Zn-Cu ratios by using stable isotope methods in young adult and elderly subjects. The American Journal of Clinical Nutrition 1989, 50, 1457-1463.

253. Sandstead, H.H. Requirements and toxicity of essential trace elements, illustrated by zinc and copper. The American Journal of Clinical Nutrition 1995, 61, 621s-624s.

254. Blakeborough, P.; Salter, D.N. The intestinal transport of zinc studied using brush-border-membrane vesicles from the piglet. The British Journal of Nutrition 1987, 57, 45-55.

255. Valberg, L.S.; Flanagan, P.R.; Chamberlain, M.J. Effects of iron, tin, and copper on zinc absorption in humans. The American Journal of Clinical Nutrition 1984, 40, 536-541.

256. Johnson, M.A.; Baier, M.J.; Greger, J.L. Effects of dietary tin on zinc, copper, iron, manganese, and magnesium metabolism of adult males. The American Journal of Clinical Nutrition 1982, 35, 1332-1338.

257. Solomons, N.W.; Jacob, R.A.; Pineda, O.; Viteri, F.E. Studies on the bioavailability of zinc in man. III. Effects of ascorbic acid on zinc absorption. The American Journal of Clinical Nutrition 1979, 32, 2495-2499.

References

142

258. Nair, K.M.; Brahmam, G.N.; Radhika, M.S.; Dripta, R.C.; Ravinder, P.; Balakrishna, N.; Chen, Z.; Hawthorne, K.M.; Abrams, S.A. Inclusion of guava enhances non-heme iron bioavailability but not fractional zinc absorption from a rice-based meal in adolescents. The Journal of Nutrition 2013, 143, 852-858.

259. Pabón, M.L.; Lönnerdal, B. Effect of citrate on zinc bioavailability from milk, milk fractions and infant formulas. Nutrition Research 1993, 13, 103-111.

260. Lönnerdal, B.; Stanislowski, A.G.; Hurley, L.S. Isolation of a low molecular weight zinc binding ligand from human milk. Journal of Inorganic Biochemistry 1980, 12, 71-78.

261. Gharibzahedi, S.M.T.; Jafari, S.M. The importance of minerals in human nutrition: Bioavailability, food fortification, processing effects and nanoencapsulation. Trends in Food Science & Technology 2017, 62, 119-132.

262. Furniss, D.E.; Vuichoud, J.; Finot, P.A.; Hurrell, R.F. The effect of Maillard reaction products on zinc metabolism in the rat. The British Journal of Nutrition 1989, 62, 739-749.

263. O'Brien, J.; Morrissey, P.A. Metal ion complexation by products of the Maillard reaction. Food Chemistry 1997, 58, 17-27.

264. Kumar, V.; Sinha, A.K.; Makkar, H.P.S.; Becker, K. Dietary roles of phytate and phytase in human nutrition: A review. Food Chemistry 2010, 120, 945-959.

265. Krȩżel, A.; Maret, W. The biological inorganic chemistry of zinc ions. Archives of Biochemistry and Biophysics 2016, 611, 3-19.

266. Quarterman, J. Metal absorption and the intestinal mucus layer. Digestion 1987, 37, 1-9.

267. Whitehead, M.W.; Thompson, R.P.; Powell, J.J. Regulation of metal absorption in the gastrointestinal tract. Gut 1996, 39, 625-628.

268. Rudzki, Z.; Baker, R.J.; Deller, D.J. The iron-binding glycoprotein of human gastric juice. II. Nature of the interaction of the glycoprotein with iron. Digestion 1973, 8, 53-67.

269. Crowther, R.S.; Marriott, C. Counter-ion binding to mucus glycoproteins. The Journal of Pharmacy and Pharmacology 1984, 36, 21-26.

270. De Domenico, I.; Ward, D.M.; Kaplan, J. Hepcidin regulation: Ironing out the details. Journal of Clinical Investigation 2007, 117, 1755-1758.

271. Smith, K.T.; cousins, R.J.; Silbon, B.L.; Failla, M.L. Zinc absorption and metabolism by isolated, vascularly perfused rat intestine. The Journal of Nutrition 1978, 108, 1849-1857.

272. Tacnet, F.; Watkins, D.W.; Ripoche, P. Studies of zinc transport into brush-border membrane-vesicles isolated from pig small-intestine. Biochimica et Biophysica Acta 1990, 1024, 323-330.

273. Tacnet, F.; Watkins, D.W.; Ripoche, P. Zinc binding in intestinal brush-border membrane isolated from pig. Biochimica et Biophysica Acta 1991, 1063, 51-59.

274. Russel, M.W.S.; Burch, R.L. The Principles of Humane Experimental Technique. Methuen: London, Great Britian, 1959.

275. Flecknell, P. Replacement, reduction and refinement. Altex 2002, 19, 73-78. 276. Shah, P.; Jogani, V.; Bagchi, T.; Misra, A. Role of Caco-2 cell monolayers in prediction

of intestinal drug absorption. Biotechnology Progress 2006, 22, 186-198. 277. Laparra, J.M.; Barbera, R.; Alegria, A.; Glahn, R.P.; Miller, D.D. Purified

glycosaminoglycans from cooked haddock may enhance Fe uptake via endocytosis in a Caco-2 cell culture model. Journal of Food Science 2009, 74, H168-173.

References

143

278. Reeves, P.G.; Briske-Anderson, M.; Johnson, L. Physiologic concentrations of zinc affect the kinetics of copper uptake and transport in the human intestinal cell model, Caco-2. Journal of Nutrition 1998, 1794-1801.

279. Jovani, M.; Barbera, R.; Farre, R.; Aguilera, E.M.d. Calcium, iron, and zinc uptake from digests of infant formulas by Caco-2 cells. Journal of Agricultural and Food Chemistry 2001, 49, 3480−3485.

280. Delie, F.; Werner, R. A human colonic cell line sharing similarities with enterocytes as a model to examine oral absorption: Advantages and limitations of the Caco-2 model. Critical Reviews in Therapeutic Drug Carrier Systems 1997, 14, 221-286.

281. Pinto, M.; Robineleon, S.; Appay, M.D.; Kedinger, M.; Triadou, N.; Dussaulx, E.; Lacroix, B.; Simonassmann, P.; Haffen, K.; Fogh, J., et al. Enterocyte-like differentiation and polarization of the human colon carcinoma cell-line Caco-2 in culture. Biology of the Cell 1983, 47, 323-330.

282. Sambuy, Y.; de Angelis, M.; Ranaldi, G.; Scarino, M.L.; Stammati, A.; Zucco, F. The Caco-2 cell line as a model for the intestinal barrier: Influence of cell and culture-related factors on Caco-2 cell functional characteristics. Cell Biology and Toxicology 2005, 21, 1-26.

283. Hayeshi, R.; Hilgendorf, C.; Artursson, P.; Augustijns, P.; Brodin, B.; Dehertogh, P.; Fisher, K.; Fossati, L.; Hovenkamp, E.; Korjamo, T., et al. Comparison of drug transporter gene expression and functionality in Caco-2 cells from 10 different laboratories. European Journal of Pharmaceutical Sciences 2008, 35, 383-396.

284. Corti, G.; Maestrelli, F.; Cirri, M.; Zerrouk, N.; Mura, P. Development and evaluation of an in vitro method for prediction of human drug absorption II. Demonstration of the method suitability. European Journal of Pharmaceutical Sciences 2006, 27, 354-362.

285. Lozoya-Agullo, I.; Araújo, F.; González-Álvarez, I.; Merino-Sanjuán, M.; González-Álvarez, M.; Bermejo, M.; Sarmento, B. Usefulness of Caco-2/HT29-MTX and Caco-2/HT29-MTX/Raji b coculture models to predict intestinal and colonic permeability compared to Caco-2 monoculture. Molecular Pharmaceutics 2017, 14, 1264-1270.

286. Chen, Q.; Guo, L.; Du, F.; Chen, T.; Hou, H.; Li, B. The chelating peptide (gpagphgppg) derived from alaska pollock skin enhances calcium, zinc and iron transport in Caco-2 cells. International Journal of Food Science & Technology 2017, 52, 1283-1290.

287. Sreenivasulu, K.; Raghu, P.; Nair, K.M. Polyphenol-rich beverages enhance zinc uptake and metallothionein expression in Caco-2 cells. Journal of Food Science 2010, 75, H123-128.

288. Sreenivasulu, K.; Raghu, P.; Ravinder, P.; Nair, K.M. Effect of dietary ligands and food matrices on zinc uptake in Caco-2 cells: Implications in assessing zinc bioavailability. Journal of Agricultural and Food Chemistry 2008, 56, 10967-10972.

289. Cheng, Z.; Tako, E.; Yeung, A.; Welch, R.M.; Glahn, R.P. Evaluation of metallothionein formation as a proxy for zinc absorption in an in vitro digestion/Caco-2 cell culture model. Food & function 2012, 3, 732-736.

290. García-Nebot, M.J.; Alegría, A.; Barberá, R.; Clemente, G.; Romero, F. Does the addition of caseinophosphopeptides or milk improve zinc in vitro bioavailability in fruit beverages? Food Research International 2009, 42, 1475-1482.

291. Cámara, F.; Barberá, R.; Amaro, M.A.; Farré, R. Calcium, iron, zinc and copper transport and uptake by Caco-2 cells in school meals: Influence of protein and mineral interactions. Food Chemistry 2007, 100, 1085-1092.

References

144

292. Iyengar, V.; Pullakhandam, R.; Nair, K.M. Dietary ligands as determinants of iron-zinc interactions at the absorptive enterocyte. Journal of Food Science 2010, 75, H260-264.

293. Tupe, R.S.; Agte, V.V. Effect of water soluble vitamins on Zn transport of Caco-2 cells and their implications under oxidative stress conditions. European Journal of Nutrition 2009, 49, 53-61.

294. Kim, E.-Y.; Pai, T.-K.; Han, O. Effect of bioactive dietary polyphenols on zinc transport across the intestinal Caco-2 cell monolayers. Journal of Agricultural and Food Chemistry 2011, 59, 3606-3612.

295. Viadel, B.; Barberá, R.; Farré, R. Uptake and retention of calcium, iron, and zinc from raw legumes and the effect of cooking on lentils in Caco-2 cells. Nutrition Research 2006, 26, 591-596.

296. Salunke, R.; Rawat, N.; Tiwari, V.K.; Neelam, K.; Randhawa, G.S.; Dhaliwal, H.S.; Roy, P. Determination of bioavailable-zinc from biofortified wheat using a coupled in vitro digestion/Caco-2 reporter-gene based assay. Journal of Food Composition and Analysis 2012, 25, 149-159.

297. Kruger, J.; Taylor, J.R.N.; Du, X.; De Moura, F.F.; Lönnerdal, B.; Oelofse, A. Effect of phytate reduction of sorghum, through genetic modification, on iron and zinc availability as assessed by an in vitro dialysability bioaccessibility assay, Caco-2 cell uptake assay, and suckling rat pup absorption model. Food Chemistry 2013, 141, 1019-1025.

298. Han, O.; Failla, M.L.; Hill, A.D.; Morris, E.R.; Smith, J.C., Jr. Inositol phosphates inhibit uptake and transport of iron and zinc by a human intestinal cell line. The Journal of Nutrition 1994, 124, 580-587.

299. Frontela, C. Effect of dephytinization on bioavailability of iron, calcium and zinc from infant cereals assessed in the Caco-2 cell model. World Journal of Gastroenterology 2009, 15, 1977.

300. Frontela, C.; Ros, G.; Martínez, C. Phytic acid content and “in vitro” iron, calcium and zinc bioavailability in bakery products: The effect of processing. Journal of Cereal Science 2011, 54, 173-179.

301. Jou, M.-Y.; Du, X.; Hotz, C.; Lönnerdal, B. Biofortification of rice with zinc: Assessment of the relative bioavailability of zinc in a Caco-2 cell model and suckling rat pups. Journal of Agricultural and Food Chemistry 2012, 60, 3650-3657.

302. Zemann, N.; Zemann, A.; Klein, P.; Elmadfa, I.; Huettinger, M. Differentiation- and polarization-dependent zinc tolerance in Caco-2 cells. European Journal of Nutrition 2011, 50, 379-386.

303. Jackson, K.A.; Valentine, R.A.; McKay, J.A.; Swan, D.C.; Mathers, J.C.; Ford, D. Analysis of differential gene-regulatory responses to zinc in human intestinal and placental cell lines. The British Journal of Nutrition 2009, 101, 1474-1483.

304. Gefeller, E.M.; Bondzio, A.; Aschenbach, J.R.; Martens, H.; Einspanier, R.; Scharfen, F.; Zentek, J.; Pieper, R.; Lodemann, U. Regulation of intracellular zn homeostasis in two intestinal epithelial cell models at various maturation time points. The Journal of Physiological Sciences 2015, 65, 317-328.

305. Lodemann, U.; Gefeller, E.M.; Aschenbach, J.R.; Martens, H.; Einspanier, R.; Bondzio, A. Dose effects of apical versus basolateral zinc supplementation on epithelial resistance, viability, and metallothionein expression in two intestinal epithelial cell lines. Journal of Biochemical and Molecular Toxicology 2015.

References

145

306. Shen, H.; Qin, H.; Guo, J. Cooperation of metallothionein and zinc transporters for regulating zinc homeostasis in human intestinal Caco-2 cells. Nutrition Research 2008, 28, 406-413.

307. Wang, X.; Valenzano, M.C.; Mercado, J.M.; Zurbach, E.P.; Mullin, J.M. Zinc supplementation modifies tight junctions and alters barrier function of Caco-2 human intestinal epithelial layers. Digestive Diseases and Sciences 2012, 58, 77-87.

308. Jou, M.Y.; Philipps, A.F.; Kelleher, S.L.; Lönnerdal, B. Effects of zinc exposure on zinc transporter expression in human intestinal cells of varying maturity. Journal of Pediatric Gastroenterology and Nutrition 2010, 50, 587-595.

309. Pieri, M.; Christian, H.C.; Wilkins, R.J.; Boyd, C.A.; Meredith, D. The apical (hPepT1) and basolateral peptide transport systems of caco-2 cells are regulated by AMP-activated protein kinase. American Journal of Physiology-Gastrointestinal and Liver Physiology 2010, 299, G136-143.

310. Hardyman, J.E.; Tyson, J.; Jackson, K.A.; Aldridge, C.; Cockell, S.J.; Wakeling, L.A.; Valentine, R.A.; Ford, D. Zinc sensing by metal-responsive transcription factor 1 (MTF1) controls metallothionein and ZnT1 expression to buffer the sensitivity of the transcriptome response to zinc. Metallomics 2016, 8, 337-343.

311. Oestreicher, P.; Cousins, R.J. Zinc uptake by basolateral membrane vesicles from rat small intestine. The Journal of Nutrition 1989, 119, 639-646.

312. Mahler, G.J.; Shuler, M.L.; Glahn, R.P. Characterization of Caco-2 and HT29-MTX cocultures in an in vitro digestion/cell culture model used to predict iron bioavailability. The Journal of Nutritional Biochemistry 2009, 20, 494-502.

313. Hilgendorf, C.; Spahn-Langguth, H.; Regardh, C.G.; Lipka, E.; Amidon, G.L.; Langguth, P. Caco-2 versus Caco-2/HT29-MTX co-cultured cell lines: Permeabilities via diffusion, inside- and outside-directed carrier-mediated transport. Journal of Pharmaceutical Sciences 2000, 89, 63-75.

314. Walter, E.; Janich, S.; Roessler, B.J.; Hilfiger, J.M.; Amidon, G.L. HT29-MTX/Caco-2 cocultures as an in vitro model for the intestinal epithelium: In vitro−in vivo correlation with permeability data from rats and humans. Journal of Pharmaceutical Sciences 1996, 85, 1070-1076.

315. Pontier, C.; Pachot, J.; Botham, R.; Lenfant, B.; Arnaud, P. HT29-MTX and Caco-2/TC7 monolayers as predictive models for human intestinal absorption: Role of the mucus layer. Journal of Pharmaceutical Sciences 2001, 90, 1608-1619.

316. Ferraretto, A.; Bottani, M.; De Luca, P.; Cornaghi, L.; Arnaboldi, F.; Maggioni, M.; Fiorilli, A.; Donetti, E. Morphofunctional properties of a differentiated Caco2/HT-29 co-culture as an in vitro model of human intestinal epithelium. Bioscience Reports 2018, 38.

317. Calatayud, M.; Vazquez, M.; Devesa, V.; Velez, D. In vitro study of intestinal transport of inorganic and methylated arsenic species by Caco-2/HT29-MTX cocultures. Chemical Research in Toxicology 2012, 25, 2654-2662.

318. Vazquez, M.; Calatayud, M.; Velez, D.; Devesa, V. Intestinal transport of methylmercury and inorganic mercury in various models of Caco-2 and HT-29-MTX cells. Toxicology 2013, 311, 147-153.

319. Laparra, J.M.; Glahn, R.P.; Miller, D.D. Different responses of Fe transporters in Caco-2/HT29-MTX cocultures than in independent Caco-2 cell cultures. Cell Biology International 2009, 33, 971-977.

320. Moreno-Olivas, F.; Tako, E.; Mahler, G.J. ZnO nanoparticles affect nutrient transport in an in vitro model of the small intestine. Food and Chemical Toxicology 2018.

References

146

321. Laparra, J.M.; Sanz, Y. Comparison of in vitro models to study bacterial adhesion to the intestinal epithelium. Letters in Applied Microbiology 2009, 49, 695-701.

322. Fleet, J.C.; Turnbull, A.J.; Bourcier, M.; Wood, R.J. Vitamine D-sensitive and quinacrine-sensitive zinc transport in human intestinal cell line Caco-2. American Journal of Physiology 1993, 264, G1037-1045.

323. Reeves, P.G.; Briske-Anderson, M.; Johnson, L. Pre-treatment of Caco-2 cells with zinc during the differentiation phase alters the kinetics of zinc uptake and transport. Journal of Nutritional Biochemistry 2001, 12, 674–684.

324. Krebs, N.F.; Hambidge, K.M. Zinc metabolism and homeostasis: The application of tracer techniques. BioMetals 2001, 14, 397-412.

325. Wastney, M.E.; Aamodt, R.L.; Rumble, W.F.; Henkin, R.I. Kinetic analysis of zinc metabolism and its regulation in normal humans. The American Journal of Physiology 1986, 251, R398-408.

326. Sandström, B.; Lönnerdal, B. Promoters and antagonists of zinc absorption. In Zinc in Human Biology, Mills, C.F., Ed. Springer London: London, 1989; pp 57-78.

327. Tran, C.D.; Gopalsamy, G.L.; Mortimer, E.K.; Young, G.P. The potential for zinc stable isotope techniques and modelling to determine optimal zinc supplementation. Nutrients 2015, 7, 4271-4295.

328. Cerchiaro, G.; Manieri, T.M.; Bertuchi, F.R. Analytical methods for copper, zinc and iron quantification in mammalian cells. Metallomics 2013, 5, 1336-1345.

329. Haase, H.; Rink, L. Zinc signals and immune function. BioFactors 2014, 40, 27-40. 330. Carter, K.P.; Young, A.M.; Palmer, A.E. Fluorescent sensors for measuring metal ions

in living systems. Chemical Reviews 2014, 114, 4564-4601. 331. Ackerman, C.M.; Lee, S.; Chang, C.J. Analytical methods for imaging metals in biology:

From transition metal metabolism to transition metal signaling. Analytical Chemistry 2017, 89, 22-41.

332. Vinkenborg, J.L.; Nicolson, T.J.; Bellomo, E.A.; Koay, M.S.; Rutter, G.A.; Merkx, M. Genetically encoded fret sensors to monitor intracellular Zn2+ homeostasis. Nature Methods 2009, 6, 737-740.

333. Aper, S.J.; Dierickx, P.; Merkx, M. Dual readout BRET/FRET sensors for measuring intracellular zinc. ACS Chemical Biology 2016, 11, 2854-2864.

334. Hessels, A.M.; Chabosseau, P.; Bakker, M.H.; Engelen, W.; Rutter, G.A.; Taylor, K.M.; Merkx, M. eZinch-2: A versatile, genetically encoded fret sensor for cytosolic and intraorganelle Zn2+ imaging. ACS Chemical Biology 2015, 10, 2126-2134.

335. Qin, Y.; Miranda, J.G.; Stoddard, C.I.; Dean, K.M.; Galati, D.F.; Palmer, A.E. Direct comparison of a genetically encoded sensor and small molecule indicator: Implications for quantification of cytosolic Zn2+. ACS Chemical Biology 2013, 8, 2366-2371.

336. Park, J.G.; Qin, Y.; Galati, D.F.; Palmer, A.E. New sensors for quantitative measurement of mitochondrial Zn2+. ACS Chemical Biology 2012, 7, 1636-1640.

337. Qin, Y.; Sammond, D.W.; Braselmann, E.; Carpenter, M.C.; Palmer, A.E. Development of an optical Zn2+ probe based on a single fluorescent protein. ACS Chemical Biology 2016, 11, 2744-2751.

338. Wallrabe, H.; Periasamy, A. Imaging protein molecules using FRET and FLIM microscopy. Current Opinion in Biotechnology 2005, 16, 19-27.

339. Boute, N.; Jockers, R.; Issad, T. The use of resonance energy transfer in high-throughput screening: BRET versus FRET. Trends in Pharmacological Sciences 2002, 23, 351-354.

References

147

340. Pfleger, K.D.G.; Seeber, R.M.; Eidne, K.A. Bioluminescence resonance energy transfer (BRET) for the real-time detection of protein-protein interactions. Nature Protocols 2006, 1, 337-345.

341. Dittmer, P.J.; Miranda, J.G.; Gorski, J.A.; Palmer, A.E. Genetically encoded sensors to elucidate spatial distribution of cellular zinc. The Journal of Biological Chemistry 2009, 284, 16289-16297.

342. Chabosseau, P.; Tuncay, E.; Meur, G.; Bellomo, E.A.; Hessels, A.; Hughes, S.; Johnson, P.R.; Bugliani, M.; Marchetti, P.; Turan, B., et al. Mitochondrial and er-targeted ecalwy probes reveal high levels of free Zn2+. ACS Chemical Biology 2014, 9, 2111-2120.

343. Qin, Y.; Dittmer, P.J.; Park, J.G.; Jansen, K.B.; Palmer, A.E. Measuring steady-state and dynamic endoplasmic reticulum and golgi Zn2+ with genetically encoded sensors. Proceedings of the National Academy of Sciences of the United States of America 2011, 108, 7351-7356.

344. Pancholi, J.; Hodson, D.J.; Jobe, K.; Rutter, G.A.; Goldup, S.M.; Watkinson, M. Biologically targeted probes for Zn2+: A diversity oriented modular "click-snar-click" approach. Chemical Science 2014, 5, 3528-3535.

345. Li, D.; Chen, S.; Bellomo, E.A.; Tarasov, A.I.; Kaut, C.; Rutter, G.A.; Li, W.-h. Imaging dynamic insulin release using a fluorescent zinc indicator for monitoring induced exocytotic release (ZIMIR). Proceedings of the National Academy of Sciences of the United States of America 2011, 108, 21063-21068.

346. Radford, R.J.; Chyan, W.; Lippard, S.J. Peptide-based targeting of fluorescent zinc sensors to the plasma membrane of live cells. Chemical Science 2013, 4, 3080-3084.

347. Ranaldi, G.; Ferruzza, S.; Canali, R.; Leoni, G.; Zalewski, P.D.; Sambuy, Y.; Perozzi, G.; Murgia, C. Intracellular zinc is required for intestinal cell survival signals triggered by the inflammatory cytokine TNF-alpha. The Journal of Nutritional Biochemistry 2013, 24, 967-976.

348. Krȩżel, A.; Maret, W. Zinc-buffering capacity of a eukaryotic cell at physiological pZn. Journal of Biological Inorganic Chemistry 2006, 11, 1049-1062.

349. Lu, Q.; Haragopal, H.; Slepchenko, K.G.; Stork, C.; Li, Y.V. Intracellular zinc distribution in mitochondria, er and the golgi apparatus. International Journal of Physiology, Pathophysiology and Pharmacology 2016, 8, 35-43.

350. Lee, S.R. Cellular toxicity of zinc can be attenuated by sodium hydrogen sulfide in neuronal SH-SY5Y cell. Molecular & Cellular Toxicology 2018, 14, 425-436.

351. Walkup, G.K.; Burdette, S.C.; Lippard, S.J.; Tsien, R.Y. A new cell-permeable fluorescent probe for Zn2+. Journal of the American Chemical Society 2000, 122, 5644-5645.

352. Burdette, S.C.; Walkup, G.K.; Spingler, B.; Tsien, R.Y.; Lippard, S.J. Fluorescent sensors for Zn2+ based on a fluorescein platform: Synthesis, properties and intracellular distribution. Journal of the American Chemical Society 2001, 123, 7831-7841.

353. Haase, H.; Hebel, S.; Engelhardt, G.; Rink, L. Flow cytometric measurement of labile zinc in peripheral blood mononuclear cells. Analytical Biochemistry 2006, 352, 222-230.

354. Dineley, K.E.; Malaiyandi, L.M.; Reynolds, I.J. A reevaluation of neuronal zinc measurements: Artifacts associated with high intracellular dye concentration. Molecular Pharmacology 2002, 62, 618-627.

References

148

355. Kenneth, V.H.; John, A.S.; Walter, C. Fetal bovine serum: A multivariate standard. Proceedings of the Society for Experimental Biology and Medicine 1975, 149, 344-347.

356. Zheng, X.; Baker, H.; Hancock, W.S.; Fawaz, F.; McCaman, M.; Pungor, E. Proteomic analysis for the assessment of different lots of fetal bovine serum as a raw material for cell culture. Part IV. Application of proteomics to the manufacture of biological drugs. Biotechnology Progress 2006, 22, 1294-1300.

357. Gstraunthaler, G. Alternatives to the use of fetal bovine serum: Serum-free cell culture. Altex 2003, 20, 275-281.

358. Bal, W.; Sokołowska, M.; Kurowskaa, E.; Faller, P. Binding of transition metal ions to albumin: Sites, affinities and rates Biochimica et Biophysica Acta 2013, 1830, 5444–5455.

359. Schömig, V.J.; Kasdorf, B.T.; Scholz, C.; Bidmon, K.; Lieleg, O.; Berensmeier, S. An optimized purification process for porcine gastric mucin with preservation of its native functional properties. RSC Advances 2016, 6, 44932-44943.

360. Svensson, O.; Arnebrant, T. Mucin layers and multilayers - physicochemical properties and applications. Current Opinion in Colloid & Interface Science 2010, 15, 395-405.

361. Araujo, F.; Sarmento, B. Towards the characterization of an in vitro triple co-culture intestine cell model for permeability studies. International Journal of Pharmaceutics 2013, 458, 128-134.

362. Beduneau, A.; Tempesta, C.; Fimbel, S.; Pellequer, Y.; Jannin, V.; Demarne, F.; Lamprecht, A. A tunable Caco-2/HT29-MTX co-culture model mimicking variable permeabilities of the human intestine obtained by an original seeding procedure. European Journal of Pharmaceutics and Biopharmaceutics 2014, 87, 290-298.

363. Hoeger, J.; Simon, T.-P.; Doemming, S.; Thiele, C.; Marx, G.; Schuerholz, T.; Haase, H. Alterations in zinc binding capacity, free zinc levels and total serum zinc in a porcine model of sepsis. BioMetals 2015, 1-8.

364. Bloxam, D.L.; Tan, J.C.; Parkinson, C.E. Non-protein bound zinc concentration in human plasma and amniotic fluid measured by ultrafiltration. Clinica Chimica Acta 1984, 144, 81-93.

365. Magneson, G.R.; Puvathingal, J.M.; Ray, W.J., Jr. The concentrations of free Mg2+ and free Zn2+ in equine blood plasma. The Journal of Biological Chemistry 1987, 262, 11140-11148.

366. Smith, K.T.; Failla, M.L.; Cousins, R.J. Identification of albumin as the plasma carrier for zinc absorption by perfused rat intestine. Biochemical Journal 1979, 184, 627-633.

367. Wang, X.; Zhou, B. Dietary zinc absorption: A play of ZIPs and ZnTs in the gut. IUBMB life 2010, 62, 176-182.

368. Messer, H.H.; Murray, E.J.; Goebel, N.K. Removal of trace metals from culture media and sera for in vitro deficiency studies. The Journal of Nutrition 1982, 112, 652-657.

369. Finamore, A.; Massimi, M.; Conti Devirgiliis, L.; Mengheri, E. Zinc deficiency induces membrane barrier damage and increases neutrophil transmigration in Caco-2 cells. The Journal of Nutrition 2008, 138, 1664-1670.

Appendix

VIII

Appendix

A. Supplemental Material of Chapter 4

Figure S4.1: Transepithelial electrical resistance (TEER) during differentiation of Caco-2 and

Caco-2-eCalwy.

Cells were grown on transwell membranes for 21d and integrity of the cell layer was monitored by measuring TEER during differentiation. Data are shown as means + SD of three independent measurements.

Appendix

IX

Figure S4.2: Localization of tight junction proteins in differentiated Caco-2 cells.

Caco2-WT (upper panel) and Caco-2-eCalwy (lower panel) were analyzed by confocal laser scanning immunofluorescence microscopy. False colored images of Caco-2-WT and -eCalwy clones after immunofluorescent staining of claudin-2 (A, E), occludin (B, F), zonola occludens-1 (ZO-1) (C, G) and E-cadherin (D, H) are shown. X-y-Sections depict an overview on the stained cell-layers, whereas x-z-scans visualize the cellular distribution of tight junction proteins together with stained nuclei (blue). Scale bars correspond to 5µm. Sections are the same as the ones shown in Figure 4.4.

Appendix

X

B. Supplemental Material of Chapter 5

Supplemental Methods

Real time PCR

qPCR-analysis for ALP and MUC5AC was performed as described in the main text, using the

primer listed in Suppl. Table S4.1.

Table S5.1: Oligonucleotide sequences used for qPCR

Primer NCBI Reference

Sequence

Sequence fwd 5'-3' Sequence rev 5'-3' Ref

ALP NM_001631.4 CCGCTTTAACCAGTGCAACA CCCATGAGATGGGTCACAGA [S1]

MUC5AC NM_001304359.1 CATCAACGGGACCCTGTACC ACAGGTCGACTGGTTCTGGT

β-actin NG_007992.1 CGCCCCAGGCACCAGGGC GCTGGGGTGTTGAAGGT [S2]

Histological staining of mucins

The mucin-secretion of Caco-2/HT-29-MTX co-cultures was investigated by histological

staining of anionic mucins, using alcianblue, and neutral mucins with the PAS (periodic acid-

Schiff)-staining. Therefore, a total of 120.000 cells was cultivated in 6-well plates for 21 d,

whereas several ratios of Caco-2 and HT-29-MTX were co-cultivated (Caco-2/HT-29-MTX:

100/0, 90/10, 75/25, 50/50, 0/100). The protocols for the alcianblue- and PAS-staining were

adapted after [S3]. Prior to the staining, cells were washed carefully with PBS and fixed using

3.7% formaldehyde in PBS. For the alcianblue-staining, cells were washed again with PBS,

incubated with 3% acetic acid for 3 min cells and directly treated with 1% alcianblue 8GX (in

3% acetic acid) for 30min. Subsequently, the staining reagent was carefully removed, cells

were washed with H2O and macroscopic pictures were taken. Prior to the PAS-staining, cells

were incubated with 5% periodic acid for 5 min, washed carefully with PBS and incubated

with Schiff-reagents for 15min. The Schiff reagent was generated following the method from

Graumann [S4] using 5% pararosanilin. After successful incubation, Schiff-reagent was

removed from the cells and pictures were taken instantly.

Alkaline phosphatase activity

The effect of cellular ratio on the activity of alkaline phosphatase (ALP) in co-cultures of

Caco-2 and HT-29-MTX was investigated, examining several Caco-2/HT-29-MTX ratios:

100/0, 90/10, 75/25, 50/50 and 0/100. Therefore, two days prior to the HT29-MTX cells,

Caco-2 cells were transferred to 96-well plates (a total of 5000 cells) and ALP-activity after 21

d of cultivation was analyzed as described [S1].

Appendix

XI

Supplemental Figures

Figure S5.1: Histological analysis of secreted mucins by Caco-2/HT-29-MTX co-cultures.

Co-cultures of various ratios of Caco-2 and HT-29-MTX cells were cultured for 21 d and mucin secretion was investigated using alcianblue- (A) and PAS-staining (B).

100 90 75 50 00

50

100

150

200

Caco-2 [%] in co-culture

Alk

alin

e p

ho

sp

hata

se

acti

vit

y [

mU

/mg

pro

tein

]

Figure S5.2: Activity of alkaline phosphatase (ALP) in Caco-2 and HT-29-MTX co-cultures.

Shown is the ALP-activity in Caco-2/HT-29-MTX co-cultures with different cellular ratios depicted as the relative amount of Caco-2 cells. Enzyme activity was measured after 21 d of cultivation using the ALP-assay and is displayed relative to cellular protein. Data are displayed as means + SD of three independent experiments.

Appendix

XII

90 75 50 00.0

0.2

0.4

0.6

0.8

Caco-2 [%] in co-culture

alp

gen

e e

xp

ressio

n

[rela

tive t

o 1

00%

Caco

-2 c

ell

s]

90 75 50 00

1000

2000

3000

4000

5000

Caco-2 [%] in co-culture

muc5ac

gen

e e

xp

ressio

n

[rela

tive t

o 1

00%

Caco

-2 c

ell

s]A B

Figure S5.3: Expression of characteristic genes for Caco-2 and HT-29-MTX in the co-cultures.

Gene expression of differentiation marker alp [S1] and muc5ac, which is one of the characteristic mucins secreted by HT-29-MTX [S5], was analyzed in Caco-2/HT-29-MTX co-cultures after 21d of cultivation using qPCR (A) Alp-expression is shown relative to Caco-2 monocultures. (B) Muc5ac-expression is depicted relative to Caco-2 monocultures. Data are shown as means + SD of three replicates.

0 25 50 1000

50

100

150

200

Added zinc [µM]

TE

ER

[%

rel. t

o t

0]

0 25 50 1000

50

100

150

200

Added zinc [µM]

TE

ER

[%

rel. t

o t

0]

0 25 50 100

1.010 -08

1.010 -06

1.010 -04

1.010 -02

1.01000

Added zinc [µM]

Pap

p [

cm

/sec]

0 25 50 100

1.010 -08

1.010 -06

1.010 -04

1.010 -02

1.01000

Added zinc [µM]

Pap

p [

cm

/sec]

Without BSA With BSA

A B

C D

Figure S5.4: Integrity of Caco-2/HT-29-MTX cell monolayers used for the transport studies

depicted in Figure 5.6 measured as TEER (A, C) and paracellular permeability (C,

D).

Shown is the TEER of cell-monolayers after the transport-experiment relative to TEER measured before incubation with zinc and 0mg mL-1 (A) or 30mg mL-1 albumin, respectively. The permeability of the cell monolayer during the transport assay without (C) or with albumin (D) is depicted as the apparent permeability (Papp) of a 20kDa FITC-Dextran. Data are shown as means + SD of three replicates.

Appendix

XIII

Table S5.2: Parameters of the non-linear regression analysis applied in the zinc cytotoxicity

study in Figure 5.3.

WST MTT NRU SRB

0% FCS 10% FCS 0% FCS 10% FCS 0% FCS 10% FCS 0% FCS 10% FCS

Best-fit values

Bottom 21,97 33,95 9,525 -42,44 0.0 0.0 3,663 1,599

Top 97,04 87,61 109,7 104,1 111,6 105,1 108,6 103,3

Hill slope -2,315 -4,216 2,473 2,881 -4,163 -4,165 -5,034 -5,012

95% Confidence Interval of LC50

162.5 to 567.7

315.4 to 761.2

247.4 to 356.3

214.1 to 2694

397.7 to 486.8

648.2 to 741.6

239.1 to 306.4

451.2 to 506.9

Goodness of Fit

Degree of Freedom

17 17 17 17 18 18 17 17

R² 0.7939 0.6740 0.9664 0.9453 0.9601 0.9761 0.9719 0.9860

Absolute Sum of Squares

4,033 3,543 1,094 1,294 1,294 456.2 1,219 402.1

Standard deviation of residuals

15.40 14.44 8.023 8.723 8.479 5.034 8.467 4.863

Shown are parameters of the applied non-linear regression using a sigmoidal dose-response curve with variable slope as a function of the logarithm of zinc concentration. Data were obtained in three independent experiments and analyzed with GraphPad Prism software version 5.01 (GraphPad Software Inc., CA, USA).

Appendix

XIV

References

[S1] M. Maares, C. Keil, S. Thomsen, D. Günzel, B. Wiesner, H. Haase, Characterization of

Caco-2 cells stably expressing the protein-based zinc probe eCalwy-5 as a model

system for investigating intestinal zinc transport, Journal of Trace Elements in

Medicine and Biology (2018).

[S2] K. Wolf, C. Schulz, G.A.J. Riegger, M. Pfeifer, Tumour necrosis factor‐α induced CD70

and interleukin‐7R mRNA expression in BEAS‐2B cells, European Respiratory Journal

20(2) (2002) 369-375.

[S3] H. Denk, H. Künzele, H. Plenk, J. Rüschoff, W. Seller, Romeis Mikroskopische Technik.

17., neubearbeitete Auflage, Urban und Schwarzenberg, München-Wien. Baltimore,

1989, pp. 439-50.

[S4] W. Graumann, Zur Standardisierung des Schiffschen Reagens, S HIRZEL VERLAG

Stuttgart, 1953, pp. 225-226.

[S5] G. Nollevaux, C. Deville, B. El Moualij, W. Zorzi, P. Deloyer, Y.J. Schneider, O. Peulen,

G. Dandrifosse, Development of a serum-free co-culture of human intestinal

epithelium cell-lines (Caco-2/HT29-5M21), BMC Cell Biology 7 (2006) 20.

Appendix

XV

C. Supplemental Material of Chapter 6

400 500 6000.0

0.1

0.2

0.3

0.4

0.5 0 µM

2 µM

4 µM

6 µM

8 µM

10 µM

12 µM

14 µM

16 µM

18 µM

20 µM

24 µM

28 µM

32 µM

36 µM

40 µM

Zn2+

[nm]

Ab

so

rpti

on

(

=4

85

nm

)

0 5 10 15 20 250.0

0.1

0.2

0.3

0.4

0.5

Added zinc [µM]

Ab

so

rpti

on

(

=485 n

m)

A) Absorptionspectra PAR-zinc complex

B) Spectrophotometric titration of PAR with zinc

Figure S6.1: Spectrophotometric titration with 4-(2-pyridylazo)resorcinol (PAR) and zinc.

Shown are absorption spectra (A) and spectrophotometric titrations (B) of 20 µM PAR with different zinc concentrations in Tris(hydroxylmethyl)aminomethan buffered saline (TBS), pH 7.4. Data are presented as means ± SD of three independent experiments.

Appendix

XVI

103 104 105 106

100

1000

100006 h

12 h

24 h

Added zinc [µM]

Reta

ined

zin

c [

µM

]

Figure S6.2: Zinc binding capacity of mucins using dialysis for different time intervals.

The zinc binding capacity was investigated after dialysis against TBS after different time intervals. Therefore, 25 mg mL-1 porcine mucin was incubated overnight with different zinc concentrations and dialysis was performed for 6 h, 12 h and 24 h. The amount of zinc retained by binding to mucins was measured using flame atomic absorption spectrometry (FAAS). Data are presented as means ± SD of three independent experiments.

Appendix

XVII

400 500 600 700 8000.0

0.1

0.2

0.3

0.4

0.5 0 µM

10 µM

20 µM

30 µM

40 µM

50 µM

60 µM

70 µM

80 µM

90 µM

100 µM

Zn2+

[nm]

Ab

so

rpti

on

(

=6

20

nm

)

400 500 600 700 8000.0

0.1

0.2

0.3

0.4

0.50 µM

5 µM

10 µM

12.5 µM

15 µM

17.5 µM

20 µM

22.5 µM

25 µM

27.5 µM

30 µM

32 µM

34 µM

35 µM

40 µM

45 µM

50 µM

55 µM

60 µM

Zn2+

[nm]

Ab

so

rpti

on

(

=6

20

nm

)

A) Absorptionspectra of Zincon-zinc-complex

B) Absorptionspectra of Zincon-zinc-complex in the presence of mucins

0 20 40 60 80 1000.0

0.1

0.2

0.3

0.4

Added zinc [µM]

Ab

so

rpti

on

(

=620 n

m)

C) Spectrophotometric titration of Zincon with zinc

Figure S6.3: Spectrophotometric titration with 2-carboxy-2′-hydroxy-5′-sulfoformazylbenzene

monosodium salt (zincon) and zinc.

Shown are the absorption spectra (A, B) and spectrophotometric titrations (C) of 50 µM zincon with different zinc concentrations in TBS, pH 7.4. Moreover, the absorption spectrum was conducted in the presence of 0 mg mL-1 (A) and 1 mg mL-1 porcine mucins (B). Data are presented as means ± SD of three independent experiments.

Appendix

XVIII

0 25 50 1000

50

100

150

200

Added zinc [µM]

TE

ER

[%

rel. t

o t

0]

0 25 50 1000

50

100

150

200

Added zinc [µM]

TE

ER

[%

rel. t

o t

0]

0 25 50 1001.010 -09

1.010 -07

1.010 -05

1.010 -03

1.010 -01

Added zinc [µM]

Pap

p [

cm

/sec]

0 25 50 1001.010 -09

1.010 -07

1.010 -05

1.010 -03

1.010 -01

Added zinc [µM]

Pap

p [

cm

/sec]

Caco-2 monoculture Caco-2/HT-29-MTX co-culture

A B

C D

Figure S6.4: Integrity of Caco-2 and Caco-2/HT-29-MTX cell monolayers used for the transport

studies measured as transepithelial electrical resistance (TEER) and paracellular

permeability.

Shown is the TEER of Caco-2 (A) or Caco-2/HT-29-MTX monolayers (B) after the transport experiment relative to TEER measured prior to incubation with zinc. The permeability of the cell monolayer during the transport assay using the monoculture (C) or the co-culture (D) is depicted as the apparent permeability (Papp) of a 20 kDa fluorescein isothiocyanate (FITC)-Dextran. Data are presented as means + SD of three independent experiments.

Appendix

XIX

Table S6.1: Exact amounts of zinc [ng cm-2] transported by Caco-2/HT-29-MTX co-cultures and

Caco-2 monocultures.

Caco-2 monocultures Caco-2/HT-29-MTX co-cultures

Added zinc [µM]

0 25 50 100 0 25 50 100

Apical zinc uptake

[ng/cm²]

- 93.8±45.4 168.8±52.7 198.4±88.5 - 89.2±18.5 193.4±16.5 332.2±137.4

Cellular zinc [ng/cm²]

59.6±14.5 76.8±21.2 86.9±23.6 95.6±25.9 37.6±2.6 67.7±13.5 101.4±31.8 130.8±38.1

Resorbed zinc [ng/cm²]

29.9±14.5 48.8±6.5 70.8±8.9 93.3±12.5 47.0±7.5 83.7±6.4 105.1±24.7 149.6±35.1

Shown are the amounts of zinc which are absorbed into the cells (zinc uptake), the cellular zinc content and which are resorbed into the basolateral compartment in ng zinc per resorption area (in cm²). Data are presented as means ± SD of three independent experiments.

Appendix

XX

D. Experimental conditions for Instrumental Zinc Quantification

Table S1: Experimental conditions for ICP-MS (Agilent 8800 ICP-QQQ)

Forward power 1550 W

Cool gas flow 15 L min-1

Auxiliary gas flow 0.9 L min-1

Nebulizer gas flow Argon, 1 L min-1

Nebulizer type MicroMist

Mode Single Quad

Collision gas flow Helium, 3 mL min-1

Quadrupole (m/z) 66 (Zn)

Limit of quantitation 0.2 µg L-1

Calibration range 1-100 µg L-1

Table S2: Experimental conditions for FAAS (Perkin Elmer AAnalyst800)

Gas flow Acetylen, 2.0 L min

-1

Oxygen 17 L min-1

Lamp Hollow Cathode Lamp

Wavelength [nm] 213.19 nm

Slit [nm] 0.7 nm

Lamp Current 18 mA

Limit of quantitation 0.02 mg L-1

Calibration range 0.05-1 mg L-1

Appendix

XXI

E. Supplemental Results of Zinc Resorption Studies with in vitro Intestinal

Models

In addition to findings of the zinc transport studies using three-dimensional in vitro intestinal

models in Chapter 5 and 6, Figure S1, Table S3 and Table S4 include supplemental results of

zinc transport studies. For a better understanding of the impact of mucins and basolateral

albumin on zinc resorption, zinc transport of Caco-2 monocultures and Caco-2/HT-29-MTX

co-cultures in presence and absence of basolateral albumin after 8 h is shown in Figure S1.

For this, data of Caco-2/HT-29-MTX was extracted from Chapter 5. To give an overview on

zinc transport using mono- or co-cultures with and without basolateral albumin, Table S3

additionally summarizes detailed quantitative data of zinc uptake into the cells, cellular zinc

content, and the amount of transported zinc to the basolateral side in Caco-2 monocultures

and Caco-2/HT-29-MTX co-cultures after incubation with zinc for 4 h and 8 h. Moreover,

fractional zinc resorption as well as zinc transport rates of mono- and co-cultures are

summarized in Table S4.

Appendix

XXII

25 50 1000

100

200

300

400

co-culture

monoculture

****

***

***

***

**

*

Added zinc [µM]

Cellu

lar

zin

c u

pta

ke

[ng

/mg

pro

tein

]

25 50 1000

100

200

300

400

co-culture

monoculture

**

**

Added zinc [µM]

Cellu

lar

zin

c u

pta

ke

[ng

/mg

pro

tein

]

0 25 50 1000

2

4

6

8

****

**

Added zinc [µM]

Zin

c t

ran

sp

ort

rate

[n

mo

l/cm

²]

0 25 50 1000

2

4

6

*

*

*****

***

Added zinc [µM]

Zin

c t

ran

sp

ort

rate

[n

mo

l/cm

²]

25 50 1000

2

4

6

8

* **

Added zinc [µM]

Fra

cti

on

al zin

c r

eso

rpti

on

[%

]

rel. t

o in

cu

bate

d z

inc

25 50 1000

2

4

6

8 *

*****

****

*****

Added zinc [µM]

Fra

cti

on

al zin

c r

eso

rpti

on

[%

]

rel. t

o in

cu

bate

d z

inc

Without BSA With BSA

A

C

E

B

D

F

Figure S1: Zinc resorption of Caco-2 mono- and Caco-2/HT-29-MTX co-cultures in the

presence or absence of albumin (after incubation for 8 h).

Shown are the cellular zinc uptake (A and B) of Caco-2 monocultures and Caco-2/HT-29-MTX co-

cultures relative to cellular protein content after subtracting basal cellular zinc content without (A)

and with 30 mg mL-1 albumin (B) in the basolateral compartment. The zinc transport rates in nmol

zinc per cm2 resorption area of mono- and co-cultures (E and F) are shown. Moreover, fractional zinc

resorption relative to the added zinc concentration of the transport-assay in the absence (E) and

presence of albumin (F) are displayed. Data are presented as means + SD of three independent

experiments. Significant differences to control cells (0 µM zinc) are indicated (*p < 0.05; **p < 0.01;

***p < 0.001; one-way ANOVA with Dunnett’s multiple comparison test). Moreover, significant

differences between Caco-2 monocultures and co-cultures within one zinc and albumin

concentration are indicated (*p < 0.05; **p < 0.01; two-way ANOVA with a Bonferroni post hoc test).

According to a two-way ANOVA with a Bonferroni post hoc test comparing the results within one

added zinc concentration and incubation time, there are significant differences between the zinc

transport rate (100 µM: p < 0.01) of Caco-2 monocultures and the fractional resorption of Caco-2 co-

cultures after zinc treatment for 8 h (25 µM: p < 0.001; 50 µM: p<0.05) with or without albumin in

the basolateral compartment.

Appendix

XXIII

Table S3: Exact amounts of zinc [ng cm-2] transported by Caco-2 monocultures and Caco-

2/HT-29-MTX co-cultures

Caco-2 monoculture

Without BSA With BSA

Added zinc [µM]

0 25 50 100 0 25 50 100

Incubation for 4 h

Data used in Chapter 6

Apical zinc uptake

[ng/cm²] - 73.4±6.6 50.6±152.8 335.9±379.4 - 93.8±45.4 168.8±52.7 198.4±88.5

Cellular zinc

[ng/cm²] 39.4±24.5 51.1±17.1 73.8±24.8 82.9±20.5 59.6±14.5 76.8±21.2 86.9±23.6 95.6±25.9

Resorbed zinc

[ng/cm²] 14.3±20.7 32.2±22.4 59.7±43.1 86.5±29.4 29.9±14.5 48.8±6.5 70.8±8.9 93.3±12.5

Incubation for 8 h

Apical zinc uptake

[ng/cm²] - 134±38.7 106.4±95.3 387.4±532.5 - 98.9±124.6 160.2±133.0 296.1±182.4

Cellular zinc

[ng/cm²] 43.9±19.4 52.9±16.6 64.5±16.7 86.2±17.1 43.6±5.4 45.1±2.1 63.0±18.7 91.6±1.8

Resorbed zinc

[ng/cm²] 35.3±20.0 44.9±16.0 75.4±7.4 100.5±29.1 33.3±14.2 74.2±19.0 128.0±12.7 180.6±2.2

Caco-2/HT-29-MTX co-culture

Incubation for 4 h

Data used in Chapter 6

Apical zinc uptake

[ng/cm²] - 14.1±27.9 150.7±9.6 220.4±52.0 - 89.2±18.5 193.4±16.5 332.2±137.4

Cellular zinc

[ng/cm²] 38.1±11.5 59.7±7.7 87.8±9.8 108.3±6.9 37.6±2.6 67.7±13.5 101.4±31.8 130.8±38.1

Resorbed zinc

[ng/cm²] 10.3±10.9 17.5±10.0 30.7±16.5 71.8±4.8 47.0±7.5 83.7±6.4 105.1±24.7 149.6±35.1

Incubation for 8 h

Data used in Chapter 5 Data used in Chapter 5

Apical zinc uptake

[ng/cm²] - 43.7±14.0 205.6±47.9 383.9±34.6 - 116.7±27.2 270.8±35.9 471.6±112.4

Cellular zinc

[ng/cm²] 35.8±5.8 65.9±6.6 84.5±11.9 121.5±25.1 36.3±0.3 48.6±7.2 60.4±20.3 117.9±33.4

Resorbed zinc

[ng/cm²] 10.9±7.6 36.1±10.6 84.2±9.5 146.1±35.9 72.0±1.8 113.7±14.3 175.1±50.1 231.6±67.2

Shown are the amounts of zinc that are absorbed into the cells (zinc uptake), the cellular zinc content and the amount resorbed into the basolateral compartment in ng zinc per resorption area (in cm²). Data are presented as means ± SD of three independent experiments. Additionally, it is indicated, which data were already shown in Chapter 5 and Chapter 6 of this thesis.

Appendix

XXIV

Table S.4 Fractional zinc resorption [%] and zinc transport rate [nmol cm-2] of Caco-2

monocultures and Caco-2/HT-29-MTX co-cultures with and without basolateral albumin

In vitro model, Incubation time

Without BSA With BSA

Fractional resorption [%] (25 -100 µM apical added zinc)

Caco-2, 4 h 1.6 ± 0.96 – 1.05 ± 0.46 1.8 ± 0.67 – 1.07 ± 0.16

Caco-2, 8 h 2.48 ± 1.12 – 1.35 ± 0.59 3.2 ± 1.2 – 2.2 ± 0.04

Caco-2/HT-29-MTX, 4 h 0.88 ± 0.6 – 0.92 ± 0.2 4.2 ± 0.44 – 1.9 ± 0.5

Caco-2/HT-29-MTX, 8 h 1.8 ± 0.7 – 1.9 ± 0.7 5.8 ± 0.97 – 2.98 ± 1.1

Zinc transport rate [nmol cm-2 ] (0 - 100 µM apical added zinc)

Caco-2, 4 h 0.29 ± 0.3 – 1.2 ± 0.43 0.3 ± 0.3 – 1.29 ± 0.2

Caco-2, 8 h 0.6 ± 0.4 – 1.59 ± 0.58 0.37 ± 0.37 – 2.62 ± 0.19

Caco-2/HT-29-MTX, 4 h 0.15 ± 0.2 – 1.09 ± 0.09 1.13 ± 0.14 – 2.28 ± 0.65

Caco-2/HT-29-MTX, 8 h 0.16 ± 0.14 – 2.2 ± 0.6 1.1 ± 0.03 – 3.54 ± 1.25

Shown is the fractional zinc resorption [%] relative to apical added zinc concentration and zinc

transport rate [nmol cm-2] of Caco-2 monocultures and Caco-2/HT-29-MTX co-cultures of this thesis. Data are presented as means ± SD of three independent experiments.

Appendix

XXV

F. Application of in vitro Caco-2 Monocultures

Table S.5 Application of in vitro Caco-2 monocultures to study zinc-dependent gene expression in enterocytes

Cell model Incubation parameter Analysis Main Outcome Reference

Caco-2 Differentiation time: 14 d 2D

Recombinant expression of myc-tagged hZnT-5B in Caco-2 cells

Addition of ZnCl2 to growth medium: Stepwise increase from 20, 50 and 100 µM each for 7d

Recombinant transfection

Gene expression: RT-PCR

Immunochemical staining

- highest expression of ZTL1 in mouse kidney, brain, duodenum and jejunum

- apical localization of hZTL1 at apical membrane of Caco-2

- hZTL1 (later named ZnT-5B) and MT expression increased in Caco-2-WT cells after prolonged zinc treatment

Cragg et al. 2002 [137]

Caco-2 Cultivation time: 14 d 2D

Human study: 25 mg ZnSO4 /d (placebo Na SO4); duration: 14 d Caco-2: 100 µM or 200 µM ZnCl2

(in DMEM + 10% FCS) for 3 d

Gene expression: RT-PCR

Protein quantification: Immunocytochemistry

- mRNA expression and protein of ZnT-1, ZnT-5, ZnT-5, ZIP-4 in enterocytes (biopsies of ileal mucosa) ↓

- znt-1 ↓ - MT mRNA increased ↑ - mRNA and protein expression in Caco-2 cells was in

agreement of human study - localization of ZnT-5 at apical membrane of human

enterocytes and Caco-2 cells

Cragg et al. 2005 [132]

Caco-2 Cultivation time: 24 h 2D

0-100 µM Zn (in serum-free DMEM) Transient transfection of Caco-2 cells with pEGFP-ZnT5B

- ZnT-5 variant b is a bidirectional zinc transporter and can operate in an efflux mode, increasing cytoplasmic zinc concentration of Caco-2 cells

- upregulation of MT-2 indicates increase of intracellular zinc content in transfected Caco-2 cells

Valentine et al. 2007 [134]

Caco-2 Cultivation time: 24 h, pre-confluent 2D

0-300 µM ZnSO4 or 0-10 µM TPEN (in n.a.) for 6 or 12 h

Gene expression: qPCR

- zinc-dependent mRNA expression of mt-1, dmt-1, zip-4 and znt-1 regulates zinc homeostasis in Caco-2 cells

- zip4 ↑ after zinc depletion with TPEN - mt1 ↑ and znt-1 with added zinc concentration

Shen et al. 2008 [306]

Caco-2 Differentiation time: 11-13 d 2D

Iron/zinc interaction 0-200 µM ZnCl2 or FeCl3, respectively, (in DMEM) for 2h

Zinc Uptake: 65

Zn

- iron uptake was inhibited dose-dependently by zinc - iron increased cellular zinc uptake - analysis suggests that iron and zinc transport by

DMT-1 is not occurring simultaneously

Iyengar et al. 2009 [309]

Appendix

XXVI

Caco-2 Cultivation time: 14 d 2D

3-100 µM ZnCl2 (in DMEM+ 10%FCS)

for 12 or 24 h

Transcriptomic study: Micro-array

Gene expression: qPCR

- zinc-regulated genes were analyzed with an micro-array

- identification of several genes which are regulated zinc-dependent (such as mt-1h, mt-2a, mt-3, mtf-1)

Jackson et al. 2009 [303]

Caco-2 Cultivation time: 21d 3D Transwell (comparison undifferentiated and differentiated cells)

100-800 µM ZnCl2 (in DMEM + 5% FCS)

apical or basolateral incubation) for 24 h

Gene expression: qPCR

- influence of polarization and differentiation of Caco-2 cells on zinc tolerance

- mRNA expression of znt-1 ↑, znt-5, zip1, zip4, mt-1a ↑, mt-1x ↑, mt-2a ↑ after exposure with higher zinc concentrations (100-800 µM; apical or basolateral, respectively)

- under physiologic zinc concentrations (apical: 100 µM; basolateral: 15 µM zinc) only mt-1a ↑

Zemann et al. 2010 [302]

Caco-2 (1) FHs 74 Int cells (2) Cultivation time (1): Undifferentiated (U) (4 d) Differentiated (D) (12 d) 2D

50 µM ZnSO4 (in serum free medium) for 15 min

Zinc Uptake: 65

Zn

Gene expression: qPCR

Western Blot

Biotinylation of surface proteins

- role of zinc exposure on intestinal cells of varying maturity;

- zinc uptake in fetal intestinal cells and undifferentiated cells was higher than in differentiated cells

- ZnT-1 protein and znt-1, znt-2 as well as mt-1 ↑, while zip-4 ↑ in U and ↓ in D Caco-2 cells

- localization of ZIP-4 and ZnT-1 at the plasma membrane of differentiated Caco-2 cells was significantly changed by zinc exposure

Jou et al. 2010 [308]

Caco-2 confluent cells 2D; 3D

0-100 µM ZnSO4 (DMEM +10% FCS)

for 7 d

Zinc Uptake: FAAS

Western blot

- cellular zinc content increased concentration-

dependent (100 µM: 0.4 µg mg-1

protein)

- expression of TJ protein claudin-2 and tricellulin decreased with added zinc concentration

- TEER increased with added zinc concentration

Wang et al. 2012 [307]

Caco-2 (1) IPEC-J2 (2) Cultivation time (1): Pre-confluent (2-3 d) Post-confluent (19-21 d) 2D

0-200 µM ZnSO4 (in DMEM +10% FCS) for 6 h and 24 h

Zinc uptake: FAAS

Gene expression: qPCR

- cellular zinc uptake increases significantly after incubating with 200 µM zinc for 24 h

- zinc incubation of post-confluent Caco-2 cells did not change zip-4 and only showed a trend in mt1a and znt-1 upregulation

- enterocytes’ zinc homeostasis is maintained by expression of these genes

Gefeller et al. 2015 [304]

Appendix

XXVII

Caco-2 (1) IPEC-J2 (2) Cultivation time (1): 21 d 3D Transwell

0-200 µM ZnSO4 (apical or basolateral side, in DMEM + 10% FCS) for 24 h

Gene expression: qPCR

- znt-1 and mt expression ↑ with higher added zinc concentrations basolaterally

- zip-4 expression did not change

Lodemann et al. 2015 [305]

Caco-2 Cultivation time: 24 h

3 or 150 µM zinc (in serum free DMEM) for 24 h

MTF-1 depletion by transient transfection with siRNA

MT-2a stable transfection

Transiently transfection with ZnT-5 promotor

Gene expression: Microarray qPCR

- zinc-dependent expression of MTF-1 dependent genes in MTF-1 depleted Caco-2 compared to CTR: znt-1 ↓ and mt-1b ↓, mt-1e ↓, mt-1g ↓, mt-1h ↓, mt-1m ↓, mt-2a ↓, mt-1a , mt-2a and mt-x did not change

- in MTF-1 depleted cells, zinc incubation changed mRNA expression of genes that are normally not affected by increased cellular zinc, indicating that MT and ZnT-1 are buffering their expression

- MT-2a overexpressed Caco-2 cells showed higher ZnT-5 promoter activity upon zinc uptake

- MTF-1 is controlling intracellular zinc homeostasis by regulating MT and ZnT-1

Hardyman et al. 2016 [310]

3D, three-dimensional; DMEM, Dulbecco’s Modified Eagles Medium; (F)AAS, (flame) atomic absorption spectrometry; FCS, fetal calf serum; HBSS, Hank's Balanced Salt Solution; ICP-MS, inductively-coupled plasma mass spectrometry; n.a., not available; PC, polycarbonate; TEER, transepithelial electrical resistance; TJ, tight junction; Zn, zinc

Appendix

XXVIII

Table S.6 Application of in vitro Caco-2 monocultures to investigate the effect of dietary factors on zinc bioavailability

Cell model Zinc added Food component or Ligand Zinc Quantification Main Outcome Reference

Caco-2 Differentiation time: 10-12 d 2D and 3D Transwell

ZnSO4

FeCl3

(apical: HEPES buffer, basolateral: DMEM + 15% FCS) for 1 h (uptake), 1-5 h (transport)

- Inositolphosphates (IP) (phytic acid): IP3, IP4, IP5, IP6

65Zn,

55Fe

- inhibition of iron and zinc transport by phytate in Caco-2

- reduction of zinc uptake and transport rate correlated with level phosphorylation (IP3 to IP6)

- cellular uptake was analyzed in 2D, transport with 3D transwell

Han et al. 1994 [298]

Caco-2 Differentiation time: 15-18 d 2D

40.22 µM ZnCl2, 88.24 µM FeCl3 or 823.53 µM CaCl2 respectively (in uptake buffer)

- infant formulas: adapted (milk based) and soy-based

- in vitro digestion model AAS

- lower zinc uptake von soy-based than from milk-based infant formulas

- cellular zinc uptake solely observed from digested infant formulas and not from liquid metal solutions

Jovani et al. 2001 [279]

Caco-2 Differentiation time: 19 – 21 d 3D Transwell (PE membrane)

sample c

(apical: soluble mineral fraction, basolateral: HBSS buffer) for 2 h

- raw legumes: white beans, chickpeas, lentils

- effect on cooking of lentils - in vitro digestion model

AAS

- chickpeas yielded the highest amount of transported zinc

- cooking process negatively affected the mineral content of lentils and the soluble zinc fraction decreased

Viadel et al. 2006 [295]

Caco-2 Differentiation time: 21 d 3D Transwell (PES membrane)

sample c

(apical: soluble mineral fraction; basolateral: HBSS buffer) for 2 h

- school meals - in vitro digestion model

FAAS

- iron, copper, zinc and calcium uptake and transport was analyzed

- protein content of meals had no influence on zinc uptake

- negative mineral interaction of iron and zinc: soluble iron decreased and transported zinc; soluble zinc and iron retention

Camara et al. 2007 [291]

Appendix

XXIX

Caco-2 Differentiation time: 14-12 d 2D

25 µM 65

ZnCl2 (in MEM) for 3 h

- phytic acid, tannic acid, tartaric acid, polyphenols (from tea extract and grape juice), wheat, arginine, methionine, histidine

- molar ratio: zinc/dietary ligands (1:1; 1:5; 1:10)

- in vitro digestion model (use of dialysis membrane for incubation of cells with digested samples)

65Zn

- zinc depletion with TPEN increased zinc uptake, but zinc repletion did not affect uptake

- zinc uptake in Caco-2 cells shows a saturable and non-saturable component depending on added zinc concentration

- tannic acid (1:50) enhanced zinc uptake from wheat- and rice-food-matrix

- histidine, phytate, tartaric acid (1:1) and methionine (1:10) resulted in decreased zinc uptake relative to control cells

Sreeniva- sulu et al. 2008 [288]

Caco-2 Differentiation time: 14-21 d 3D Transwell (PE membrane)

sample c

(in salt buffer) for 2 h

- influence of caseinophosphopeptides (CPPs) and milk on zinc uptake from fruit beverages

- in vitro digestion model

AAS - zinc retention, transport and uptake was

higher for milk-containing fruit beverages than for CPPs-based fruit beverages

García-Nebot et al. 2009 [290]

Caco-2 Differentiation time: 21 d 3D Transwell (PC membrane)

sample c

(apical: HEPES,MES, glucose, basolateral: HBSS) for 3 h

- cereals and dephytinized cereals (phytase)

- in vitro digestion model FAAS

- effect of dephytinization on zinc, iron and calcium bioavailability in Caco-2 cells

- zinc and iron solubility and fractional zinc and iron resorption increased after dephytinization of cereals

Frontela et al. 2009 [299]

Caco-2 Differentiation time: 11-13 d 2D

50 µM 65

ZnCl2 Iron-zinc interactions: Zn:Fe (1:1) (in DMEM) for 2 h

- ascorbic acid (1 mM) and phytic acid, tannic acid, tartaric acid, cysteine, histidine, methionine (each 500 µM)

65Zn

- ascorbic acid, tartaric acid and tannic acid increased zinc uptake

- phytic acid and histidine decreased cellular zinc uptake

- increase of iron uptake in presence of methionine, increased also zinc uptake

- without added ligands, zinc inhibited iron uptake into Caco-2

- ligands can modulate iron : zinc-interaction

Iyengar et al. 2010 [292]

Appendix

XXX

Caco-2 Differentiation time: 12-14 d 2D

25 µM 65

ZnCl2 (in MEM) for 3 h

- polyphenol-rich beverages: red wine, green tea, red grape juice

- tannic acid, quercetin, gallic acid, caffeic acid (each 250µM)

- in vitro digestion model (including a rice matrix)

65Zn

- polyphenol-rich beverages increased cellular zinc uptake from digested rice matrix

- tannic acid and quercetin enhanced zinc uptake

Sreeniva- sulu et al. 2010 [287]

Caco-2 Differentiation time: 21-28 d 3D Transwell (PET-HD membrane)

50 µM zinc (in HEPES buffer) for 1 h

- water soluble vitamins: folic acid, nicotinic acid, ascorbic acid, riboflavin, thiamine, pyridoxine

- effect of oxidative species on vitamin-dependent zinc uptake was analyzed

- phytic acid and histidine

FAAS

- zinc transport was slightly enhanced by nicotinic acid and slightly decreased by thiamine, riboflavin, and pyridoxine

- phytic acid significantly decreased zinc uptake compared to control cells, where histidine resulted in a slight increase of zinc uptake

Tupe et al. 2010 [293]

Caco-2 Differentiation time: 21 d 3D Transwell (PC membrane)

sample c

(in apical and basolateral HBSS) for 1 h

- samples from each stage of processing: wheat flour, whole wheat flour; fermented and final product: white bread, whole wheat bread, muffin

- in vitro digestion model

FAAS

- effect of ‘processing’ of baking products on bioavailability of calcium, iron and zinc in Caco-2 cells

- no differences in zinc uptake from fermented dough and after baking

Frontela et al. 2011 [300]

Caco-2 Differentiation time: 21 d 2D

0 - 25 µM ZnCl2

(in minimum essential medium) for 24 h

- varying Zn: PA ratios - in vitro digestion model (red beans,

fish samples)

ICP-MS

MT quantification a

- MT formation was investigated as a proxy for zinc uptake

- PA significantly decreased zinc uptake and MT formation for Zn:PA ratios < 1:5

Cheng et al. 2011 [289]

Caco-2 Differentiation time: 17 d 3D Transwell (collagen coated)

10 µM 65

ZnCl2

(apical: HBSS buffer, basolateral: DMEM) for 1-3 h

- bioactive dietary polyphenols: epigallocatechin-3-gallate (EGCG), green tea extract (GT), and grape seed extract (GSE) (each 46 mg/L)

- phytate (100 µM)

65Zn

- GSE decreased zinc resorption by inhibiting cellular zinc uptake, similar to phytate

- EGCG and GT did not reduce zinc resorption

Kim et al. 2011 [294]

Appendix

XXXI

Caco-2 Differentiation time:14 d 2D

sample c

(in medium) for 6 h

- different rice varieties and zinc biofortified rice (polished or parboiled samples)

- in vitro digestion model

65Zn

- comparison of zinc uptake from biofortified rice with in vivo rat pups

- biofortified rice yielded significant higher net absorption in vitro and in vivo

- net absorption was smaller in Caco-2 but showed the same correlation between different rice samples

Jou et al. 2012 [301]

Caco-2 Differentiation time: 13 d 2D

sample c

for 2 h followed by additional incubation for 10 h

- biofortified wheat (low-phytate mutants with varying zinc content)

- in vitro digestion model

reporter gene

assay b

- positive and negative correlations for zinc bioavailability dependent on zinc : phytate-ratios

Salunke et al. 2012 [296]

Caco-2 Differentiation time: 2D

sample c

(in MEM + 3% FCS) for 6 h

- reduction of phytate content in sorghum(genetic modification)

- in vitro digestion model

65Zn,

59Fe

- comparison of fractional resorption in Caco-2 and in vivo suckling rat pup model

- phytate reduction significantly increased zinc bioavailability in Caco-2 cells comparable to in vivo analysis

Kruger et al. 2013 [297]

Caco-2 Transwell (PE) Differentiation time: 21 d 3D Transwell (PE membrane)

250 µM ZnSO4

(in DPBS) for 2 h

GPAGPHGPPG peptide (derived from Alaska pollock)

AAS

- influence of GPAGPHGPPG peptide on zinc, iron and calcium transport

- GPAGPHGPPG peptide significantly increased mineral transport.

Chen et al. 2017 [286]

Caco-2 Differentiation time: 10 d 2D

50 µM ZnCl2

(in PBS) for 30 min

- amino acids (AAs): glutamate (Glu), lysine (Lys), methionine (Met)

- ZnAAs complexes: ZnGlu, ZnMet, ZnLys

Fluorescent zinc sensor Zinpyr-1

- ZnAAs are probably absorbed by AAs transporters

- zinc uptake into Caco-2 cells is not enhanced by ZnAAs complexes

- results suggest that ZnAAs represent a more efficient war for zinc supplementation that zinc salts; especially for AE patients

Sauer et al. 2017 [157]

3D, three-dimensional; BSA, bovine serum albumin; DMEM, Dulbecco’s Modified Eagles Medium; FAAS, flame atomic absorption spectrometry; FCS, fetal calf serum; HBSS, Hank's Balanced Salt

Solution; HD, high density; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; ICP-MS, inductively-coupled plasma mass spectrometry; IP, inositolphosphate; MEM, minimum essential

medium; n.a., not available; PC, polycarbonate; PE, polyethylene; PES, polyester; Zn, zinc; a

MT formation was analyzed using a cadmium/hemoglobin assay; b reporter gene assay based on the

metal response element (MRE)-binding transcription factor-1 (MTF-1) and MRE luciferase, c mineral bioavailability from the sample solely was examined; no extra zinc added.

Appendix

XXXII

G. Author contributions

This thesis is based on three peer-reviewed publications. As stated before, I designed the

concept of the studies with support of my supervisors Prof. Dr. Dr. Hajo Haase and Dr.

Claudia Keil. I performed most of the experiments, conducted the analysis and wrote the

manuscripts. In the following detailed contributions of the other co-authors from the

respective studies are listed in detail.

Chapter 4:

The transfection of Caco-2 cells as well as preparation and verification of purchased peCalwy

plasmids was planned and executed by me. The following selection process was supervised

by me and executed by Susanne Thomsen, our technician who picked the cell clones during

this process. I designed and performed the following examination of proper cellular

differentiation as well as changes in zinc homeostasis compared to Caco-2-WT cells.

Together with Dr. Claudia Keil, I planned the morphological characterization of cell clones,

which was analyzed with TEM and REM, executed at the Electron Microscopy Core Facility,

Charité Universitätsmedizin Berlin. Immunochemical analyses of tight junction proteins were

done in corporation with Prof. Dr. Dorothee Günzel from the Institute of Clinical Physiology,

Charité Universitätsmedizin Berlin. For this, I prepared and analyzed the cell samples at the

Institute of Clinical Physiology according to the protocol of Prof. Dr. Dorothee Günzel. FRET-

measurements, using laser scanning microscopy, were conducted in close cooperation with

Dr. Burkhard Wiesner from the core facility for cellular imaging at the Leibniz

Forschungsinstitut für molekulare Pharmakologie (FmP) in Berlin. Together with Dr.

Burkhard Wiesner, I designed and coordinated the FRET and FLIM analysis. I conducted

FRET-measurements with LSM and multiphoton LSM myself and undertook the subsequent

data analysis together with Prof. Dr. Dr. Hajo Haase. The published manuscript was written

by me and reviewed by Prof. Dr. Dr. Hajo Haase, Prof. Dr. Dorothee Günzel and Dr. Burkhard

Wiesner.

Chapter 5:

Ayşe Duman executed the impact of FCS on long-term zinc uptake in Caco-2 cells as well as

on expression of important proteins of zinc homeostasis as part of her diploma thesis, which

was designed and supervised by me. All other analysis were planned and performed by me.

ICP-MS measurements were conducted by me in close cooperation with Prof. Dr. Tanja

Schwerdtle from the University of Potsdam. The published manuscript was written by me

and reviewed by Prof. Dr. Dr. Hajo Haase, Prof. Dr. Tanja Schwerdtle and Dr. Claudia Keil.

Chapter 6:

Sophia Straubing performed the visualization of extracellular mucins of HT-29-MTX cells with

FITC-dextran and immunochemical detection of these mucins as part of her diploma thesis.

This diploma thesis was planned and supervised by Dr. Claudia Keil and me. Jenny Koza

analyzed the binding capacity of mucins as well as the influence of zinc saturation of mucins

Appendix

XXXIII

on short-time zinc uptake into Caco-2 cells as part of her diploma thesis, which was designed

and supervised by me. All the other analysis were planned and performed by me. I

performed ICP-MS measurements in close cooperation with Prof. Dr. Tanja Schwerdtle from

the University of Potsdam. The published manuscript was written by me and reviewed by

Prof. Dr. Dr. Hajo Haase, Prof. Dr. Tanja Schwerdtle and Dr. Claudia Keil.

List of Publications

XXXIV

List of Publications

Published Peer-Reviewed Articles

Maares, M. and Haase, H. (2016). "Zinc and immunity: An essential interrelation." Archives of

Biochemistry and Biophysics 611(Supplement C): 58-65.

https://doi.org/10.1016/j.abb.2016.03.022

Maares, M., Keil, C., Thomsen, S., Günzel, D., Wiesner, B., Haase, H. (2018).

"Characterization of Caco-2 cells stably expressing the protein-based zinc probe

eCalwy-5 as a model system for investigating intestinal zinc transport." Journal of

Trace Elements in Medicine and Biology 49: 296-304.

https://doi.org/10.1016/j.jtemb.2018.01.004

Bulut, A., Maares, M., Atak, K., Zorlu, Y., Çoşut, B., Zubieta, J., Beckmann, J., Haase, H.,

Yücesan, G. (2018). "Mimicking cellular phospholipid bilayer packing creates

predictable crystalline molecular metal–organophosphonate macrocycles and cages."

CrystEngComm 20(15): 2152-2158.

https://doi.org/10.1039/c8ce00072g

Sumarokova, M., Iturri, J., Weber, A., Maares, M., Keil, C., Haase, H., Toca-Herrera, J. L.

(2018). "Influencing the adhesion properties and wettability of mucin protein films by

variation of the environmental pH." Scientific Reports 8(1): 9660.

https://doi.org/10.1038/s41598-018-28047-z

Maares, M., Duman, A., Keil, C., Schwerdtle, T., Haase, H. (2018). "The impact of apical and

basolateral albumin on intestinal zinc resorption in the Caco-2/HT-29-MTX co-culture

model." Metallomics 10(7): 979-991.

https://doi.org/10.1039/C8MT00064F

Maares, M., Keil, C., Koza, J., Straubing, S., Schwerdtle, T., Haase, H. (2018). "In vitro Studies

on Zinc Binding and Buffering by Intestinal Mucins." International Journal of

Molecular Sciences 19(9): 2662.

https://doi.org/10.3390/ijms19092662

Oral Presentations

Zinc-UK/Zinc-net Conference; 21st-22nd November 2016 in Belfast, United Kingdom

Maares, M., Keil, C., Haase, H.

In vitro-studies to investigate the impact of mucins on intestinal zinc resorption

Biomarker in Prevention and Nutrition; 6thDezember 2016 in Potsdam, Germany

Maares, M., Keil, C., Haase, H.

In vitro-Intestinalmodell zur Untersuchung der Bioverfügbarkeit von Zink

List of Publications

XXXV

Congress of the regional association of German Food Chemists Nord-East (LChG, GdCh);

6th March 2018 in Berlin, Germany

Maares, M, Keil, C., Haase, H.

In vitro-Studien zur Untersuchung der Zink-Bioverfügbarkeit und intestinalen Zink-

Resorption

47th Congress of the German Society of Food Chemists (LChG, GdCh); 17th-19th September

2018 in Berlin, Germany;

Maares, M, Keil, C., Haase, H.

Analytische Ansätze zum Monitoring der intestinalen Zinkaufnahme mit Zink-

Biosonden

Poster Presentations

44th Congress of the German Society of Food Chemists (LChG, GdCh); 14th-16th September

2015 in Karlsruhe, Germany

Maares, M., Keil, C., Haase, H.

In vitro-Enterozytenmodell zur Untersuchung der intestinalen Bioverfügbarkeit von

Zink

32nd Annual Conference of the German Society of Minerals and Trace Elements; 13th – 15th

October 2016 in Berlin, Germany

Maares, M., Duman, A., Keil, C., Haase, H.

Enterocyte vs. goblet cell: Investigation of the intestinal bioavailability of zinc

5th Meeting of the International Society for Zinc in Biology (ISZB), in collaboration with Zinc-

Net, 18th- 22nd June 2017 in Cyprus, Greece

Maares, M., Duman, A., Straubing, S., Keil, C., Haase, H.

Zinc-buffering by intestinal glycoproteins - in vitro-studies to investigate the role of

mucins in zinc-resorption (Poster Prize)

33h Annual Conference of the German Society of Minerals and Trace Elements; 28th – 30th

September 2018 in Aachen, Germany

Maares, M., Thomsen, S., Keil, C., Günzel, D., Wiesner, B., Haase, H.

Stable expression of eCalwy-FRET sensors in Caco-2 cells: a practical tool to monitor

free zinc in enterocytes

34th Annual Conference of the German Society of Minerals and Trace Elements; 7th – 9th June

2018 in Jena, Germany

Maares, M., Duman, A., Keil, C., Schwerdtle, T., Haase, H.

The ambiguous role of albumin on zinc resorption in a human in vitro intestinal cell

model

List of Publications

XXXVI

47th Congress of the German Society of Food Chemists (LChG, GdCh); 17th-19th September

2018 in Berlin, Germany;

Maares, M., Keil, C., Koza, J., Straubing, S., Haase, H.

Relevanz von Muzinen für die intestinale Zinkresorption

47th Congress of the German Society of Food Chemists (LChG, GdCh); 17th-19th September

2018 in Berlin, Germany;

Maares, M., Löher, L., Keil, C., Haase; H., Toca-Herrera, J., Iturri, J.

Strukturelle Untersuchungen von intestinalen Becherzellen mittels

Rasterkraftmikroskopie

11th‎ “Doktorandensymposium”‎and‎DRS‎ Seminar for Presentations in Biomedical Sciences;

21th September 2018 in Berlin, Germany

Maares, M.; Keil, C.; Haase, H.

In vitro studies on luminal and basolateral factors influencing intestinal zinc

resorption by Caco-2/HT-29-MTX co-cultures

20th International Congress on In vitro Toxicology; 15th- 18th October in Berlin, Germany

Maares, M., Keil, C., Haase, H.

In vitro intestinal model Caco-2/HT-29-MTX - Investigation of luminal and basolateral

factors influencing intestinal zinc resorption

Grants and Prizes

Zinc Net Travel Grant for the Zinc-UK/Zinc-net Conference2016;

21st-22nd November 2016 in Belfast, United Kingdom

Travel Grant from the Berlin Institute of Technology for the 5th Meeting of the International

Society for Zinc in Biology (ISZB) in collaboration with Zinc-Net;

18th- 22nd June 2017 in Cyprus, Greece

Metallomics Poster Prize

5th Meeting of the International Society for Zinc in Biology (ISZB) in collaboration with

Zinc-Net; 18th- 22nd June 2017 in Cyprus, Greece

Trainee Grant from COST Action TD1304 for the 33th Annual Conference of the German

Society of Minerals and Trace Elements;

28th - 30th September2018 in Aachen, Germany

Travel Grant from the Berlin Institute of Technology for the 20th International Congress on In

Vitro Toxicology;

15th- 18th October in Berlin, Germany

Acknowledgements

XXXVII

Acknowledgements

Ein besonderer Dank geht an Prof. Dr. Dr. Hajo Haase für die Unterstützung und Betreuung

während meiner gesamten Promotionszeit. Danke für die Eröffnung dieses spannenden

Themas, die Möglichkeit in deiner Arbeitsgruppe zu arbeiten und natürlich die vielen

konstruktiven Diskussionen und Ratschläge. Ich durfte in der Zeit als Doktorandin bei dir

einige interessante Tagungen besuchen und Vorträge halten - Vielen Dank für die stetige

Förderung und Motivation!

Ich danke Prof. Dr. Anna Kipp für die Begutachtung dieser Arbeit.

Dr. Claudia Keil möchte ich für die wertvollen Diskussionen, sowie die stetige Hilfe bei

praktischen Fragen und technischen Anwendungen danken. Vielen Dank für deine

motivierenden Worte und die vielen schönen und Schokoladenreichen Stunden in unserem

Büro. Nicht zu vergessen, unsere gemeinsamen Tagungen, bei denen wir einige amüsante

Abende miteinander verbracht haben.

Ich möchte außerdem unseren Kooperationspartnerinnen und -partnern für ihre

Unterstützung und Beteiligung an den jeweiligen Publikationen, auf denen diese Arbeit

beruht, danken. Ich danke Prof. Dr. Dorothee Günzel vom Institut für Klinische Physiologie,

Charité Universitätsmedizin Berlin, für die praktische Hilfe und fachliche Beratung bei den

immunchemischen Analysen der Tight junction Proteine. Prof. Dr. Tanja Schwerdtle danke

ich für die Ermöglichung und das Vertrauen, die ICP-MS-Messungen in ihrem Arbeitskreis an

der Universität Potsdam durchzuführen. Besonderer Dank geht auch an Dr. Talke Marschall

und Dr. Sören Meyer für die technische Unterstützung bei den Messungen. Dr. Burkhard

Wiesner, von der Core Facility für Cellular Imaging am Leibniz Forschungsinstitut für

molekulare Pharmakologie (FmP) in Berlin, möchte ich für die Unterstützung bei der FRET-

Messung des Caco-2-eCalwy Klons und die hilfreichen Diskussionen über die FRET und FLIM-

Analysen danken.

Ich möchte mich an dieser Stelle bei meinen Kolleginnen und Kollegen für das gute

Arbeitsklima, die schöne Zeit und die zahlreichen gemeinsam verspeisten Kuchen bedanken.

Ich denke gerne an die letzten vier Jahre zurück. Conny Richter danke ich für die Hilfe bei

jeglichen elementar-analytischen Fragen, Susanne Thomsen für die Unterstützung bei der

Selektion der Caco-2-eCalwy Klone, Timo Neubert für den verlässlichen Beistand bei kleinen

und großen technischen Problemen. Die lustigen Geschichten von Thomas Bernhardt haben

die Arbeit im Labor erleichtert. Von dir habe ich außerdem gelernt, wie man einen

Autoreifen wechselt! Ein großes Dankeschön geht an Wiebke Alker für die aufheiternden

Gespräche, die emotionale Unterstützung und die stetigen Versuche ein gemeinsames

Mittagessen zu organisieren!

Ich danke außerdem meinen Diplomandinnen, Ayşe Duman, Stefanie Haberecht, Sofia

Straubing, Cathrin Schröder, sowie Diplomanden, Tobias Hensel und Leif Löher, die ich

während meiner Promotionszeit betreuen durfte. Es hat mir unglaublich viel Spaß gemacht,

Acknowledgements

XXXVIII

mit euch zusammenzuarbeiten. Danke für die vielen guten Ergebnisse und schönen

Erinnerungen.

Meinen Freundinnen Viktoria Ganß, Olga Wesker und Jessica Dietrich möchte ich für den

emotionalen Rückhalt und die Ablenkung vom Laboralltag in den letzten vier Jahren danken.

Ein besonderer Dank gilt meiner Familie. Meinen Eltern danke ich für die stetige und

bedingungslose Unterstützung, die Ermöglichung meines gesamten Studiums und das viele

Korrekturlesen dieser Arbeit. Ich danke meinen Geschwistern Phoebe und Seymour Maares

für ihren andauernden Glauben in mich und dass sie mich immer zum Lachen bringen.

Meiner Schwester möchte ich außerdem für die vielen (fachlichen) Diskussionen, die Hilfe

beim Korrigieren und Gegenlesen meiner Texte und die gemütliche Herberge während

meiner Zeit in Wien danken. Christoph Leo danke ich für den Beistand, die Rücksichtnahme

und die tägliche Aufheiterung während der Anfertigung dieser Arbeit.