is there an association between vitamin d status …
TRANSCRIPT
IS THERE AN ASSOCIATION BETWEEN VITAMIN D STATUS
DURING PREGNANCY AND CHILD`S RISK FOR DEVELOPING
TYPE 1 DIABETES
L. Reinert, May 2010
Department of Food and Environmental Sciences
Linnea Reinert
Master’s thesis
University of Helsinki
HEBIOT
Biotechnology
May 2010
HELSINGIN YLIOPISTO HELSINGFORS UNIVERSITET UNIVERSITY OF HELSINKI Tiedekunta/Osasto Fakultet/Sektion Faculty
Faculty of Agriculture and Forestry
Laitos Institution Department
Department of Food and Environmental Sciences HEBIOT
Tekijä Författare Author
Linnea Reinert Työn nimi Arbetets titel Title
Is there an Association between Vitamin D Status During Pregnancy and Child`s Risk for Developing Type 1 Diabetes
Oppiaine Läroämne Subject
Biotechnology Työn laji Arbetets art Level
Master’s Thesis
Aika Datum Month and year
May 2010
Sivumäärä Sidoantal Number of pages
61
Tiivistelmä Referat Abstract
Vitamin D is either obtained through synthesis in the skin due to UVB-light (290-315 nm) or from the diet. The hydroxylased metabolite 25-hydroxyvitamin D (25(OH)D) is the metabolite to measure when vitamin D status wants to be determined. The active form of vitamin D is 1,25-dihydroxyvitamin D (1,25(OH) D) which interacts with a large set of tissue cells (especially bone) through its nuclear receptor the vitamin D receptor (VDR). Vitamin D deficiency can lead to rickets in children and osteoporosis or osteomalacia in adults.
Type 1 Diabetes (T1D) is an autoimmune disease which is caused by the destruction of the pancreatic -cells. The disease has genetic and environmental features but the whole mechanism of disease development is still unknown. The prevalence of T1D is constantly growing in the whole world. Therefore it is important to study possible environmental factors that can eventually serve as pathogenesis modifiers. Vitamin D and T1D have been associated among others because there is a seasonal and geographical variation in T1D incidence; more cases have been identified in the North and during winter. The aim of this study was to investigate if the serum 25(OH)D status during first trimester of pregnancy is associated with T1D development in the offspring. The subjects where mothers of T1D children (N=310) and the controls were mothers of healthy children (N=310). Serum samples were obtained from the Finnish Maternity Cohort (FMC) and analyzed for S-25(OH)D. S-25(OH)D measurement was performed with an indirect enzyme immunoassay (EIA).
No significant (p>0.05) difference was seen between S-25(OH)D mean concentrations in cases and controls. The mean concentration of cases was 43.3 ± 15.9 nmol/l and 43.0 ± 15.5 nmol/l (mean ± standard deviation (SD)) of controls. Insufficient and deficient S-25(OH)D status was seen in 72% of the whole study population. As a result of this study it has been shown that the S-25(OH)D status during first trimester of pregnancy is not associated with T1D development in the offspring. Samples from later stages of pregnancy could be analyzed to determine if the overall status during pregnancy has an effect on T1D development in the offspring. Considering the possible health outcomes of vitamin D insufficiency, recommended vitamin D supplementation should be raised to improve maternal and fetal health. Avainsanat Nyckelord Keywords
25-hydroxyvitamin D, Type 1 Diabetes, Pregnancy, EIA, Vitamin D deficiency Säilytyspaikka Förvaringsställe Where deposited
Department of Food and Environmental Sciences Muita tietoja Övriga uppgifter Further information
Supervisor and responsible professor: Christel Lamberg-Allardt and Teemu Teeri
HELSINGIN YLIOPISTO HELSINGFORS UNIVERSITET UNIVERSITY OF HELSINKI Tiedekunta/Osasto Fakultet/Sektion Faculty
Maatalous-metsätieteellinen tiedekunta
Laitos Institution Department
Elintarvike- ja ympäristötieteiden laitos HEBIOT
Tekijä Författare Author
Linnea Reinert Työn nimi Arbetets titel Title
Is there an Association between Vitamin D Status During Pregnancy and Child`s Risk for Developing Type 1 Diabetes
Oppiaine Läroämne Subject
Biotekniikka Työn laji Arbetets art Level
Pro Gradu tutkielma
Aika Datum Month and year
Toukokuu 2010
Sivumäärä Sidoantal Number of pages
61
Tiivistelmä Referat Abstract
D-vitamiinia saadaan joko ravinnosta tai UVB-säteilyn (290-315 nm) avulla kehon omasta tuotannosta. Kun kehon D-vitamiinitaso halutaan määrittää, mitattava metaboliitti on 25-hydroksivitamiini D (25(OH)D). D-vitamiinin aktiivinen metaboliitti, joka vaikuttaa laajaan määrään eri kudoksia (erityisesti luustoon) tumareseptorinsa D-vitamiini reseptorin (vitamin D receptor (VDR)) kautta, on 1,25-dihydroksivitamiini D (1,25(OH) D). D-vitamiinin puutos voi johtaa lapsilla riisitautiin ja aikuisilla osteoporoosiin tai osteomalasiaan.
Tyypin 1 diabetes (T1D) on autoimmuunisairaus, joka aiheutuu haiman beetasolujen tuhoutumisesta. Taudilla on sekä perinnöllisiä tekijöitä että ympäristötekijöitä, mutta kaikkia taudin syntymekanismin vaiheita ei vieläkään tunneta täysin. Tyypin 1 diabeteksen esiintyvyys lisääntyy koko ajan ympäri maailmaa. Tästä syystä on tärkeä kartoittaa mahdollisia ympäristötekijöitä, jotka saattavat toimia taudin kehityksen muokkaajina. D-vitamiinin ja T1D:n yhteyttä on tutkittu muun muassa siksi, että T1D:n esiintyvyydessä on todettu vuodenaikaisvaihtelua sekä maantieteellistä vaihtelua; tapauksia on todettu enemmän pohjoisessa sekä talvisin. Tutkimuksen tarkoitus oli selvittää onko raskauden ensimmäisen kolmanneksen aikaisella 25(OH)D tasolla yhteyttä siihen sairastuuko lapsi tyypin 1 diabetekseen. Tapauksina toimivat äidit, joilla on diabeetikkolapsi (N=310) ja verrokkeina äidit, joilla on terve lapsi (N=310). Seeruminäytteet ovat Finnish Maternity Cohort (FMC) seerumipankista ja niistä määritettiin seerumin 25-hydroksivitamiini D (S-25(OH)D) konsentraatio epäsuoran entsyymi-immuunimäärityksen (enzyme immunoassay (EIA)) avulla.
Tilastollisesti merkittävää (p<0.05) eroa seerumin 25(OH)D pitoisuuksissa tapausten ja verrokkien välillä ei ollut. Tapausten S-25(OH)D pitoisuuden keskiarvo oli 43.3 ± 15.9 nmol/l ja verrokkien oli 43.0 ± 15.5 nmol/l (keskiarvo ± keskihajonta). Riittämätön sekä puutteellinen S-25(OH)D pitoisuus tavattiin 72%:lla kaikista tutkittavista. Tämä tutkimus osoittaa, että raskauden ensimmäisen kolmanneksen S-25(OH)D pitoisuus ei ole yhteydessä siihen, sairastuuko lapsi tyypin 1 diabetekseen. Jotta saataisiin määritettyä, onko raskauden aikaisella D-vitamiinitasolla yleisesti yhteyttä T1D:n kehittymiseen lapsessa, näytteitä raskauden eri vaiheista voisi analysoida. Koska D-vitamiinin puutoksella on mahdollisia terveydelle haitallisia vaikutuksia, suosituksia tulisi nostaa parantamaan sekä äidin, että sikiön terveyttä. Avainsanat Nyckelord Keywords
25-hydroksivitamiini D, tyypin 1 diabetes, raskaus, EIA, D-vitamiinin puutos Säilytyspaikka Förvaringsställe Where deposited
Elintarvike- ja ympäristötieteiden laitos Muita tietoja Övriga uppgifter Further information
Supervisor and responsible professor: Christel Lamberg-Allardt and Teemu Teeri
4
TABLE OF CONTENTS
ABBREVIATIONS .............................................................................................. 5
1 INTRODUCTION ............................................................................................. 9
1.1 Vitamin D ................................................................................................... 9 1.1.1 General definition................................................................................... 9 1.1.2 Metabolic Pathway of Vitamin D ........................................................... 14 1.1.3 The biological functions of vitamin D ..................................................... 16 1.1.4 Vitamin D Deficiency ........................................................................... 26 1.1.5 Vitamin D recommendations in Finland .................................................. 28 1.1.6 Vitamin D status in the healthy Finnish population .................................. 29 1.1.7 Vitamin D status and disease in the Finnish population ............................ 34
1.2 Type 1 Diabetes mellitus............................................................................ 36 1.2.1 T1D pathogenesis and genetics .............................................................. 36 1.2.2 Prevalence and trend in Finland ............................................................. 37
1.3 Vitamin D and Type 1 Diabetes ................................................................. 38 1.3.1 Genetic association between vitamin D and Type 1 Diabetes .................... 39 1.3.2 Animal studies ..................................................................................... 39 1.3.3 Human studies ..................................................................................... 40 1.3.3.1 Maternal vitamin D intake and status................................................... 40 1.3.3.2 Early life vitamin D intake and status .................................................. 42 1.3.3.3 Vitamin D status in adults ................................................................... 44
1.4 Vitamin D Measurement methods ............................................................. 45 1.4.1 Different methodology .......................................................................... 45 1.4.2 The International Quality Assessment Scheme for Vitamin D metabolites (DEQAS) .................................................................................................... 48
2 OBJECTIVES................................................................................................. 50
2.1 Aim of this study ....................................................................................... 50
3 MATERIALS AND METHODS ...................................................................... 51
3.1 Finnish Maternity Cohort ......................................................................... 51 3.2 Samples .................................................................................................... 51 3.3 25-hydroxyvitamin D measurement ........................................................... 52 3.4 Laboratory equipment .............................................................................. 56 3.5 Statistical analysis ..................................................................................... 56
4 RESULTS ....................................................................................................... 56
4.1 Vitamin D deficiency in the whole study population ................................... 56 4.2 Vitamin D status comparison between mothers with healthy offspring and mothers with T1D offspring ............................................................................ 57
5 DISCUSSION ................................................................................................. 58
6 CONCLUSIONS ............................................................................................. 60
7 ACKNOWLEDGEMENTS ............................................................................. 61
9 REFERENCES ............................................................................................... 61
5
Abbreviations
1,25(OH) D 1,25-dihydroxyvitamin D
1-OHase 25-hydroxyvitamin D-1 -hydroxylase, CYP27B1
24-OHase 25-hydroxyvitamin D-24-hydroxylase, CYP24A1
25(OH)D 25-hydroxyvitamin D
25-OHase vitamin D-25-hydroxylase, CYP27A1, CYP2R1, CYP3A4,
CYP2J3
7-DHC 7-dehydrocholesterol
ALTM all-laboratory trimmed mean
AMI acute myocardial infarction
APC antigen presenting cell
API-ES atmospheric pressure ionization-electro spray
BMI body mass index
BUA broadband ultrasound attenuation
CALEX girl-cohort for calcium and vitamin D intervention
CD4 cluster of differentiation 4
CD8 cluster of differentiation 8
CPBA competitive protein binding assay
CTLA4 cytotoxic T-lymphocyte associated protein 4
CV coefficient of variance
D vitamin D
D vitamin D
DAISY diabetes autoimmunity study in the young
DBP vitamin D binding protein
DC dendritic cell
DEQAS International Assessment Scheme for vitamin D metabolites
DIPP type 1 diabetes prediction and prevention
DISS diabetes incidence study in Sweden
EIA enzyme immunoassay
EURODIAB cross-sectional study of T1D patients in Europe
FasL Fas ligand
Fc constant fragment
FFQ food frequency questionnaire
6
FGF-23 fibroblast growth factor 23
FINRISK population survey on risk factors of chronic, noncommunicable
diseases
FMC Finnish Maternity Cohort
FOXP3 forkhead box P3
FPG fasting plasma glucose
FPI fasting plasma insulin
GADA glutamate decarboxylase autoantibody
GDM gestational diabetes mellitus
GM-CSF granulocyte-macrophage colony stimulating factor
HCl hydrochloric acid
HIV human immunodeficiency virus
HLA human leukocyte antigen
hnRNP heterogeneous nuclear ribonucleoprotein
HOMA homeostatic model assessment
HPLC high performance liquid chromatography
HRP horseradish peroxidase
HSA human serum albumin
IA islet autoimmunity
IA2 insulin autoantigen 2
IAA insulin autoantibody
ICA islet cytoplasmic autoantibody
ICA512/IA2A islet cell autoantibody 512/protein tyrosine phosphatase IA2
IDBP intracellular vitamin D binding protein
IDDM insulin dependent diabetes mellitus
IFG impaired fasting glycaemia
IFN interferon
Ig immunoglobulin
IGT impaired glucose tolerance
IL interleukin
INS insulin
ISI insulin sensitivity index
LBD ligand binding domain
LC-MS liquid chromatography-mass spectrometry
LDL low-density lipoprotein
7
LNCaP prostate cancer cell line (lymph node carcinoma of the prostate)
LPS lipopolysaccharide
MAPK mitogen-activated protein kinase
MCF-7 breast cancer cell line (Michigan Cancer Foundation-7)
MED minimal erythemal dose
Met methionine
MHC major histocompability complex
MM method mean
MyD88 Myeloid differentiation primary response gene 88
NF- B nuclear factor- B
NOD non-obese diabetic
OPTIFORD project to study vitamin D fortification
PABC pregnancy induced breast cancer
pre-mRNA pre-messenger RNA
PTH parathyroid hormone
RANTES regulated upon activation, normal T-cell expressed and secreted
RIA radioimmunoassay
RXR retinoic X receptor
SD standard deviation
Ser serine
S-iPTH serum intact PTH
SOS speed of sound
SPF sun protecting factor
SPSS Statistical Package for the Social Sciences
STRIP intervention project for coronary heart disease risk factors
(sepelvaltimotaudin riskitekijöiden interventioprojekti)
T1D type 1 diabetes
T2D type 2 diabetes
Th T helper
TLR4 toll like receptor 4
TMB tetramethylbenzidine
TNF- tumor necrosis factor-
TR thyroid receptor
Tyr tyrosine
UVB ultraviolet light at wavelength 290-315 nm
8
VDR vitamin D receptor
VDRE vitamin D responsive element
VNTR variable number of tandem repeats
VRN National Nutrition Council (valtion ravitsemusneuvottelukunta)
ZnTA zinc transporter antibody
9
1 Introduction
1.1 Vitamin D
1.1.1 General definition
Vitamin D is a pro-hormone which is either synthesized in the skin due to sunlight or
artificial ultraviolet light (UVB=290-315 nanometers) exposure, or obtained from the
diet (reviewed in Holick, 2003). Its major roles are in calcium homeostasis where it
increases the absorption of calcium from the intestine. When the calcium concentration
is too low, vitamin D signals through osteoblasts for osteoclast precursors to dissolve
calcium (reviewed in Holick, 2003). The most important role of vitamin D is to regulate
the metabolism of calcium and phosphate (reviewed in Barnes et al., 2006). It enhances
the absorption of calcium and phosphorus from the small intestine into the enterocytes
(see Deluca and Cantorna, 2001). When parathyroid hormone is present vitamin D
mobilizes calcium from the bone and improves the renal reabsorption of calcium (see
Deluca and Cantorna, 2001). Other tissues are also affected by vitamin D. It has been
shown that vitamin D also regulates cell growth, proliferation and differentiation (see
Holick, 2005). Cells regulated by vitamin D are among others enterocytes, osteoblasts,
keratinocytes, parathyroid cells, monocytes and lymphocytes (see Deluca and Cantorna,
2001). Other cells which are thought to be regulated by vitamin D are islet cells of the
pancreas, endocrine cells, ovarian cells, developing myoblasts, skin fibroblasts and
aortic endothelial cells (see Deluca and Cantorna, 2001).
Vitamin D is a 9,10-secosteroid and it has two main forms; vitamin D (D ) and
vitamin D (D ). D is derived from a plant sterol ergosterol and D is formed from 7-
dehydrocholesterol (Figure 1 and Figure 2). D differs from D in the way that it has
one more double bond and one more methyl group than D .
10
A) B)
Figure 1. Cholesterol (A) and Cholecalciferol (Vitamin D ) (B).
A) B)
Figure 2. Ergosterol (A) and Ergocalciferol (Vitamin D ) (B).
Vitamin D has two major metabolites; 25-hydroxyvitamin D (25(OH)D) and 1 ,25-
dihydroxyvitamin D (1,25(OH) D) (Figure 3) (Blunt et al., 1968, Haussler et al., 1968).
25(OH)D is an interesting metabolite to measure because it represents the overall
vitamin D status (see Bouillion et al., 1998). 1,25(OH) D is the active form of vitamin
D and its activity mainly goes via the nuclear vitamin D receptor (VDR), which is
present in many tissues (reviewed in Deluca and Cantorna, 2001).
A) B)
Figure 3. 1 ,25-dihydroxyvitamin D (1,25(OH) D ) (A) and 25-hydroxyvitamin D (25(OH)D )
(B).
11
Naturally vitamin D is found in a few foods like fish and fish liver oils. The form of
vitamin D In fish is D . In mushrooms on the other hand the main form found is D
(Table 1). The estimated vitamin D content of different foods varies from author to
author and there are no exact values for that. Holick (2007) estimated the content of
vitamin D in 100 grams (g) of sardine to be 7.5 micrograms (µg) whereas Bouillion et
al. (1998) estimated the value to be as high as 40.0 µg. Egg yolk has believed to contain
much vitamin D but now it has been estimated to content only 2.2 µg per 100 g (Table
1). Eel, lamprey and sander seem to contain more vitamin D that other fish. Chanterelle
seems to be the best plant source for vitamin D. Fish roe, which is a very popular food
in Finland also contains a good amount of vitamin D (9.5 µg/100 g). The best meat
source for vitamin D is chicken with a low content of 0.7 µg/100 g.
Table 1. Vitamin D content in different foods in Finland. Source: www.fineli.fi.
Food Concentration µg / 100 g
Fish liver oil (Möller, 100 ml) (D ) 200.0
Eel (D ) 25.6
Lamprey (D ) 25.6
Sander (D ) 24.6
Lavaret (D ) 22.1
Baltic Herring (D ) 18.7
Bream (D ) 14.0
European Cisco (D ) 13.0
Chanterelle (D ) 12.8
Fish average, Baltic Herring/European Cisco/Perch/Pike (D ) 10.3
Roach (D ) 10.0
Baby`s milk formula, mean, powder (D ) 9.6
Fish roe, mean (D ) 9.5
Rainbow trout file, low fat, oven baked (D ) 9.1
Salmon file, fried (D ) 8.7
Anchovy (D ) 5.7
Rufous milk-cap (D ) 5.5
False morel (D ) 4.4
Mushroom, boletus, russula (D ) 4.4
Pike (D ) 3.0
Penny bun bolet (D ) 2.9
Egg-butter (D , D ) 2.8
12
Fried egg (D , D ) 2.5
Boiled egg (D , D ) 2.2
Trout (D ) 2.1
Flounder (D ) 0.7
Chicken (D ) 0.7
French salad dressing, oil, egg yolk, sugar, salt (D , D ) 0.5
Many foods have been fortified with vitamin D (Table 2.) due to the seasonal variation
in vitamin D synthesis in the skin and the growing prevalence of vitamin D deficiency
and insufficiency in the population all over the world. Chen et al. (2007) studied the
vitamin D content of fortified milk samples from the United States and British
Columbia, Canada and found out that 49% of them contained less than 80% of the
labeled content and 14% did not contain any vitamin D. This elicits the question if
consumers who rely on fortified products really get adequate vitamin D requirement. In
Finland milk products and spreads are fortified with vitamin D except for organic
products which are not fortified (www.fineli.fi). The products in Finland are fortified
with vitamin D . Spreads and margarines usually contain vitamin D in concentrations
between 5.3 µg/100 g and 10 µg/100 g. Milk products on the other hand contain vitamin
D concentrations of 0.5 µg/100 g. Milk products and spreads are important vitamin D
sources for the Finnish population.
Table 2. Vitamin D content in different fortified foods in Finland. Source: www.fineli.fi.
Fortified foods Concentration µg/100 g
Spread 23% becel pro-activ extra light 10.0
Spread 30% keiju 10.0
Spread 35% becel pro-activ 10.0
Spread70 % mean 10.0
Spread 70% keiju lactose free 10.0
Margarine 40% keiju light lactose free 10.0
Margarine 60 % kultarypsi 10.0
Margarine 60% keiju normal salty 10.0
Margarine, flora light 40 10.0
Spread mixture 30 % oivariini balansia 10.0
Baking margarine 80% liquid, mean 9.2
13
Margarine 40 % rainbow/eldorado 9.2
Margarine 40 % mean 9.2
Margarine 60 % mean 8.8
Spread mixture 30% kevyt levi 5.3
Baby`s milk formula, mean, ready to use 1.2
Acidophilus sour milk, 2,5% fat, a-sour milk/ab-sour milk 0.5
Gefilus-sour milk 1% fat 0.5
Yogurt, unseasoned, non-fat, low lactose 0.5
Yogurt, unseasoned, a+, fat 2,5 %, low lactose 0.5
Semi-skimmed milk 1,5% fat 0.5
Semi-skimmed milk 1,5% fat, low lactose 0.5
Buttermilk 0,3% fat 0.5
Milk, calcium, skimmed 0.5
Milk drink, lactose free, fat 1,5 g 0.5
Milk drink, lactose free, skimmed 0.5
Berry-/fruit-yogurt, fat 2 %, low lactose 0.5
Skimmed milk 0.5
Skimmed milk, low lactose 0.5
Skimmed sour milk 0.5
Skimmed milk, low lactose 0.5
Whole milk 3,5% fat 0.5
One-milk 1% fat 0.5
Vitamin D supplements are usually recommended during the winter in Finland and
different concentrations are available. Capsules containing 2.5 µg to 25 µg are available
without recipe and capsules containing 200 µg are available when prescripted (Table
3.).
Table 3. Vitamin D content in supplements in Finland. Source: www.yliopistonapteeki.fi.
Supplements Concentration (µg)
Vitamin D capsules 2.5, 5.0, 7.5, 10.0, 20.0, 25.0, 200
Vitamin D liquid supplement (ml)
61.0
Multivitamin capsules 5.0, 7.5, 10.0
14Table 4. Vitamin D content from sun exposure. Modified from Holick, 2007.
Other Concentration (µg)
Exposure to sunlight, UVB radiation 0.5 minimal erythemal dose* (D3) 75.00
*5-10 min exposure of the arms and legs to direct sunlight, depends on time of day, season, latitude, skin
sensitivity
The amount of vitamin D obtained from supplements and UVB exposure is illustrated in
Table 3 and Table 4. Interestingly Chen et al. (2007) also found out, when studying the
vitamin D composition in farmed and wild fish, that farmed salmon also contained
vitamin D . This can be due to feed which is used, but it can also elicit the question if
the vitamin D content of fish and other foods has to be reevaluated.
1.1.2 Metabolic Pathway of Vitamin D
The two forms of vitamin D are metabolized almost similarly in the human body. The
metabolic pathway of D starts (when synthesized in the skin) with the conversion of 7-
dehydrocholesterol (7-DHC) to previtamin D in the lipid bilayer of epidermal
keratinocytes and dermal fibroblasts through UVB radiation (Holick et al., 1977,
Holick, 2002). Due to additional UVB radiation previtamin D is converted to inactive
vitamin D metabolites as lumisterol and tachysterol (see Wolpowitz and Gilchrest,
2006). This is the bodys precaution for preventing intoxication. Previtamin D is heat-
induced converted to vitamin D which no longer is sterically acceptable for the plasma
membrane and gets ejected to the extracellular space (see Holick, 2002).
Vitamin D and D are metabolized nearly similar in the human body, but more
is known about vitamin D . Differences have been found in vitamin D binding protein
affinity (see Houghton and Vieth, 2006). One difference in metabolic activity is seen in
the metabolites 1,24,25(OH) D and 1,24,25(OH) D . The D metabolite has the ability
to bind to the vitamin D receptor but the D metabolite is an inactive metabolite unable
to bind (see Houghton and Vieth, 2006). Vitamin D obtained from the diet is transported
from the intestine via lymph veins to the liver in chylomicrons (see Holick and Chen,
2008, see Bouillon et al., 1998). In the blood circulation vitamin D and its metabolites
are bound mainly to vitamin binding protein (DBP) (25(OH)D and 1,25(OH) D 80-
15
90%) or albumin (25(OH)D and 1,25(OH) D 10-20%) and lipoproteins (see Zerwekh,
2008). Vitamin D can be stored in adipose tissue and bound to DBP it can be absorbed
by cells via endocytosis where DBP is degraded (see Holick, 2007, see Zerwekh, 2008).
According to Bouillon et al. (1998) the cell surface lipoprotein-like receptor involved in
the process is megalin and is at least present in kidney cells. This process was proved by
Nykjaer et al. (1999). The half life of vitamin D is only ~24 h and similar to that the half
life of 1,25(OH) D is ~4 h (see Zerwekh, 2008). The amount of 1,25(OH) D is also
tightly regulated by several factors discussed further. In contrast the half life of
25(OH)D is ~3 weeks and it is not significantly regulated, which makes it clear why to
measure this metabolite from the blood. Vitamin D metabolites are rapidly absorbed by
cells when they are not bound to DBP.
Once vitamin D is in the circulation it is transported to the liver, where it is
converted to 25(OH)D by vitamin D-25-hydroxylase (CYP27, CYP2D25) (see Holick,
2007, van Etten and Mathieu, 2005). CYP27 is a sterol-27-hydroxylase (cytochrome P-
450) with many functions and it is encoded by a gene located on human chromosome 2
(see Bouillon et al., 1998). It may have up to four different isoforms which act in
vitamin D metabolism (CYP27A1, CYP2R1, CYP3A4 and CYP2J3) (Prosser and
Jones, 2004). Besides 25- or 24-hydroxylation of vitamin D it can additionally, as a
main function, do side-chain hydroxylations, especially 27-hydroxylation of bile acid
biosynthesis metabolites and cholesterol. It is mainly found on the mitochondrial
membrane of hepatocytes (reviewed by Jones et al., 1998). This enzyme is also found in
other tissue cells like the duodenum, the adrenal gland, the lung and macrophages (see
Bouillon et al., 1998). The synthesis of the enzyme in the liver is apparently only
loosely regulated (see Jones et al., 1998). 25(OH)D is not the main active metabolite
and must be converted into 1,25(OH) D. 25(OH)D can also be converted to
24R,25(OH) D by 25-hydroxyvitamin D-24-hydroxylase (encoded by CYP24A1 gene)
(see Wolpowitz and Gilchrest, 2006, Lou et al., 2009).
The hydroxylation of 25(OH)D into 1,25(OH) D is catalyzed by the enzyme 25-
hydroxyvitamin D-1 -hydroxylase (1-OHase encoded by the CYP27B1 gene) in the
kidney mainly. CYP27B1 is located on chromosome 12q13.1-q13.3 (reviewed by Bailey
et al., 2007). Other tissues which produce 1,25(OH) D are keratinocytes,
immunological cells like monocytes, glial cells, bone and pneumocytes (Bouillion et al.,
1998). The synthesis of 1,25(OH) D is tightly regulated. Some regulative factors are
16
serum phosphorus, calcium (and therefore PTH) and fibroblast growth factor (FGF-23)
(see Holick, 2007). The metabolite also decreases its own synthesis and increases its
degradation by increasing the production of 25-hydroxyvitamin D-24-hydroxylase (24-
OHase or CYP24) (see Holick, 2003). The metabolic pathway is illustrated in Figure 4.
The water-soluble calcitroic acid is excreted in the bile (see Holick, 2007).
Figure 4. The metabolic pathway of vitamin D.
1.1.3 The biological functions of vitamin D
Vitamin D and its metabolites are mostly bound to their vitamin D binding protein
(DBP) in the bloodstream. When vitamin D, or its active metabolite 1,25(OH) D, is
absorbed into the cells DBP is degraded and 1,25(OH) D can act on transcription
through its nuclear receptor the vitamin D receptor (VDR). The VDR has one ligand
binding and one DNA binding domain. The main role of vitamin D is in calcium
homeostasis through parathyroid hormone (PTH) but vitamin D has been shown to
affect many other tissues and cells too for example the immune system.
17
The vitamin D binding protein (DBP) and vitamin D transport in the bloodstream
The vitamin D binding protein was discovered by Thomas et al. (1959) when they
studied the serum half-life of vitamin D and related metabolites. They suggested that the
fact that vitamin D was stable in serum but not in aqueous solutions must be due to a
macro-molecule to which it is bound in serum. Daiger et al. (1975) suggested that a
protein identified in the late 1950`s called group-specific protein was the same protein
that bound and carried vitamin D (DBP) in serum. They considered that both were
synthesized in the liver and had similar electrophoretic mobility. The complete amino
acid structure was determined in 1986 by Schoentgen et al. (1986). They conducted that
DBP was a 458 amino acid 58 kilodalton (kDa) globulin protein. The gene encoding
DBP (Gc) is located on chromosome 4q11-q13. Uniprot defines DBP as a globular,
multifunctional protein found in plasma, ascitic fluid, cerebrospinal fluid, and urine and
on the surface of many cell types. According to Uniprot DBP, besides carrying vitamin
D metabolites in the bloodstream, also prevents polymerization of actin by binding its
monomers. DBP can bind to membrane-bound immunoglobulin (Ig) on the surface of
B-lymphocytes and associate with IgG constant fragment (Fc) receptor on the
membranes of T-lymphocytes (www.uniprot.org). Verboven et al. (2002) determined
the structure of DBP and its vitamin D binding site by bioinformatic means (Figure 5).
They solved the structure to 2.3 Å resolution and defined that the DBP has a -helical
structure and three similar domains. The vitamin D binding site is located in domain one
and consists of helices 1 to 6. The reason why DBP is able to bind vitamin D is by
Verboven et al. (2002) due to its first domains helical structure differences compared
with human serum albumin (HSA). The fourth helix of HSA is closer to helices two and
three and therefore leaves no space for vitamin D to bind. They considered that the
possibility to bind actin is also due to DBPs different structural orientation. They
characterized the vitamin D binding site as a hydrophobic region where the 25-OH
group forms a hydrogen bond with the Tyr 32 and the 3-OH group forms hydrogen
bonds with the Ser 76 and Met 107. The amount in that the site was occupied by
25(OH)D was estimated by them to be 55%.
18
Figure 5. Vitamin D binding protein bound to 25(OH)D (Verboven et al., 2002).
Interestingly the binding site in DBP is not similar to the binding site in the vitamin D
receptor (VDR) which may be the reason for their different affinity to vitamin D
metabolites. Haddad et al. (1993) showed that DBP was a specific carrier for newly skin
synthesized vitamin D rather than vitamin D obtained from the diet. They suggested that
endogenously synthesized vitamin D was longer available in the bloodstream than
vitamin D obtained from the diet which to their opinion was rapidly transported to the
liver for swift metabolism. This resulted according to them to a faster clearance of
vitamin D from the body due to appearance of water-soluble conjugates. They also
showed that DBP was the main carrier of vitamin D and its metabolites in the blood
circulation. When vitamin D metabolites are free of DBP they enter the cells
immediately through diffusion due to their lipophilic character.
The endocytotic uptake of vitamin D into cells
The mechanism which enables the metabolic pathway of vitamin D to continue in the
kidney was characterized by Nykjaer et al. (1999). They used megalin knock-out mice
to identify that megalin was the receptor for DBP-vitamin D complex in kidney cells.
Megalin is a multifunctional receptor similar to the low-density-lipoprotein (LDL)
receptor. Reviewed by Nykjaer et al. (1999) megalin is expressed in the
neuroepithelium and in proximal tubular cells of the kidney and its ligands are
lipoproteins, proteases and protease inhibitors. They found that 25(OH)D-DBP
complexes were filtered through the glomerulus and absorbed into the epithelial cell by
endocytosis. In the cell the complex is delivered to lysosomal compartments and DBP is
degraded. The free 25(OH)D is then either further metabolised or released to the
interstitial fluid where it can again bind DBP. Nykjaer et al (1999) proved that megalin
19
is essential for the uptake of 25(OH)D-DBP in megalin deficient mice. These developed
vitamin D deficiency and bone disease due to the absence of megalin.
Later it has been proven that megalin is not the only receptor for DBP. Verroust
et al. (2002) reviewed that another important endocytic receptor is cubilin. Cubilin and
megalin function together in the endocytic absorption of vitamin D and minimize the
urinal loss of vitamin D (see Verroust et al., 2002). Cubilin is a 460 kDa glycoprotein
receptor with no transmembrane part (see Verroust et al., 2002). It is believed that
cubilin facilitates the megalin-mediated absorption of vitamin D metabolite-DBP
complexes. Driel et al. (2006) showed that megalin and cubilin are not only expressed in
the endocytic apparatus of epithelial cells of the renal proximal tubule but also in other
cells like bone cells.
Free vitamin D metabolites are also thought to enter cells directly by diffusion
(see Verroust et al., 2002). When further metabolized vitamin D metabolites are
secreted at the basolateral membrane they are either secreted bound to newly
synthesized carrier proteins or as free metabolites which first in plasma bind carrier
molecules (see Verroust et al., 2002).
The active metabolite 1,25(OH) D
The active metabolite of vitamin D was determined by Haussler et al. (1968) by silicic
acid column chromatography of whole mucosa and mucosa chromatin from rachitic
chicks which had been administrated with radioactively labelled vitamin D . They
identified at least three possible metabolites from which one was especially biologically
active and located in the chromatin (1,25(OH) D). Another metabolite with some
biological activity was identified but it was not localized in the nucleus. They assumed
it to be an intermediate between vitamin D and the active compound (possibly
25(OH)D). A third peak was characterized by Haussler et al. (1968) as a heterogenous
collection of other and former mentioned metabolites. Tsoukas et al. (1984) showed that
1,25(OH) D has a function as an immunosuppressing agent by suppressing interleukin-
2 (IL-2). They also showed that 1,25(OH) D inhibited the growth of mitogen-activated
lymphocytes. According to Deluca and Cantorna (2001) the main proven target cells for
1,25(OH) D are the intestinal enterocytes, osteoblasts, distal renal cells, keratinocytes
of the skin, promyelocytes and monocytes, lymphocytes, colon enterocytes, shell glands
and chick chorioallantoic membrane. They suggested that putative target cells are islet
20
cells in the pancreas, endocrine cells in the stomach, pituitary cells, ovarian cells,
placenta cells, epididymis cells, cells in the hypothalamus, developing myoblasts, aortic
endothelial cells and fibroblasts. According to Holick (2005) the main actions of
1,25(OH) D in calcium homeostasis are to increase the intestinal calcium absorption
and to mobilize calcium stores from the skeleton. Holick (2005) pointed out that studies
have shown that 1,25(OH) D helps to regulate cell growth and maturation, stimulates
insulin secretion, inhibits renin production and affects T- and B-lymphocytes as well as
macrophages.
The Vitamin D receptor
The vitamin D receptor was first found and isolated from the chromatin fraction of
rachitic chick intestinal mucosa cells by Haussler and Norman (1969). They
characterized it as a receptor with significant amounts of RNA and many possible forms
with the molecular weight varying from 50 kDa to over 200 kDa. They confirmed it to
bind the active form and other forms of vitamin D.
The active form of vitamin D, 1,25(OH) D, interacts with cells mainly through
its nuclear receptor the vitamin D receptor (VDR). The VDR was cloned by Baker et al.
(1988) and they concluded that the receptor is part of the superfamily of trans-acting
transcriptional regulatory factors. Other receptors which belong to this family are
steroid and thyroid hormone receptors. The human VDR gene is located on
chromosome 12q12-q14 and the human VDR has 427 amino acids and a molecular
mass of ~50 kDa. VDR is a homodimer when vitamin D is not bound to it and a
heterodimer (together with Retinoic X receptor) when vitamin D binds
(www.uniprot.org). These receptors have a highly conserved DNA-binding domain with
two zinc fingers at their amino terminal which is rich in cysteine, lysine and arginine
residues (see Jones et al., 1998, Baker et al., 1988, Brown et al., 1999). The ligand-
binding domain is located at the carboxyl terminal. The DNA-binding domain between
VDR and steroid and thyroid hormone receptors is 40% conserved according to Baker
et al. (1988). They also suggested that this domain is harboring both DNA binding and
transcriptional activities. In their study they confirmed that the VDR affinity lowers in
following order; 1,25(OH) D, 25(OH)D, 24R,25(OH) D, 1 ,24(R),25(OH) D,
(OH) D and vitamin D. The different domains are illustrated in Figure 6 and Figure
7.
21
DNA binding domain Ligand binding domain
Figure 6. Vitamin D receptor.
Norman et al. (1999) designed a three-dimensional structure of the VDR ligand binding
domain (LBD) with sequence alignment to the LBD of the rat -helical thyroid receptor
(TR) using X-ray crystallography. The human VDR LBD residues 142-427 were
aligned with the rat TR LBD residues 157-410 manually. The model of the human VDR
LBD created was an elongated globular protein with 12 -helical elements linked by
short loops forming an anti-parallel helical triple sandwich.
Figure 7. VDR DNA binding domain (A) bound to DNA and vitamin D binding domain (B).
Norman et al. (1999) showed that the interior surface of the LBD was comprised by
hydrophobic residues (H305, S306, Y401, S405 and T415) which interact with
427
192 21 96
22
1,25(OH) D groups 1 -OH, 3 -OH and 25-OH to form hydrogen bonds. According to
them the natural mutation in H305Q of VDR, results in 80% decreased ligand affinity.
The reason for this might be that this residue is particularly important to a specific
portal opening. They concluded that the conformation of helix 12 in the receptors
docking site is changed conformationally when the ligand binds to the receptor. The
helix 12 creates a portal where it is rotated out and down in an unoccupied receptor and
up to interact with helixes 3 and 5 in an occupied receptor to close the portal. They
showed that it facilitates the entrance through the portal if the ligand is in optimal
conformation. They considered that the conformation favoured for entrance could be a
“slim” 1 ,25(OH)2-6-s-trans-D form while the optimal form for binding could be
closer to the classic “pudgy” form 1 ,25(OH)2-6-s-cis-D . It is known that the
conformational flexibility of 1,25(OH) D is due to its A-ring and the rotation around the
6,7 carbon single bond. The conformational changes are possible due to the fact that
vitamin D and its metabolites do not have a 9,8 carbon bond between the A- and the B-
ring. Norman et al. (1999) considered that it was not clear if the volume of the VDR
LBD interior (~620 ų) was enough for the ligand (375 ų) to change conformation
from trans-form to cis-form. The LBD is also responsible for dimerization with the
retinoic X receptor (RXR) after binding 1,25(OH) D. This heterodimeric receptor then
binds to vitamin D responsive elements (VDRE) and induces transcription of target
genes with the help of histone release and transcription factors (reviewed by Jones et al.,
1998). The mechanism is illustrated in Figure 8.
23Figure 8. The activating mechanism of vitamin D in the nucleus.
As reviewed by Jones et al. (1998) the RXR is required as a cofactor to establish VDR-
RXR binding to VDREs. It has been shown that VDREs are usually composed of two
repeats of the sequence AGGTCA separated by one to several non-specific bases. The
most common VDREs from genes up-regulated by VDR have three separating bases
between the half-sites (see Jones et al., 1998). The first VDRE isolated was from the rat
osteocalcin gene (Demay et al., 1990). Jones et al. (1998) reminded that the RXR
usually binds to the 5`-half-site of the VDRE and the VDR to the 3`-half-site. Binding
of the VDR-RXR to the VDRE, results in bending of the VDRE (Kimmel-Jehan et al.,
1999). Jones et al. (1998) speculated that this bending and the release of histones may
be the combination that makes the promoter of target genes more reachable for
transcription factors. VDR has been shown to interact with other cofactors too but it is
not clear according to Jones et al. (1998) if it binds them also.
Lou et al. (2009) showed that VDR also binds 25(OH)D. The intermediate
metabolite showed direct anti-proliferate effects on human LNCaP prostate cancer cells
and gene regulatory properties in primary mouse kidney, skin, prostate and human
MCF-7 breast cancer cells therefore indicating that not only 1,25(OH) D has gene
regulatory and anti-proliferative properties. This could be a promising result considering
therapeutic applications. The administration of 1,25(OH) D has showed some severe
side effects in patients like hypercalcemia and hypercalciuria and therefore the use of
25(OH)D instead would be a possible approach (Lou et al., 2009).
One study also showed that there are intracellular proteins that enhance vitamin
D mediated transcriptional activation (Adams et al., 2004). Two types of proteins were
identified were the first family was considered to be called heterogeneous nuclear
ribonucleoproteins (hnRNPs) and the second to be named intracellular vitamin D
binding proteins (IDBPs). The hnRNPs were identified as pre-mRNA interacting
proteins which show cis-acting, trans-dominant regulation of vitamin D dependent gene
regulation.
Vitamin D and Calcium homeostasis
The major role of vitamin D in calcium homeostasis is mediated through the active
metabolite and hormone 1,25(OH) D. Low levels of plasma calcium stimulate the
production of PTH from the parathyroid glands which then bind to osteoblasts and
24
kidney cells to enhance the production of 1,25(OH) D through 1 -hydroxylase
(reviewed in Deluca and Cantorna, 2001). In the intestine 1,25(OH) D induces the
expression of calbindin D-9K, an epithelial calcium channel (see Holick, 2005). Added
to that 1,25(OH) D also regulates several other calcium related genes and their protein
products. These include according to Bouillion et al. (1998) osteocalcin, osteopontin,
collagen type 1, matrix gla protein, 3 integrin and VDR in the bone, calbinding D-
9K and 24-hydroxylase in the kidney and PTH and calcitionin in the endocrine cells and
parathyroid cells.
Vitamin D deficiency caused secondary hyperparathyroidism results in wasting
of skeleton leading to osteoporosis or osteomalacia (see Holick, 2005). The bone
density is decreased which is called osteopenia and this is investigated by for example
X-ray bone density measurement. The loss of mineralization is caused by loss of
phosphorus into the urine which results in lower serum phosphorus levels. This leads to
inadequate calcium-phosphorus products which causes poor or defective mineralization
of the bone (see Holick, 2005). This can lead to rickets in children, where the poorly
mineralized skeleton deforms under the body weight and classic bony rachitic
deformities like bowed legs or knocked knees can form (see Holick, 2005). According
to Holick (2005) the results in adults can lead to osteomalacia. This is a condition where
the newly formed osteoid can not be properly mineralized and becomes hydrated. The
calciumhydroxyapatite can not be formed in the matrix. This leads to a state where the
osteoid is not able to provide support for sensory fibers in the periosteal covering and
this causes either global or isolated throbbing, aching bone pain.
The excess of vitamin D can lead to hypervitaminosis D and then to
hypercalcemia where often symptoms like nausea, vomiting and polyuria are observed.
The amount of calcium in the serum in hypercalcemia is thought to be over 2.6 mmol/l
(reviewed in Holick, 2007). It has been estimated that S-25(OH)D concentrations over
374 nmol/l can cause vitamin D intoxification (reviewed in Holick, 2007).
Vitamin D and PTH
Both 1,25(OH) D and PTH play an important role in calcium homeostasis as discussed
before. Secondary hyperparathyroidism is a condition where serum intact PTH (iPTH)
concentration is elevated and the reason for this must not necessarily be hypercalcemia
(Freaney et al., 1993). It can be caused by renal insufficiency in 1- -hydroxylation,
privational hypovitaminosis D or intestinal diseases with calcium malabsorption
25
(Freany et al., 1993, Kauppinen-Mäkelin et al., 2001). Freany et al. (1993) investigated
the PTH concentrations in elderly subjects and concluded that secondary
hyperparathyroidism linked to aging is caused by the decline in 1- -hydroxylase activity
and hypovitaminosis D, and can be moderated by vitamin D supplementation.
Hypovitaminosis D can be defined as the S-25(OH)D concentration where S-iPTH
begins to rise. This has varied in different studies (25 nmol/l (Freaney et al., 1993), 37.5
nmol/l (Thomas et al., 1998) 50 nmol/l (Kauppinen-Mäkelin et al., 2001), 78 nmol/l
(Välimäki et al., 2004)).
Vitamin D and the immune system
The role of 1,25(OH) D in the immune system has been studied by Veldman et al.
(2000). They found out that the VDR is specially expressed in T lymphocyte CD8 cells.
Significant amounts were also expressed in CD4 cells. They suggested that the major
target of 1,25(OH) D in the immune system are CD8 cells. Jeffrey et al. (2009) studied
the effect of 1,25(OH)2D on human CD4+CD25- T–cells. They found out that
1,25(OH) D inhibited production of interferon gamma (IFN- ), IL-17 and IL-21. T-cell
division in contrast was not affected. Jeffrey et al. (2009) observed that 1,25(OH) D had
an enhancing effect on CTLA-4 and FOXP3 production. T-cells which were treated
with 1,25(OH) D suppressed the proliferation of normally responsive T-cells. Hewison
et al. (2003) showed in their study that 1,25(OH) D is capable in inhibiting the
differentiation, maturation and activation of monocyte-derived dendritic cells. They also
showed that 1,25(OH) D is produced locally in dendritic cells and macrophages.
Mathieu et al. (1994) showed that 1,25(OH) D is able to decrease T lymphocyte helper
cell (Th1 cell) infiltration in the pancreas. Equils et al. (2006) showed that 1,25(OH) D
may also be a potential adjuvant in the treatment of gram-negative bacterial infections.
They showed that 1,25(OH) D modulates the immune response of human microvessel
endothelial cells and inhibits TLR4 agonist LPS-induced MyD88-dependent NF- B
activation. They also showed that 1,25(OH) D inhibits proinflammatory cytokines, IL-6
and IL-8, and CC-chemokine RANTES release. The VDR is presented constitutively on
antigen-presenting cells (APCs) like macrophages and dendritic cells (DCs) and
inducibly on lymphocytes following activation (van Etten and Mathieu, 2005). The
26
25(OH)D -1- -hydroxylase is regulated differently in the kidney than in the
macrophages according to van Etten and Mathieu (2005). They showed that in the
kidney it is regulated by PTH and 1,25(OH) D, and in contrast by immune signals like
IFN- in the macrophages. In DCs the enzyme is regulated by the maturation of these
cells associated with p38 MAPK- and NF- B. Th 1 cells secrete IFN- and IL-2. IFN-
is important for APCs and T lymphocyte recruitment. IL-2 is a growth factor for T
lymphocytes and promotes their activation and proliferation. When 1,25(OH) D inhibits
these, the profile of T helper cells shifts from Th1 to Th2, which secrete IL-4, IL-5, IL-
13, which have a more regulatory effect.
The activation of an inflammatory response to infection is induced by
granulocyte-macrophage colony-stimulating factor (GM-CSF), which activates mature
granulocytes and macrophages. Van Etten and Mathieu (2005) summarized that
1,25(OH) D down-regulates this and also inhibits expression of the FasL ligand for Fas
receptor, which together induce T cell apoptosis.
1.1.4 Vitamin D Deficiency
There are no global recommendations for sufficient serum vitamin D status. The
concentration considered to indicate deficient status varies from under 20 nmol/l to
under 50 nmol/l (table 5). The borders for insufficient status have been very different
depending on source. Lamberg-Allardt and Viljakainen (2006) thought that sufficient
status was between 50 and 79.9 nmol/l (table 5). Almost same values were suggested by
Holick (2007) to indicate insufficiency. Wolpowitz and Gilchrest (2006) reviewed
sources where insufficient status was either <37.5-50 or <80-100 nmol/l. Optimal status
varies between over 50 nmol/l and over 100 nmol/l (table 5). The concentrations I will
use in my work to characterize the results are the ones estimated by Lamberg-Allardt
and Viljakainen (2006).
Table 5. Different opinions on adequate and inadequate vitamin D status.
Concentration Status Reference
<50 nmol/l insufficient Lamberg-Allardt and Viljakainen, 2006
50-79.9 nmol/l sufficient
80 nmol/l optimal
50 nmol/l (Lips) optimal Dawson-Hughes et al., 2005
70 nmol/l (Vieth) optimal
2775 nmol/l (Holick) optimal
75 nmol/l (Meurnier) optimal
80 nmol/l (Heany) optimal
80 nmol/l (Dawson-Hughes) optimal
>374 nmol/l intoxication Holick, 2007
<50 nmol/l deficient
51-74 nmol/l insufficient
<20-25 deficient Reviewed by Wolpowitz and Gilchrest, 2006
<37.5-50 or <80-100 nmol/l insufficient
above insufficient sufficient
Factors that effect vitamin D status are age, clothing, skin pigmentation, sunscreen use,
time of year and day, geographical location, and metabolic disorders. Obesity has also
been associated with vitamin D deficiency. Hyppönen and Power (2006) investigated
the relation between S-25(OH)D and body mass index (BMI). They concluded that high
BMI was associated to low S-25(OH)D in a 45-year old population of 7198 subjects in
Britain. They also suggested that the impact of S-25(OH)D on glucose metabolism may
be dependent on body mass. The fact that vitamin D production in the skin is lower, or
absent, in the winter in many regions, due to lower or absent UVB light, has been
known for long. Kull et al. (2009) studied the seasonal variation of S-25(OH)D in
Estonia in 367 randomly selected individuals (age ranged between 25 and 70 years).
Their results showed that, mean S-25(OH)D concentration during summer was 59
nmol/l and during winter 44 nmol/l. They compared their results to other studies where
concentrations had been 67 nmol/l and 44 nmol/l (New Zealand, Rockell et al., 2006),
79 nmol/l and 38 nmol/l (Japan, Ono et al., 2005), 84 nmol/l and 43 nmol/l (Italy,
Carnevale et al., 2001) and 67-70 nmol/l and 40-45 nmol/l respectively (Germany,
Scharla and Scheidt-Nave, 1996). These results show that there is a significant variation
between winter and summer 25(OH)D status and these results can be dependably
compared due to the fact that S-25(OH)D concentrations have been assessed with the
same method (RIA, Diasorin). When the vitamin D production in the body due to UVB
radiation is compared in older individuals it has been shown that a 70-year old person
synthesizes only 25% of the amount a 20-year old person synthesizes vitamin D,
exposed to the same amount of sunlight (see Holick, 2004). This is due to the fact that
the elderly have lower concentrations of 7-dehydrocholesterol in the skin. According to
Holick (2004) a sunscreen of sun protecting factor (SPF) 15 reduces the skins capacity
of producing vitamin D3 by over 98%. Chen et al. (2007) showed that the skin of a
28
person with type II converts about 5 to 10 times more 7-DHC to previtamin D than the
skin of a person with type V. Vitamin D deficiency in children causes rickets by
inadequate calcium-phosphate products which cause mineralization defects (Holick
2004). Symptoms of vitamin D deficiency can include bone and muscle pain and muscle
weakness which often indicate the mineralization defect condition osteomalacia (see
Holick, 2003). Secondary hyperparathyroidism is often associated with osteomalacia
and it can increase the mobilization of the collagen matrix and mineral from the
skeleton (see Holick, 2003). This can then lead to osteoporosis. Holick (2003)
recommends one minimal erythemal dose (MED, 54 mJ/cm) two to three times a week
by exposing face, arms and legs in order to receive adequate vitamin D requirement and
storage in the body fat.
According to van Groningen et al. (2010) the vitamin D loading dose required
to reach the serum 25(OH)D target level of Holick and Meurnier (75 nmol/l) can be
calculated as follows:
Dose (IU) = 40 x (75 - serum 25(OH)D ) x body weight.
1.1.5 Vitamin D recommendations in Finland
The vitamin D status is especially from October to May really low in Finnish people
due to the absence of UVB light during this time (Hirvonen et al., 2004). Even in April
and May the UVB light is very low in Finland (Lamberg-Allardt, 1984). In Finland
vitamin D is specially recommended for infants and toddlers (0 to 3 years), pre-school-
and school-aged children (3 to 15 years) and adults over 60 years (www.evira.fi). It is
also recommended during pregnancy and lactation. The recommended time for
additional vitamin D consumption from supplements is from the beginning of October
to the end of March. For elderly people, in institutional care or homecare, who are not
getting any outside exercise, vitamin D consumption is recommended all year-round
(www.evira.fi).
The main food sources for vitamin D for Finns are foods of fish (45%), dietary
fats (21%) and liquid milk products. Dietary fats and liquid milk products are fortified
with vitamin D (usually D ) (www.evira.fi, www.fineli.fi). The National Nutrition
Council has recently suggested that the amount of vitamin D should be doubled in
liquid milk products (from 0.5 µg/100 g to 1 µg/100 g) (www.evira.fi).
29
Dietary fats have been fortified with vitamin D for the first time in Finland in
1941. In the 1950`s the milk for Finnish school pupils was fortified with vitamin D but
this was only a short term solution. Before 1964 the recommendation was 4000-5000 IU
(100-125 g) per day. In 1964 the dose was lowered to 2000 IU (50 µg) and in 1975 to
1000 IU (25 g) per day. The dose of 250 IU (5 µg) per day was suggested in 1992.
Because many studies indicated that the vitamin D status of Finns was too low during
the 1990`s, the National Nutrition Council (VRN; valtion ravitsemusneuvottelukunta)
suggested in 2002 that the amount of fortified vitamin D into liquid milk products
should be raised. In 2003 the new vitamin D fortification of liquid milk-products
started. The recommendations for vitamin D intake have been changed in 2004 after that
by the Ministry of Social Affairs and Health. It is estimated that new recommendations
are soon to be made hence to the raising number in vitamin D deficiency all over the
world. (www.evira.fi)
The current recommendation for vitamin D intake in Finland is 7,5 µg per day
for 3 to 60 year-olds when it has earlier been 5 µg per day. The recommended dose for
children under 3 years and pregnant or lactating women is 10 µg per day. The
recommendation for the elderly over 60 years has just been raised (30.03.2010) to 20 µg
per day. Interestingly labels of fortified products (milk) still seem to have old
recommendations (5 µg). The highest recommended dose for individuals under 10 years
is 25 µg per day, and for young adults and adults 50 µg per day (www.evira.fi). If one
eats fish at least twice a week, drinks half a liter milk per day and uses vitamin D
fortified dietary fats in cooking and on bread, there is no need to use vitamin D
supplements to reach an adequate status of serum vitamin D (Lamberg-Allardt and
Viljakainen, 2006). But if one of vitamin D sources mentioned is lacking, the use of
supplements is needed.
1.1.6 Vitamin D status in the healthy Finnish population
Kauppi et al. (2009) studied the vitamin D status in Finnish healthy subjects and the
association to ultrasound parameters considering bone health (broadband ultrasound
attenuation (BUA) and speed of sound (SOS)). The number of subjects was 6035 (2736
men and 3299 women) and the subjects were between 30 and 79 years of age.
Concentration lower than 20 nmol/l was said to indicate vitamin D deficiency and
concentration between 20 and 37.5 nmol/l was to indicate moderate hypovitaminosis D.
Serum 25(OH)D (S-25(OH)D) status was measured by Radioimmunoassay (RIA). The
seasonal variation was taken into account by considering the period August-October to
30
represent autumn and November-March to represented winter. The mean concentration
for men was 45.1 ± 0.53 (range 5-132) nmol/l and for women 45.2 ± 0.46 (7-143)
nmol/l. Hypovitaminosis D was more common in the subjects examined in the winter
than in those examined in autumn. Moderate hypovitaminosis D was found in 42.2% of
men and 39.4% of women examined in the winter. Vitamin D deficiency was found in
5.2% and 3.4% respectively. In the autumn period the numbers dropped to 25.2% and
2.7% for men and 28.2% and 2.9% for women. Low S-25(OH)D was common and
correlated with low BUA and SOS values. Low S-25(OH)D was also related to higher
body fat mass.
The relationship between S-25(OH)D and risk of stress fractures was studied by
Välimäki et al. (2005). The study consisted of 179 Finnish military recruits who were
aged between 18 and 20 years. Of those 15 experienced a stress fracture during military
service. The rest was considered to be the control group. S-25(OH)D was measured
with RIA. Concentration lower than 20 nmol/l indicated vitamin D deficiency and
concentrations below 37.5 nmol/l indicated hypovitaminosis D. The S-25(OH)D was
measured once during summer and once during winter. The mean concentration for
stress facture subjects was 44 (range 26-64) nmol/l and for none-stress facture also 44
(18-107) nmol/l. Vitamin D deficiency was common in the wintertime when 39% of all
men had it. Hypovitaminosis D was recognized in 95% of all subjects in the winter. In
the summer prevalence of vitamin D deficiency was only 1% to 2% and that of
hypovitaminosis D 27% to 30%. Low S-25(OH) was common but not related to stress
fracture incident.
The S-25(OH)D concentrations in healthy subjects from the Helsinki-Vantaa
region and Turku-Loimaa region (both ~60ºN) were analyzed in 2001 (Lamberg-Allardt
et al., 2001). Samples were taken in March 1998. The aim of the study was to determine
if recommended vitamin D intake (5µg/d) lead to a sufficient vitamin D status. Data was
obtained from 202 women and 126 men aged between 31 and 43 years. Deficiency was
determined by two different aspects. S-25(OH)D concentration below 25 nmol/l was
considered to indicate vitamin D deficiency. On the other hand the S-25(OH)D
concentration where serum intact parathyroid hormone (S-iPTH) began to rise was
considered to indicate insufficiency. S-25(OH)D concentrations were measured by RIA.
Based on the first consideration for vitamin D deficiency 26.2% of the women and
28.6% of the men represented this status. Based on the relation between S-25(OH)D and
S-iPTH (80 nmol/l in women and 40 nmol/l in men) 86% of the women and 56% of the
31
men had insufficient vitamin D status. Vitamin D status was low in a big part of the
study population even when vitamin D intake met the recommendations.
Five different Finnish studies were analyzed for vitamin D status in the study of
Lamberg-Allardt and Viljakainen (2006). The overall amount of serum samples was
nearly 1500. The aim was to investigate if vitamin D status had improved from 2002 to
2004 due to new fortification of liquid dairy products. All measurements were done
with the enzyme immunoassay (EIA). Subjects were from the study Type 1 Diabetes
Prediction and Prevention (DIPP) (n=107), the intervention project for coronary heart
disease risk factors (STRIP) (n=101), the girl-cohort for calcium and vitamin D
intervention (CALEX) (n=100), the FINRISK study (n=249) and the OPTIFORD study
(n=100). In the end 634 subjects in 2002 and 630 subjects in 2004 were analyzed.
Vitamin D insufficiency was characterized as S-25(OH)D concentration below 50
nmol/l and concentrations between 50 and 79.9 nmol/l were considered to be sufficient.
Concentrations above 80 nmol/l were considered to be optimal. In 2002 37.7% of all
subjects had insufficient vitamin D status whereas the amount was 21.3% in 2004. The
amount of sufficient status subjects was respectively 38.1% and 36.7%. The amount of
optimal status subjects grew the most from 24.3% to 41.7%. The study subjects were
not optimal to reflect the normal Finnish population, due to vitamin D supplement use
and healthier food consumption.
Välimäki et al. (2004) wanted to assess the prevalence of vitamin D deficiency
and the connection between this and peak bone mass in young Finnish men. The
subjects were between 18.3 and 20.6 years old and the number of subjects was 220.
Vitamin D measurement was done once in the summer (July 2000 n=220), once in the
winter (January 2001 n=167) and again in the summer (July 2001 n=94). Vitamin D
deficiency was defined as a concentration 20 nmol/l. Measurement of S-25(OH)D
concentration was done with RIA. In July 2000 only 0.9% of subjects had vitamin D
deficiency. The mean concentration was 44 (15-107) nmol/l. In January 2001 38.9% of
the subjects suffered from vitamin D deficiency. There the mean concentration was 24
(6.0-66) nmol/l. When subjects were analysed six months later in the summer again
only 2.1% had vitamin D deficiency and the mean concentration was 41 (18-72) nmol/l.
They suggested that vitamin D deficiency may be detrimental to acquisition of maximal
peak bone mass.
The vitamin D status of Finnish pregnant women was assessed in 1984 by
Lamberg-Allardt et al. (1984). They measured the S-25(OH)D concentration from the
maternal and cord blood at delivery with a competitive protein binding assay. They
32
studied 192 mothers of which 99 received a 12,5 µg/d supplement. The measurement
time periods were summer (June-August) and autumn (September-November). Their
results showed that there was a clear variation between S-25(OH)D concentration
means in the summer and autumn in the supplemented and not supplemented group (75
nmol/l vs. 52 nmol/l and 103 nmol/l vs. 73 nmol/l). It became clear that the S-25(OH)D
concentration means were lower in the cord blood samples than the maternal blood
samples (46 nmol/l vs. 59 nmol/l and 60 nmol/l vs. 87 nmol/l for both groups
respectively).
The studies of Kauppi et al. (2009), Välimäki et al. (2005), Lamberg-Allardt et
al. (2001) and Välimäki et al. (2004) are summarized in Table 6 for their good
comparability. As a conclusion it can be said that vitamin D status in the healthy
Finnish population is generally low which is mostly due to the short period of UVB
light exposure.
33
Table 6. Vitamin D status in the healthy Finnish population.
Cases Controls Method CV % (mean) Conclusion Reference
Subjects Age (mean) Number Sex
S-25(OH)D (nmol/l) (Mean)
Subjects Age (mean) Number Sex
S-25(OH)D (nmol/l) mean
intra-assay
inter-assay
Healthy adults 54.5 (30-79)
2736 Male 45.1
RIA (Incstar*) 3.5 6.9 low S-25(OH)D concentrations were
associated with (BUA) and (SOS) Kauppi et al., 2009
3299 Female 45.2
Stress fractured military recruits
19 (18-20) 15 Male 44
Not stress fractured military recruits
19 (18-20) 164 Male 44 RIA
(Diasorin) 8.9 12.8 S-25(OH)D concentrations not lower in stressed subjects
Välimäki et al., 2005
Young men baseline (July)
19.6 (18.3-20.6)
220 Male
44
RIA (Diasorin) 7.4 10.8
almost 40% of the subjects had vitamin d deficiency during the winter, S-25(OH)D concentration correlated positively with bone mineral content
Välimäki et al., 2004
Young men 6 months from baseline (winter)
24
Young men 12 months from baseline
41
Healthy adults
38 202 Female 47
RIA (Incstar) 10.1 14.9
86% of women and 56% of men had insufficient vitamin d status (based on the relation to S-iPTH)
Lamberg-Allardt et al., 2001 37 126 Male 45
*Diasorin was formerly known as Incstar
34
1.1.7 Vitamin D status and disease in the Finnish population
Toriola et al. (2010) performed a study where the relation between S-25(OH)D and risk
for ovarian cancer was assessed. Finnish subjects were diagnosed with ovarian cancer
10 years after serum sampling and the number was 201. Two control sets were matched
to subjects, one for age, parity and sampling season (± 4 weeks) (always 2 per subject n
= 398) and one for age, parity and opposite sampling season (6 months ± 4 weeks) (n =
199). The age raged from 17.5 to 44 years. The S-25(OH)D was measured with RIA. S-
25(OH)D below 50 nmol/l was considered to indicate vitamin D deficiency, 50-75
nmol/l was considered to indicate insufficient status and concentration above 75 nmol/l
were to indicate sufficient status. Median values for cases, same season controls and
opposite season controls were 35.0, 35.8 and 34.3 nmol/l respectively. Sufficient status
was observed in 4 cases of 200 and 21 of 398 controls. Insufficient status was very
common both in cases and controls.
The connection between vitamin D status and breast cancer or pregnancy
induced breast cancer (PABC) has been studied in Finland (Agborsangaya et al., 2010).
The study subjects were identified from the Finnish Cancer Registry and samples were
drawn from the Finnish Maternity Cohort for both cases and controls. Breast cancer
patients with at least two singleton pregnancies were selected. The sample size of breast
cancer patients (follow-up time 10 years) was 100 and the same amount of controls
was matched to them for blood withdrawal time and year (±1) (summer=May-August,
winter=December-March), age at sampling and parity (±1). The number of PABC cases
was 111 and the same amount of controls was matched for age (±1), parity (±1) and
date of blood sampling (±15 d). The S-25(OH)D concentration was measured with RIA.
The concentration results were divided into quintiles (breast cancer 1. pregnancy: 27.5,
27.6-37.5, 37.6-48.5, 48.6-55.5, 61.6 nmol/l, 2. pregnancy: 27.8, 27.9-36.9, 37.0-
46.5, 46.6-55.5, 56.6 nmol/l and PABC: 25.8, 25.9-34.9, 35.0-44.7, 44.8-64.0, 64.1
nmol/l). They considered that conducting to their results vitamin D concentration was
not associated to breast cancer in general. Interestingly they considered higher (2nd to 5th
quintiles compared to lowest quintile) vitamin D concentrations to be a risk for PABC.
The relation between S-25(OH)D and acute myocardial infarction (AMI) and
stroke in elderly subjects from the city of Turku and surrounding areas in southwestern
Finland was investigated by Marniemi et al. (2005). The age ranged between 65 and 99
years. The study was a 10-year follow-up. The AMI group consisted of 65 men and 65
women, and the stroke group consisted of 34 men and 36 women. The control group for
35
the AMI group consisted of 255 men and 304 women who had experienced no coronary
event. The control group for the stroke cases consisted of 282 men and 308 women with
similar characteristics. The S-25(OH)D concentration was measured by RIA. The mean
± standard deviation (SD) S-25(OH)D concentration for AMI and stroke cases were
28.4 ± 17.0 and 29.6 ± 17.9 nmol/l. The results for the respective control groups were
30.7 ± 19.7 and 31.3 ± 19.4 nmol/l. S-25(OH)D concentration was lower in the case-
groups than in the control groups.
Nurmi et al. (2005) studied the concentration of S-25(OH)D in southeastern
Finnish hip fracture patients. Cases were recognized from two Finnish hospitals and the
number was 223. Both men (n=65) and women (n=158) took part in the study. Age
varied from 47-96 in women and 38-91 in men. The cases were divided into groups
depending on place of residence (actual home, residential home or institution). The
seasonal variation of S-25(OH)D concentration was followed. Measurement was done
by RIA. Severe vitamin D deficiency was defined as a concentration <20 nmol/l and
moderate vitamin D deficiency as a concentration between 20 and 37.4 nmol/l. Out of
all patients 52% of women and 55% of men suffered from moderate vitamin D
deficiency. In 9% of all patients vitamin D deficiency was severe. Only 3% of all
patients had vitamin D concentrations above 74 nmol/l. The mean concentration among
women was 38.1 nmol/l and among men 37 nmol/l. Concentrations were lower in
patients living in a residential home or institution than patients living at home, 55%,
61% and 50% respectively.
The amount of vitamin D deficiency in in- and outpatients in the capital region
of Finland was assessed in 2001 (Kauppinen-Mäkelin et al. 2001). Inpatients consisted
of 57 women and 49 men who were between 45 and 58 years. Outpatients consisted of
48 women and 51 men with the age range of 42 to 46 years. Measurement was done in
October 1998 (31 inpatients were reanalysed in March or April 1999). The method used
to determine S-25(OH)D was RIA. Severe vitamin D deficiency was considered to be
the case when patients had S-25(OH)D concentrations below 20 nmol/l. When S-
25(OH)D concentration was 50 nmol/l or less S-iPTH concentration began to rise
(hypovitaminosis D). Overall 61% of male hospital patients had S-25(OH)D
concentrations 37.5 nmol/l. The amount of female hospital patients who had the
preceding concentration was 70%. When outpatients where considered the amounts
were 37 and 44% respectively. Severe vitamin D deficiency was found in 20% of male
inpatients and 26% of female inpatients. The amounts for outpatients were 6% and 2%
respectively.
36
In conclusion it can be noted that the vitamin D status generally in different
disease groups is low. Low vitamin D was observed in all studies, but it was not
necessarily associated to the disease when compared with controls. There are
suggestions that low vitamin D status can be associated to heart conditions and
institutional care. The latter is obviously due to absence of sunlight exposure.
1.2 Type 1 Diabetes mellitus
1.2.1 T1D pathogenesis and genetics
T1D is caused by the autoimmune destruction of the beta-cells in the pancreas. The
beta-cells in the islets of Langerhans are infiltrated by dendritic cells, macrophages and
T lymphocytes. Usually onset of the disease is at young age from 1 year old to 18 year
olds (see Pociot and McDermott, 2002).
The autoimmune process is characterised by specific autoantibodies developed
against T1D antigens in the beta-cells (Wenzlau et al., 2007). According to Ounissi-
Benkalha et al., (2008) and Atkinson and Eisenbarth (2001) the main autoantibodies are
insulin autoantibody (IAA), islet cytoplasmic autoantibody (ICA), glutamate
decarboxylase antibody (GADA), insulin autoantigen 2 (IA2) and islet cell autoantibody
512/protein tyrosine phosphatase IA2 (ICA512/IA2A). Wenzlau et al. (2007) described
another autoantibody in 2007; a zinc transporter antibody (ZnTA). According to
Eisenbarth (1986) the development of T1D is divided into six stages. The disease begins
with the first stage; the genetic susceptibility. Some genetically susceptible individuals
develop active autoimmunity (stage 3) through a hypothetical triggering event (stage 2).
Stage four is characterised as a condition where blood glucose level is still normal but
insulin secretion is progressively lost. During stage five some insulin is still secreted but
it is followed by stage six; the complete destruction of beta-cells in the pancreas.
T1D is a complex autoimmune disease with inheritance related and
environmental features. The most important chromosomal regions associated to T1D are
the human leukocyte antigen (HLA) region at chromosome 6p21.3 and the insulin (INS)
gene region at 11p15.5 (see Pociot and McDermott, 2002). These regions have been
named IDDM1 and IDDM2, where IDDM1 is localised in the major histocompability
complex (MHC) on 6p21 consisting of at least HLA-DQB1 and HLA-DRB1 loci and
probably also HLA-DPB1 (Cox et al., 2001). The genes of the MHC are divided into
four classes I, II, III and IV and the strongest association to T1D risk has been
connected with class II genes (see Pociot and McDermott, 2002). HLA related
37
susceptibility may account for less than 50% of the inherited disease risk according to
Pociot and McDermott (2002). IDDM2 susceptible loci is a variable number of tandem
repeats (VNTR) which consists of a repetition of 14-15 bp oligonucleotide sequence and
can be repeated 26 to 200 times (see Pociot and McDermott, 2002). The VNTR alleles
are divided into three classes (I, II, and III) which vary in repeat size from 26 to 63,
mean of 80 and 141 to 209 repeats, where strongest association to increased T1D risk
has been for class I alleles and protective association has been linked to class III alleles
(reviewed by Pociot and McDermott, 2002). About 20 non-HLA genomic intervals with
linkage evidence to T1D have been discovered in different studies but only the function
of IDDM2 and IDDM12 has been identified (reviewed by Pociot and McDermott, 2002,
Atkinson and Eisenbarth, 2001). IDDM12 is located on chromosome 2q33 and
apparently has some effect on the cytotoxic T-lymphocyte-associated protein 4 (CTLA-
4) which has an effect in the modulation of immune responsiveness (Nistico et al.,
1996).
No environmental factor has been identified to have an indefinite effect on T1D
development (see Atkinson and Eisenbarth, 2001). The main environmental factors that
have been suggested to affect the T1D risk incidence are divided into three groups: viral
infections (for example coxsackievirus and cytomegalovirus), early infant diet (the
relation of breast feeding versus early introduction of cow`s milk) and toxins (like N-
nitroso derivatives) (reviewed by Atkinson and Eisenbarth, 2001). Other factors
suggested are vaccine administration, psychological stress and climatic influences and
environmental factors. These are believed to serve as modifiers rather than disease
pathogenesis triggers (see Atkinson and Eisenbarth, 2001). Findings that multiple
infections during the first years of life are associated to a decreased risk of T1D
somehow explain the rapid growing prevalence in a world where sanitation, health care
access and vaccination have improved.
1.2.2 Prevalence and trend in Finland
Harjutsalo et al. (2008) studied the prevalence of T1D between 1980 and 2005 in
Finland. The number of T1D cases diagnosed before the age of 15 was 10737 (5816
boys and 4921 girls). The number of incidence grew from 31.4/100 000 per year in
1980 to 64.2/100 000 per year in 2005. The predicted number for new cases between
2006 and 2020 was 10 800 according to Harjutsalo et al. (2008). The trend from 1965 to
2005 is illustrated in Figure 9.
38
Figure 9. The number of T1D incidents in Finland between 1965-2005 (Harjutsalo, 2007).
1.3 Vitamin D and Type 1 Diabetes
Vitamin D and T1D have been linked together due to the fact that there is a
geographical variation in T1D prevalence and vitamin D synthesis in the skin. Both
environmental and genetic factors have been studied. Insulin secretion is dependent on
vitamin D status. Vitamin D deficiency reduces secretion of insulin directly and
indirectly by increasing intracellular calcium which decreases insulin secretion
(Penckofer et al., 2009). Vitamin D may reduce the risk of infections that can be
associated to autoimmune process of beta-cell destruction and vitamin D may directly
reduce the autoimmune process of beta-cell destruction (see Holick, 2008). Vitamin D
deficiency increases insulin resistance via TNF- (Boucher, 1998).
Maternal vitamin D deficiency and its role on foetal development have been
assessed in different studies. In the study of Fronzak et al. (2003) low maternal vitamin
D intake via food was associated to higher risk in islet autoimmunity in the offspring.
Stene et al. (2000) concluded that the maternal intake of cod liver oil reduced the risk of
the offspring to develop T1D either due to vitamin D or the n-3 fatty acids
eicosapentaenoic acid or docosahexaenoic acid. No study has been conducted before
where the maternal S-25(OH)D status has been assessed and connected to T1D
development in the offspring to my knowledge.
39
1.3.1 Genetic association between vitamin D and Type 1 Diabetes
The reason that there is a seasonal variation (more incidents in autumn and winter
months) and a geographical gradient (more incidents in the north) in T1D diagnosis
suggests that sunshine and T1D are connected. The genetic associations between T1D
and vitamin D have therefore been investigated. Bailey et al. (2007) showed that gene
polymorphisms in CYP27B1 (gene that codes for vitamin D 1 -hydroxylase) are
associated to T1D. They identified the polymorphisms as -1260 and +2838. They did
not find any other T1D associated vitamin D related genes in their study. VDR receptor
polymorphisms have been associated to T1D (see Zella and DeLuca, 2003). One is
associated with glucose intolerance and another with insulin resistance. Four common
VDR variants have been described (see Zella and DeLuca, 2003). They differ in their
restriction enzyme sites (FokI, BsmI, ApaI and TaqI sites, polymorphisms at exon 2,
intron between exon 8 and 9 and exon 9). The BsmI restriction site alleles have been
associated to T1D susceptibility in Indian Asian, Taiwanese and German populations
whereas the FokI restriction site alleles have been associated to Japanese T1D
populations (see Zella and DeLuca, 2003).
1.3.2 Animal studies
Mathieu et al. (1994) showed in their study that 1,25(OH) D treatment prevented the
development of clinical diabetes in none-obese diabetic (NOD) mice which serve as an
animal model for human autoimmune diabetes. Mice were fed with 5 g 1,25(OH) D
per kg twice a day for 200 days. This led in the significant reduction of insulitis, a
leukocytic infiltration of the pancreatic islets, which leads to development of diabetes.
The mechanism by which 1,25(OH) D affects this is not quite clear. Mathieu et al.
(1994) suggested that due to the fact that 1,25(OH) D inhibits effector T-cell
proliferation and lymphokine (especially IL-2, IFN- and TNF- ) production, stimulates
monocyte differentiation, macrophage function and suppressor function in vitro, it
probably restores the defective suppressor mechanisms in NOD mice (Bhalla et al.,
1984, Rigby et al., 1987, Reichel et al., 1987, Meehan et al., 1992). They also concluded
that the problem with 1,25(OH) D when used in vivo, is the major effect on calcium
metabolism. The case mice did not develop hypercalcaemia, but the effect was seen on
bone turnover markers. Mathieu et al. (1994) therefore suggested that the treatment
40
should be done with an analogue of 1,25(OH) D which has decreased effects on
calcium metabolism.
NOD mice were partially protected from diabetes by sufficient vitamin D status
alone in one study (Zella et al., 2003). There 88% of vitamin D deficient female mice
were diabetic at 200 d age, whereas only 46% of vitamin D sufficient female mice were
diabetic by the same time. In the male mice the amounts were 44% and 0% respectively.
Zella et al. (2003) also reported the onset of diabetes much earlier in vitamin D deficient
mice than in vitamin D sufficient mice. Additionally they found that mice that where
fed 50 ng 1,25(OH) D per day were completely protected from diabetes regardless of
vitamin D status. They studied if a dose of 10 or 20 ng would be enough but received no
or only partial positive results.
The effect of early life vitamin D deficiency on diabetes onset in NOD mice by
breeding test mice parents and test mice in a ultra-violet free environment and feeding
them with vitamin D depleted diet for 100 days was studied by Giulietti et al. (2004).
All mice were followed 250 days when 35% of male vitamin D deficient mice and 66%
of female vitamin D deficient mice were diabetic compared to 15% and 45% of the
control mice. They showed that at 100 days, more vitamin D deficient mice were
glucose intolerant than control mice. The IL-1 concentration in islets of vitamin D
deficient mice was higher and the cytokine profile of their peritoneal macrophages was
abnormal, showing low IL-1 and IL-6, and high IL-15 concentrations. The vitamin D
deficient mice also had lower concentrations of L-selectin expressing T-lymphocyte
(CD4+CD62L+) cells in their thymus and lymph nodes. Giulietti et al. (2004)
concluded that the disturbed cytokine profile might contribute to the earlier beta-cell
destruction in NOD mice.
In conclusion, it has been shown that development of T1D in NOD mice has
been prevented with either vitamin D metabolite supplementation or sufficient vitamin
D status alone.
1.3.3 Human studies
1.3.3.1 Maternal vitamin D intake and status
The study of Maghbooli et al. (2007) showed that there is a relation between vitamin D
deficiency and insulin resistance in GDM patients. They studied 741 pregnant women
of which 52 developed GDM. Severe vitamin D deficiency (<12.5 nmol/l) was more
41
common in GDM patients than in normoglycaemic pregnancies. The insulin resistance
was calculated by this formula:
IShoma= (FPG*FPI)/22.5
where FPG=fasting plasma glucose and FPI=fasting plasma insulin. Homeostatic model
assessment index (HOMA index) values below 3.0 are normal and show no sign of
insulin resistance. The HOMA values were significantly higher in GDM patients. The
25(OH)D measurements where done with RIA. They defined 25(OH)D statuses as
severe deficiency when the concentration was <12.5 nmol/l, moderate deficiency when
it was between 12.5-24.9 nmol/l, mild deficiency when it was between 25-34.9 nmol/l
and normal when it was above 34.9 nmol/l. Their results showed that 70.6% of all
participants had vitamin D deficiency (<25 nmol/l). Insulin resistance was higher in
vitamin D deficient participants.
One study investigated if maternal dietary vitamin D intake had influence on
islet autoimmunity (IA) in offspring (Fronczak et al., 2003). The results suggested that
vitamin D through food during pregnancy may have a protective effect on the
appearance of IA in the offspring. The intake of vitamin D was evaluated from the third
trimester of pregnancy with a questionnaire (Willet Food Frequency Questionnaire
(FFQ)). The number of subjects was 233. Mothers with T1D children were identified
through the Diabetes Autoimmunity Study in the Young (DAISY). Children were
recruited to DAISY by screening for diabetes-susceptibility HLA-alleles from all
children born in the St. Joseph Hospital in Denver, Colorado or through identifying
first-degree relatives with T1D. The interesting results showed that vitamin D intake via
supplements was not related to a protective effect on IA in the offspring, only vitamin D
through food.
The effect of maternal intake of cod liver oil or multivitamin administration on
T1D development in the offspring was studied by Stene et al. (2000). The results were
obtained by a questionnaire. The study population consisted of 85 diabetic participants
and 1071 control subjects. The diabetic subjects were identified from the Norwegian
National Childhood Diabetes Registry and living in the county of Vest-Agder. Patients
born between 1982 and 1998 were selected. Controls were selected randomly from the
official population register of the county. Results showed that maternal intake of cod
liver oil strongly correlated negatively with the risk of T1D in the children. They found
42
out that there was no significant association between maternal multivitamin
administration and risk of T1D in the child.
To conclude, it can be said that vitamin D deficiency during pregnancy was
associated to insulin resistance in GDM patients. It seems that maternal vitamin D
intake via food or cod liver oil has a risk lowering effect on T1D development in the
offspring whereas vitamin D intake through supplements or multivitamins has no effect
on T1D development.
1.3.3.2 Early life vitamin D intake and status
The EURODIAB Substudy 2 (1999) investigated the association of early life vitamin D
intake with the risk of T1D in seven European study centres (Austria, Bulgaria, Latvia,
Lithuania, Luxembourg, Romania and Northern Ireland). They collected the results with
questionnaires and interviews. The study population consisted of 746 diabetic
participants and 2188 control subjects. They conducted that there was a clear negative
correlation between vitamin D intake during infancy and risk of diabetes.
The relation between cod liver oil intake in the first year of life or during
maternity and risk of T1D was assessed by Stene et al., 2003. Children diagnosed with
T1D between 1997 and 2000 and born between 1985 and 1999 were selected and the
number of participants was 545 patients. The control subjects were randomly selected
but they had to be born in the same time of year. The number of controls was 1668. The
results were obtained by questionnaire addressed to the mother. In contrast to their study
in 2000, they concluded that the maternal cod liver oil intake had no effect on the risk of
T1D in the offspring. On the other hand they found out that the intake during first year
of life was associated to lower risk of T1D.
The study of Hyppönen et al. (2001) was conducted to assess the relation
between early-life vitamin D intake and risk of T1D. The study population was
conducted of children who were born in Oulu and Lapland during 1966. 10 821 children
were followed up until end of 1997 for the possibility of T1D development. Of these
10 821, 81 children were diagnosed with diabetes. The vitamin D supplementation and
frequency of rickets was obtained by questionnaires to the mothers. The study resulted
that the development of T1D was associated with low vitamin D intake and rickets
during the first year of life. The vitamin D recommendations just changed in 1964 when
the dose was reduced from 4000-5000 IU (100-125 g) to 2000 IU (50 g).
Differences in vitamin D status between T1D children and non-diabetic children
were investigated in Qatar (Bener et al., 2009). The study population consisted of 170
43
T1D and control subjects below 16 years. The study was conducted during a period
between August and December 2007. Bener et al. (2009) defined that vitamin D
deficiency was a condition with serum 25(OH)D levels below 30 ng/ml (75 nmol/l) and
an optimal status was obtained with concentrations between 30 and 80 ng/ml (75 nmol/
and 200 nmol/l). The concentration of S-25(OH)D was measured with RIA. Their
results showed that vitamin D deficiency was higher in T1D children (90.6%) than in
healthy children (85.3%) although the prevalence was high in both groups. They also
studied by questionnaire if vitamin D supplementation along with breast feeding was
associated to lower T1D risk. They found out that vitamin D supplementation was poor
in both groups (46.5% and 51.8%) and again a little poorer in the T1D group. The
people in Qatar cover their whole body with clothing except their faces.
Svoren et al. (2009) studied the vitamin D status in young adults with T1D in
the North-eastern United States. The subjects were identified from the Joslin Diabetes
Centre, Boston and 128 T1D subjects participated in the study. S-25(OH)D was
measured with RIA. They defined the concentration 30 ng/ml (75 nmol/l) to indicate
sufficient status, 21 to 29 ng/ml (52.5-72.5 nmol/l) to indicate insufficient status and
20 ng/ml ( 50 nmol/l) to indicate deficient status. Their results showed that 61% had
insufficient S-25(OH)D status, 15% deficient and 24% sufficient status. Similar results
were found in the study of Pozzilli et al. (2005) where they studied the S-25(OH)D
status differences in newly diagnosed T1D patients and healthy controls in the Lazio
region in Italy. The S-25(OH)D concentrations were significantly lower in the T1D
group than in the control group.
Greer et al. (2007) studied the same hypothesis in Queensland, Australia. Their
study consisted of 47 T1D subjects and 94 control subjects. The S-25(OH)D
concentration was measured by RIA. The concentration they determined to indicate
vitamin D deficiency was 50 nmol/l. The amount of vitamin D deficient patients was
significantly higher in the T1D group (43%) than in the control group (18%). Samples
were stored samples which were taken at any time of the year.
To conclude, vitamin D intake during early life can be associated to lower risk
of T1D development later in life based on data obtained from questionnaires. This data
however is considered not to be highly reliable due to difficulties in remembering
vitamin D intake from years ago. Another aspect which can be concluded, is the fact
that S-25(OH)D concentrations are lower in young T1D subjects when compared with
healthy subjects.
44
1.3.3.3 Vitamin D status in adults
In 2006 one study investigated if plasma 25(OH)D levels were lower in young adult
T1D subjects than controls in Sweden (Littorin et al., 2006). The subjects were
identified from the Diabetes Incidence Study in Sweden (DISS) during 1987-1988. The
number of T1D patients was 138 and the number of controls was 208. The 25(OH)D
concentration was measured by the Nichols Advantage 25OHD assay. The mean
concentration of 25(OH)D was lower in T1D patients than controls (82.5 vs. 96.7
nmol/l). After a 8-year follow-up the concentrations were lower when the whole study
population was considered (86.3 vs. 81.5 nmol/l). Littorin et al. (2006) also showed that
the 25(OH)D concentration was significantly lower in diabetic men than women (77.9
vs. 90.1 nmol/l).
Borissova et al. (2003) showed that treating T2D patients with vitamin D (1332
IU=33.3 µg daily for one month) increased first phase insulin secretion by 34.3%. Ten
T2D women served as patients and 17 healthy women as controls. At baseline 70% of
the patients were deficient in 25(OH)D concentration whereas 70% had sufficient status
after treatment.
The possibility that subjects with hypovitaminosis D are at higher risk for
insulin resistance and metabolic syndrome was investigated by Chiu et al. (2004). They
studied 126 healthy glucose-tolerant subjects in Los Angeles, United States, for
25(OH)D concentration, glucose-concentration and insulin sensitivity. The results
showed that 47 subjects had 25(OH)D concentrations below 20 ng/ml (50 nmol/l). Low
25(OH)D concentration was associated with higher 60-, 90-, and 120-min post
challenge plasma glucose-concentrations. A positive correlation was found between
25(OH)D concentration and insulin sensitivity index (ISI). They concluded that low
25(OH)D concentration must therefore have an effect on beta-cell function and insulin
response. They found that 30% of the hypovitaminosis D subjects were at risk for
metabolic syndrome whereas the amount was only 11% in the non-hypovitaminosis D
subjects.
Bierschenk et al. (2009) compared the S-25(OH)D status in 415 study subjects.
The amount of healthy controls was 153, new-onset T1D subjects 46, established T1D
subjects (samples 5 months from diagnosis) 110 and first-degree relatives of the
diabetic patients 106. The age ranged considering all groups from 1.0 to 65.1 years. The
study was conducted in a solar-rich environment in Florida, United States. Measurement
was done with an EIA. Vitamin D deficiency was defined as a concentration 20 ng/l
45
(50 nmol/l), insufficiency as 21-30 ng/ml (52.5-75 nmol/l) and sufficiency as >30 ng/l
(>75 nmol/l). Even though their study was conducted in a solar-rich environment mean
S-25(OH)D concentrations in all groups were under a sufficient (>30 ng/l/75 nmol/l)
concentration. The groups did not differ significantly in their S-25(OH)D
concentrations. The concentrations ranged in healthy controls between 13.0 and 37.4
ng/ml (32.5-93.5 nmol/l), in new-onset T1D patients between 12.2 and 30.2 ng/l (30.5-
60.4 nmol/l), established T1D patients between 13.8 and 33.9 ng/l (34.5-84.8 nmol/l)
and first-degree relatives between 12.7 and 33.1 ng/ml (31.8-82.8 nmol/l). The group of
new-onset T1D patients had little lower concentrations than the other groups but not in a
significant amount.
To conclude the vitamin D status in adults, it can be said, that some studies
showed lower S-25(OH)D concentrations in T1D subjects when compared to controls
and others did not show differences between T1D and control subjects. Lower S-
25(OH)D concentration was interestingly found in T1D men than women. It can also be
concluded that vitamin D status can be associated to T2D, where sufficient vitamin D
status can prevent insulin resistance and metabolic syndrome. Vitamin D
supplementation can also enhance insulin secretion.
1.4 Vitamin D Measurement methods
1.4.1 Different methodology
For a laboratory measuring 25(OH)D it is highly important to participate in the
International Quality Assessment Scheme for Vitamin D metabolites (DEQAS) survey
(described in 1.4.2) for accurate measurement data. Circulating 25(OH)D is very stable
when stored in plasma or serum. Hollis (2008) reported that his laboratory has used
same pooled human 25(OH)D internal controls for over 10 years stored at -20 ºC. They
noticed none detectable degradation of the compound in the samples during that time.
The samples provided by the DEQAS are also shipped by ground post to participants
and no degradation has been seen there either. Wielders and Wijnberg (2008) reported
their results on 25(OH)D stability in 2008. They showed that a mean decrease of
25(OH)D concentration of 2.3% was seen when whole blood samples were left on the
bench for 72 h. When serum samples were left on the bench a mean decrease of 3.4%
(24 h) and 8.5% (7 d) was observed. About same results were obtained when serum
samples were left on the bench in the dark (4.5% and 8.1%). After 7 days of storage in
46
the refrigerator mean decrease of 1.8% was measured. Wielders and Wijnberg (2008)
also tested the effect of repeated freeze-thaw cycles. After 4 cycles of freeze-thawing
the mean increase of 25(OH)D concentrations was 2.6%. The cis-triene structure of
25(OH)D is responsible for the molecules possible oxidation, UV induced
conformational changes, heat-induced conformational changes and possibility of attacks
by free radicals (Hollis, 2008). Hollis (2008) made an experiment where he exposed
crystalline 25(OH)D in ethanol on an open Petri dish to UV light. The molecule was
destroyed in a few minutes. When the molecule was exposed to UV light in serum or
plasma there was no degradation detected after 2 days of UV exposure. Two reasons for
this could be that first UV light penetrates aqueous liquids poorly and second vitamin D
and its metabolites are mostly bond to DBP or other proteins in serum and plasma. This
complex seems to protect the degradation of 25(OH)D.
It is important to find simple, high-throughput methods for measuring 25(OH)D
from patient samples because of the rising need in diagnosing vitamin D deficiency, and
other vitamin D related metabolic statuses (Wagner et al., 2009). Vitamin D nutritional
status (represented by the serum/plasma status of 25(OH)D)) is important for
monitoring bone health and vitamin D deficiency and intoxication. It is also important
for diagnosing rickets, secondary hyperparathyroidism and osteoporotic fracture (see
Kimball and Vieth, 2007). Vitamin D nutritional status is measured by determining the
serum concentration of 25(OH)D. In all species where the conversion of vitamin D to
25(OH)D has been measured the change has been rapid. 25(OH)D has a much longer
shelf-life than vitamin D and is therefore a good metabolite to measure vitamin D
nutritional status (see Bouillion et al., 1998).
When vitamin D wants to be measured different approaches are used to define
vitamin adequate and insufficient status. The relation between 25(OH)D and PTH can
be studied to find some 25(OH)D concentrations where PTH concentrations begin to
rise over appropriate healthy level. The optimal 25(OH)D concentration for optimal
intestinal absorption of calcium can be estimated and studied or the relation between
25(OH)D concentrations and frequency of various diseases can be the site of interest.
On the other had studying the relation of mean concentrations of 25(OH)D and positive
effects, like disease prevention can be the sight of aspect (Cavalier et al., 2009). As
mentioned before the measuring of vitamin D is challenging. The most difficult
problems of the molecule are that it is strongly bound to binding protein (DBP or other)
in the blood, that it is present in low (nanomolar) concentrations, and in two different
forms (D and D ) (Wagner et al., 2009). The particular molecule 25(OH)D is
47
challenging to measure because of its lipophilic character. It is really prone to protein
binding assays matrix effects (something in the sample tubes, which is not in the
standard tubes causes matrix effects, which alter the ability of binding agent, antibody,
or binding protein to associate with 25(OH)D from the way of which they associate with
the standard) (see Hollis, 2008).
The two approaches to measure 25(OH)D are competitive immunoassays and
methods based on chromatographic separation followed by direct detection (see Wallace
et al., 2010). The first method to measure 25(OH)D was a competitive protein binding
assay (CPBA) by Haddad in 1971. The assay measures circulating 25(OH)D by using
DBP as a primary binding agent and Human-25(OH)D as a reporter. The method is
challenging because, the sample has to be extracted with organic solvent, dried with
nitrogen and purified using column chromatography. The procedure requires also
sample 25(OH)D concentration estimation, because of endogenous loss during the
method. The CPBA of Haddad has been tried to modify towards an easier procedure
involving no chromatographic purification, but those attempts had other problems like
matrix problems due to ethanolic sample extraction (see Hollis, 2008). Today CPBA is
a rarely used method because of its cumbersome character. In the early 1980s the group
of Hollis came up with a radioimmunoassay (RIA) where an antigen is used that
generates an antibody specific for 25(OH)D and 25(OH)D . It is a simple non-
chromatographic quantification. In 1992 125-I-labeled reporter and calibrators as well
as standards for mass assessment were introduced. The method is good because of its
accuracy in detecting the whole circulating 25(OH)D content of the sample. A few
manufacturers are on the market with RIAs. Some of them detect both 25(OH)D and
25(OH)D (see Hollis, 2008). The unfavourable aspect of the RIA is the nature of the
radioactivity where special trained personal has to be performing the method. The
strengths of RIAs are the relative inexpensiveness, the minimization of matrix effects
after solvent extraction and the ease of handling. The drawbacks are the solvent
extraction, creation of radioactive waste and possible lot variation (see Wallace et al.,
2010).
High performance liquid chromatography (HPLC) is considered to be the
standard method to measure 25(OH)D status at the moment. The method separates and
quantificates 25(OH)D and D individually. The method is good because it is highly
repeatable. The down side of the HPLC is that it is cumbersome. It requires large
48
sample size and a radioactive internal standard. The sample throughput is slow and a
technician has to be highly trained and dedicated to complete the task (see Hollis, 2008).
Liquid chromatography-mass spectrometry (LC-MS) quantificates D and D
metabolites separately. It is a good method because it is really accurate and sample
throughput is larger than in HPLC. Its down sides are that it is really expensive and it
needs an internal standard for each component. Also LC-MS needs highly trained
personal to perform the analysis. It also has a special problem: it does not distinguish
between 25(OH)D and its inactive form 3-epi-25(OH)D (see Hollis, 2008). Still it is
thought to become the new gold standard method for 25(OH)D measurement because of
its accuracy. Enzyme immunoassays (EIA) are also used to determine 25(OH)D status
from serum or plasma. The advantages of the EIA are the ease of handling, the relative
inexpensiveness and the safety due to no radioactivity. Most EIAs also measure both
25(OH)D and 25(OH)D . The drawbacks however are the susceptibility to matrix
effects, the possible variation between lots and the possible underestimation of
25(OH)D (see Wallace et al., 2010).
1.4.2 The International Quality Assessment Scheme for Vitamin D metabolites
(DEQAS)
The International Quality Assessment Scheme for Vitamin D metabolites (DEQAS) was
founded in 1989. It was established to control the reliability of 25(OH)D measurement
assays (see Kimball and Vieth, 2007). In 1997 it expanded to control also the quality of
1,25(OH) D measurement methods. The number of participating laboratories has grown
from 25 in 1989 to over 100 in 18 countries in 2004 (see Hollis, 2004). The number of
participants has in 2010 reached almost 700 laboratories (Figure 5).
The participating laboratories receive five normal human serum samples for
total 25(OH)D measurement every three months and have about five weeks to report
their results for the statistical analysis (Carter et al., 2004, www.deqas.org). Samples
usually contain only 25(OH)D but occasionally also 25(OH)D . The serum samples
are posted in January, April, July and October. Before distributing them to participants
the samples are screened for Human Immunodeficiency Virus (HIV), Hepatitis B and C.
The “true” target value is created using gas chromatography-mass spectrometry
(see Zerwekh, 2008). When the statistical analysis has been made all participants
49
receive a report via e-mail or post with the All-Laboratory Trimmed Mean (ALTM),
Standard Deviation (SD) and Coefficient of Variance (CV) for every sample. The
accuracy of each result is measured by the percentage amount it differs from the ALTM.
All results are also grouped by their method and therefore create a method mean (MM).
The overall method accuracy is counted by this formula:
{[(MM – ALTM)/ALTM] x 100} (Carter et al., 2004).
All results can be found on the Internet (since 2004) if the laboratory is participating in
the survey. Results are not available for third part individuals. The report also includes
the ALTM, SD and CV by major method groups and individual histograms. When
participants return their results via Internet, the updated statistics and histograms are
immediately available to the participant (www.deqas.org).
The methods which correlated best with the ALTM in the four-year comparison
from 2000 to 2004 were the RIA from Diasorin and the competitive protein binding
assay. The method which varied the most from the mean ALTM was the Nichols
automated chemiluminescence method (Carter et al., 2004). The Figure 10 shows the
number of participated laboratories illustrated together with the number of each method
used in January 2010. The number of participating laboratories has grown from 609 in
October 2009 to 689 in January 2010. More and more laboratories seem to recognise the
importance of adequate quality control monitoring in 25(OH)D measurement methods.
0
50
100
150
200
250
300
Automate
d IDS E
IA
Biosourc
e
DiaSorin R
IAHPLC
IDS EIA
IDS R
IA
IDS-iS
YSLC
-MS
Roche
25OHD3
Unknown
CLBA
Diasori
n Liaison
Method
Num
ber
of L
abor
ator
ies
50Figure 10. Number of laboratories participating in the survey in January 2010 dealt to different
blocks depending on method.
The popularity of immunoassays is clearly seen in the reports of the DEQAS assessment
group. The number of different methods in January 2010 is clearly dominated by
immunoassays (automated IDS EIA, Diasorin RIA, IDS EIA, IDS RIA, IDS-ISYS,
Roche 25OHD3 and Diasorin Liason) (Figure 10). The most popular method was the
Diasorin Liason method and the second most used was the manual EIA from IDS.
If one wishes to check the accuracy of ones method but can not take part in the
DEQAS survey, one can purchase samples when needed. For 25(OH)D the amount for
ordering is GBP150 per year. After reporting results a full statistical analysis is made by
the DEQAS.
The DEQAS advisory panel consists of scientists with expertise in vitamin D
methodology and/or quality assessment schemes. If a scientist has problems regarding
sample preparation and distribution, vitamin D methodology, statistical analysis and
data presentation or setting of performance targets the panel is happy to help with these
questions. When a laboratory has met the performance target set by the DEQAS
Advisory panel it will receive annually a proficiency certificate from the panel
(www.deqas.org). Another quality control assessment scheme is the EQAS for vitamin
D measurement assays. As the DEQAS is located in the United Kingdom this one is
located in Finland (www.labquality.fi). Twice a year they send two serum samples for
participants and provide a statistical analysis of the results. The true value here is
generated using HPLC atmospheric pressure ionization electro spray (API-ES) MS
(www.labquality.fi).
2 Objectives
Hypothesis
Low serum vitamin D status during first trimester of pregnancy increases the risk of
T1D development in the offspring.
2.1 Aim of this study
The aim of this study was to solve if vitamin D is one environmental factor associated to
T1D development, especially if vitamin D status of the mother during the first trimester
of pregnancy has any effect on T1D prevalence in the offspring. Serum 25(OH)D
51
measurements were done to compare the status of mothers with T1D offspring and
mothers with healthy offspring. This study is part of a more comprehensive study.
3 Materials and methods
3.1 Finnish Maternity Cohort
The Finnish Maternity Cohort is a serum bank, which in the end of 2006, hold up over
1.3 million samples collected from 720 000 female donators (Toriola et al., 2009). It is
therefore the world`s largest serum bank for female donors. It was established in 1983
by the National Institute for Health and Welfare. Over 98% of pregnant women in
Finland have given a sample to the serum bank. If withdrawal has occurred after
1.5.2001 informed consent is needed (www.thl.fi). Samples withdrawn before that are
stored and used for research aiming for population based health promotion, due to the
law considering this, which came into force 1.5.2001. The samples are withdrawn at a
municipal maternity clinic during the first trimester of pregnancy. The reason for taking
these serum samples is to screen for infections. When the screening is done, the
remaining sample (1-3 ml) is stored at the laboratory of maternity serology at the
National Institute for Health and Welfare in Oulu, Finland (www.thl.fi). The samples
are stored at -25 ºC in polypropylene cryo vials. The samples are used for assessing the
association of various biomarkers to diverse diseases of either the mother or the child
(www.thl.fi, Toriola et al., 2009).
3.2 Samples
The study was designed as a case-control study where cases were matched with controls
for serum sample withdrawal date. The cases were identified from the National Institute
for Health and Welfare T1D database by identifying all in or after 1994 born T1D
offspring and therefore the whole family and especially the mother. Samples from 310
women with T1D offspring and 310 women with healthy offspring were analysed for S-
25(OH)D. The samples were all-year-round samples, where the control was always
matched with the case for sample withdrawal date. The contact information was
provided by the Population Register Centre of Finland. Written consent was collected
from all participants. Serum samples were stored at -20 ºC until analysed.
52
3.3 25-hydroxyvitamin D measurement
The method, to measure the serum 25(OH)D concentration, used was the IDS OCTEIA
enzyme immunoassay (EIA) (Immunodiagnostic Systems Ltd.). This method is done on
a 96-well microtiter plate, which has a specific 25(OH)D-antibody attached to the wells.
It is an indirect detection method, meaning that the 25(OH)D content of the samples is
not measured directly but through a competing labelled 25(OH)D. The sensitivity,
meaning the concentration corresponding to the mean minus two standard deviations of
ten replicates of the zero calibrator, of the kit is <6 nmol/l (IDS Ltd.). The specificity of
the kit regarding vitamin D metabolites is shown in Table 7. Nevertheless the overall
amount of further hydroxylated metabolites in the circulation is quite small ( 6%) and
can be seen as insignificant (see Zerwekh, 2008). The CVs that should be reached for
intra-assay and interassey recommended by the manufacturer are <8% and <10%
respectively. The assay detects a S-25(OH)D concentration range between 6 and 360
nmol/l. The samples that can be analysed by this method are serum or plasma samples
and the amount needed for single determinations is 12.5 µl. When samples are analysed
by single determination 76 samples can be analysed on one plate.
Table 7. Cross-reactivity to vitamin D metabolites
Metabolite Cross-Reactivity
25(OH)D 100 %
25(OH)D 75 %
24,25(OH) D 100%
D <0.01%
D <0.30%
Controls and calibrators provided by the manufacturer are lyophilised, buffered human
serum samples containing 25(OH)D and <1% sodium azide. Samples, kit controls and
calibrators are first diluted with biotin labelled 25(OH)D (lyophilised buffer containing
25(OH)D labelled with biotin and stabilizers). To 12.5 µl of sample 500 µl of 25(OH)D-
biotin solution is added. When diluting controls and calibrators, 25 µl of
calibrator/control is added to 1 ml of 25(OH)D-biotin solution. The solution is added to
all to ensure the dissociation of vitamin D from its binding proteins. The mixture is
vortexed thoroughly and 200 µl of each is added to the anti-25(OH)D coated (25(OH)D
sheep polyclonal antibody) plate where the biotin labelled 25(OH)D and the samples
53
25(OH)D compete to bind the antibody. The plate is then incubated at room temperature
for two hours. After incubation the plate is washed trice with the plate washer (300 µl of
phosphate buffered saline containing Tween per well per cycle) to remove any unbound
and excess biotin labelled 25(OH)D. To detect bound 25(OH)D-biotin conjugate, 200 µl
of avidin horseradish peroxidase (HRP) is added to each well and incubated for 30 min
at room temperature. After incubating the washing step is repeated. The chromogenic
substrate used here for avidin-HRP is tetramethylbenzidine (TMB and hydrogen
peroxide as an oxidizing agent) which forms a blue product when reacting with
peroxidase enzymes. The reaction mixture is again incubated for 30 min and then
stopped by adding 100 µl of 0.5 M Hydrochlorid acid (HCl). This turns the product
yellow and the intensity can be read at 450 nm (650 nm). The absorbance is inversely
proportional to the concentration of 25(OH)D in the samples. The more intense the
colour formation the more biotin labelled 25(OH)D and the less 25(OH)D originating
from the samples is present. The method is illustrated in Figure 11. The results are
calculated by the standard curve composed of the calibrators 0-6 which have 25(OH)D
concentrations of 0 nmol/l, 6.8 nmol/l, 14.0 nmol/l, 27.0 nmol/l, 67.0 nmol/l, 179.0
nmol/l and 380 nmol/l. The results were processed computationally, to create a B/Bo
standard curve and therefore calculate the S-25(OH)D concentrations of the samples.
54
Add 12,5 µl sample Add 500 µl/1 ml 25(OH)D-Biotin Add 200 µl of Solution mixture to plate
Add 200 µl Avidin HRP Wash 3x Incubate 2 hours RT
Incubate 30 min RT Wash 3x
Incubate 30 min Add 200 µl TMB substrate to plate
Add 100 µl stop reagentto plate
read absorbance at 450 nm
Figure 11. Serum-25(OH)D measurement method with EIA.
To improve repeatability and reliability all samples were analysed using a single batch
of assay kits and analysed in the same laboratory (University of Helsinki, Department of
Food and Environmental Sciences, Finland). Cases and matched controls were always
analyzed in the same intra-assay. The repeatability and reliability was inwardly assessed
with controls provided for each kit by the manufacturer and one own (Pool) sample of
the laboratory. The CVs for intra- and interassay were 3.57% and 3.68% and therefore
remained in the recommended values. The quality control was further strengthened by
55
participating in the DEQAS during this work (processing of DEQAS samples in
October 2009). The own results seemed to be slightly higher than the ALTM and the
MM (Figure 12). The CV for the IDS EIA was 12.9 %. The bias from the ALTM and
MM are shown in Table 8. This method gives higher values than both the ALTM and
the MM. The method usually gives lower concentrations than the ALTM (Method
accuracy [(MM – ALTM)/ALTM] x 100 = -2,67).
Table 8. Bias from ALTM and MM in October 2009.
Sample Bias from ALTM (nmol/l) Bias from MM (nmol/l) 361 20.9 25.1 362 2.8 10.2 363 4.2 8.5 364 13.2 14.2 365 24.6 22.7
0
10
20
30
40
50
60
70
80
90
361 362 363 364 365
Sample
Conc
entra
tion
(nm
ol/l)
ALTMMMOwn Result
Figure 12. Comparison of own results to ALTM and MM in October 2009.
S-25(OH)D concentration below 25 nmol/l was considered to indicate vitamin D
deficiency, concentrations between 25 and 50 nmol/l to indicate insufficiency and
concentrations between 50 and 80 nmol/l to indicate sufficiency. Concentrations over 80
nmol/l were considered to indicate optimal level of S-25(OH)D.
56
3.4 Laboratory equipment
The samples were diluted for analyses in polypropylene tubes to prevent the 25(OH)D
in sticking to the tubes. Precision pipetting device to deliver 12.5 µl, 25 µl, 200 µl and 1
ml were used. Repeating pipetting devices to deliver 500 µl and 1 ml were used. To add
samples and reagents to microwells, precision pipetting device to deliver 200 µl and
multichannel pipette to deliver 200 µl and 100 µl were used. Highly purified water was
used for solubilisation of controls and calibrators. Plates were washed with an automatic
plate washer dispensing 300 µl per well and aspirating during three cycles per washing
step. The absorbance was measured by a photometric plate reader at 450 nm and 650
nm. The results were calculated with a curve-fit software.
3.5 Statistical analysis
Statistical analysis was carried out using SPSS 15.0 for Windows (SPSS Inc., Chicago,
USA). The variables were tested for normality and a paired t-test was performed. A
two-sided p < 0.05 was considered statistically significant. Normally distributed data
was presented as means and standard deviation.
4 Results
4.1 Vitamin D deficiency in the whole study population
When considering the whole study population 72% (N=445) had insufficient or
deficient S-25(OH)D status (<50 nmol/l) during the first trimester of pregnancy. Only
26% (N=162) had sufficient (50-80 nmol/l) status and 2% (N=13) optimal (>80 nmol/l)
status (Figure 13).
57
72 %
26 %
2 %
Deficiency and Insufficiency (<50nmol/l)
Sufficiency (50-79.9 nmol/l)
Optimal (>80 nmol/l)
N=162
N=13
N=445
Figure 13. Vitamin D status in Finnish pregnant women. The situation of the whole study population.
4.2 Vitamin D status comparison between mothers with healthy offspring and
mothers with T1D offspring
The S-25(OH)D concentrations did not differ between cases and controls (p=0.748)
(Figure 14, Table 9). The percentage of women having a S-25(OH)D below 25 nmol/l,
meaning deficient status was 9.7% in the case group and 8.7% in the control group (Fig.
9). The biggest group in both study groups was the one with S-25(OH)D concentrations
between 25 nmol/l and 50 nmol/l which is characterized as insufficient status. There the
percentage of women from the case group was 61.0% and from the control group 64.1%
(Figure 14). Sufficient S-25(OH)D status (50-80 nmol/l) was reached by 27.4% and
28.7% of cases and controls respectively (Figure 14). Optimal status of S-25(OH)D
(>80 nmol/l) was measured in 1.9% of cases and 2.3% of controls (Figure 14). Mean S-
25(OH)D concentration in cases was 43.3 ± 15.9 nmol/l (Mean ± SD) and 43.0 ± 15.5
nmol/l in controls.
58
0
10
20
30
40
50
60
70
80
90
100
0-25 25-50 50-80 >80
Concentration (nmol/l)
Per
cent
age
(%)
Control
Case
N=27 N=30
N=199N=189
N=77 N=85
N=7 N=6
Figure 14. Serum-25(OH)D concentration between control and case groups measured from samples
during first trimester of pregnancy.
Table 9. Statistical variables for different S-25(OH)D concentration groups.
Cases Controls
S-25(OH)D N Range Mean Median SD N Range Mean Median SD
0-25 nmol/l 30 15.7-25.0 21.99 22.69 2.81 27 14.3-25.0 22.20 23.48 2.82
25-50 nmol/l 189 25.5-49.7 37.04 36.59 7.02 199 25.5-49.9 37.20 36.79 6.96
50-80 nmol/l 85 50.7-77.7 61.31 58.87 8.33 77 50.1-77.6 60.74 58.91 7.49
>80 nmol/l 6 82.7-100.5 92.09 92.46 6.53 7 81.5-114.9 92.00 89.50 11.47
Total 310 15.7-100.5 43.30 39.88 15.93 310 14.3-25.0 42.98 40.85 15.50
5 Discussion
The presented results show a high incident of vitamin D deficiency in pregnant women
in Finland. This is the first study to assess this during first trimester of pregnancy in
Finland. The mean concentrations in both control and case groups were barely half of
the concentration considered as a lowest value for optimal status. This is a result
presented by numerous studies (Holmes et al., 2009, Toriola et al., 2009). Low statuses
of S-25(OH)D may be due to the reason that recommended doses for pregnant women
59
for vitamin D are quite low (10µg per day) and they are not followed. The results are an
overall mean for the whole year. Holmes et al. (2009) showed in their study that vitamin
D supplementation of pregnant women did neither prevent vitamin D deficiency (<25
nmol/l) during winter and spring time nor prevent insufficiency (<50 nmol/l) during the
whole year.
No significant relationship between low S-25(OH)D during pregnancy and risk
of T1D development in the offspring was observed. The study Fronczak et al. (2003)
and the study of Stene et al. (2000) both showed that association between risk of T1D in
the offspring was associated to vitamin D intake via natural sources but not
supplements. This could indicate that the beneficial effect of vitamin D on T1D
prevention in the offspring is only achieved with the help of some other metabolites
(vitamin A, fatty acids) together with vitamin D. Also in the study of Hyppönen et al.
(2001) it is difficult to say if the positive effect was due to vitamin D, the combined
effect of vitamin A and D or just vitamin A supplementation, due to the fact that
vitamin D and A supplementation was combined in one supplement at the time of the
study (Ala-Houhala et al., 1995). Another possible reason for the natural vitamin D to
be more efficient in T1D prevention is the fact that it is metabolized differently in the
body than synthetic forms found in supplements. Former studies where the association
between vitamin D status and T1D development in the offspring has been show might
also be due to false positive results. In this study no association was observed and also
impossible to detect for the reason that all concentrations were low. It might be possible
that some Finnish mothers still avoid fish and mushrooms in their diet, due to the
warnings that fish might contain to much mercury and mushrooms might contain
radioactive compounds resulting from the Chernobyl disaster in 1989. The mothers
hence consume vitamin D not from natural sources but from supplements if they follow
the recommendations. Lamberg-Allardt et al. (1984) investigated the S-25(OH)D
concentrations at time of delivery from maternal and cord blood and also the vitamin D
intake of mothers in Finland. They showed that the intake (5.5 µg/d) did not meet the
recommendations of the Food and Nutrition board at that time (10 µg). They also
showed that mothers who consumed vitamin D supplements had higher S-25(OH)D
concentrations than mothers who did not consume supplements (87 nmol/l vs. 59
nmol/l). Studies where there has been a beneficial effect of mothers S-25(OH)D during
pregnancy and lower risk of T1D in the offspring have not been reported. It might be
that low vitamin D status is associated to T1D development in the offspring only when
genetic factors are prevalent also. These genetic factors can be readily investigated in
60
this study population due to the fact that the S-25(OH)D concentrations are so similar in
cases and controls.
Narchi et al. (2010) showed in their study that S-25(OH)D concentrations of
pregnant women lowered during pregnancy. They measured the concentrations once at
antenatal visit, once after birth and once 6 months after birth. The amount of women
having adequate status fell from 31% to 23% and finally 17% respectively. This could
suggest that if possible serum samples from later stages of pregnancy and possibly close
to child birth should be analysed for S-25(OH)D to see if this decrease in S-25(OH)D
concentration is also obtained in the present study. Unfortunately samples from later
stages of pregnancy are not available. Holmes et al. (2009) indicated that the increased
maternal calcium absorption together with increased renal calcium loss during
pregnancy leads to significant changes in vitamin D metabolism. They measured the S-
25(OH)D statuses during pregnancy at 12 weeks (1st trimester), 20 weeks and 35 weeks
gestation. Vitamin D deficient (<25 nmol/l) were 35%, 44% and 16% and insufficient
(<80 nmol/l) were 96% (12 and 20 weeks) and 75% at these measurement points. This
indicates that the S-25(OH)D status was lowest at 20 weeks of gestation. It could be
possible that the foetal demands for calcium and vitamin D are highest at this stage of
pregnancy and hence this could be a critical point for maternal vitamin D status during
pregnancy.
Strengths of this study include the population-based nature and the prospective
nature. The FMC contains samples of 98% of Finnish women and the samples were
collected before offspring T1D diagnosis and invitation to this study. Study subjects
have not been possible to change their eating habits or intake of vitamin D supplements
for this study. A limitation is the absence of samples during later pregnancy and time of
child birth which could give further information about the possible effects of vitamin D
on T1D development. But these samples are not available.
6 Conclusions
In conclusion S-25(OH)D status during pregnancy 1st trimester is not associated to T1D
development in the offspring. Other samples from during pregnancy and close to birth
should be analysed to assess if the S-25(OH)D concentration during whole pregnancy is
associated to T1D development in the offspring. This unfortunately is impossible due to
the fact that samples from later stages of pregnancy are not available. This study
showed for the first time in Finland that the prevalence of vitamin D deficiency and
61
insufficiency is high in pregnant women also at the first trimester of pregnancy. The
vitamin D recommendations should be raised in order to provide adequate statuses for
the whole population. The following of the recommendations should be monitored
somehow and vitamin D supplements could be given for free to pregnant women during
their first visits at maternity clinics.
7 Acknowledgements
The experimental part of this work was done in the Calcium Research Unit of Docent
Christel Lamberg-Allardt, in the division of Nutrition at the Department of Food and
Environment Sciences of the University of Helsinki. I thank my supervisors Docent
Christel Lamberg-Allardt and M. Sc. Maija Miettinen for the guidance and advice for
the experimental and written work. I wish to thank Anu Heiman-Lindh for the excellent
guidance for the experimental part of my work. I would also like to thank Ph. D. Heli
Viljakainen for the additional advice during my work. Finally a special thanks to all the
other people who helped me with my experimental or written work either in Viikki or
Meilahti.
This research was part of the joint study Vitamin D and Type 1 Diabetes
between the National Institute for Health and Welfare and the University of Helsinki.
The study was funded by the the Finnish Diabetes Association and the Academy of
Finland.
9 References
Adams, J. S., Chen, H., Chun, R., Gacad, M. A., Encinas, C., Ren, S., et al. (2004). Response element binding proteins and intracellular vitamin D binding proteins: Novel regulators of vitamin D trafficking, action and metabolism. The Journal of Steroid Biochemistry and Molecular Biology, 89-90(1-5), 461-465.
Agborsangaya, C. B., Surcel, H. M., Toriola, A. T., Pukkala, E., Parkkila, S., Tuohimaa, P., et al. (2010). Serum 25-hydroxyvitamin D at pregnancy and risk of breast cancer in a prospective study. European Journal of Cancer (Oxford, England : 1990), 46(3), 467-470.
Ala-Houhala, M., Sorva, R., Pelkonen, A., Johansson, C., Stählberg, M-R., Hakulinen, A., Lautala, P., Visakorpi, J., Perheentupa, J. (1995). Riisitaudin uusi tulemine-esiintyvyys, diagnostiikka ja hoito. Duodecim, 111(4):337.
Atkinson, M. A., & Eisenbarth, G. S. (2001). Type 1 diabetes: New perspectives on disease pathogenesis and treatment. Lancet, 358(9277), 221-229.
Atkinson, M. A., & Maclaren, N. K. (1994). The pathogenesis of insulin-dependent diabetes mellitus. The New England Journal of Medicine, 331(21), 1428-1436.
62
Bailey, R., Cooper, J. D., Zeitels, L., Smyth, D. J., Yang, J. H., Walker, N. M., et al. (2007). Association of the vitamin D metabolism gene CYP27B1 with type 1 diabetes. Diabetes, 56(10), 2616-2621.
Baker, A. R., McDonnell, D. P., Hughes, M., Crisp, T. M., Mangelsdorf, D. J., Haussler, M. R., et al. (1988). Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proceedings of the National Academy of Sciences of the United States of America, 85(10), 3294-3298.
Barnes, M., Robson, P., Bonham, M., Strain, J., Wallace, J. (2006). Effect of vitamin D supplementation on vitamin D status and bone turnover markers in young adults. European Journal of Clinical Nutrition, 60, 727-733.
Bener, A., Alsaied, A., Al-Ali, M., Al-Kubaisi, A., Basha, B., Abraham, A., et al. (2009). High prevalence of vitamin D deficiency in type 1 diabetes mellitus and healthy children. Acta Diabetologica, 46(3), 183-189.
Bhalla, A. K., Amento, E. P., Serog, B., & Glimcher, L. H. (1984). 1,25-dihydroxyvitamin D3 inhibits antigen-induced T cell activation. Journal of Immunology (Baltimore, Md.: 1950), 133(4), 1748-1754.
Bierschenk, L., Alexander, J., Wasserfall, C., Haller, M., Schatz, D., & Atkinson, M. (2009). Vitamin D levels in subjects with and without type 1 diabetes residing in a solar rich environment. Diabetes Care, 32(11), 1977-1979.
Blunt, J. W., DeLuca, H. F., & Schnoes, H. K. (1968). 25-hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry, 7(10), 3317-3322.
Borissova, A. M., Tankova, T., Kirilov, G., Dakovska, L., & Kovacheva, R. (2003). The effect of vitamin D3 on insulin secretion and peripheral insulin sensitivity in type 2 diabetic patients. International Journal of Clinical Practice, 57(4), 258-261.
Boucher, B. J. (1998). Inadequate vitamin D status: Does it contribute to the disorders comprising syndrome 'X'? The British Journal of Nutrition, 79(4), 315-327.
Bouillon, R., Carmeliet, G., Daci, E., Segaert, S., & Verstuyf, A. (1998). Vitamin D metabolism and action. Osteoporosis International : A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA, 8 Suppl 2, S13-9.
Brown, A. J., Dusso, A., & Slatopolsky, E. (1999). Vitamin D. The American Journal of Physiology, 277(2 Pt 2), F157-75.
Carnevale, V., Modoni, S., Pileri, M., Di Giorgio, A., Chiodini, I., Minisola, S., et al. (2001). Longitudinal evaluation of vitamin D status in healthy subjects from southern italy: Seasonal and gender differences. Osteoporosis International : A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA, 12(12), 1026-1030.
Carter, G. D., Carter, R., Jones, J., & Berry, J. (2004). How accurate are assays for 25-hydroxyvitamin D? data from the international vitamin D external quality assessment scheme. Clinical Chemistry, 50(11), 2195-2197.
Cavalier, E., Rozet, E., Gadisseur, R., Carlisi, A., Monge, M., Chapelle, J. P., et al. (2009). Measurement uncertainty of 25-OH vitamin D determination with different commercially available kits: Impact on the clinical cut offs. Osteoporosis International : A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA,
Chen, H., Hewison, M., Hu, B., & Adams, J. S. (2003). Heterogeneous nuclear ribonucleoprotein (hnRNP) binding to hormone response elements: A cause of vitamin D resistance. Proceedings of the National Academy of Sciences of the United States of America, 100(10), 6109-6114.
63
Chen, T., C., Chimeh, F., Lu, Z., Mathieu, J., Person, K., S., Zhang, A., et al. (2007). Factors that influence the cutaneous synthesis and dietary sources of vitamin D. Arch.Biochem.Biophys, 460, 213.
Chiu, K. C., Chu, A., Go, V. L., & Saad, M. F. (2004). Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction. The American Journal of Clinical Nutrition, 79(5), 820-825.
Cox, N. J., Wapelhorst, B., Morrison, V. A., Johnson, L., Pinchuk, L., Spielman, R. S., et al. (2001). Seven regions of the genome show evidence of linkage to type 1 diabetes in a consensus analysis of 767 multiplex families. American Journal of Human Genetics, 69(4), 820-830.
Daiger, S. P., Schanfield, M. S., & Cavalli-Sforza, L. L. (1975). Group-specific component (gc) proteins bind vitamin D and 25-hydroxyvitamin D. Proceedings of the National Academy of Sciences of the United States of America, 72(6), 2076-2080.
Dawson-Hughes, B., Heaney, R. P., Holick, M. F., Lips, P., Meunier, P. J., & Vieth, R. (2005). Estimates of optimal vitamin D status. Osteoporosis International : A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA, 16(7), 713-716.
Deluca, H. F., & Cantorna, M. T. (2001). Vitamin D: Its role and uses in immunology. The FASEB Journal : Official Publication of the Federation of American Societies for Experimental Biology, 15(14), 2579-2585.
Demay, M. B., Gerardi, J. M., DeLuca, H. F., & Kronenberg, H. M. (1990). DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proceedings of the National Academy of Sciences of the United States of America, 87(1), 369-373.
DEQAS. International Quality Assessment Scheme for Vitamin D metabolites. Retrieved 02/04, 2010, from www.deqas.org.
Eisenbarth, G. S. (1986). Type I diabetes mellitus. A chronic autoimmune disease. The New England Journal of Medicine, 314(21), 1360-1368.
EQAS. Finnish Quality Assessment Scheme for Vitamin D metabolites. Retrieved 02/04, 2010, from www.labquality.fi.
Equils, O., Naiki, Y., Shapiro, A. M., Michelsen, K., Lu, D., Adams, J., et al. (2006). 1,25-dihydroxyvitamin D inhibits lipopolysaccharide-induced immune activation in human endothelial cells. Clinical and Experimental Immunology, 143(1), 58-64.
EURODIAB, Substudy, & 2. (1999). Vitamin D supplement in early childhood and risk for type I (insulin-dependent) diabetes mellitus. the EURODIAB substudy 2 study group. Diabetologia, 42(1), 51-54.
EVIRA. Finnish Food Safety Authority. Retrieved 02/04, 2010, from www.evira.fi. FINELI. Database for Food ingredients in Finland. Retrieved 04/21, 2010, from
www.fineli.fi. Freaney, R., McBrinn, Y., & McKenna, M. J. (1993). Secondary hyperparathyroidism
in elderly people: Combined effect of renal insufficiency and vitamin D deficiency. The American Journal of Clinical Nutrition, 58(2), 187-191.
Fronczak, C. M., Baron, A. E., Chase, H. P., Ross, C., Brady, H. L., Hoffman, M., et al. (2003). In utero dietary exposures and risk of islet autoimmunity in children. Diabetes Care, 26(12), 3237-3242.
Giulietti, A., Gysemans, C., Stoffels, K., van Etten, E., Decallonne, B., Overbergh, L., et al. (2004). Vitamin D deficiency in early life accelerates type 1 diabetes in non-obese diabetic mice. Diabetologia, 47(3), 451-462.
64
Greer, R. M., Rogers, M. A., Bowling, F. G., Buntain, H. M., Harris, M., Leong, G. M., et al. (2007). Australian children and adolescents with type 1 diabetes have low vitamin D levels. The Medical Journal of Australia, 187(1), 59-60.
Haddad, J. G., Matsuoka, L. Y., Hollis, B. W., Hu, Y. Z., & Wortsman, J. (1993). Human plasma transport of vitamin D after its endogenous synthesis. The Journal of Clinical Investigation, 91(6), 2552-2555.
Harjutsalo, V. (2007). Familial Aggregation of Type 1Diabetes and diabetic Nephropathy in Finland. Publications of the National Institute for Health and Welfare A 1/2007.
Harjutsalo, V., Sjoberg, L., & Tuomilehto, J. (2008). Time trends in the incidence of type 1 diabetes in finnish children: A cohort study. Lancet, 371(9626), 1777-1782.
Haussler, M. R., & Norman, A. W. (1969). CHROMOSOMAL RECEPTOR FOR A VITAMIN D METABOLITE. Proceedings of the National Academy of Sciences of the United States of America, 62(1), 155-162.
Haussler, M. R., Myrtle, J. F., & Norman, A. W. (1968). The association of a metabolite of vitamin D3 with intestinal mucosa chromatin in vivo. Journal of Biological Chemistry, 243(15), 4055-4064.
HELSINGIN SANOMAT. Retrieved 04/23, 2010, from www.hs.fi. Hirvonen T, Tapanainen H, Valsta L, Virtanen M, Aro A, Pietinen P.
Elintarvikkeidentäydentäminen D-vitamiinilla ja kalsiumilla. Kansanterveyslaitoksenjulkaisuja B19/2004. Retrieved 02/04 2010, from http://www.ktl.fi.
Hitman, G. A., Mannan, N., McDermott, M. F., Aganna, E., Ogunkolade, B. W., Hales, C. N., et al. (1998). Vitamin D receptor gene polymorphisms influence insulin secretion in bangladeshi asians. Diabetes, 47(4), 688-690.
Holick, M. F. (2002). Sunlight and vitamin D: Both good for cardiovascular health. Journal of General Internal Medicine, 17(9), 733-735.
Holick, M. F. (2003). Vitamin D: A millenium perspective. Journal of Cellular Biochemistry, 88(2), 296-307.
Holick, M. F. (2005). The influence of vitamin D on bone health across the life cycle. The Journal of Nutrition, 135(11), 2726S-7S.
Holick, M. F. (2007). Vitamin D deficiency. The New England Journal of Medicine, 357(3), 266-281.
Holick, M. F. (2008). The vitamin D deficiency pandemic and consequences for nonskeletal health: Mechanisms of action. Molecular Aspects of Medicine, 29(6), 361-368.
Holick, M. F., & Chen, T. C. (2008). Vitamin D deficiency: A worldwide problem with health consequences. The American Journal of Clinical Nutrition, 87(4), 1080S-6S.
Holick, M. F., Frommer, J. E., McNeill, S. C., Richtand, N. M., Henley, J. W., & Potts, J. T.,Jr. (1977). Photometabolism of 7-dehydrocholesterol to previtamin D3 in skin. Biochemical and Biophysical Research Communications, 76(1), 107-114.
Hollis, B. W. (2004). Editorial: The determination of circulating 25-hydroxyvitamin D: No easy task. The Journal of Clinical Endocrinology and Metabolism, 89(7), 3149-3151.
Hollis, B. W. (2008). Measuring 25-hydroxyvitamin D in a clinical environment: Challenges and needs. The American Journal of Clinical Nutrition, 88(2), 507S-510S.
Holmes, V. A., Barnes, M. S., Alexander, H. D., McFaul, P., & Wallace, J. M. (2009). Vitamin D deficiency and insufficiency in pregnant women: A longitudinal study. The British Journal of Nutrition, 102(6), 876-881.
Houghton, L. A., & Vieth, R. (2006). The case against ergocalciferol (vitamin D2) as a vitamin supplement. The American Journal of Clinical Nutrition, 84(4), 694-697.
65
Houghton, L. A., & Vieth, R. (2006). The case against ergocalciferol (vitamin D2) as a vitamin supplement. The American Journal of Clinical Nutrition, 84(4), 694-697.
Hypponen, E., & Power, C. (2006). Vitamin D status and glucose homeostasis in the 1958 british birth cohort: The role of obesity. Diabetes Care, 29(10), 2244-2246.
Hypponen, E., Laara, E., Reunanen, A., Jarvelin, M. R., & Virtanen, S. M. (2001). Intake of vitamin D and risk of type 1 diabetes: A birth-cohort study. Lancet, 358(9292), 1500-1503.
Jeffery, L. E., Burke, F., Mura, M., Zheng, Y., Qureshi, O. S., Hewison, M., et al. (2009). 1,25-dihydroxyvitamin D3 and IL-2 combine to inhibit T cell production of inflammatory cytokines and promote development of regulatory T cells expressing CTLA-4 and FoxP3. Journal of Immunology (Baltimore, Md.: 1950), 183(9), 5458-5467.
Jones, G., Strugnell, S., A., & DeLuca, H., F. (1998). Current understanding of the molecular actions of vitamin D.78(4), 1193--1231.
Kauppi, M., Impivaara, O., Maki, J., Heliovaara, M., Marniemi, J., Montonen, J., et al. (2009). Vitamin D status and common risk factors for bone fragility as determinants of quantitative ultrasound variables in a nationally representative population sample. Bone, 45(1), 119-124.
Kauppinen-Mäkelin, R., Tähtelä, R., Löyttyniemi, E., Kärkkäinen, J., & Välimäki, M., J. (2001). A high prevalence of hypovitaminosis D in finnish medical in- and outpatients.(249), 559-563.
Kimball, S. M., & Vieth, R. (2007). A comparison of automated methods for the quantitation of serum 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D. Clinical Biochemistry, 40(16-17), 1305-1310.
Kimmel-Jehan, C., Darwish, H. M., Strugnell, S. A., Jehan, F., Wiefling, B., & DeLuca, H. F. (1999). DNA bending is induced by binding of vitamin D receptor-retinoid X receptor heterodimers to vitamin D response elements. Journal of Cellular Biochemistry, 74(2), 220-228.
Kull, M.,Jr, Kallikorm, R., Tamm, A., & Lember, M. (2009). Seasonal variance of 25-(OH) vitamin D in the general population of estonia, a northern european country. BMC Public Health, 9, 22.
Lamberg-Allardt, C. (1984). Vitamin D intake, sunlight exposure and 25-hydroxyvitamin D levels in the elderly during one year. Annals of Nutrition & Metabolism, 28(3), 144-150.
Lamberg-Allardt, C. J., & Viljakainen, H. (2006). D-vitamiinitilanteen seurantatutkimus 2002-2004Sosiaali- ja terveysministeriö.
Lamberg-Allardt, C. J., Outila, T. A., Karkkainen, M. U., Rita, H. J., & Valsta, L. M. (2001). Vitamin D deficiency and bone health in healthy adults in finland: Could this be a concern in other parts of europe? Journal of Bone and Mineral Research : The Official Journal of the American Society for Bone and Mineral Research, 16(11), 2066-2073.
Lamberg-Allardt, C., Larjosto, M., Schultz, E. (1984). 25-hydroxyvitamin D concentrations in maternal and cord blood at delivery and in maternal blood during lactation in Finland. Human Nutrition: Clinical Nutrition, 38C, 261-268.
Littorin, B., Blom, P., Scholin, A., Arnqvist, H. J., Blohme, G., Bolinder, J., et al. (2006). Lower levels of plasma 25-hydroxyvitamin D among young adults at diagnosis of autoimmune type 1 diabetes compared with control subjects: Results from the nationwide diabetes incidence study in sweden (DISS). Diabetologia, 49(12), 2847-2852.
Lou, Y., Molnár, F., Peräkylä, M., Qiao, S., Kalueff, A., V., St-Arnaud, R., et al. (2009). 25-hydroxyvitamin D3 is an agonistic vitamin D receptor ligand.
66
Maghbooli, Z., Hossein-Nezhad, A., Shafaei, A. R., Karimi, F., Madani, F. S., & Larijani, B. (2007). Vitamin D status in mothers and their newborns in iran. BMC Pregnancy and Childbirth, 7, 1.
Marniemi, J., Alanen, E., Impivaara, O., Seppanen, R., Hakala, P., Rajala, T., et al. (2005). Dietary and serum vitamins and minerals as predictors of myocardial infarction and stroke in elderly subjects. Nutrition, Metabolism, and Cardiovascular Diseases : NMCD, 15(3), 188-197.
Mathieu, C., Waer, M., Laureys, J., Rutgeerts, O., & Bouillon, R. (1994). Prevention of autoimmune diabetes in NOD mice by 1,25 dihydroxyvitamin D3.37, 552-558.
Meehan, M. A., Kerman, R. H., & Lemire, J. M. (1992). 1,25-dihydroxyvitamin D3 enhances the generation of nonspecific suppressor cells while inhibiting the induction of cytotoxic cells in a human MLR. Cellular Immunology, 140(2), 400-409.
Narchi, H., Kochiyil, J., Zayed, R., Abdulrazzak, W., & Agarwal, M. (2010). Maternal vitamin D status throughout and after pregnancy. Journal of Obstetrics and Gynaecology : The Journal of the Institute of Obstetrics and Gynaecology, 30(2), 137-142.
Nistico, L., Buzzetti, R., Pritchard, L. E., Van der Auwera, B., Giovannini, C., Bosi, E., et al. (1996). The CTLA-4 gene region of chromosome 2q33 is linked to, and associated with, type 1 diabetes. belgian diabetes registry. Human Molecular Genetics, 5(7), 1075-1080.
Norman, A.W., D. Adams, E.D. Collins, W.H. Okamura and R.J. Fletterick (1999). Three-dimensional model of the ligand binding domain of the nuclear receptor for 1a,25-dihydroxy-vitamin D3. Journal of Cellular Biochemistry 74:323-333
Nurmi, I., Kaukonen, J. P., Luthje, P., Naboulsi, H., Tanninen, S., Kataja, M., et al. (2005). Half of the patients with an acute hip fracture suffer from hypovitaminosis D: A prospective study in southeastern finland. Osteoporosis International : A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA, 16(12), 2018-2024.
Nykjaer, A., Dragun, D., Walther, D., Vorum, H., Jacobsen, C., Herz, J., et al. (1999). An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell, 96(4), 507-515.
Ono, Y., Suzuki, A., Kotake, M., Zhang, X., Nishiwaki-Yasuda, K., Ishiwata, Y., et al. (2005). Seasonal changes of serum 25-hydroxyvitamin D and intact parathyroid hormone levels in a normal japanese population. Journal of Bone and Mineral Metabolism, 23(2), 147-151.
Ounissi-Benkalha, H., & Polychronakos, C. (2008). The molecular genetics of type 1 diabetes: New genes and emerging mechanisms. Trends in Molecular Medicine, 14(6), 268-275.
Penckofer, S., Kouba, J., Wallis, D. E., & Emanuele, M. A. (2008). Vitamin D and diabetes: Let the sunshine in. The Diabetes Educator, 34(6), 939-40, 942, 944 passim.
Pociot, F., & McDermott, M. F. (2002). Genetics of type 1 diabetes mellitus. Genes and Immunity, 3(5), 235-249.
Pozzilli, P., Manfrini, S., Crino, A., Picardi, A., Leomanni, C., Cherubini, V., et al. (2005). Low levels of 25-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 in patients with newly diagnosed type 1 diabetes. Hormone and Metabolic Research = Hormon- Und Stoffwechselforschung = Hormones Et Metabolisme, 37(11), 680-683.
Prosser, D., & Jones, G. (2004). Enzymes involved in the activation and inactivation of vitamin D. Trends in Biochemical Sciences, 29 (12), 664-673.
67
Reichel, H., Koeffler, H. P., Tobler, A., & Norman, A. W. (1987). 1 alpha,25-dihydroxyvitamin D3 inhibits gamma-interferon synthesis by normal human peripheral blood lymphocytes. Proceedings of the National Academy of Sciences of the United States of America, 84(10), 3385-3389.
Rigby, W. F., Denome, S., & Fanger, M. W. (1987). Regulation of lymphokine production and human T lymphocyte activation by 1,25-dihydroxyvitamin D3. specific inhibition at the level of messenger RNA. The Journal of Clinical Investigation, 79(6), 1659-1664.
Rockell, J. E., Skeaff, C. M., Williams, S. M., & Green, T. J. (2006). Serum 25- hydroxyvitamin D concentrations of new zealanders aged 15 years and older. Osteoporosis International : A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA, 17(9), 1382-1389.
Scharla, S. H., Scheidt-Nave, C., Leidig, G., Woitge, H., Wuster, C., Seibel, M. J., et al. (1996). Lower serum 25-hydroxyvitamin D is associated with increased bone resorption markers and lower bone density at the proximal femur in normal females: A population-based study. Experimental and Clinical Endocrinology & Diabetes : Official Journal, German Society of Endocrinology [and] German Diabetes Association, 104(3), 289-292.
Schoentgen, F., Metz-Boutigue, M. H., Jolles, J., Constans, J., & Jolles, P. (1986). Complete amino acid sequence of human vitamin D-binding protein (group-specific component): Evidence of a three-fold internal homology as in serum albumin and alpha-fetoprotein. Biochimica Et Biophysica Acta, 871(2), 189-198.
Stene, L. C., Joner, G., & Norwegian Childhood Diabetes Study Group. (2003). Use of cod liver oil during the first year of life is associated with lower risk of childhood-onset type 1 diabetes: A large, population-based, case-control study. The American Journal of Clinical Nutrition, 78(6), 1128-1134.
Stene, L. C., Ulriksen, J., Magnus, P., & Joner, G. (2000). Use of cod liver oil during pregnancy associated with lower risk of type I diabetes in the offspring. Diabetologia, 43(9), 1093-1098.
Svoren, B. M., Volkening, L. K., Wood, J. R., & Laffel, L. M. (2009). Significant vitamin D deficiency in youth with type 1 diabetes mellitus. The Journal of Pediatrics, 154(1), 132-134.
THL. National Institute for Health and Welfare. Retrieved 02/04, 2010, www.thl.fi Thomas, M. K., Lloyd-Jones, D. M., Thadhani, R. I., Shaw, A. C., Deraska, D. J., Kitch,
B. T., et al. (1998). Hypovitaminosis D in medical inpatients. The New England Journal of Medicine, 338(12), 777-783.
Thomas, W. C.,Jr, Morgan, H. G., Connor, T. B., Haddock, L., Bills, C. E., & Howard, J. E. (1959). Studies of antiricketic activity in sera from patients with disorders of calcium metabolism and preliminary observations on the mode of transport of vitamin D in human serum. The Journal of Clinical Investigation, 38(7), 1078-1085.
Toriola, A. T., Surcel, H. M., Agborsangaya, C., Grankvist, K., Tuohimaa, P., Toniolo, P., et al. (2010). Serum 25-hydroxyvitamin D and the risk of ovarian cancer. European Journal of Cancer (Oxford, England : 1990), 46(2), 364-369.
Tsoukas, C. D., Provvedini, D. M., & Manolagas, S. C. (1984). 1,25-dihydroxyvitamin D3: A novel immunoregulatory hormone. Science (New York, N.Y.), 224(4656), 1438-1440.
UNIPROT. Protein knowledgebase. Retrieved 02/04 2010, from www.uniprot.org. Valimaki, V. V., Alfthan, H., Lehmuskallio, E., Loyttyniemi, E., Sahi, T., Stenman, U.
H., et al. (2004). Vitamin D status as a determinant of peak bone mass in young finnish men. The Journal of Clinical Endocrinology and Metabolism, 89(1), 76-80.
68
Valimaki, V. V., Alfthan, H., Lehmuskallio, E., Loyttyniemi, E., Sahi, T., Suominen, H., et al. (2005). Risk factors for clinical stress fractures in male military recruits: A prospective cohort study. Bone, 37(2), 267-273.
van Driel, M., Koedam, M., Buurman, C., Hewison, M., Chiba, H., Uitterlinden, A., Pols, H., van Leeuwen, J. (2006). Evidence for auto/paracrine actions of vitamin D in bone: 1 -hydroxylase expression and activity in human bone cells. The FASEB Journal, 20, 1811-1819.
van Etten, E., & Mathieu, C. (2005). Immunoregulation by 1,25-dihydroxyvitamin D3: Basic concepts. The Journal of Steroid Biochemistry and Molecular Biology, 97(1-2), 93-101.
van Groningen, L., Opdenoordt, S., van Sorge, A., Telting, D., Giesen, A., & de Boer, H. (2010). Cholecalciferol loading dose guideline for vitamin D deficient adults. European Journal of Endocrinology / European Federation of Endocrine Societies
Veldman, C. M., Cantorna, M. T., & DeLuca, H. F. (2000). Expression of 1,25-dihydroxyvitamin D(3) receptor in the immune system. Archives of Biochemistry and Biophysics, 374(2), 334-338.
Verboven, C., Rabijns, A., De Maeyer, M., Van Baelen, H., Bouillon, R., & De Ranter, C. (2002). A structural basis for the unique binding features of the human vitamin D-binding protein. Nature Structural Biology, 9(2), 131-136.
Verroust, P., Birn, H., Nielsen, R., Kozyraki, R., Christensen, E. (2002). The tandem endocytic receptors megalin and cubilin are important proteins in renal pathology. Kidney International, 62, 745-756.
Wagner, D., Hanwell, H. E., & Vieth, R. (2009). An evaluation of automated methods for measurement of serum 25-hydroxyvitamin D. Clinical Biochemistry, 42(15), 1549-1556.
Wallace, A. M., Gibson, S., de la Hunty, A., Lamberg-Allardt, C., & Ashwell, M. (2010). Measurement of 25-hydroxyvitamin D in the clinical laboratory: Current procedures, performance characteristics and limitations. Steroids,
Wenzlau, J. M., Juhl, K., Yu, L., Moua, O., Sarkar, S. A., Gottlieb, P., et al. (2007). The cation efflux transporter ZnT8 (Slc30A8) is a major autoantigen in human type 1 diabetes. Proceedings of the National Academy of Sciences of the United States of America, 104(43), 17040-17045.
Wielders, J. P., & Wijnberg, F. A. (2009). Preanalytical stability of 25(OH)-vitamin D3 in human blood or serum at room temperature: Solid as a rock. Clinical Chemistry, 55(8), 1584-1585.
Wolpowitz, D., & Gilchrest, B. A. (2006). The vitamin D questions: How much do you need and how should you get it? Journal of the American Academy of Dermatology, 54(2), 301-317.
YA. Yliopiston apteekki. Retrieved 05/12 2010, from www.yliopistoapteekki.fi. Zella, J. B., McCary, L. C., & DeLuca, H. F. (2003). Oral administration of 1,25-
dihydroxyvitamin D3 completely protects NOD mice from insulin-dependent diabetes mellitus. Archives of Biochemistry and Biophysics, 417(1), 77-80.
Zerwekh, J. E. (2008). Blood biomarkers of vitamin D status. The American Journal of Clinical Nutrition, 87(4), 1087S-91S.