phd thesis glenting j

Lactic Acid Bacteria in Vaccine Development

PhD thesis by Jacob Glenting

Bioneer A/S

And

Biocentrum

Technical University of Denmark

June 2007

Preface

This thesis describes the results of a PhD study initiated in December 2003 and

finished in May 2007. The study was done mainly at Bioneer A/S, but with close

interactions with Statens Serum Institute (SSI), Danish Toxicology Center (now

DHI), ALK-Abello (ALK), and the Allergy Clinic, National University Hospital,

DK. The PhD study was interrupted for 4 months because of a time consuming

project work at Bioneer.

Bioneer and the Danish Ministry of Science, Technology and Innovation have

financially supported the work. I wish to thank the management of Bioneer for

being supportive.

Hanne Frøkiær has been my supervisor at DTU Biocentrum. I wish to thank Hanne

for helpful discussions and teaching me how to work with dendritic cells. Hans

Israelsen was my supervisor at Bioneer in the beginning of the study to whom I owe

many thanks for teaching me how to do high quality science. However, as Hans left

Bioneer, Søren Madsen has been taking over and contributed with ideas, assistance,

and great discussions. Anders Fomsgaard (SSI) has been supervising the work on

the gene vaccines and allowed me a peek into the world of virology and

immunology. Thank you Anders for our collaboration.

The work of this thesis overlaps with very different biological disciplines. Therefore

I have relied on the expertise and help from several excellent researchers. Thanks to

Mercedes Ferreras and Jens Brimnes (ALK), Ann Detmer and Stephen Wesssels

(DHI), Gregers Gram (SSI), Lars K. Poulsen (Allergy Clinic), Bjørn Holst, Peter

Ravn, Helle Wium and Simon Jensen (Bioneer). Outstanding technical assistance

has been given by Ulla Poulsen, Pernille Smith, Annemette Brix, and Anne Cathrine

Steenbjerg (Bioneer).

Finally I wish to thanks Vera for being patient during the writing phase, cooking

dinner, and her warm love!

Jacob Glenting

Hørsholm, June 2007

Table of contents

Abstract(s) UK and DK version ……………………………………………………………………………………….…………….1 Outline of thesis…………………………………………………………………………………………………. 7 Chapter 1 General introduction…………………………………………………………………………………………………….8 Chapter 2 Immunological analysis of a Lactococcus lactis based DNA vaccine expressing HIV gp120…………………39 Chapter 3 Cell surface display of Bet v 1 on immunomodulatory lactobacilli: potential oral delivery vehicle for treatment of birch pollen allergy…………………………………………………………………………………..51 Chapter 4 DNA inversion Controls Expression of a Mannose Specific Adhesin from Lactobacillus plantarum …………70 Chapter 5 Recombinant Production of Immunological Active Peanut Allergen Ara h 2 using Lactococcus lactis……….94 Chapter 6 Conclusions and concluding remarks……………………………………………………………………………….114 Appendix A plasmid selection system in Lactococcus lactis and its use for gene expression in L. lactis and human kidney fibroblasts……………………………………………………………………………………….117

Lactic Acid Bacteria in Vaccine Development

Abstract

This PhD study is focused on the use of lactic acid bacteria (LAB) in development of

vaccines and therapeutics. The applications of LAB in strategies to promote health

and prevent diseases are several: (i) the bacterium can be used as a therapeutic itself,

(ii) gene engineered LAB are suited for the delivery of medical components by the

mucosal route, (iii) LAB are attractive microbial cell factories of heterologous

proteins and pharmaceutical plasmid DNA. This thesis analysed these applications of

LAB with the aim to develop novel vaccines and learn more about the interactions of

LAB with the human mucosal surfaces and immune system.

Three types of LAB based vaccines were developed and tested including a plasmid

DNA vaccine, a live recombinant vaccine vehicle and a subunit protein vaccine.

Because vaccines most often are given to healthy people, and therefore a minimum of

risk is accepted, I felt compelled to analyse the safety aspects of vaccines and to give

some suggestions for future development of safer vaccines. The result was two

reviews focusing on either live bacterial vaccines or plasmid DNA vaccines. Both

reviews include a discussion on the safety aspects of the vaccine technologies and

give suggestions of aspects to consider in the early phases of vaccine development.

For reasons of efficiency Escherichia coli is used today as the microbial factory for

production of plasmid DNA vaccines. To avoid hazardous antibiotic resistance genes

and endotoxins from plasmid systems used nowadays, we have developed a system

based on the food-grade Lactococcus lactis and a plasmid without antibiotic

resistance genes. The L. lactis system was compared to a traditional one in E. coli

using identical vaccine constructs encoding the gp120 of HIV-1. Although the

plasmid DNA vaccines encode similar antigens their immune effect differs. This

provides information about the role of the “silent” plasmid backbone of DNA

vaccines and the immune activating effect of DNA from human commensals.

Antigen surface display on bacteria is an attractive strategy to co-present the antigen

and the adjuvant effect of the bacterium. However, surface display of proteins can

lower the access to vaccine epitopes by steric hindrance and change the surface

1

architecture of LAB and thereby affect their immune modulating activity. New

protein anchors were isolated from lactobacilli and compared for their efficiency in

protein display. By using a C-terminal anchor in combination with a long spacer the

display of an active enzyme and a birch pollen allergen with preserved

immunereactivity was obtained. Although surface molecules of LAB are key factors

in the activation of the immune system, changing the cell wall by display of an

allergen, did not alter their adjuvant properties. The developed allergen displaying

LAB may represent a promising oral vaccine delivery vehicle for treatment of birch

pollen allergy.

An important feature of live LAB vaccines is their interaction with the mucosal

surfaces. As mannose covers the mucosal surfaces we analysed the molecular factors

mediating mannose adhesion in lactobacilli. A mannose specific adhesin was isolated

and identified to be responsible for the binding to intestinal epithelial cells.

Interestingly, the expression of the adhesion-gene was regulated by a flip-flop

inversion of a DNA element present in the untranslated leader of the gene encoding

the adhesin. The findings represent a new all-or-nothing transcriptional control in

lactobacilli, which is also observed in other bacteria like Escherichia coli that reside

in the human body.

Lactococcus lactis is also an attractive microorganism for use in the production of

protein therapeutics. L. lactis is considered food grade, free of endotoxins, and is able

to secrete the heterologous product together with few other native proteins.

Hypersensitivity to peanut represents a serious allergic problem. Some of the major

allergens in peanut have been described. However, for therapeutic usage more

information about the individual allergenic components is needed. In this thesis

recombinant production of the Ara h 2 peanut allergen was tested using L. lactis.

L. lactis could offer high yields of secreted, full length and immunologically active

allergen. The L. lactis expression system can support recombinant allergen material

for immunotherapy and component resolved allergen diagnostics. Furthermore, using

the L. lactis expression system makes it relatively simple to engineer and screen

allergen variants of Ara h 2 with reduced binding to IgE.

2

The experiments presented in this thesis suggest new LAB based vaccine candidates:

(i) a live bacterial birch-pollen-allergen vaccine (ii) a plasmid DNA vaccine encoding

a HIV-1 surface molecule (iii) a subunit peanut allergen vaccine. Furthermore, the

developed LAB based vaccines are important tools to study the cross talk between

commensals and the human body.

3

Anvendelse af Mælkesyrebakterier til Vaccinefremstilling

Resume

Dette PhD studie fokuserer på brugen af mælkesyrebakterier (LAB) til udvikling af

vacciner og terapeutika. Anvendelserne af LAB til at promovere sundhed og

forebygge sygdomme er flere: (i) bakterierne kan benyttes som terapeutika i sig selv,

(ii) genmodificerede LAB er velegnet til aflevering af medicinske komponenter via

den mukosale rute, (iii) LAB er attraktive mikrobielle cellefabrikker af heterologe

proteiner og pharmaceutisk plasmid DNA. Denne afhandling analyserede disse

applikationer af LAB med målet at udvikle nye vacciner og lære mere om

interaktionerne mellem LAB og den humane mukosale overflade samt

immunsystemet.

Tre typer af LAB-baserede vacciner blev udviklet og testet. Disse inkluderer en

plasmid DNA vaccine, en levende LAB-rekombinant vaccine og en subunit protein

vaccine. Vacciner er ofte givet til raske personer. Derfor er den accepterede risiko ved

vaccination meget lav. I denne afhandling er der derfor også fokuseret på

sikkerhedsaspekterne af de udviklede vacciner. Resultatet var to reviews, der

fokuserer på levende vacciner og DNA vacciner. Begge reviews inkluderer en

diskussion af sikkerhedsaspekterne ved de to typer af vacciner og giver forslag til

hvilke aspekter, der kan behandles i den tidlige udviklingsfase af vacciner.

På grund af effektiviteten er Escherichia coli benyttet i dag som mikrobiel fabrik af

plasmid DNA vacciner. For at undgå antibiotika resistens gener og endotoxin i disse

anvendte produktionssystemer er her udviklet et system som er baseret på en sikker

organisme, Lactococcus lactis, og som ikke anvender antibiotika. Dette system blev

sammenlignet med et traditionelt E. coli baseret system ved brug af identiske

vaccinekonstrukter, der koder for gp120 proteinet fra HIV-1. På trods af at plasmid

vaccinerne koder for identiske antigener var det inducerede immunrespons forskelligt.

Dette giver informationer om den ikke-kodende del af DNA vacciner og hvad DNA

kompositionen betyder for adjuvanseffekten i DNA vacciner. Dette giver også

information om den immunaktiverende effekt af DNA fra mælkesyrebakterier til stede

i den humane bakterieflora.

4

Overflade display af antigener på bakterier er en attraktiv strategi til co-præsentation

af antigen og adjuvans fra bakterien. Men overflade display af protein kan inhibere

fremvisningen af vaccine epitoper til immunsystemet ved sterisk hindring og

samtidigt ændre overflade arkitekturen af LAB, der derved ændrer den vigtige

immunmodulerende aktivitet. Nye proteinankre blev isoleret fra laktobaciller og deres

effektivitet mht. overflade display blev sammenlignet. Ved brug af et C-terminalt

anker i kombination med en lang ”arm” kunne et aktivt enzym og et birkepollen

allergen med konserveret immunreaktivitet immobiliseres til celle overfladen. På

trods af at overfladekomponenter på LAB er nøglefaktorer i aktivering af

immunsystemet ændrede display af birkepollen-allergenet ikke adjuvanseffekten af

bakterien. Den udviklede allergen vaccine er kandidat til en ny oral behandling af

birkepollen allergi.

En vigtig evne af levende LAB vacciner er deres interaktion med den mukosale

overflade. Mannose er en vigtig bestanddel af den mukosale overflade. Derfor

analyserede vi de molekylære faktorer bag laktobacillers evne til at binde mannose. Et

mannose specifikt adhesin blev identificeret som værende en central faktor i

bindingen til epitel celler. Ekspression af adhesinet blev analyseret og var reguleret af

en flip-flop mekanisme, hvor et DNA element opstrøms adhesinet inverteres. Denne

opdagelse er en ny alt eller intet transskriptions reguleringsmekanisme i laktobaciller,

som også er observeret i andre mave-tarm associerede bakterier som E. coli.

Lactococcus lactis er også attraktiv til produktionen af heterologe proteiner. L. lactis

er anerkendt som sikker og producerer ikke endotoxiner, samt sekreterer det

heterologe produkt til det ekstracellulære miljø sammen med få andre native

proteiner. Hypersensitivitet til peanuts er en alvorlig allergi. Nogle af allergenerne i

peanuts er beskrevet. Men til terapeutisk brug er der brug for mere information om de

enkelte allergener. I denne afhandling er rekombinant ekspression af Ara h 2

allergenet testet ved brug af L. lactis. Her opnåedes produktion af høje mængder af

allergen med konserveret immunreaktivitet. L. lactis systemet kan benyttes til at

producere Ara h 2 til immunterapi og til diagnostik af hypersensitivitet ved brug af

isolerede allergen komponenter. Yderligere er det forholdsvist simpelt at udvikle Ara

h 2 varianter med reduceret IgE binding.

5

Eksperimenterne præsenteret i denne afhandling foreslår nye LAB baserede vaccine

kandidater: (i) en levende mælkesyrebakterie til terapeutisk behandling af birkepollen

allergi (ii) en plasmid DNA vaccine udtrykkende gp120 fra HIV-1 (iii) en peanut

allergen vaccine. Ud over at være lovende vaccineteknologier kan de anvendes som

vigtige redskaber til analyse af interaktionen mellem den residerende flora og humane

krop.

6

Outline of thesis The thesis begins with an introducing chapter, which is divided in two parts: (i) An

overview of lactic acid bacteria (LAB) and their applications in vaccine development,

(ii) two published reviews that give a more thorough description of the subject. The

first review deals with the use of LAB as live microbial vehicles of vaccines and

therapeutics. The other is a mini review and describes plasmid DNA vaccines and

some aspects of their production. In both reviews the use of LAB in vaccine

production is described and compared to alternative organisms. In addition to review

the published literature the manuscripts discuss and give suggestion for the

development of safer vaccines for the future.

The second part of the thesis contains the experimental studies. Here a LAB and non-

antibiotic based plasmid DNA vaccine was developed and compared to a routinely

used Escherichia coli based gene vaccine. The use of Lactococcus lactis as new and

antibiotic-free microbial factory of pharmaceutical plasmid DNA is discussed. The L.

lactis host-vector plasmid selection system, which is the backbone of the suggested

DNA vaccine, was developed before initiating this PhD study. However, as the

plasmid and host-strain constructions indeed are relevant for this thesis I have

attached my publication from 2002 in the appendix.

In chapter 3, genetic elements were analysed to construct a live allergy vaccine with

immunomodulatory activity. Chapter 4 represents a time consuming part of the thesis.

Here the mucosal adhesive phenotype of lactobacilli was investigated. In chapter 5 L.

lactis was used for recombinant production of a peanut allergen. This manuscript is

submitted to Microbial Cell Factories.

The summarising chapter 6 extracts the most important findings of the study and

gives suggestions for future directions.

7

Chapter 1

General Introduction

8

Introduction

Lactic acid bacteria (LAB) are a functionally related group of organisms known

primarily from their role in bioprocessing of food products. LAB are gram-positive,

anaerobic, with low G+C content, and acid tolerant. Acidification is important during

food processing. However, LAB also contributes to the flavour, texture and nutritional

level in the end product. The central role of LAB in industrial fermentation of food

and beverages has driven the research on genetics and metabolisms of these bacteria.

Today state of the art research on LAB in the industry develops and selects tailored

strains with special metabolic characters. Although LAB are considered generally

regarded as safe (GRAS) the manipulation of genes confer new challenges to this

definition. To avoid labelling as genetically modified a mutagenesis strategy is

usually employed. However, random mutagenesis of LAB, by use of chemicals or

radiation, results in a relatively large strain library and isolation of the clone with

proper gene modification can be a challenge. High throughput screening technology

and the availability of genome sequences facilitate the selection and characterisation

of the strain. Specific mutagenesis strategies using integration systems have been

developed and optimized towards food grade status. These systems use the native

LAB gene elements to obtain knock out or over expression mutants and in some cases

alleviate the need for laborious screening activities. Although, gene modifications

using site directed integration and gene manipulations ensure a fully characterised

strain, the EU regulations demands GMO labelling, whereas a random mutagenisised

strain escapes this process.

Although a century has past by since the Noble prize awarded Elya Metchnikoff

(1845-1916) proposed that LAB could promote health, the clinical and molecular data

behind the acclaimed heath effects has been recently established. Today’s availability

of genetic tools, appropriate in vitro and animal models, and clinical data allows for

critical evaluation of this life-promoting effect of LAB. These “new” applications of

LAB have fuelled the research activity and are by some researchers called the “LAB

renaissance”.

9

The potential of LAB to survive through the gastro-intestinal tract, adhere to mucosal

surfaces, and activate the immune responses make them attractive as transporters of

vaccines and therapeutics. Today LAB has been used as delivery vehicles of

cytokines [Steidler et al., 2001], therapeutic enzymes [Kiatpapen et al., 2001],

antimicrobial peptides [Freitas et al., 2005], antigens [Pouwells et al., 1998],

allergens [Daniel et al., 2006], hormones [Yao et al., 2006], antagonists [Ricci et al.,

2003], and antibody fragments [Krüger et al., 2002]. Although obvious risks are

associated with their recombinant status and non-controllable in situ antigen synthesis

the scientific progress is promising. Especially after the positive outcome of the

clinical trial with interleukin 10 secreting Lactococcus lactis [Braat et al., 2006].

Alongside the development of LAB as vaccine carriers several groups have focused

on LAB as microbial cell factories of recombinant proteins or metabolic precursors.

LAB as live mucosal vaccines

Needle free and mucosal administration of vaccines is becoming increasingly relevant

as the importance of mucosal immunity is acknowledged. In addition, non-parenteral

administration avoids the risk of contaminated needles and need for a healthcare

infrastructure. Live vaccines based on bacteria demands a less complicated down

stream processing compared to subunit vaccines based on purified protein

components. Some strains of LAB are attractive as live mucosal vaccines. Their

GRAS status, ability to survive through the GI tract, adhesive properties, and

immunomodulatory effect make them suitable for vaccine vehicles. However, their

recombinant and live status adds certain issues that should be addressed (Table 1). A

functional live vaccine based on LAB includes two basic elements: (i) the bacterial

strain, and (ii) the recombinant expression unit that drives antigen synthesis. An

overview of these is given below.

Physical and immunological properties of LAB as vaccine vehicles

Several physical properties make LAB interesting microbial vehicles of vaccine

components. Especially antigens from pathogenic bacteria can be presented with close

mimicry. Indeed, induction of protective immunity against Helicobacter pylori and

Streptococcus pneumonia was obtained by immunization with L. lactis expressing the

Cag12 membrane protein [Kim et al., 2006] and the PspA pneumococcal surface

protein [Hannify et al., 2007], respectively. Although speculative, the size of LAB

10

allows for uptake through M-cells of the Peyer´s patches of the GI-tract and opens for

their subsequent distribution to the mucosal associated lymphatic tissue. Indeed, L.

plantarum expressing green fluorescent protein and given orally to mice was shown

embedded in the mucus and in close contact with epithelial cells [Geoffroy et al.,

2000]. This study also showed that L. plantarum was phagocytized by

bronchoalveolar macrophages following nasal administration.

LAB responds to the harsh milieu of the GI-tract by induction of genes that encode

components to resist bile salts [Pfeiler et al., 2007], stress and metabolic changes

[Bron et al., 2004]. Although concerns of prolonged persistency of the recombinant

vaccine strain has been raised, the colonising capacity of some LAB may play a

central role in their ability to induce an immune response. Bacterial colonisation

requires adhesion of bacteria to the mucosal surfaces. Indeed, some strains of LAB

express specific cell wall components or adhesins that mediate their adherence to the

extracellular matrix (ECM) of the host. These molecular adhesion factors have been

investigated using tissue samples, cell lines and components of the ECM [Miyoshi et

al., 2006, Adlerberth et al., 1996, Granato et al., 1999, Greene et al., 1994, Henriksson

et al., 1991, Henriksson et al., 1992, Henriksson et al., 1996, Hynonen et al., 2002,

Rojas et al., 2002, Sillanpaa et al., 2000, Toba et al., 1995]. The ligands of these

adhesins have been identified as sugar components [Adlerberth et al., 1996], and

ECM proteins like fibronectin [Hynonen et al., 2002], mucin [Granato et al., 2004],

and collagen [Sillanpaa et al., 2000]. The chemical identity of adhesion factors

include both protein and non-protein components of the bacterial cell surface. Most

often cell-surface-adhesins are proteins with signal sequences for their secretion and

mechanisms for covalent anchoring to bacterial cell wall. One important mechanism

anchors the carboxyl terminal via an LPXTG motif and a surface located sortase that

catalyzes the covalent linkage of the adhesion [reviewed by Navarre & Schneewind,

1999]. However, adhesins without signal peptides and anchoring domains like the

LPXTG motif have also been identified [Chhatwal et al., 2002]. The elongation factor

EF-TU, normally involved in protein synthesis and without apparent signal peptide or

cell wall anchoring motif, was identified as a cell surface protein mediating adhesion

to intestinal cells [Granato et al., 2004]. Surprisingly was also the GroEL heat shock

protein found on the surface of the same bacterial strain and identified as an adhesion

factor [Bergonzelli et al., 2006]. Non-proteianous cell surface molecules like

11

lipoteichoic acid of L. johnsonii have also been shown to participate in the adhesion to

intestinal cells [Granato et al., 1999]. The diversity of adhesins and their complex in

vivo regulation illustrates the challenge associated with analysing interactions of LAB

with the host.

A central component of vaccines is adjuvant, which augments the induced response of

both the innate and adaptive immune system. For most vaccines an exogenous added

adjuvant is necessary. However, some strains of LAB have intrinsic adjuvant

properties. Because the adjuvant effect of LAB differs from strain to strain and that

both pro and anti-inflammatory strains have been isolated the term

immunomodulatory is more appropriate. This effect has been evaluated in animal

studies [Matsuzaki et al., 1998]. But more recently in vitro co-incubation with LAB

and dendritic cells is used to analyse their immuneregulatory effect [Christensen et al.,

2002, Mohamadzadeh et al., 2005, Zeuthen et al., 2006]. DNA, lipoteichoic acid, and

bacterial surface proteins has been suggested as the molecular factors responsible for

the immune activating effect of LAB [Pisetsky et al., 1999, Matsuguchi et al., 2003,

Gram et al., 2007]. Although the DC model may provide new information on the

communication between bacteria and the immune system the in vivo correlation may

be questionable. Indeed, the metabolism and surface architecture of LAB changes

considerable when bacteria are transferred from the laboratory to the environments in

the GI-tract [Bron et al., 2004].

Genetic engineering of LAB for vaccine delivery

The expression unit encoding the passenger protein can be episomal as plasmid DNA

or integrated into the chromosome. Usually plasmid based gene expression support

higher product yield due to the higher gene doses. However, plasmid systems adds to

the associated with horizontal gene transfer to the indigenous flora. The risk of

plasmid transfer can be lowered using narrow host range replicons or even replicons

that are active only in a specific mutant strain. Gene units integrated on the

chromosome are less promiscuous and were tested in L. lactis encoding IL10 [Steidler

et al., 2003]. Here, the IL10 expression cassette was inserted into the thyA gene

creating an auxotroph strain with a growth-requirement for external added thymine or

thymidine.

12

The heterologous passenger protein can be targeted to three compartments:

intracellular accumulation, cell wall associated, or secreted in a free form to the extra

cellular milieu. Intracellular accumulation can protect the antigen during passage

through the GI-tract. Protective immune reactions have been induced using

intracellular accumulated tetanus toxin fragment C (TTFC) in L. lactis [Wells et al.,

1993]. TTFC is a highly potent antigen and less immunogenic proteins may require a

more efficient display. Antigen leaking mutants of LAB have therefore been

developed. The alanine racemase mutants of L. lactis and L. plantarum contain a

fragile cell wall when grown in absence of D-alanine and were more immunological

potent using TTFC and the nasal route, than their wild type counterparts [Grangette et

al., 2004]. Secretion of free form proteins has also induced immune responses using

mucosal vaccination of LAB secreting and a variety of different proteins [Enouf et al.,

2001, Chatel et al., 2001, Yao et al., 2006].

Bacterial surface display is often preferred to co-present adjuvant and antigen in close

proximity to each other. Several display systems exist for association of a

recombinant passenger protein to the surface of gram positive bacteria [Navarre &

Schneewind, 1999]. Some involve interactions with the cytoplasmic membrane,

residues of the lipotheicoic acid, whereas others are covalently linked to the cell wall.

The sortase-mediated linkage is dictated by a sorting signal made of LPXTG followed

by 20 hydrophobic aa residues and a tail of positively charged aa. Protein anchor

signals using LPXTG from the Streptococcus pyogenes M6 protein was effective as

surface display system in different lactobacilli but less so in L. lactis [Dieye et al.,

2001]. However, other groups showed that the protein anchor of M6 could efficiently

immobilize the L7/L12 Brucella abortus antigen to the surface of L. lactis [Ribeiro et

al., 2002]. The lactococcal surface protease PrtP is also anchored by the LPXTG

mechanism and was used as surface display system of chimeric malaria antigen Msa2

in L. lactis [Ramasamy et al., 2006]. Anchoring mechanisms that not relies on

LPXTG has been identified in the autolysin AcmA of L. lactis, which is a non-

covalently surface attached enzyme [Raha et al., 2005]. The C-terminal anchor

domain of AcmA successfully targeted and immobilized the E. coli fimbrial F18

adhesin to the surface of L. lactis [Lindholm et al., 2004]. Although surface display

ensures maximum exposure to the immune system it may lead to degradation by

proteases present in the GI-tract. The mechanism of maintaining a non-degraded and

13

functional cell surface protein is unknown. But may simply be an increase in the rate

of turn over of the surface proteins. Recently, non-recombinant but antigen displaying

LAB has been developed. Here the protein anchor domain of AcmA is used to attach

chimeric antigens to the peptidoglycan layer of wild type LAB [van Roosmalen et

al.,2006]. Although the technology avoids the GMO issues the complexity of the

technology may be problematic for large-scale vaccine manufacturing.

LAB as microbial cell factories

Besides applications in food processing, probiotics, and live vaccines, LAB are

interesting microbial factories of industrial relevant metabolites and heterologous

proteins. The scientific progress within metabolic engineering opens for

bioproduction of specific chemical enantiomers like L-alanine [Hols et al., 1999] and

L-lactate [Okano et al., 2007], which can be difficult to produce by chemical

synthesis. Their GRAS status and lack of enodotoxins make LAB attractive producers

of medical important components. In addition contain gram positive bacteria a cell

wall mono layer and therefore absence of periplasmic space. This enables full

secretion of the heterologous product to the culture medium simplifying the down

stream purification steps.

The genetic elements of heterologous expression systems

For increased gene dosage plasmid based expression system are preferred. Several

expression plasmids have been developed for various LAB and supports either a

constitutive or regulated expression of proteins. The genetic elements of expression

plasmids are similar and include an expression unit that drives the synthesis of the

Table 1 Advantages and drawbacks of LAB as live vaccines

Pros/Cons Description Pros

Non-pathogenic status No risk of reversion to pathogenic status Mimicry of infection Bacterial antigens can be displayed in close resemble to native state Mucosal immunity Induction of mucosal immune response Mucosal administration Needle free vaccine administration Manufacturing process Established fermentation technology, simple down stream processing

Cons GMO status Release of GMO in nature Dosage control In situ antigen synthesis may be difficult to control Undesired immune reactions Antigens may induce an immune reaction to the bacterium itself Prolonged persistency High stability of LAB in vivo is undesired Induction of tolerance Immunomodulatory effect of LAB in vivo is unclear

14

heterologous protein, a plasmid replication unit, and a selectable marker for plasmid

maintenance during bacterial growth.

A signal sequence is placed in translational fusion with the heterologous gene to allow

secretion of the protein. Proteins that are targeted for secretion by the Sec-dependant

pathway include a signal peptide of 25-35 aa in size, which is cleaved off by the

signal peptidase during secretion. Several signal sequences have been identified in

LAB using enzyme reporters of secretion like nuclease [Poquet et al., 1998] or ß-

lactamase [Sibakov et al., 1991]. Furthermore, secretion efficiency can be enhanced

using synthetic derivatives of signal sequences [Ravn et al., 2003], addition of a

synthetic propeptide sequence (LEISSTCDA) to the N-terminal of the mature protein

[Hazebrouck et al., 2007], and by co-expressing chaperones [Lindholm et al., 2006].

Highly active promoters are used for efficient transcription of the gene of interest.

Promoters active in LAB usually contain a core region with a -10 region (TATAAT)

and a -35 region (TTGACA) often spaced by 17 bp [Hawley & McCLure, 1982].

Regulatable promoters are preferred for high yield protein production. These are

controlled by adding external components, by environmental conditions or the growth

phase. The gene regulatory elements of the nisin gene cluster of L. lactis has been

used for heterologous and regulatable expression in L. lactis [de Ruyter et al., 1996]

and lactobacilli [Pavan et al., 2000]. Here expression is activated by addition of the

peptide nisin to the growth medium. Genetic elements responsible for regulation and

expression of the bacteriocin sakacin have been isolated from L. sakei [Axelsson et

al., 1993] and used for development of an inducible expression system [Axelsson et

al., 2003]. Here the SapA and SapI promoters are induced by addition of a peptide

pheromone. By placing the genes encoding the response regulator and histidine kinase

to the expression vector LAB without the sakacin operon can be used as hosts. A few

studies describe the use of genetic components from the lac operon in heterologous

production [deVos & Gasson, 1989]. Here the lacA promoter is repressed during

growth on glucose but induced by a shift to medium with lactose as carbon source.

Elements from the lac operon have been combined with elements from the E. coli

bacteriophage T7 [wells et al., 1993]. Here, Wells et al. placed the T7 RNA

polymerase under control of the lactose promoter. Growth on lactose induced T7

RNA polymerase expression, which in turn transcribes the heterologous gene via the

15

T7 promoter. Environmentally regulated promoters avoid the use of exogenous added

inducers. In L. lactis the P170 promoter is induced in the transition to the stationary

growth phase and also affected by the lactate concentration [Madsen et al., 1999].

Although most LAB are generally regarded as safe, their status can be compromised

by the introduction of foreign DNA necessary for synthesis of recombinant proteins.

Usually high copy number plasmids are used for high level expression of recombinant

proteins. A simple way to prevent plasmid loss is to use plasmid-encoded antibiotic

resistance markers and grow the bacteria in the presence of antibiotics. The chief

drawbacks of this approach are the potential loss of selective pressure as a result of

antibiotic degradation (as in the case of β-lactamase) and contamination of the

biomass or purified protein by antibiotics and resistance genes, which is unacceptable

from a medical point of view.

Alternative genetic markers have been developed especially for L. lactis. Depending

on the type of selection, they can be placed in two groups: resistance and

complementation markers. Examples of resistance markers that confer immunity to an

added agent such as nisin [Froseth et al., 1991] or the metal ions cadmium (Cd++) [Liu

et al., 1996] and copper (Cu++) [Liu et al., 2002] have been designed for plasmid

maintenance. Although some strains of LAB are naturally resistant to nisin and metal

ions, the dominant nature of resistance markers make them versatile as they can be

used in different lactococcal strains.

The use of auxotrophic markers is based on complementation of a mutation or

deletion in the host chromosome and is therefore strain-specific. In L. lactis, the first

example was based on complementation of a lacF– strain deficient in lactose

utilization [MacCormick et al., 1995]. In two other systems, auxotrophic markers

complement purine and pyrimidine-auxotrophic strains using genes encoding

nonsense tRNA suppressors [Dickely et al., 1995, Sørensen et al., 2000]. In these

systems, expression of the plasmid-borne suppressor tRNA gene allows read-through

of nonsense mutation(s) in the genes encoding purine or pyrimidine biosynthetic

enzymes. Both systems permit selection in milk or other media that contain small or

no amounts of purines or pyrimidines. Furthermore, amino acid-auxotrophic strains

16

with a requirement for either threonine or D-alanine has been constructed and

complemented with the relevant genes on plasmid [Glenting et al., 2002] or on the

chromosome [Bron et al., 2002].

Choice of a suitable cell factory

Several key parameters must be addressed for the choice of particular protein

production system. Posttranslational modifications like glycosylation or disulphide

bridges may be essential for activity of the recombinant product. Here eukaryotes,

rather than prokaryotes, should be used a cell factory. However, the fermentation

costs and the production time are lowered using a bacterial production system. A

major challenge facing biomanufacturing of proteins is down stream processing.

Secretion of the recombinant product simplifies purification and can be achieved by

eukaryotic and gram-positive bacterial systems. Furthermore, the relative expression

level compared to the contaminants is important. Here, the complete lack of

endotoxins in gram-positive organisms is an advantage as LPS often is co-purified

with the target protein purified by ion exchange principles. LPS is a major challenge

in production of pharmaceutical plasmid DNA as it is co-purified with the negatively

charged DNA [Petsch & Anspach 2000]. The use of Gram-positive bacteria as

plasmid DNA factories can avoid LPS-contamination, but may be problematic in

terms of DNA yield [Gram et al., 2007].

Summary

The new applications of LAB as gene engineered vehicles of mucosal vaccines and

cell factories of pharmaceutical protein and plasmid DNA are promising. With the

increasing knowledge on the interplay of LAB with the human body, specific strains

with desired immune activity and adhesive properties can be selected. The

biotechnological advantages of using LAB in vaccine development rely partly on the

GRAS status and the good name of these bacteria. The challenge for the future

vaccine development lies in harnessing the unique features of LAB, while maintaining

their GRAS status.

17

References

Adlerberth I, Ahrne S, Johansson ML, Molin G, Hanson LA, Wold AE: A mannose-specific adherence mechanism in Lactobacillus plantarum conferring binding to the human colonic cell line HT-29. Appl Environ Microbiol 1996, 62: 2244-2251.

Axelsson L, Holck A, Birkeland SE, Aukrust T, Blom H: Cloning and nucleotide sequence of a gene from Lactobacillus sake Lb706 necessary for sakacin A production and immunity. Appl Environ Microbiol 1993, 59: 2868-2875.

Axelsson L, Lindstad G, Naterstad K: Development of an inducible gene expression system for Lactobacillus sakei. Lett Appl Microbiol 2003, 37: 115-120.

Bergonzelli GE, Granato D, Pridmore RD, Marvin-Guy LF, Donnicola D, Corthesy-Theulaz IE: GroEL of Lactobacillus johnsonii La1 (NCC 533) is cell surface associated: potential role in interactions with the host and the gastric pathogen Helicobacter pylori. Infect Immun 2006, 74: 425-434.

Braat H, Rottiers P, Hommes DW, Huyghebaert N, Remaut E, Remon JP et al.: A phase I trial with transgenic bacteria expressing interleukin-10 in Crohn's disease. Clin Gastroenterol Hepatol 2006, 4: 754-759.

Bron PA, Benchimol MG, Lambert J, Palumbo E, Deghorain M, Delcour J et al.: Use of the alr gene as a food-grade selection marker in lactic acid bacteria. Appl Environ Microbiol 2002, 68: 5663-5670.

Bron PA, Marco M, Hoffer SM, Van ME, de Vos WM, Kleerebezem M: Genetic characterization of the bile salt response in Lactobacillus plantarum and analysis of responsive promoters in vitro and in situ in the gastrointestinal tract. J Bacteriol 2004, 186: 7829-7835.

Chatel JM, Langella P, del-Patient K, Commissaire J, Wal JM, Corthier G: Induction of mucosal immune response after intranasal or oral inoculation of mice with Lactococcus lactis producing bovine beta-lactoglobulin. Clin Diagn Lab Immunol 2001, 8: 545-551.

Chhatwal GS: Anchorless adhesins and invasins of Gram-positive bacteria: a new class of virulence factors. Trends Microbiol 2002, 10: 205-208.

Christensen HR, Frokiaer H, Pestka JJ: Lactobacilli differentially modulate expression of cytokines and maturation surface markers in murine dendritic cells. J Immunol 2002, 168: 171-178.

Daniel C, Repa A, Wild C, Pollak A, Pot B, Breiteneder H et al.: Modulation of allergic immune responses by mucosal application of recombinant lactic acid bacteria producing the major birch pollen allergen Bet v 1. Allergy 2006, 61: 812-819.

de Ruyter PG, Kuipers OP, de Vos WM: Controlled gene expression systems for Lactococcus lactis with the food-grade inducer nisin. Appl Environ Microbiol 1996, 62: 3662-3667.

de Vos WM, Gasson MJ: Structure and expression of the Lactococcus lactis gene for phospho-beta-galactosidase (lacG) in Escherichia coli and L. lactis. J Gen Microbiol 1989, 135: 1833-1846.

Dickely F, Nilsson D, Hansen EB, Johansen E: Isolation of Lactococcus lactis nonsense suppressors and construction of a food-grade cloning vector. Mol Microbiol 1995, 15: 839-847.

Dieye Y, Usai S, Clier F, Gruss A, Piard JC: Design of a protein-targeting system for lactic acid bacteria. J Bacteriol 2001, 183: 4157-4166.

Enouf V, Langella P, Commissaire J, Cohen J, Corthier G: Bovine rotavirus nonstructural protein 4 produced by Lactococcus lactis is antigenic and immunogenic. Appl Environ Microbiol 2001, 67: 1423-1428.

Freitas DA, Leclerc S, Miyoshi A, Oliveira SC, Sommer PS, Rodrigues L et al.: Secretion of Streptomyces tendae antifungal protein 1 by Lactococcus lactis. Braz J Med Biol Res 2005, 38: 1585-1592.

Froseth BR, McKay LL: Molecular characterization of the nisin resistance region of Lactococcus lactis subsp. lactis biovar diacetylactis DRC3. Appl Environ Microbiol 1991, 57: 804-811.

Geoffroy MC, Guyard C, Quatannens B, Pavan S, Lange M, Mercenier A: Use of green fluorescent protein to tag lactic acid bacterium strains under development as live vaccine vectors. Appl Environ Microbiol 2000, 66: 383-391.

Glenting J, Madsen SM, Vrang A, Fomsgaard A, Israelsen H: A plasmid selection system in Lactococcus lactis and its use for gene expression in L. lactis and human kidney fibroblasts. Appl Environ Microbiol 2002, 68: 5051-5056.

18

Gram GJ, Fomsgaard A, Thorn M, Madsen SM, Glenting J: Immunological analysis of a Lactococcus lactis-based DNA vaccine expressing HIV gp120. Genet Vaccines Ther 2007, 5: 3.

Granato D, Perotti F, Masserey I, Rouvet M, Golliard M, Servin A et al.: Cell surface-associated lipoteichoic acid acts as an adhesion factor for attachment of Lactobacillus johnsonii La1 to human enterocyte-like Caco-2 cells. Appl Environ Microbiol 1999, 65: 1071-1077.

Granato D, Bergonzelli GE, Pridmore RD, Marvin L, Rouvet M, Corthesy-Theulaz IE: Cell surface-associated elongation factor Tu mediates the attachment of Lactobacillus johnsonii NCC533 (La1) to human intestinal cells and mucins. Infect Immun 2004, 72: 2160-2169.

Grangette C, Muller-Alouf H, Hols P, Goudercourt D, Delcour J, Turneer M et al.: Enhanced mucosal delivery of antigen with cell wall mutants of lactic acid bacteria. Infect Immun 2004, 72: 2731-2737.

Greene JD, Klaenhammer TR: Factors involved in adherence of lactobacilli to human Caco-2 cells. Appl Environ Microbiol 1994, 60: 4487-4494.

Hanniffy SB, Carter AT, Hitchin E, Wells JM: Mucosal delivery of a pneumococcal vaccine using Lactococcus lactis affords protection against respiratory infection. J Infect Dis 2007, 195: 185-193.

Hawley DK, McClure WR: Mechanism of activation of transcription initiation from the lambda PRM promoter. J Mol Biol 1982, 157: 493-525.

Hazebrouck S, Pothelune L, Azevedo V, Corthier G, Wal JM, Langella P: Efficient production and secretion of bovine beta-lactoglobulin by Lactobacillus casei. Microb Cell Fact 2007, 6: 12.

Henriksson A, Szewzyk R, Conway PL: Characteristics of the adhesive determinants of Lactobacillus fermentum 104. Appl Environ Microbiol 1991, 57: 499-502.

Henriksson A, Conway PL: Adhesion to porcine squamous epithelium of saccharide and protein moieties of Lactobacillus fermentum strain 104-S. J Gen Microbiol 1992, 138: 2657-2661.

Henriksson A, Conway PL: Adhesion of Lactobacillus fermentum 104-S to porcine stomach mucus. Curr Microbiol 1996, 33: 31-34.

Hols P, Kleerebezem M, Schanck AN, Ferain T, Hugenholtz J, Delcour J et al.: Conversion of Lactococcus lactis from homolactic to homoalanine fermentation through metabolic engineering. Nat Biotechnol 1999, 17: 588-592.

Hynonen U, Westerlund-Wikstrom B, Palva A, Korhonen TK: Identification by flagellum display of an epithelial cell- and fibronectin-binding function in the SlpA surface protein of Lactobacillus brevis. J Bacteriol 2002, 184: 3360-3367.

Kiatpapan P, Yamashita M, Kawaraichi N, Yasuda T, Murooka Y: Heterologous expression of a gene encoding cholesterol oxidase in probiotic strains of Lactobacillus plantarum and Propionibacterium freudenreichii under the control of native promoters. J Biosci Bioeng 2001, 92: 459-465.

Kim SJ, Jun dY, Yang CH, Kim YH: Expression of Helicobacter pylori cag12 gene in Lactococcus lactis MG1363 and its oral administration to induce systemic anti-Cag12 immune response in mice. Appl Microbiol Biotechnol 2006, 72: 462-470.

Kruger C, Hu Y, Pan Q, Marcotte H, Hultberg A, Delwar D et al.: In situ delivery of passive immunity by lactobacilli producing single-chain antibodies. Nat Biotechnol 2002, 20: 702-706.

Lindholm A, Smeds A, Palva A: Receptor binding domain of Escherichia coli F18 fimbrial adhesin FedF can be both efficiently secreted and surface displayed in a functional form in Lactococcus lactis. Appl Environ Microbiol 2004, 70: 2061-2071.

Lindholm A, Ellmen U, Tolonen-Martikainen M, Palva A: Heterologous protein secretion in Lactococcus lactis is enhanced by the Bacillus subtilis chaperone-like protein PrsA. Appl Microbiol Biotechnol 2006, 73: 904-914.

Liu CQ, Leelawatcharamas V, Harvey ML, Dunn NW: Cloning vectors for lactococci based on a plasmid encoding resistance to cadmium. Curr Microbiol 1996, 33: 35-39.

Liu CQ, Charoechai P, Khunajakr N, Deng YM, Widodo, Dunn NW: Genetic and transcriptional analysis of a novel plasmid-encoded copper resistance operon from Lactococcus lactis. Gene 2002, 297: 241-247.

MacCormick CA, Griffin HG, Gasson MJ: Construction of a food-grade host/vector system for Lactococcus lactis based on the lactose operon. FEMS Microbiol Lett 1995, 127: 105-109.

19

Madsen SM, Arnau J, Vrang A, Givskov M, Israelsen H: Molecular characterization of the pH-inducible and growth phase-dependent promoter P170 of Lactococcus lactis. Mol Microbiol 1999, 32: 75-87.

Matsuguchi T, Takagi A, Matsuzaki T, Nagaoka M, Ishikawa K, Yokokura T et al.: Lipoteichoic acids from Lactobacillus strains elicit strong tumor necrosis factor alpha-inducing activities in macrophages through Toll-like receptor 2. Clin Diagn Lab Immunol 2003, 10: 259-266.

Matsuzaki T, Yamazaki R, Hashimoto S, Yokokura T: The effect of oral feeding of Lactobacillus casei strain Shirota on immunoglobulin E production in mice. J Dairy Sci 1998, 81: 48-53.

Miyoshi Y, Okada S, Uchimura T, Satoh E: A mucus adhesion promoting protein, MapA, mediates the adhesion of Lactobacillus reuteri to Caco-2 human intestinal epithelial cells. Biosci Biotechnol Biochem 2006, 70: 1622-1628.

Mohamadzadeh M, Olson S, Kalina WV, Ruthel G, Demmin GL, Warfield KL et al.: Lactobacilli activate human dendritic cells that skew T cells toward T helper 1 polarization. Proc Natl Acad Sci U S A 2005, 102: 2880-2885.

Navarre WW, Schneewind O: Surface proteins of gram-positive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol Mol Biol Rev 1999, 63: 174-229.

Okano K, Kimura S, Narita J, Fukuda H, Kondo A: Improvement in lactic acid production from starch using alpha-amylase-secreting Lactococcus lactis cells adapted to maltose or starch. Appl Microbiol Biotechnol 2007.

Pavan S, Hols P, Delcour J, Geoffroy MC, Grangette C, Kleerebezem M et al.: Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: a tool to study in vivo biological effects. Appl Environ Microbiol 2000, 66: 4427-4432.

Petsch D, Anspach FB: Endotoxin removal from protein solutions. J Biotechnol 2000, 76: 97-119.

Pfeiler EA, zcarate-Peril MA, Klaenhammer TR: Characterization of a novel bile-inducible operon encoding a two component regulatory system in Lactobacillus acidophilus. J Bacteriol 2007.

Pisetsky DS: The influence of base sequence on the immunostimulatory properties of DNA. Immunol Res 1999, 19: 35-46.

Poquet I, Ehrlich SD, Gruss A: An export-specific reporter designed for gram-positive bacteria: application to Lactococcus lactis. J Bacteriol 1998, 180: 1904-1912.

Pouwels PH, Leer RJ, Shaw M, Heijne den Bak-Glashouwer MJ, Tielen FD, Smit E et al.: Lactic acid bacteria as antigen delivery vehicles for oral immunization purposes. Int J Food Microbiol 1998, 41: 155-167.

Raha AR, Varma NR, Yusoff K, Ross E, Foo HL: Cell surface display system for Lactococcus lactis: a novel development for oral vaccine. Appl Microbiol Biotechnol 2005, 68: 75-81.

Ramasamy R, Yasawardena S, Zomer A, Venema G, Kok J, Leenhouts K: Immunogenicity of a malaria parasite antigen displayed by Lactococcus lactis in oral immunisations. Vaccine 2006, 24: 3900-3908.

Ravn P, Arnau J, Madsen SM, Vrang A, Israelsen H: Optimization of signal peptide SP310 for heterologous protein production in Lactococcus lactis. Microbiology 2003, 149: 2193-2201.

Ribeiro LA, Azevedo V, Le LY, Oliveira SC, Dieye Y, Piard JC et al.: Production and targeting of the Brucella abortus antigen L7/L12 in Lactococcus lactis: a first step towards food-grade live vaccines against brucellosis. Appl Environ Microbiol 2002, 68: 910-916.

Ricci S, Macchia G, Ruggiero P, Maggi T, Bossu P, Xu L et al.: In vivo mucosal delivery of bioactive human interleukin 1 receptor antagonist produced by Streptococcus gordonii. BMC Biotechnol 2003, 3: 15.

Rojas M, Ascencio F, Conway PL: Purification and characterization of a surface protein from Lactobacillus fermentum 104R that binds to porcine small intestinal mucus and gastric mucin. Appl Environ Microbiol 2002, 68: 2330-2336.

Sibakov M, Koivula T, von WA, Palva I: Secretion of TEM beta-lactamase with signal sequences isolated from the chromosome of Lactococcus lactis subsp. lactis. Appl Environ Microbiol 1991, 57: 341-348.

Sillanpaa J, Martinez B, Antikainen J, Toba T, Kalkkinen N, Tankka S et al.: Characterization of the collagen-binding S-layer protein CbsA of Lactobacillus crispatus. J Bacteriol 2000, 182: 6440-6450.

Sorensen KI, Larsen R, Kibenich A, Junge MP, Johansen E: A food-grade cloning system for industrial strains of Lactococcus lactis. Appl Environ Microbiol 2000, 66: 1253-1258.

20

Steidler L: In situ delivery of therapeutic cytokines by genetically engineered Lactococcus lactis. Meded Rijksuniv Gent Fak Landbouwkd Toegep Biol Wet 2001, 66: 551, 553-551, 556.

Steidler L, Neirynck S, Huyghebaert N, Snoeck V, Vermeire A, Goddeeris B et al.: Biological containment of genetically modified Lactococcus lactis for intestinal delivery of human interleukin 10. Nat Biotechnol 2003, 21: 785-789.

Toba T, Virkola R, Westerlund B, Bjorkman Y, Sillanpaa J, Vartio T et al.: A Collagen-Binding S-Layer Protein in Lactobacillus crispatus. Appl Environ Microbiol 1995, 61: 2467-2471.

van Roosmalen ML, Kanninga R, El KM, Neef J, Audouy S, Bosma T et al.: Mucosal vaccine delivery of antigens tightly bound to an adjuvant particle made from food-grade bacteria. Methods 2006, 38: 144-149.

Wells JM, Wilson PW, Norton PM, Gasson MJ, Le Page RW: Lactococcus lactis: high-level expression of tetanus toxin fragment C and protection against lethal challenge. Mol Microbiol 1993, 8: 1155-1162.

Wells JM, Wilson PW, Norton PM, Le Page RW: A model system for the investigation of heterologous protein secretion pathways in Lactococcus lactis. Appl Environ Microbiol 1993, 59: 3954-3959.

Yao XY, Yuan MM, Li DJ: Mucosal inoculation of Lactobacillus expressing hCGbeta induces an anti-hCGbeta antibody response in mice of different strains. Methods 2006, 38: 124-132.

Zeuthen LH, Christensen HR, Frokiaer H: Lactic acid bacteria inducing a weak interleukin-12 and tumor necrosis factor alpha response in human dendritic cells inhibit strongly stimulating lactic acid bacteria but act synergistically with gram-negative bacteria. Clin Vaccine Immunol 2006, 13: 365-375.

21

BioMed CentralMicrobial Cell Factories

ss
Open AcceReviewLive bacterial vaccines – a review and identification of potential hazardsAnn Detmer*1 and Jacob Glenting2
Address: 1Danish Toxicology Centre, Hørsholm, Denmark and 2Bioneer A/S, Hørsholm, Denmark

Email: Ann Detmer* - [email protected]; Jacob Glenting - [email protected]

* Corresponding author

AbstractThe use of live bacteria to induce an immune response to itself or to a carried vaccine componentis an attractive vaccine strategy. Advantages of live bacterial vaccines include their mimicry of anatural infection, intrinsic adjuvant properties and their possibility to be administered orally.Derivatives of pathogenic and non-pathogenic food related bacteria are currently being evaluatedas live vaccines. However, pathogenic bacteria demands for attenuation to weaken its virulence.The use of bacteria as vaccine delivery vehicles implies construction of recombinant strains thatcontain the gene cassette encoding the antigen. With the increased knowledge of mucosalimmunity and the availability of genetic tools for heterologous gene expression the concept of livevaccine vehicles gains renewed interest. However, administration of live bacterial vaccines posessome risks. In addition, vaccination using recombinant bacteria results in the release of liverecombinant organisms into nature. This places these vaccines in the debate on application ofgenetically modified organisms. In this review we give an overview of live bacterial vaccines on themarket and describe the development of new live vaccines with a focus on attenuated bacteria andfood-related lactic acid bacteria. Furthermore, we outline the safety concerns and identify thehazards associated with live bacterial vaccines and try to give some suggestions of what to considerduring their development.

BackgroundLive vaccines have played a critical role from the begin-ning of vaccinology. Indeed, the very first vaccinationexperiment in the Western world was Jenner's inoculationof a boy with the milder cowpox virus to protect againstthe deadly smallpox. Although effective the technologyhas safety problems associated with the risk of reversionto a virulent organism and the possibility of causing dis-ease in immune compromised individuals. Within the last20 years the concept of live vaccines gains renewed inter-est due to our increased immunological understandingand the availability of molecular techniques making the

construction of safer live vaccines possible. This opens forthe development of new live bacterial vaccines that canavoid the downsides of parenterally administered vaccinebecause it (i) mimics the route of entry of many patho-gens and stimulate the mucosal immune response (ii) canbe administered orally or nasally avoiding the risk associ-ated with contaminated needles and need for a profes-sional healthcare infra structure (iii) has a simple downstream processing. Broadly, live bacterial vaccines can beclassified as a self-limiting asymptomatic organism stimu-lating an immune response to one or more expressed anti-gens.

Published: 23 June 2006

Microbial Cell Factories 2006, 5:23 doi:10.1186/1475-2859-5-23

Received: 25 April 2006Accepted: 23 June 2006

This article is available from: http://www.microbialcellfactories.com/content/5/1/23

© 2006 Detmer and Glenting; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Page 1 of 12(page number not for citation purposes)

22

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16796731

http://www.microbialcellfactories.com/content/5/1/23

http://creativecommons.org/licenses/by/2.0

http://www.biomedcentral.com/

http://www.biomedcentral.com/info/about/charter/

Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23

Furthermore, live bacterial vaccines can be designed toinduce an immune response to itself or to a carried heter-ologous antigen. A non-virulent or attenuated derivativeof the pathogen is used to induce a response to the bacte-rium itself. When used as a vaccine vehicle the bacteriumexpresses an antigen from another species. Most com-monly, these vaccine vehicles are based on either attenu-ated pathogens or bacteria used in the food industry. Bothclasses of bacteria deliver the vaccine component to theimmune system whereby immunization may benefit fromthe bacterium's intrinsic adjuvant. The vaccine compo-nent to be delivered can be either protein or DNA. In addi-tion, the vaccine component may be a classical antigenbut may also be allergens or therapeutics. A recent devel-opment is the use of invasive bacteria for the delivery ofplasmid DNA vaccines to mammalian cells obtaining invivo synthesis of the plasmid-encoded antigen. As such,the applications of live bacterial vaccines are extensiveand has lead to more than 2000 published papers. How-ever, only very few of the promising candidates have sur-vived the licensing process and become registered [1]illuminating the difficulty in developing a commerciallive vaccine. One typhoid vaccine (Ty21a) contains liveattenuated Salmonella typhi and is administered orallyeither as a liquid or as acid resistant capsules. Both formu-lations require three doses within one week to give immu-nity. The other registered vaccine based on live bacteria isagainst cholera and is given orally as a single dose of atten-uated Vibrio cholerae (CVD 103-HgR) in liquid formula-tion. This vaccine is used in a lower dose (5 × 108 livebacteria) for travellers from non-endemic regions and aone log higher dose for residents in endemic regions (5 ×109 live bacteria). The very few examples of live bacterialvaccines on the market may be due to lack of success inclinical trials. However, we believe that the safety of thesevaccines is another issue. Indeed, prophylactic vaccinesare given to healthy people and despite excellent safetyrecord they remain targets of un-substantiated allegationsby anti vaccine movements. Furthermore, future live vac-cines will most likely be either targeted mutagenised orequipped with foreign antigens and therefore consideredrecombinant. As such, they fall into the debate on releas-ing genetically modified organisms into nature. The feasi-bility of this new vaccine strategy will therefore inparticular depend on considerations of safety issues. Webelieve that considering safety issues alongside the scien-tific consideration early in vaccine development will facil-itate its public acceptance and its entrance to the market.We therefore felt compelled to outline a review about livevaccines and their safety aspects.

Attenuated pathogens as vaccines and vaccine vehiclesLindberg [2] has excellently reviewed the history of livebacterial vaccines. The first use of a live bacterial vaccinewas in Spain in 1884 and consisted of a subcutaneous

injection of weakened Vibrio cholerae. This study was fol-lowed a few years later by field trials in India with a moreefficacious V. cholerae vaccine, however still parenteral.The first live oral V. cholerae vaccine candidate did notappear until the 1980s. Later the V. cholerae strain CVD103 Hg-R has been found to be both safe and immuno-genic after a single oral dose. In 1996 a bivalent vaccinewaspresented including two strains of V. cholerae calledCVD 103 Hg-R and CVD 111 [3]. However, later on prob-lems with attenuation of strain CVD 111 appeared [4].The development of the other registered live bacterial vac-cine began Hg-in the early 1970s using various live atten-uated S. typhi to vaccinate against typhoid fever. Oneproposed strain was made streptomycin-dependent, butfailed to be efficacious in freeze-dried formulation [5].Furthermore, the strain was genetically unstable andreverted to virulence. Another S. typhi strain (Ty21a) witha defect galE gene, as well as other not defined mutations,requires an external source of galactose. This strain wasextensively evaluated in several field trials and has shownexcellent safety record [6]. Later, other auxotrophic strainsunable to synthesise essential compounds like aromaticamino acids were developed and tested on human volun-teers with variable safety and immunogenicity results [7-10]. Attenuated live vaccines to prevent shigellosis havealso been proposed. Both genetically engineered orselected mutants of Shigella have been tried but showedside effects in clinical trials and points to the need of addi-tional attenuation without hampering immunogenicity[11-13]. Kotloff et al attenuated the guanine auxotrophicShigella flexneri 2a further by deleting two genes encodingenterotoxins [14]. In a phase 1 trial this strain with inacti-vated enterotoxin genes was better tolerated but stillimmunogenic compared to the guanine auxotrophicstrain that contain active entoroxins.

Recombinant Shigella has also been proposed as a vaccinevehicle [15]. Pathogenic Shigella has a virulence plasmidencoding proteins involved in thesecretion apparatus andproteins necessary for the entry process into human cells.This invasive capacity can be used to deliver plasmid DNAvaccines into mammalian cells [16]. Here, the deliveredplasmid DNA encodes an antigen, which is expressed bythe protein synthesis apparatus of the infected cells.Diaminopimelate Shigella auxotrophs undergo lysisunless diaminopimelate is present in the growth media[16]. Human cells contain low amounts of diami-nopimelate and upon entry the Shigella mutant lyse mak-ing the delivery of vaccine components more effective.Other attenuated bacteria have also been tested as vaccinevehicles of various proteins and plasmid DNA (Table 1).In conclusion, the mimicry of natural infection makesattenuated bacteria effective. The ability to deliver vaccinecomponents of different origins like e.g., HIV [15,17,18]or piece of parasitic DNA [19] or gamete specific antigen


23


[20] make attenuated bacteria a versatile vaccinology tool.However, in spite of the efforts in constructing attenuatedpathogens for use as bacterial vaccine vehicles none ofthem has reached the market yet.

Lactic acid bacteria as vaccine vehiclesThe potential of using lactic acid bacteria (LAB) for thedelivery of vaccine components is less exploited thanattenuated pathogens. Due to their safe status and theavailability of genetic tools for recombinant gene expres-sion LAB are attractive for use as vaccine vehicles. Further-more, their non-pathogenic status circumvents the needto construct attenuated mutants. However, LAB are non-invasive and the vaccine delivery to antigen presentingcells may be less effective than invasive bacteria. Still, anti-gen specific immune responses have been obtained withseveral LAB (Table 2). Geoffroy et al [21] used a green flu-orescent protein to visualize the phagocytosis of Lactoba-cillus plantarum by macrophages in vitro and in mice.Macrophages act as antigen presenting cells and this canexplain a possible way to at least elicit a ClassII MHCreceptor presentation of the antigen. Even though thetransit time of Lactococcus lactis through the intestine is

less than 24 h in mice [22], a potent immune response hasbeen obtained with several antigens including tetanustoxin fragment C (TTFC). Surprisingly, a similar responsewas induced using dead or alive Lactococcus suggestingthat in situ antigen synthesis is not essential [23]. A slightlybetter result was in the same study obtained with L.plantarum, but also here a similar response was inducedfrom living or UV-light inactivated cells.

Active vaccination using LABThe prospect of using live LAB as vaccine carriers has beenreviewed [24,25]. The most frequently used model anti-gen is TTFC in which good results have been obtainedboth in intranasal and oral mice models using strains of L.plantarum and L. lactis [23,26]. Grangette et al [27] testedcytoplasmic expression of TTFC antigen in both L.plantarum and L. lactis showing protective effect in an oralmouse model. Shaw et al [28] tested both cytoplasmic andsurface associated expression of same TTFC antigen andfound that cytoplasmic expression was superior to surfaceexposed TTFC in L. lactis. In contrast, Bermúdez-Humaránet al [29] tested human papillomavirus type 16 E7 antigensorted in different cellular compartments and found cell

Table 1: Attenuated bacteria as vaccine vehicles

Vaccine strain Attenuation Foreign insert Model Ref.

Shigella flexneri Δasd pCMVβ Guinea pig, in vitro [80]Δasd CS3 and LTB/STm Mouse [81]ΔrfbF HIV-1 SF2Gag Mouse [17]ΔdapA ΔdapB β-gal, gene vaccine In vitro [16]ΔaroA ΔiscA gp120, gene vaccine Mouse [15]

Salmonella enterica ΔaroA pCMVβ, pCMVactA and pCMVhly

In vitro, mouse [82]

ΔaroA ΔaroD C. tetani TTFC Mouse [83]ΔaroA ΔhtrA TTFC Mouse [83]ΔaroA+others GFP+cytokines Mouse [84]Δcya Δcrp Δasd SP10 cDNA Mouse [20]GalE + unspecified H. pylori, ureAB Human [85]

Yersinia enterocolitica pYV- B. abortus, P39 Mouse [86]pYV- Ova Mouse [87]

Listeria monocytogenes ΔactA Leichmania major Mouse [88]ΔactA LCM virus Mouse [89]Δdal Δdat HIV-1 gag gene vaccine Mouse [90]Δ2 M. bovis gene vaccine Mouse [91]

Bordetella bronhiseptica ΔaroA TTFC Mouse [92]

Erysipelotrix rhusiopatie Tn916- M. hyopneumonie Mouse, pig [93]

Mycobacterium bovis unspecified P. falciparum, CSP Mouse [94]

Brucella abortus Rough mutant (O-) lacZ or HSP65 Mouse [95]


24


wall-anchored antigen to induce the most potent immuneresponse. The different outcome of these experiments maybe explained by different stability of surface exposed TTFCand E7 antigen. Intracellular expression of a labile antigencan protect it from proteolytic degradation and environ-mental stress encountered at the mucosal surfaces.Genetic modification of the LAB cell wall rendering thestrain more permeable increases the in vivo release of cyto-plasmic TTFC antigen and was tested by Grangette et al[27]. When administered orally these alanin racemasemutants were more immunogenic than their wild typecounterparts. One explanation could be that oral immu-nization is very dependant on a sufficiently large dose ofthe antigen [27].

The use of live LAB as carriers of DNA vaccines has untilnow not been an option as they are non-invasive andtherefore inefficiently deliver the plasmid DNA to thecytoplasma of antigen presenting cells. Recently Guima-rães et al [30] developed L. lactis expressing cell wall-anchored internalin from Listeria monocytogenes. This L.lactis inlA+ strain has been shown to enter eukaryotic cellsin vitro, but also in vivo using an oral guinea pig model. To

determine the tropism of recombinant invasive strainsCritchley-Thorne el al used a perfusion bath with murineileal tissue and tested an invasive E. coli vaccine candidate[31]. Although change of tropism of a bacterial carrieropens for targeted delivery it introduces new safety issuesthat should be addressed by persistence and distributionstudies of the bacterial strain after vaccination.

Active vaccination using recombinant L. johnsonii to treatallergy has been suggested [32]. IgE epitopes was fused toproteinase PrtB and cell wall-anchored. Subcutaneousand intranasal immunization of mice induced a systemicIgG response against human IgE. As such, allergy-induc-ing IgE may be cleared by IgG antibodies induced by therecombinant L. johnsonii. However, it remains to beproven if these antibodies are protective in humanpatients.

In conclusion, LAB has been successfully used for activevaccination of animals like rodents (Table 2). WhetherLAB will be effective as a mucosal vaccine in humans canonly be answered by clinical trials. Furthermore, as thedose of recombinant LAB needed to elicit immune

Table 2: LAB as vaccine vehicles

Vaccine strain Foreign insert Model Ref.

Lactococcus lactis C. tetani TTFC Mouse [23,96]TTFC+IL-2 or IL-6 Mouse [97]Human IL-10 Mouse [39]H. pylori ureB Mouse [98]B. abortus L7/L12 Mouse [99]S. pneumonie CPS Mouse [100]Rotavirus vp7 Mouse [101]B-lactoglobulin Mouse [102]HIV-1 gp120 Mouse [103]Malaria MSP-1 Mouse [104]SARS Nucleocapsid protein In vitro [105]E. rhusiopathiae SpaA Mouse [106]

Lactobacillus plantarum TTFC Mouse [107]Allergen Der p1 Mouse [36]H. pylori (ureB) Mouse [108]

Streptococcus gordonii Antibody Rat [34]Hornet venom Ag5.2 Mouse [109]TTFC Mouse [110]

Lactobacillus casei B. anthracis (protective Ag) In vitro [111]SARS spike protein Mouse [112]Human papillomavirus L1 In vitro [113]Coronavirus S glycoprotein Mouse [114]S. pneumonie PsaA PspA In vitro [115]

Lactobacillus zeae Antibody Rat [33]

Lactobacillus johnsonii TTFC mimotope Mouse [116]


25


responses in animals is high it is unknown if the necessarydose for use in humans will be feasible and cost effective.

Passive immunization using LABProtection by preformed antibodies or antibody frag-ments is called passive vaccination. The pioneer experi-ments were based on injection of antisera produced byimmunized animals like horse or sheep to combat forexample rattlesnake venom. Recently, passive immunitywas delivered using lactobacilli that secretes single-chainantibodies [33]. In a rat caries model, colonisation of themouth with a L. zeae expressing a single-chain antibodyfragment recognizing the adhesion molecule of Streptococ-cus mutans decreased the number of S. mutans and reducedthe development of caries. Recombinant Streptococcus gor-donii displaying a microbiocidal single-chain antibody(H6) has been used to treat vaginal candidiasis in a ratmodel [34]. Although passive immunity has limits in itstemporary nature, these results suggest that LAB elegantlycan be used for the delivery of neutralising antibodies atmucosal sites.

Allergy vaccines using LAB expressing allergensFor a normal vaccination against an infectious disease,induction of tolerance to the infectious agent is consid-ered a side effect. This side effect is more prone to happenwhen vaccinating early in life [35]. However, induction oftolerance can have positive clinical implications when thepurpose is to treat allergy. In a mouse model the use of arecombinant L. plantarum expressing the house dust miteallergen Der p1 as a fusion protein in the cytoplasm inhib-ited house dust mite-specific T-cell responses [36]. In thisstudy mice were sensitized by immunization with thehouse dust mite peptide and then given either L.plantarum expressing Der p1 or L. plantarum without Derp1. Both strains inhibited IFN-γ production by T cells. Butthe decrease in production of-5 was only seen for the L.plantarum expressing the Der p1 peptide antigen. Thisindicates that the lactobacilli strain expressing Der p1 cansuppress the cytokine milieu promoting the Th2 allergicresponse. Another example of the strain specific effect ofLAB on induction and maintenance of oral tolerance hasbeen shown using ί-lactoglobulin and gnotobiotic mice[37]. In this study L. paracasei (NCC 2461) was more effec-tive to induce and maintain oral tolerance in gnotobioticmice than was L. johnsonii (NCC 533). The allergen canalso be co-administered instead of recombinant expressedby the LAB. Mucosal co-application of L. plantarum or L.lactis together with birch pollen allergen Bet v1 shifted theimmune response towards an anti-allergic Th1 responseboth in sensitized and un-sensitized animals [38]. Recom-binant strains expressing immune polarizing cytokineslike IL-10 have also been developed and in vivo effects inboth mice [39] and pigs [40] have been observed. Moreknowledge on the mechanisms behind skewing the

immune response is however needed to select the properstrain with anti allergic immune polarization. Further-more, the immune regulatory effect of one strain of LABmay differ in allergic and non-allergic individuals. A downregulation in allergic persons and an immune stimulatingeffect in normal persons was observed when using samestrain of LAB [41].

Immune stimulatory effects of LABAmong LAB's effect on the immune system there is a straindependent induction of cytokines. Different LAB strainsinduce distinct mucosal cytokine profiles in BALB/c mice[42] pointing at the importance of using one strain forimmune induction and another for induction of toleranceor a partial down regulation of the immune system. Thesame authors [43] also indicate growth phase dependentdifferences of orally administered LAB strains on the IgG1(Th2)/IgG2a (Th1) antibody ratio in mice further compli-cating the process of choosing the proper strain for spe-cific modulation of the immune response. Adding to thecomplexity of these observations, a human study hasshown that non-specific immune modulation by a givenstrain of L. rhamnosus (GG, ATCC 53103) differs inhealthy and allergic subjects. In healthy persons the strainwas immune stimulatory whereas in allergic persons itdown-regulated an inflammatory response [44]. Interac-tions between different LAB strains can also interfere withthe in vitro production of cytokines by dendritic cells [45].As is shown in another study [46], two different lactoba-cilli with similar probiotic properties in vitro were shownto elicit divergent patterns of colonisation and immuneresponse in germfree mice. Further evidence for animmune modulating effect is seen when either L. lactis orL. plantarum was used in a mouse model of birch pollenallergy [38]. In combination with birch pollen allergenBet v1 both strains skewed the immune response fromTh2 to Th1 in sensitised mice as indicated by the IFN-γ/IL-5 ratio. The immune polarizing effect of LAB has also beenobserved in humans. A clinical trial showed a straindependent immune modulation of two different LABstrains when administered together with an oral S. typhivaccine (Ty21a) [47]. Here, thirty healthy volunteers wererandomised into three groups receiving L. rhamnosus GG,L. lactis or placebo for 7 days. On days 1, 3 and 5 theTy21a vaccine was given orally. Analysis showed a highernumber of specific IgA-secreting cells in the group receiv-ing L. rhamnosus GG and a higher CR3 receptor expressionon neutrophils in the group receiving L. lactis. A partialdown regulation of the immune system has also beenobserved. Atopic children receiving 2 × 1010 L. rhamnosusGG daily for 30 days enhanced their IL-10 production insera as well as in mitogen-induced peripheral bloodmononuclear cells [41].


26


It can be concluded that immune polarization towardseither a Th2 or a Th1 response can be obtained using dif-ferent LAB. As such the intrinsic immune modulatorycapacity of the LAB must be evaluated and selected to fitthe purpose of vaccination.

Safety concerns of the bacterial vaccine strainSeveral safety concerns of the bacterial vaccine strain havebeen raised (Table 3). Before using pathogenic bacteria forvaccination purposes, its pathogenicity must be weakenedvia attenuation. Attenuation usually involves deletion ofessential virulence factors or mutation of genes encodingmetabolic enzymes whose function is essential for sur-vival outside the laboratory. Inactivation of a metabolicgene has the advantage that the bacteria still express viru-lence determinants important to elicit a protectiveimmune response. Appropriate stable auxotrophic strainsare usually not able to replicate in the human body andcan safely be used even in immune compromised individ-uals. Defined deletions of at least two metabolic essentialgenes are usually used [2] and decrease the probability ofreversion to virulence. To reduce the risk of spreading for-eign genetic material to the environment the antigenencoding gene cassette can be inserted into the chromo-some replacing the metabolic essential gene. If the bacte-rium acquires the deleted gene it will automatically loosethe antigen-encoding cassette. The use of antibiotic resist-ance genes as marker genes in vaccines is not encouragedas these genes can transfer to in the end humans and thus

hamper the use of therapeutic antibiotics. Different alter-natives to antibiotic resistance marker genes have beenpublished and should be used as soon as possible in thedevelopmental process of a vaccine [48-50].

Another concern using live bacterial vaccines is the onsetof autoimmune responses like arthritis especially inpatients with the HLA-B27 tissue type [51]. However, therisk is certainly lower than after natural infection. Theoccurrence of such side effects can best be followed bypost launch monitoring and must always be evaluatedagainst the health risks associated to the disease itself. Atheoretical side effect of vaccines is the possible inductionof autoimmune reactions. However, there is no recom-mendation to avoid vaccination of people with an ongo-ing autoimmune disease like rheumatoid arthritis orsystemic lupus erythematosus if vaccination otherwise ismotivated [52]. In contrast, immune-compromised hostscan have difficulties in handling replicating live attenu-ated vaccines and should therefore not be vaccinated withsuch vaccines. However, new ways of further attenuatingbacteria like combining auxotrophy with deletions of vir-ulence genes [14] may open for the use of live vaccines toimmune-compromised hosts. In addition, immune-com-promised people close to hosts vaccinated with live atten-uated vaccines should be aware of the risk of crosscontamination with the vaccine strain.

Table 3: Safety concerns of the vaccine strain

Systemic disturbance Systemic infectionConversion from avirulent to virulent bacteriumTranslocation to organsDisturbance of digestive processesInhibition of bacterial production of nutrients

Immune system Absorption of allergens through the intestinal epitheliumInduction of tolerance to pathogen instead of immunityInduction or potentiation of autoimmunityBacterial mimicry of self-antigen

Metabolites Production of harmful/undesired metabolites including enzymatic activitiesBreakdown of chemicals to toxic metabolites

Implications for natural flora in GI tract Permanent colonisation of cell substrate in the intestineGene/plasmid transfer to host's indigenous flora"Competitive exclusion" of indigenous flora

Unintentional transferral of cell substrate Unintentional transfer to other individualsUnintentional transfer to and viability/propagation in environments other than the intestines

Contamination Extraneous or perceived adventitious DNA components should be removed (possibility of oncogenicity).


27


Adverse examples of human live vaccine strains causingdeath and illness among domesticated animals are rarebut existent. In Mongolia in autumn 1979 the H1N1influenza A vaccine virus may have caused a severe influ-enza epizootic among camels [53]. No examples ofhuman bacterial vaccines causing problems among ani-mals have been found in the literature but the possibilityexists and has to be both tested and evaluated beforerelease of a live bacterial vaccine. In general the spread oflive bacterial vaccines to the environment is a concern.However, attenuated human pathogens are usually notadapted to live outside its host. Therefore survival in theenvironment is usually short. Vaccines based on recom-binantLAB may result in the release of these bacteria innature, as LAB are more suited to survive in the nature.Also here the use of auxotrophic mutants unable to repli-cate in the environment may be the answer. Releasinggene-modified organisms into the environment can causedebate and precautions to eliminate its spread are essen-tial. To avoid escape into the environment of the geneti-cally modified organism, Steidler et al. [40] replaced thethyA gene with the expressioncassette for human IL-10. Asa consequence, the L. lactis mutant is dependant on thy-midine or thymine for growth, which is present in lowamounts in nature and in the human body. Furthermore,acquirement of an intact thyA gene would recombine thetransgene out of the genome, resulting in reversion to itswild type state.

Safety concerns of the antigen encoding sequenceIn live bacterial vaccines the antigen-encoding gene iseither plasmid located or integrated in to the chromo-some. In both cases several safety concerns can be raised(Table 4). For plasmid-encoded antigens the fate of theplasmid in the vaccinee must be evaluated. The use of aprokaryote plasmid replication unit of narrow host rangecan limit the horizontal plasmid transfer to other bacteriapresent in the vaccinated individual and prevent unde-sired persistence of the plasmid. In particular for plasmidDNA vaccines a study should identify which cells take upand/or express the DNA and what is the fate of the DNAwithin those cells as well as for how long the DNA persistsin the cells [54]. Nasal administration of a naked DNA-

vaccine in mice led to some accumulation of plasmidDNA in the brain [55] illustrating the diffusion of theplasmid after immunization. The amount of accumulat-ing plasmid that is acceptable outside the target cellsneeds to be further clarified.

The recombinant plasmid harboured by bacterial vaccinevehicles may integrate in the genome of the recipient andpotentially cause oncogenesesis. Concerns about thepotential oncogenicity of biological products like contin-uous cell line products (CCL), DNA vaccines and genetherapy products have been raised [54]. In CCLs foreignDNA should be avoided in the final product and a limithas been defined as for maximal residual amount perhuman dose. In DNA vaccines DNA is obvious present butinsertion of DNA should be avoided. Finally in the genetherapy product DNA is both present and inserted butinsertional oncogenesis should be avoided. Integration offoreign DNA into the host genome is by definition inser-tional mutagenesis and can induce oncogenesis. There arethree ways the extraneous DNA can lead to transforma-tion [54]: insertion of an active oncogene, insertional acti-vation of a host proto-oncogene, and by insertionaldeactivation of a host suppressor gene. The mechanismbehind DNA integration into the chromosome is either byrandom integration, homologous recombination or retro-viral insertion [56]. The most probable cause of unwantedintegration is by random integration which occurs at a fre-quency of approximately 10-4 [54]. Unwanted integrationby homologous recombination and retroviral insertioncan be avoided by omission of sequences necessary forinsertion [57]. Analysing the antigen encoding unit car-ried by the bacteria for human homologous sequencesand eliminating these can limit the integrative possibility.Although not similar to vaccination with bacteria the clin-ical trials using retroviral therapy can give some indica-tions of the hazards of DNA integration [58]. Indeed,activation of oncogenes is a risk associated with retroviralvaccination [59]. The report of adverse effects in a Frenchgene therapy study, where 2 out of 10 patients developedleukaemia within 3 years of [60,61], illustrates occurrenceof such a transformation event by activation of a proto-oncogene. Calculation of the probability of a harmful

Table 4: Safety concerns of the antigen encoding sequence

For protein and DNA vaccines Transfer of undesired genes via plasmidTransfer of vector to indigenous floraOpen reading frames coding for injurious peptides (allergens)Imprecise transcription and translation

Specifically for DNA-vaccines Persistence of DNAPermanent expression of the foreign antigenFormation of anti-DNA antibodiesTransformation eventSpread of antibiotic resistance genes


28


effect due to integration of foreign DNA into host genomehas been performed and was found to be less than 10-16 to10-19 per DNA molecule [62]. This frequency must be putin relation to the spontaneous mutation frequency whichhas been estimated in humans to occur at the rate of 1 inevery 50 million nucleotides incorporated during DNAreplication. This means that a human cell with 6 × 109

base pairs will contain 120 new mutations [63].

Possible insertions into the chromosome can be tested byPCR techniques [64,65]. However, insertion due to ran-dom integration can be difficult to detect this way [64].Furthermore, insertion of foreign DNA can effect geneactivity at sites remote from insertion [66]. Different ani-mal trial has foreseen possible adverse effects like in thefollowing two examples. Foreign DNA ingested by micehas been shown to be covalently linked to mouse DNA[67]. Foreign DNA has also been shown in associationwith chromosomes in fetuses born by mice fed orally withbacteriophage M13 DNA [68]. There is however, no evi-dence for a germ line transmission of ingested foreignDNA [66]. The de novo methylation that frequently occurswith integrated foreign DNA has been suggested as beinga natural defence mechanism [69].

In conclusion, integration of the plasmid harboured bybacterial vaccine vehicles is a potential hazard. Integrationof gene therapy vectors has been observed, but omittingsequences driving the insertion may limit the possibilityfor integration of the plasmid carried by the bacterium.Plasmids for heterologous gene expression are usuallypreferred due to its multi copy nature and higher genedosage. However, placing the antigen encoding genes onto the bacterial chromosome may limit the spread of thegenes. The route of administration of the vaccine may alsobe important when evaluating hazards. As live bacterialvaccines is fit for mucosal administration one mustremember that ingestion of foreign DNA does occur everyday with our food and is as such not new.

Peptides can be absorbed through the mucosa and somemay induce an allergic reaction. The existence of genes inthe bacterial vaccine coding for such potential allergens orother injurious peptides can be checked beforehandsearching for homologies to known allergens, as the fullsequence of the bacteria and plasmid should be known.

Vaccination using live bacterial vaccines or exposure tothe natural infections can lead to the formation of autoreactive antibodies, especially in people prone to autoim-mune diseases. However, the half life of the induced autoantibodies is usually short [70] and their specificity usu-ally polyclonal [71]. Several authors have tried to eluci-date the possibility of a link between autoimmunity andvaccination [70,72-77] and much controversy in this mat-

ter is still existing. However, convincing data establishinga link between vaccination and autoimmunity in man arestill not presented. In a mouse model a difference in clin-ical outcome was observed in two different mouse strainsin relation with auto-antibodies induced by vaccinationwith dendritic cells loaded with apoptotic thymocytes[78]. In normal BALB/c mice the presence of post vaccina-tion autoantibodies was not associated with any clinicalor histological sign of autoimmunity. However, in miceprone to autoimmunity (NZBxNZW) F1 a severe pathol-ogy attributed to autoimmunity was observed. This differ-ence in outcome attributed to the difference in genotypehas also been observed in humans and it can be con-cluded that susceptibility to autoimmunity is determinedmore by genetic factors than by vaccine challenge despitethe formation of post vaccination auto-antibodies [77]. Avaccination or treatment with adjuvant can also activateregulatory T cells and can thus be used as a method to pre-vent autoimmune disease if applied at the right time [79].In the future tailor-made vaccines might be the solutionfor individuals with a genetic profile prone to autoimmu-nity.

ConclusionBoth attenuated bacteria like salmonella and food relatedlactic acid bacteria have been developed as live vaccinessuitable for oral administration. Today, live vaccinesbased on attenuated S. typhi and V. cholerae are available.The development of bacterial vaccine vehicles carrying aheterologous gene or a DNA vaccine is more problematicand none has yet reached the market. Several bacteriahave been suggested as vaccine vehicles and especially lac-tic acid bacteria are promising. Their safe status andimmune modulating capacity have been tested usingdiverse vaccine components like antigens from infectiousdiseases, allergy promoting proteins and therapeutic anti-bodies. However, considerable safety issues against livevaccine vehicles can be raised. Their recombinant naturecalls for a bio containment strategy and auxotrophmutants may be the answer. The bacterial host must befully sequenced and evaluated using bioinformatics toolsfor the production of allergy inducing peptides. The anti-gen encoding gene cassette must be sequenced andhomologies to self proteins or allergy inducing proteinsshould be addressed. Especially bacteria carrying recom-binant plasmids the probability of horizontal gene trans-fer to other bacteria present should be avoided by usinghost restricted replication units. Furthermore, the plas-mids should be evaluated for sequences facilitating inte-gration into the human genome.

Authors' contributionsThe authors contributed equally to this work.

Acknowledgements


29


This work was partly financed by the Danish Ministry of Science, Technol-ogy and Innovation. The authors thank Dr. Anders Permin for valuable comments to the manuscript.

References1. Dietrich G, Griot-Wenk M, Metcalfe IC, Lang AB, Viret JF: Experi-

ence with registered mucosal vaccines. Vaccine 2003,21:678-683.

2. Lindberg AA: The history of live bacterial vaccines. Dev BiolStand 1995, 84:211-219.

3. Taylor DN, Tacket CO, Losonsky G, Castro O, Gutierrez J, Meza R,Nataro JP, Kaper JB, Wasserman SS, Edelman R, Levine MM, Cryz SJ:Evaluation of a bivalent (CVD 103-HgR/CVD 111) live oralcholera vaccine in adult volunteers from the United Statesand Peru. Infect Immun 1997, 65:3852-3856.

4. Taylor DN, Sanchez JL, Castro JM, Lebron C, Parrado CM, JohnsonDE, Tacket CO, Losonsky GA, Wasserman SS, Levine MM, Cryz SJ:Expanded safety and immunogenicity of a bivalent, oral,attenuated cholera vaccine, CVD 103-HgR plus CVD 111, inUnited States military personnel stationed in Panama. InfectImmun 1999, 67:2030-2034.

5. Levine MM, DuPont HL, Hornick RB, Snyder MJ, Woodward W, Gil-man RH, Libonati JP: Attenuated, streptomycin-dependent Sal-monella typhi oral vaccine: potential deleterious effects oflyophilization. J Infect Dis 1976, 133:424-429.

6. Engels EA, Falagas ME, Lau J, Bennish ML: Typhoid fever vaccines:a meta-analysis of studies on efficacy and toxicity. BMJ 1998,316:110-116.

7. Tacket CO, Hone DM, Curtiss RIII, Kelly SM, Losonsky G, Guers L,Harris AM, Edelman R, Levine MM: Comparison of the safety andimmunogenicity of delta aroC delta aroD and delta cya deltacrp Salmonella typhi strains in adult volunteers. Infect Immun1992, 60:536-541.

8. Sztein MB, Wasserman SS, Tacket CO, Edelman R, Hone D, LindbergAA, Levine MM: Cytokine production patterns and lymphopro-liferative responses in volunteers orally immunized withattenuated vaccine strains of Salmonella typhi. J Infect Dis1994, 170:1508-1517.

9. Edwards MF, Stocker BA: Construction of delta aroA his deltapur strains of Salmonella typhi. J Bacteriol 1988, 170:3991-3995.

10. Tacket CO, Sztein MB, Wasserman SS, Losonsky G, Kotloff KL,Wyant TL, Nataro JP, Edelman R, Perry J, Bedford P, Brown D, Chat-field S, Dougan G, Levine MM: Phase 2 clinical trial of attenuatedSalmonella enterica serovar typhi oral live vector vaccineCVD 908-htrA in U.S. volunteers. Infect Immun 2000,68:1196-1201.

11. Lindberg AA: Vaccination against enteric pathogens: from sci-ence to vaccine trials. Current opinion in microbiology 1998,1:116-124.

12. Kotloff KL, Taylor DN, Sztein MB, Wasserman SS, Losonsky GA,Nataro JP, Venkatesan M, Hartman A, Picking WD, Katz DE, Camp-bell JD, Levine MM, Hale TL: Phase I evaluation of delta virGShigella sonnei live, attenuated, oral vaccine strain WRSS1in healthy adults. Infect Immun 2002, 70:2016-2021.

13. Kotloff KL, Noriega F, Losonsky GA, Sztein MB, Wasserman SS,Nataro JP, Levine MM: Safety, immunogenicity, and transmissi-bility in humans of CVD 1203, a live oral Shigella flexneri 2avaccine candidate attenuated by deletions in aroA and virG.Infect Immun 1996, 64:4542-4548.

14. Kotloff KL, Pasetti MF, Barry EM, Nataro JP, Wasserman SS, SzteinMB, Picking WD, Levine MM: Deletion in the Shigella entero-toxin genes further attenuates Shigella flexneri 2a bearingguanine auxotrophy in a phase 1 trial of CVD 1204 and CVD1208. J Infect Dis 2004, 190:1745-1754.

15. Shata MT, Hone DM: Vaccination with a Shigella DNA vaccinevector induces antigen-specific CD8(+) T cells and antiviralprotective immunity. J Virol 2001, 75:9665-9670.

16. Courvalin P, Goussard S, Grillot-Courvalin C: Gene transfer frombacteria to mammalian cells. C R Acad Sci III 1995,318:1207-1212.

17. Xu F, Hong M, Ulmer JB: Immunogenicity of an HIV-1 gag DNAvaccine carried by attenuated Shigella. Vaccine 2003,21:644-648.

18. Vecino WH, Morin PM, Agha R, Jacobs WRJ, Fennelly GJ: MucosalDNA vaccination with highly attenuated Shigella is superior

to attenuated Salmonella and comparable to intramuscularDNA vaccination for T cells against HIV. Immunol Lett 2002,82:197-204.

19. Soussi N, Milon G, Colle JH, Mougneau E, Glaichenhaus N, GoossensPL: Listeria monocytogenes as a short-lived delivery systemfor the induction of type 1 cell-mediated immunity againstthe p36/LACK antigen of Leishmania major. Infect Immun2000, 68:1498-1506.

20. Srinivasan J, Tinge S, Wright R, Herr JC, Curtiss RIII: Oral immuni-zation with attenuated Salmonella expressing human spermantigen induces antibodies in serum and the reproductivetract. Biol Reprod 1995, 53:462-471.

21. Geoffroy MC, Guyard C, Quatannens B, Pavan S, Lange M, MercenierA: Use of green fluorescent protein to tag lactic acid bacte-rium strains under development as live vaccine vectors. ApplEnviron Microbiol 2000, 66:383-391.

22. Gruzza M, Fons M, Ouriet MF, Duval-Iflah Y, Ducluzeau R: Study ofgene transfer in vitro and in the digestive tract of gnotobioticmice from Lactococcus lactis strains to various strainsbelonging to human intestinal flora. Microb Releases 1994,2:183-189.

23. Robinson K, Chamberlain LM, Schofield KM, Wells JM, Le Page RW:Oral vaccination of mice against tetanus with recombinantLactococcus lactis. Nat Biotechnol 1997, 15:653-657.

24. Mercenier A, Muller-Alouf H, Grangette C: Lactic acid bacteria aslive vaccines. Curr Issues Mol Biol 2000, 2:17-25.

25. Seegers JF: Lactobacilli as live vaccine delivery vectors:progress and prospects. Trends Biotechnol 2002, 20:508-515.

26. Grangette C, Muller-Alouf H, Geoffroy M, Goudercourt D, TurneerM, Mercenier A: Protection against tetanus toxin after intra-gastric administration of two recombinant lactic acid bacte-ria: impact of strain viability and in vivo persistence. Vaccine2002, 20:3304-3309.

27. Grangette C, Muller-Alouf H, Hols P, Goudercourt D, Delcour J,Turneer M, Mercenier A: Enhanced mucosal delivery of antigenwith cell wall mutants of lactic acid bacteria. Infect Immun2004, 72:2731-2737.

28. Shaw DM, Gaerthe B, Leer RJ, Van Der Stap JG, Smittenaar C, HeijneDBG, Thole JE, Tielen FJ, Pouwels PH, Havenith CE: Engineeringthe microflora to vaccinate the mucosa: serum immu-noglobulin G responses and activated draining cervicallymph nodes following mucosal application of tetanus toxinfragment C-expressing lactobacilli. Immunology 2000,100:510-518.

29. Bermudez-Humaran LG, Langella P, Cortes-Perez NG, Gruss A,Tamez-Guerra RS, Oliveira SC, Saucedo-Cardenas O, Montes OL, LeLoir Y: Intranasal immunization with recombinant Lactococ-cus lactis secreting murine interleukin-12 enhances antigen-specific Th1 cytokine production. Infect Immun 2003,71:1887-1896.

30. Guimaraes VD, Gabriel JE, Lefevre F, Cabanes D, Gruss A, Cossart P,Azevedo V, Langella P: Internalin-expressing Lactococcus lactisis able to invade small intestine of guinea pigs and deliverDNA into mammalian epithelial cells. Microbes Infect 2005,7:836-844.

31. Critchley-Thorne RJ, Stagg AJ, Vassaux G: RecombinantEscherichia coli Expressing Invasin Targets the Peyer'sPatches: the Basis for a Bacterial Formulation for Oral Vac-cination. Mol Ther 2006.

32. Scheppler L, Vogel M, Marti P, Muller L, Miescher SM, Stadler BM:Intranasal immunisation using recombinant Lactobacillusjohnsonii as a new strategy to prevent allergic disease. Vac-cine 2005, 23:1126-1134.

33. Kruger C, Hu Y, Pan Q, Marcotte H, Hultberg A, Delwar D, van DalenPJ, Pouwels PH, Leer RJ, Kelly CG, van Dollenweerd C, Ma JK, Ham-marstrom L: In situ delivery of passive immunity by lactobacilliproducing single-chain antibodies. Nat Biotechnol 2002,20:702-706.

34. Beninati C, Oggioni MR, Boccanera M, Spinosa MR, Maggi T, Conti S,Magliani W, De Bernardis F, Teti G, Cassone A, Pozzi G, Polonelli L:Therapy of mucosal candidiasis by expression of an anti-idio-type in human commensal bacteria. Nat Biotechnol 2000,18:1060-1064.

35. Mor G, Yamshchikov G, Sedegah M, Takeno M, Wang R, HoughtenRA, Hoffman S, Klinman DM: Induction of neonatal tolerance by


30




























































































plasmid DNA vaccination of mice. J Clin Invest 1996,98:2700-2705.

36. Kruisselbrink A, MJ HBG, Havenith CE, Thole JE, Janssen R: Recom-binant Lactobacillus plantarum inhibits house dust mite-spe-cific T-cell responses. Clin Exp Immunol 2001, 126:2-8.

37. Prioult G, Fliss I, Pecquet S: Effect of probiotic bacteria on induc-tion and maintenance of oral tolerance to beta-lactoglobulinin gnotobiotic mice. Clin Diagn Lab Immunol 2003, 10:787-792.

38. Repa A, Grangette C, Daniel C, Hochreiter R, Hoffmann-Sommergru-ber K, Thalhamer J, Kraft D, Breiteneder H, Mercenier A, Wieder-mann U: Mucosal co-application of lactic acid bacteria andallergen induces counter-regulatory immune responses in amurine model of birch pollen allergy. Vaccine 2003, 22:87-95.

39. Steidler L, Hans W, Schotte L, Neirynck S, Obermeier F, Falk W, FiersW, Remaut E: Treatment of murine colitis by Lactococcus lac-tis secreting interleukin-10. Science 2000, 289:1352-1355.

40. Steidler L, Neirynck S, Huyghebaert N, Snoeck V, Vermeire A, God-deeris B, Cox E, Remon JP, Remaut E: Biological containment ofgenetically modified Lactococcus lactis for intestinal deliveryof human interleukin 10. Nat Biotechnol 2003, 21:785-789.

41. Pessi T, Sutas Y, Hurme M, Isolauri E: Interleukin-10 generationin atopic children following oral Lactobacillus rhamnosusGG. Clin Exp Allergy 2000, 30:1804-1808.

42. Maassen CB, Holten-Neelen C, Balk F, Bak-Glashouwer MJ, Leer RJ,Laman JD, Boersma WJ, Claassen E: Strain-dependent inductionof cytokine profiles in the gut by orally administered Lacto-bacillus strains. Vaccine 2000, 18:2613-2623.

43. Maassen CB, Boersma WJ, Holten-Neelen C, Claassen E, Laman JD:Growth phase of orally administered Lactobacillus strainsdifferentially affects IgG1/IgG2a ratio for soluble antigens:implications for vaccine development. Vaccine 2003,21:2751-2757.

44. Pelto L, Isolauri E, Lilius EM, Nuutila J, Salminen S: Probiotic bacte-ria down-regulate the milk-induced inflammatory responsein milk-hypersensitive subjects but have an immunostimula-tory effect in healthy subjects. Clin Exp Allergy 1998,28:1474-1479.

45. Christensen HR, Frokiaer H, Pestka JJ: Lactobacilli differentiallymodulate expression of cytokines and maturation surfacemarkers in murine dendritic cells. J Immunol 2002, 168:171-178.

46. Ibnou-Zekri N, Blum S, Schiffrin EJ, von der WT: Divergent pat-terns of colonization and immune response elicited fromtwo intestinal Lactobacillus strains that display similar prop-erties in vitro. Infect Immun 2003, 71:428-436.

47. Fang H, Elina T, Heikki A, Seppo S: Modulation of humoralimmune response through probiotic intake. FEMS ImmunolMed Microbiol 2000, 29:47-52.

48. Glenting J, Madsen SM, Vrang A, Fomsgaard A, Israelsen H: A plas-mid selection system in Lactococcus lactis and its use forgene expression in L. lactis and human kidney fibroblasts.Appl Environ Microbiol 2002, 68:5051-5056.

49. Tauch A, Gotker S, Puhler A, Kalinowski J, Thierbach G: The alanineracemase gene alr is an alternative to antibiotic resistancegenes in cloning systems for industrial Corynebacteriumglutamicum strains. J Biotechnol 2002, 99:79-91.

50. Morona R, Yeadon J, Considine A, Morona JK, Manning PA: Con-struction of plasmid vectors with a non-antibiotic selectionsystem based on the Escherichia coli thyA+ gene: applicationto cholera vaccine development. Gene 1991, 107:139-144.

51. Adachi JA, D'Alessio FR, Ericsson CD: Reactive arthritis associ-ated with typhoid vaccination in travelers: report of twocases with negative HLA-B27. J Travel Med 2000, 7:35-36.

52. Cohen AD, Shoenfeld Y: Vaccine-induced autoimmunity. JAutoimmun 1996, 9:699-703.

53. Yamnikova SS, Mandler J, Bekh-Ochir ZH, Dachtzeren P, Ludwig S,Lvov DK, Scholtissek C: A reassortant H1N1 influenza A viruscaused fatal epizootics among camels in Mongolia. Virology1993, 197:558-563.

54. Robertson JS: Safety considerations for nucleic acid vaccines.Vaccine 1994, 12:1526-1528.

55. Oh YK, Kim JP, Hwang TS, Ko JJ, Kim JM, Yang JS, Kim CK: Nasalabsorption and biodistribution of plasmid DNA: an alterna-tive route of DNA vaccine delivery. Vaccine 2001,19:4519-4525.

56. Coffin JM: Molecular mechanisms of nucleic acid integration.J Med Virol 1990, 31:43-49.

57. VandenDriessche T, Collen D, Chuah MK: Biosafety of onco-ret-roviral vectors. Curr Gene Ther 2003, 3:501-515.

58. McTaggart S, Al Rubeai M: Retroviral vectors for human genedelivery. Biotechnol Adv 2002, 20:1-31.

59. Reiprich S, Gundlach BR, Fleckenstein B, Uberla K: Replication-competent chimeric lenti-oncovirus with expanded host celltropism. J Virol 1997, 71:3328-3331.

60. Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, Wulf-fraat N, Leboulch P, Lim A, Osborne CS, Pawliuk R, Morillon E,Sorensen R, Forster A, Fraser P, Cohen JI, de Saint BG, Alexander I,Wintergerst U, Frebourg T, Aurias A, Stoppa-Lyonnet D, Romana S,Radford-Weiss I, Gross F, Valensi F, Delabesse E, Macintyre E, SigauxF, Soulier J, Leiva LE, Wissler M, Prinz C, Rabbitts TH, Le Deist F,Fischer A, Cavazzana-Calvo M: LMO2-associated clonal T cellproliferation in two patients after gene therapy for SCID-X1.Science 2003, 302:415-419.

61. Hacein-Bey-Abina S, Von Kalle C, Schmidt M, Le Deist F, Wulffraat N,McIntyre E, Radford I, Villeval JL, Fraser CC, Cavazzana-Calvo M,Fischer A: A serious adverse event after successful gene ther-apy for X-linked severe combined immunodeficiency. N EnglJ Med 2003, 348:255-256.

62. Temin HM: Overview of biological effects of addition of DNAmolecules to cells. J Med Virol 1990, 31:13-17.

63. Kimball's Biology Pages http://users.rcn.com/jkim-ball.ma.ultranet/BiologyPages/M/Muta-tions.html#Frequency_of_Mutations 2004 [http:users.rcn.com/jkimball.ma.ultranet/BiologyPages/M/Mutations.html#Frequency_of_Mutations].

64. Haworth R, Pilling AM: The PCR assay in the preclinical safetyevaluation of nucleic acid medicines. Hum Exp Toxicol 2000,19:267-276.

65. Ledwith BJ, Manam S, Troilo PJ, Barnum AB, Pauley CJ, Griffiths TG,Harper LB, Schock HB, Zhang H, Faris JE, Way PA, Beare CM, BagdonWJ, Nichols WW: Plasmid DNA vaccines: assay for integrationinto host genomic DNA. Dev Biol (Basel) 2000, 104:33-43.

66. Doerfler W, Hohlweg U, Muller K, Remus R, Heller H, Hertz J: For-eign DNA integration--perturbations of the genome--onco-genesis. Ann N Y Acad Sci 2001, 945:276-288.

67. Schubbert R, Renz D, Schmitz B, Doerfler W: Foreign (M13) DNAingested by mice reaches peripheral leukocytes, spleen, andliver via the intestinal wall mucosa and can be covalentlylinked to mouse DNA. Proc Natl Acad Sci U S A 1997, 94:961-966.

68. Schubbert R, Hohlweg U, Renz D, Doerfler W: On the fate oforally ingested foreign DNA in mice: chromosomal associa-tion and placental transmission to the fetus. Mol Gen Genet1998, 259:569-576.

69. Doerfler W: Patterns of DNA methylation--evolutionary ves-tiges of foreign DNA inactivation as a host defense mecha-nism. A proposal. Biol Chem Hoppe Seyler 1991, 372:557-564.

70. Abu-Shakra M, Press J, Sukenik S, Buskila D: Influenza virus vacci-nation of patients with SLE: effects on generation of autoan-tibodies. Clin Rheumatol 2002, 21:369-372.

71. Tomer Y: Anti-thyroglobulin autoantibodies in autoimmunethyroid diseases: cross-reactive or pathogenic? Clin ImmunolImmunopathol 1997, 82:3-11.

72. Wahlberg J, Fredriksson J, Vaarala O, Ludvigsson J: Vaccinationsmay induce diabetes-related autoantibodies in one-year-oldchildren. Ann N Y Acad Sci 2003, 1005:404-408.

73. Classen JB, Classen DC: Clustering of cases of type 1 diabetesmellitus occurring 2-4 years after vaccination is consistentwith clustering after infections and progression to type 1 dia-betes mellitus in autoantibody positive individuals. J PediatrEndocrinol Metab 2003, 16:495-508.

74. Huang YP, Gauthey L, Michel M, Loreto M, Paccaud M, Pechere JC,Michel JP: The relationship between influenza vaccine-inducedspecific antibody responses and vaccine-induced nonspecificautoantibody responses in healthy older women. J Gerontol1992, 47:M50-M55.

75. Shoenfeld Y, Aron-Maor A: Vaccination and autoimmunity-'vac-cinosis': a dangerous liaison? J Autoimmun 2000, 14:1-10.

76. Rose NR: Immunologic hazards associated with vaccinationof humans. J Autoimmun 2000, 14:11-13.

77. Belloni C, Avanzini MA, De Silvestri A, Martinetti M, Pasi A, CoslovichE, Autelli M, Masanti ML, Cuccia M, Tinelli C, Rondini G, Lorini R: Noevidence of autoimmunity in 6-year-old children immunized


31




































































http://users.rcn.com/jkimball.ma.ultranet/BiologyPages/M/Mutations.html#Frequency_of_Mutations








































at birth with recombinant hepatitis B vaccine. Pediatrics 2002,110:e4.

78. Bondanza A, Zimmermann VS, Dell'Antonio G, Dal Cin E, Capobi-anco A, Sabbadini MG, Manfredi AA, Rovere-Querini P: Cuttingedge: dissociation between autoimmune response and clini-cal disease after vaccination with dendritic cells. J Immunol2003, 170:24-27.

79. Singh B: Stimulation of the developing immune system canprevent autoimmunity. J Autoimmun 2000, 14:15-22.

80. Sizemore DR, Branstrom AA, Sadoff JC: Attenuated Shigella as aDNA delivery vehicle for DNA-mediated immunization. Sci-ence 1995, 270:299-302.

81. Zheng JP, Zhang ZS, Li SQ, Liu XX, Yuan SL, Wang P, Zhan DW,Wang LC, Huang CF: Construction of a novel Shigella live-vec-tor strain co-expressing CS3 and LTB/STm of enterotoxi-genic E.coli. World J Gastroenterol 2005, 11:3411-3418.

82. Darji A, Guzman CA, Gerstel B, Wachholz P, Timmis KN, WehlandJ, Chakraborty T, Weiss S: Oral somatic transgene vaccinationusing attenuated S. typhimurium. Cell 1997, 91:765-775.

83. Roberts M, Chatfield S, Pickard D, Li J, Bacon A: Comparison ofabilities of Salmonella enterica serovar typhimurium aroAaroD and aroA htrA mutants to act as live vectors. InfectImmun 2000, 68:6041-6043.

84. Yuhua L, Kunyuan G, Hui C, Yongmei X, Chaoyang S, Xun T, DamingR: Oral cytokine gene therapy against murine tumor usingattenuated Salmonella typhimurium. Int J Cancer 2001,94:438-443.

85. Bumann D, Metzger WG, Mansouri E, Palme O, Wendland M, Hur-witz R, Haas G, Aebischer T, von Specht BU, Meyer TF: Safety andimmunogenicity of live recombinant Salmonella entericaserovar Typhi Ty21a expressing urease A and B from Helico-bacter pylori in human volunteers. Vaccine 2001, 20:845-852.

86. Al Mariri A, Tibor A, Lestrate P, Mertens P, De BX, Letesson JJ: Yers-inia enterocolitica as a vehicle for a naked DNA vaccineencoding Brucella abortus bacterioferritin or P39 antigen.Infect Immun 2002, 70:1915-1923.

87. Wiedig CA, Kramer U, Garbom S, Wolf-Watz H, Autenrieth IB:Induction of CD8+ T cell responses by Yersinia vaccine car-rier strains. Vaccine 2005, 23:4984-4998.

88. Saklani-Jusforgues H, Fontan E, Soussi N, Milon G, Goossens PL:Enteral immunization with attenuated recombinant Listeriamonocytogenes as a live vaccine vector: organ-dependentdynamics of CD4 T lymphocytes reactive to a Leishmaniamajor tracer epitope. Infect Immun 2003, 71:1083-1090.

89. Tvinnereim AR, Hamilton SE, Harty JT: CD8(+)-T-cell response tosecreted and nonsecreted antigens delivered by recom-binant Listeria monocytogenes during secondary infection.Infect Immun 2002, 70:153-162.

90. Rayevskaya M, Kushnir N, Frankel FR: Safety and immunogenicityin neonatal mice of a hyperattenuated Listeria vaccinedirected against human immunodeficiency virus. J Virol 2002,76:918-922.

91. Miki K, Nagata T, Tanaka T, Kim YH, Uchijima M, Ohara N, Naka-mura S, Okada M, Koide Y: Induction of protective cellularimmunity against Mycobacterium tuberculosis by recom-binant attenuated self-destructing Listeria monocytogenesstrains harboring eukaryotic expression plasmids for antigen85 complex and MPB/MPT51. Infect Immun 2004, 72:2014-2021.

92. Stevenson A, Roberts M: Use of Bordetella bronchiseptica andBordetella pertussis as live vaccines and vectors for heterol-ogous antigens. FEMS Immunol Med Microbiol 2003, 37:121-128.

93. Shimoji Y, Oishi E, Kitajima T, Muneta Y, Shimizu S, Mori Y: Erysip-elothrix rhusiopathiae YS-1 as a live vaccine vehicle for het-erologous protein expression and intranasal immunizationof pigs. Infect Immun 2002, 70:226-232.

94. Zheng C, Xie P, Chen Y: Recombinant Mycobacterium bovisBCG producing the circumsporozoite protein of Plasmo-dium falciparum FCC-1/HN strain induces strong immuneresponses in BALB/c mice. Parasitol Int 2002, 51:1-7.

95. Vemulapalli R, He Y, Boyle SM, Sriranganathan N, Schurig GG: Bru-cella abortus strain RB51 as a vector for heterologous pro-tein expression and induction of specific Th1 type immuneresponses. Infect Immun 2000, 68:3290-3296.

96. Norton PM, Le Page RW, Wells JM: Progress in the developmentof Lactococcus lactis as a recombinant mucosal vaccinedelivery system. Folia Microbiol (Praha) 1995, 40:225-230.

97. Steidler L, Robinson K, Chamberlain L, Schofield KM, Remaut E, LePage RW, Wells JM: Mucosal delivery of murine interleukin-2(IL-2) and IL-6 by recombinant strains of Lactococcus lactiscoexpressing antigen and cytokine. Infect Immun 1998,66:3183-3189.

98. Lee MH, Roussel Y, Wilks M, Tabaqchali S: Expression of Helico-bacter pylori urease subunit B gene in Lactococcus lactisMG1363 and its use as a vaccine delivery system against H.pylori infection in mice. Vaccine 2001, 19:3927-3935.

99. Pontes DS, Dorella FA, Ribeiro LA, Miyoshi A, Le Loir Y, Gruss A,Oliveira SC, Langella P, Azevedo V: Induction of partial protec-tion in mice after oral administration of Lactococcus lactisproducing Brucella abortus L7/L12 antigen. J Drug Target 2003,11:489-493.

100. Gilbert C, Robinson K, Le Page RW, Wells JM: Heterologousexpression of an immunogenic pneumococcal type 3 capsu-lar polysaccharide in Lactococcus lactis. Infect Immun 2000,68:3251-3260.

101. Perez CA, Eichwald C, Burrone O, Mendoza D: Rotavirus vp7 anti-gen produced by Lactococcus lactis induces neutralizingantibodies in mice. J Appl Microbiol 2005, 99:1158-1164.

102. Adel-Patient K, Ah-Leung S, Creminon C, Nouaille S, Chatel JM,Langella P, Wal JM: Oral administration of recombinant Lacto-coccus lactis expressing bovine beta-lactoglobulin partiallyprevents mice from sensitization. Clin Exp Allergy 2005,35:539-546.

103. Xin KQ, Hoshino Y, Toda Y, Igimi S, Kojima Y, Jounai N, Ohba K,Kushiro A, Kiwaki M, Hamajima K, Klinman D, Okuda K: Immuno-genicity and protective efficacy of orally administeredrecombinant Lactococcus lactis expressing surface-boundHIV Env. Blood 2003, 102:223-228.

104. Zhang ZH, Jiang PH, Li NJ, Shi M, Huang W: Oral vaccination ofmice against rodent malaria with recombinant Lactococcuslactis expressing MSP-1(19). World J Gastroenterol 2005,11:6975-6980.

105. Pei H, Liu J, Cheng Y, Sun C, Wang C, Lu Y, Ding J, Zhou J, Xiang H:Expression of SARS-coronavirus nucleocapsid protein inEscherichia coli and Lactococcus lactis for serodiagnosis andmucosal vaccination. Appl Microbiol Biotechnol 2005, 68:220-227.

106. Cheun HI, Kawamoto K, Hiramatsu M, Tamaoki H, Shirahata T, IgimiS, Makino SI: Protective immunity of SpaA-antigen producingLactococcus lactis against Erysipelothrix rhusiopathiaeinfection. J Appl Microbiol 2004, 96:1347-1353.

107. Grangette C, Muller-Alouf H, Goudercourt D, Geoffroy MC, TurneerM, Mercenier A: Mucosal immune responses and protectionagainst tetanus toxin after intranasal immunization withrecombinant Lactobacillus plantarum. Infect Immun 2001,69:1547-1553.

108. Corthesy B, Boris S, Isler P, Grangette C, Mercenier A: Oral immu-nization of mice with lactic acid bacteria producing Helico-bacter pylori urease B subunit partially protects againstchallenge with Helicobacter felis. J Infect Dis 2005,192:1441-1449.

109. Medaglini D, Pozzi G, King TP, Fischetti VA: Mucosal and systemicimmune responses to a recombinant protein expressed onthe surface of the oral commensal bacterium Streptococcusgordonii after oral colonization. Proc Natl Acad Sci U S A 1995,92:6868-6872.

110. Medaglini D, Ciabattini A, Spinosa MR, Maggi T, Marcotte H, OggioniMR, Pozzi G: Immunization with recombinant Streptococcusgordonii expressing tetanus toxin fragment C confers pro-tection from lethal challenge in mice. Vaccine 2001,19:1931-1939.

111. Zegers ND, Kluter E, van Der SH, van Dura E, van Dalen P, Shaw M,Baillie L: Expression of the protective antigen of Bacillusanthracis by Lactobacillus casei: towards the development ofan oral vaccine against anthrax. J Appl Microbiol 1999,87:309-314.

112. Lee JS, Poo H, Han DP, Hong SP, Kim K, Cho MW, Kim E, Sung MH,Kim CJ: Mucosal immunization with surface-displayed severeacute respiratory syndrome coronavirus spike protein onLactobacillus casei induces neutralizing antibodies in mice. JVirol 2006, 80:4079-4087.

113. Aires KA, Cianciarullo AM, Carneiro SM, Villa LL, Boccardo E, Perez-Martinez G, Perez-Arellano I, Oliveira ML, Ho PL: Production ofhuman papillomavirus type 16 L1 virus-like particles by


32








































































































Publish with BioMed Central and every scientist can read your work free of charge

"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."

Sir Paul Nurse, Cancer Research UK

Your research papers will be:

available free of charge to the entire biomedical community

peer reviewed and published immediately upon acceptance

cited in PubMed and archived on PubMed Central

yours — you keep the copyright

Submit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.asp

BioMedcentral

recombinant Lactobacillus casei cells. Appl Environ Microbiol2006, 72:745-752.

114. Ho PS, Kwang J, Lee YK: Intragastric administration of Lactoba-cillus casei expressing transmissible gastroentritis coronavi-rus spike glycoprotein induced specific antibody production.Vaccine 2005, 23:1335-1342.

115. Oliveira ML, Monedero V, Miyaji EN, Leite LC, Lee HP, Perez-Mar-tinez G: Expression of Streptococcus pneumoniae antigens,PsaA (pneumococcal surface antigen A) and PspA (pneumo-coccal surface protein A) by Lactobacillus casei. FEMS Micro-biol Lett 2003, 227:25-31.

116. Scheppler L, Vogel M, Zuercher AW, Zuercher M, Germond JE, Mie-scher SM, Stadler BM: Recombinant Lactobacillus johnsonii as amucosal vaccine delivery vehicle. Vaccine 2002, 20:2913-2920.


33










http://www.biomedcentral.com/info/publishing_adv.asp


BioMed CentralMicrobial Cell Factories

ss
Open AcceReviewEnsuring safety of DNA vaccinesJacob Glenting*1 and Stephen Wessels2
Address: 1Bioneer A/S, DK-2970 Hørsholm, Denmark and 2Danish Toxicology Centre, DK-2970 Hørsholm, Denmark

Email: Jacob Glenting* - [email protected]; Stephen Wessels - [email protected]


AbstractIn 1990 a new approach for vaccination was invented involving injection of plasmid DNA in vivo,which elicits an immune response to the encoded protein. DNA vaccination can overcome mostdisadvantages of conventional vaccine strategies and has potential for vaccines of the future.However, today 15 years on, a commercial product still has not reached the market. One possibleexplanation could be the technique's failure to induce an efficient immune response in humans, butsafety may also be a fundamental issue. This review focuses on the safety of the genetic elementsof DNA vaccines and on the safety of the microbial host for the production of plasmid DNA. Wealso propose candidates for the vaccine's genetic elements and for its microbial production hostthat can heighten the vaccine's safety and facilitate its entry to the market.

IntroductionVaccination with purified plasmid DNA involves injectionof the plasmid into the patient to elicit an immuneresponse to a protein that is encoded on the plasmid [1].This mini-review focuses upon several aspects of safety ofthe DNA molecule itself and of the microorganism used tomanufacture the DNA. The review is not exhaustive butdoes raise very important safety issues to be kept in mindearly in the development of DNA vaccines.

DNA vaccination was described in a study in 1990 thatdemonstrated the induction of gene expression followingdirect intramuscular injection of plasmid DNA in mice[2]. Since then our understanding of the immunologicalmechanisms behind this unexpected result has increased.This includes identification of immune stimulatory DNAsequences (ISS) that could explain how DNA vaccines canevoke an immune response without an adjuvant [3]. Theadvantages of DNA vaccines over the traditional attenu-ated or subunit vaccines are their capacity to induce abroad spectrum of cellular and humoral immune

responses, their flexible genetic design and low cost ofproduction in a microbial host. Almost two thousandpapers have been published, and several clinical trialshave been conducted testing DNA vaccines against infec-tious diseases such as HIV-1 [4], Ebola virus [5] andmalaria [6], or to generate protective immunity againsttumors [7]. Despite this extensive research, a commercialproduct has yet to come to the market. One reason for thismay be their failure to induce a strong immune responsein higher animals like primates [8]. Another reason fortheir absence from the market may be related to theirsafety. Indeed, international regulatory groups haverecently questioned the safety of certain existing DNA vac-cine constructs and their production systems [9]. Whilethe main focus of research has previously been on theirfunctionality and immunological mechanisms, work onsafety aspects most often is put off until later in develop-ment. By then, making fundamental changes to the DNAvaccine to improve its safety can be extremely costly andtime-consuming.

Published: 06 September 2005

Microbial Cell Factories 2005, 4:26 doi:10.1186/1475-2859-4-26

Received: 25 August 2005Accepted: 06 September 2005

This article is available from: http://www.microbialcellfactories.com/content/4/1/26

© 2005 Glenting and Wessels; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


34


http://www.microbialcellfactories.com/content/4/1/26





In the following we propose some basic choices related tosafety to be made during the development of DNA vac-cines. We highlight safety issues that can be addressed bythe appropriate choice of the vaccine's genetic elements,of its microbial production host and of the conditions ofmanufacture. Special focus will be put on the use of food-grade host-vector systems that are based on our experiencewith the lactic acid bacterium Lactococcus lactis.

The vaccine's genetic elementsThe organization of the genetic elements of a DNA vaccinereflects the plasmid's functionality, its bulk manufactureand its clinical use in the patient. Thus, the plasmid con-tains one unit responsible for its propagation in themicrobial host and another unit that drives the expressionof the vaccine gene in the cells of the patient. The geneticelements of the vaccine are shown in Figure 1, and partic-ular safety concerns are listed in Table 1.

The unit responsible for plasmid propagation in themicrobial host contains a replication region and a selecta-ble marker. The replication region allows the mainte-nance of multiple copies of the plasmid per host cell anda stable inheritance of the plasmid during bacterialgrowth. Furthermore, the replication region also deter-mines the plasmid's host-range. Because DNA vaccinationinvolves injection of milligram quantities of plasmid, rep-lication regions with a narrow host-range can reduce theprobability for spread of the plasmid to the patient's ownflora. A replication region dependent on chromosomallyencoded factors restricts the replication to a single hoststrain. One such bio-containment system has been devel-oped in E. coli based on trans-complementation of a repA-

plasmid replication region by a repA+ host strain [10].Here, the pWV01-derived vectors cannot replicate in theabsence of the replication factor RepA and thus relies on arepA+ helper strain. Addition of another ori (origin of rep-lication) region that is active in mammalian cells allowsprolonged persistence and expression of the vaccine genein the transfected tissue. However, uncontrolled expres-sion of the vaccine gene may induce immunological toler-ance. Furthermore, persistence and increased spread ofthe plasmid may lead to germline transmission as a resultof transfection of sperm cells or oocytes [11]. In fact, PCRstudies have detected vaccine plasmid in the gonads ofvaccinated fetuses and in offspring of these fetuses [12]. Aliterature study has identified non-replicating plasmids asa factor that reduces risk of germline transmission [13].Accordingly, only prokaryotic and narrow host range rep-lication regions should be present on vaccine plasmids.

Selectable markers ensure stable inheritance of plasmidsduring bacterial growth (Fig. 1). Most vaccine plasmidsrely for this on resistance to antibiotics. Although a pow-erful selection, resistance genes to antibiotics are discour-

aged by regulatory authorities [14]. The concern is that theplasmid may transform the patient's microflora andspread the resistance genes (Table 1). Indeed, there ismuch international scientific and regulatory focus on thisissue [15-19]. A non-antibiotic-based marker on vaccineplasmids for use in E. coli has been developed. This systemis based on the displacement of repressor molecules fromthe chromosome to the plasmid, allowing expression ofan essential gene [20]. A selection marker developed inour laboratory uses an auxotrophic marker in L. lactis[21,22]. Here, genes encoded on the plasmid relieve thehost's threonine requirement. This selection system is effi-cient and precludes the use of antibiotics.

The nature of the DNA between the functional genes invaccine plasmids is also a safety concern. Specific DNAsequences or methylation patterns can induce anti-DNAantibodies and lead to the autoimmune disease systemiclupus erythematosus [23]. Gilkeson et al. showed thatamongst various organisms bacterial DNA induced thehighest level of DNA-specific antibodies [24]. Therefore, areasonable strategy is to minimize the non-functionalsequences in the vaccine plasmid (Table 1). Vaccine plas-

Genetic elements of a plasmid DNA vaccineFigure 1Genetic elements of a plasmid DNA vaccine. Plasmid DNA vaccines consists of a unit for propagation in the microbial host and a unit that drives vaccine synthesis in the eukaryotic cells. For plasmid DNA production a replication region and a selection marker are employed. The eukaryotic expression unit comprises an enhancer/promoter region, intron, signal sequence, vaccine gene and a transcriptional terminator (poly A). Immune stimulatory sequences (ISS) add adjuvanticity and may be localized in both units.

Promoter

Intron

Vaccine gene

Poly A

ISS

Replicationregion

Selection marker

Plasmid propagation Vaccine synthesis

Signal sequence


35


mids have been developed which omit the prokaryoticbackbone using an integrase-mediated recombinationtechnology [25]. In addition, these mini-circles showedhigher in vivo gene expression than a standard plasmid.Alternatively, we have used a plasmid backbone derivedentirely from food-grade bacterial DNA [26].

The vaccine expression unit consists of the elements nec-essary for high-level expression and targeting of the vac-cine component (Figure 1). Most DNA vaccines harborpromoters and enhancer regions from pathogenic virusessuch as cytomegalo virus (CMV), simian virus 40, ormurine leukaemia virus. For instance, plasmid vaccineswith the CMV promoter have been in clinical trials and areversatile due to the promoter's activity in a variety of tis-sues and animal models [27]. As more than 50% of thepopulation in USA is infected with CMV and as the virusremains in the body throughout life [28], the use of itsexpression signals on vaccine plasmids may inducerecombination events and form new chimeras of CMV.Promoters and enhancer regions have also been suggestedfrom housekeeping genes encoding the mouse phosphoe-nolpyruvate carboxykinase and phosphoglycerate kinase[29]. However, due to the risk of insertional mutagenesisand oncogenesis, highly inter-species-conservedsequences like these should be avoided. This risk can bereduced by the use of novel synthetic promoters selectedby bioinformatic tools to have a low homology tosequences potentially present in the recipient. To augmentthe promoter activity, introns are introduced, which havea beneficial effect on the in vivo expression of the vaccinegene [30]. Most often the intron A from CMV is used.Here, too, bioinformatics can aid in the design of syn-

thetic introns thereby avoiding sequences already presentin CMV-infected individuals.

For secretion of the vaccine peptide to the extra-cellularmilieu, a signal sequence is positioned in front of the vac-cine gene. This codes for a signal peptide of about 20–40amino acids, often derived from bovine proteins such asthe plasminogen activator [31]. However, the fusion ofbovine peptides to an immunogen may induce an immu-nological cross-reaction. Signal peptides can themselvesinduce protective immunity against a microbial pathogenwhen administered as a gene vaccine [32]. Apparently, toavoid undesired immune responses, the nature of the sig-nal peptide should be considered (Table 1). Statisticalmethods like the hidden Markov model have been used topredict and generate artificial signal peptide sequences foruse in human cells [33]. Such a strategy could be appliedto DNA vaccine development to create more appropriatesignal peptides.

To enhance the potency of a DNA vaccine, ISS's are addedto the plasmid (Figure 1). These are nucleotide hexamersthat interact with Toll-like receptors and add adjuvanticity[34]. The function of the ISS is independent of its locationon the plasmid and may be present in the prokaryoticbackbone. In fact, Klinman eliminated ISS from the plas-mid backbone and could partially restore the immuno-genicity of the plasmid by exogenously added ISS DNA[35]. Therefore, changing the vector backbone or editingplasmid components may influence the immuneresponse due to deletion of the ISS. This, too, emphasizesthe importance of the proper selection of expression vec-tor early in vaccine development.

Table 1: The safety concerns and possible solutions for plasmid DNA vaccines and their production hosts. A priori each safety concern should be addressed as early in development as possible.

Safety concern Possible solution

Genetic elements Transfer of plasmid to host flora Narrow host-range replication regionNon-antibiotic plasmid marker

Germline integration Avoidance of mammalian replication regionInsertional mutagenesis and oncogenesis Artificial DNA for promoter, intron, and signal sequence

Avoidance of human-homologous DNAAdverse effects of encoded peptide(s) Artificial signal sequences

Avoidance of mammalian replication regionEvaluation of vaccine peptide case-by-case

Induction of autoimmune reactions Minimized plasmids

Production host Endotoxins and biogenic amines Use of gram-positive organismTransferable antibiotic resistance genes Determination of minimal inhibitory concentrations (MIC's)

Screening for transferabilityGenetic instability Analysis of plasmid population by sequencing and mass

spectrometryPathogenicity Use of food-grade organism


36


The microbial host and production of bulk purified plasmidThe characteristics of the microbial host affect the qualityof the purified DNA [36]. A number of safety concernshave been advanced concerning the microbial host. Asexplained in the following, these include production oftoxins and biogenic amines, transferable antibiotic resist-ances, and genetic instability, including prophage-induced promiscuity and rearrangement of plasmid DNA(Table 1).

For reasons of efficiency, E. coli is usually chosen today asthe production host, with its concomitant benefits anddrawbacks. The benefits include a high DNA yield andwell-established procedures for down-stream processingof the plasmid. However, as a gram-negative bacterium, E.coli contains highly immunogenic endotoxin, or lipopol-ysaccharides (LPS), in its outer membrane. Because of thenet negative charge of both LPS and DNA, these moleculesmay be co-purified by the ion exchange principle used inthe purification of plasmid DNA, although commercialkits do exist that can exclude LPS. On the other hand, theuse of gram-positive hosts, none of which produce LPS,eliminate this dependency on the absolute efficiency ofLPS-removing kits. Although not as efficient for plasmidproduction, L. lactis, as a gram-positive, produces neitherendotoxin nor biogenic amines [37]. Assay for transfera-ble antibiotic resistances in lactic acid bacteria is today aroutine procedure; common L. lactis research strains arealso genetically robust; and their prophages are of narrowhost-range [38,39].

For large-scale plasmid production, often in about a thou-sand liters, the fermentation medium must sustain a high-level production of biomass and of plasmid DNA. At thesame time the medium should be chemically defined andwithout components of animal origin that may containviruses or prions [40]. Growth in a synthetic medium formany organisms results in low biomass and low plasmidyield. Indeed, switching microbial host to increase yield iscomplicated as it may lead to unexpected immunologicalresults because of different DNA methylation patterns.Consequently, the production strain should be evaluatedin synthetic media at an early point in development. Alsohere, L. lactis may be the host of choice due to its efficiencyof growth in chemically defined media [41,42]. Finally,the genetic integrity of bulk purified plasmid molecules istoday primarily monitored by sequence analysis. How-ever, to reveal minor populations of molecules such asmultimers or molecules with deletions and insertions,mass spectrometry should be considered [43].

ConclusionPlasmid DNA vaccines could be the next generation ofvaccines. As yet, research has focused on building func-tional DNA vaccines. Therefore, focus on safety has been

limited. In this review we have mentioned some safetyissues to be addressed early in vaccine development.Using bioinformatic tools, safe eukaryotic expression sig-nals can be devised in synthetic DNA sequences. Safetymay also be heightened by non-antibiotic plasmid selec-tion markers, plasmid replication functions with narrowhost-ranges, and minimized plasmids. Using a bio-con-tainment strategy will also increase the safety of the micro-bial production host, as will avoidance of toxic substanceslike endotoxins. Synthetic growth media should be con-sidered early in development and will influence choice ofproduction host. Indeed, it can be easier to address severalof these safety concerns early in vaccine development bybasing the strategy on food-grade bacteria and their DNA,such as L. lactis and its DNA. Finally, the very availabilityof safe host-vector systems will most probably facilitatethe overall acceptance of DNA vaccines.

Authors' contributionsThe author(s) contributed equally to this work.

AcknowledgementsWe thank Søren M. Madsen for critical reading of the manuscript. This work was partially financed by the Danish Ministry of Science, Technology and Innovation.

References1. Azevedo V, Levitus G, Miyoshi A, Candido AL, Goes AM, Oliveira SC:

Main features of DNA-based immunization vectors. Braz JMed Biol Res 1999, 32:147-153.

2. Wolff JA, Malone RW, Williams P, Chong W, Acsadi G, Jani A, et al.:Direct gene transfer into mouse muscle in vivo. Science 1990,247:1465-1468.

3. Sato Y, Roman M, Tighe H, Lee D, Corr M, Nguyen MD, et al.: Immu-nostimulatory DNA sequences necessary for effective intra-dermal gene immunization. Science 1996, 273:352-354.

4. Mwau M, Cebere I, Sutton J, Chikoti P, Winstone N, Wee EG, et al.:A human immunodeficiency virus 1 (HIV-1) clade A vaccinein clinical trials: stimulation of HIV-specific T-cell responsesby DNA and recombinant modified vaccinia virus Ankara(MVA) vaccines in humans. J Gen Virol 2004, 85:911-919.

5. Vastag B: Ebola vaccines tested in humans, monkeys. JAMA2004, 291:549-550.

6. Wang R, Doolan DL, Le TP, Hedstrom RC, Coonan KM, CharoenvitY, et al.: Induction of antigen-specific cytotoxic T lymphocytesin humans by a malaria DNA vaccine. Science 1998,282:476-480.

7. Liu M, Acres B, Balloul JM, Bizouarne N, Paul S, Slos P, et al.: Gene-based vaccines and immunotherapeutics. Proc Natl Acad SciUSA 2004, 101(Suppl 2):14567-14571.

8. Babiuk LA, Pontarollo R, Babiuk S, Loehr B, van Drunen Littel-van denHurk: Induction of immune responses by DNA vaccines inlarge animals. Vaccine 2003, 21:649-658.

9. European Medicines Agency: Report from the ad hoc meeting of CPMPGene Therapy Expert Group 2003. 23–24 January 2003. EMEA-5382-03-Final

10. Law J, Buist G, Haandrikman A, Kok J, Venema G, Leenhouts K: Asystem to generate chromosomal mutations in Lactococcuslactis which allows fast analysis of targeted genes. J Bacteriol1995, 177:7011-7018.


12. Gallot D, Seifer I, Lemery D, Bignon YJ: Systemic diffusion includ-ing germ cells after plasmidic in utero gene transfer in therat. Fetal Diagn Ther 2002, 17:157-162.


37


































BioMedcentral


14. European Medicines Agency: Note for Guidance on the Quality,Preclinical and Clinical Aspects of Gene Transfer MedicinalProducts. 2001. CPMP/BWP/3088/99

15. European Commission: Communication from the Commissionon a community strategy against antimicrobial resistance.2001. COM(2001) 333 final

16. European Food Safety Authority: Working Document on theupdating of the criteria used in the assessment of bacteriafor resistance to antibiotics of human or veterinaryimportance. 2004.

17. European Medicines Agency: Note for Guidance on the Quality,Preclinical and Clinical Aspects of Gene Transfer MedicinalProducts. 2001. CPMP/BWP/3088/99

18. Martinez JL, Baquero F: Interactions among strategies associ-ated with bacterial infection: pathogenicity, epidemicity, andantibiotic resistance. Clin Microbiol Rev 2002, 15:647-679.

19. US Food and Drug Administration: Guidance for Industry #52.Assessment of the effects of antimicrobial drug residuesfrom food of animal origin on the human intestinal flora.2004.

20. Cranenburgh RM, Hanak JA, Williams SG, Sherratt DJ: Escherichiacoli strains that allow antibiotic-free plasmid selection andmaintenance by repressor titration. Nucleic Acids Res 2001,29:E26.

21. Madsen SM, Albrechtsen B, Hansen EB, Israelsen H: Cloning andtranscriptional analysis of two threonine biosynthetic genesfrom Lactococcus lactis MG1614. J Bacteriol 1996,178:3689-3694.


23. Pisetsky DS: Immune response to DNA in systemic lupuserythematosus. Isr Med Assoc J 2001, 3:850-853.

24. Gilkeson GS, Grudier JP, Pisetsky DS: The antibody response ofnormal mice to immunization with single-stranded DNA ofvarious species origin. Clin Immunol Immunopathol 1989,51:362-371.

25. Darquet AM, Cameron B, Wils P, Scherman D, Crouzet J: A newDNA vehicle for nonviral gene delivery: supercoiledminicircle. Gene Ther 1997, 4:1341-1349.


27. Wright A, Semyonov A, Dawes G, Crameri A, Lyons R, Stemmer WP,et al.: Diverse plasmid DNA vectors by directed molecularevolution of cytomegalovirus promoters. Hum Gene Ther 2005,16:881-892.

28. National Center for Infectious Diseases [http://www.cdc.gov/ncidod/diseases/cmv.htm]

29. Cheng L, Ziegelhoffer PR, Yang NS: In vivo promoter activity andtransgene expression in mammalian somatic tissues evalu-ated by using particle bombardment. Proc Natl Acad Sci USA1993, 90:4455-4459.

30. Brinster RL, Allen JM, Behringer RR, Gelinas RE, Palmiter RD:Introns increase transcriptional efficiency in transgenicmice. Proc Natl Acad Sci USA 1988, 85:836-840.

31. Ashok MS, Rangarajan PN: Protective efficacy of a plasmid DNAencoding Japanese encephalitis virus envelope protein fusedto tissue plasminogen activator signal sequences: studies in amurine intracerebral virus challenge model. Vaccine 2002,20:1563-1570.

32. Jiang C, Magee DM, Ivey FD, Cox RA: Role of signal sequence invaccine-induced protection against experimentalcoccidioidomycosis. Infect Immun 2002, 70:3539-3545.

33. Barash S, Wang W, Shi Y: Human secretory signal peptidedescription by hidden Markov model and generation of astrong artificial signal peptide for secreted proteinexpression. Biochem Biophys Res Commun 2002, 294:835-842.

34. Tudor D, Dubuquoy C, Gaboriau V, Lefevre F, Charley B, Riffault S:TLR9 pathway is involved in adjuvant effects of plasmidDNA-based vaccines. Vaccine 2005, 23:1258-1264.

35. Klinman DM: CpG DNA as a vaccine adjuvant. Expert RevVaccines 2003, 2:305-315.

36. ICH Topic Q5D: Quality of biotechnological products: derivation and char-acterisation of cell substrates use for production of biotechnological/biologi-cal products. Geneva: International Conference on Harmonisation ofTechnical Requirements for Registration of Pharmaceuticals for Human Use(ICH) 1997.

37. Straub BW, Kicherer M, Schilcher SM, Hammes WP: The formationof biogenic amines by fermentation organisms. Z Lebensm Unt-ers Forsch 1995, 201:79-82.

38. Rakonjac J, O'Toole PW, Lubbers M: Isolation of lactococcal pro-late phage-phage recombinants by an enrichment strategyreveals two novel host range determinants. J Bacteriol 2005,187:3110-3121.

39. European Food Safety Authority: Working Document on theupdating of the criteria used in the assessment of bacteriafor resistance to antibiotics of human or veterinaryimportance. 2004.

40. ICH Topic Q5D: Quality of biotechnological products: derivation and char-acterisation of cell substrates use for production of biotechnological /biolog-ical products. Geneva: International Conference on Harmonisation ofTechnical Requirements for Registration of Pharmaceuticals for Human Use(ICH) 1997.

41. Mierau I, Leij P, van SI, Blommestein B, Floris E, Mond J, et al.: Indus-trial-scale production and purification of a heterologous pro-tein in Lactococcus lactis using the nisin-controlled geneexpression system NICE: the case of lysostaphin. Microb CellFact 2005, 4:15.

42. Mierau I, Olieman K, Mond J, Smid EJ: Optimization of the Lacto-coccus lactis nisin-controlled gene expression system NICEfor industrial applications. Microb Cell Fact 2005, 4:16.

43. Marvin LF, Roberts MA, Fay LB: Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in clinicalchemistry. Clin Chim Acta 2003, 337:11-21.


38
























http://www.cdc.gov/ncidod/diseases/cmv.htm

http://www.cdc.gov/ncidod/diseases/cmv.htm





































Chapter 2

Immunological analysis of a Lactococcus lactis based DNA vaccine

expressing HIV gp120

39

BioMed CentralGenetic Vaccines and Therapy

ss
Open AcceResearchImmunological analysis of a Lactococcus lactis-based DNA vaccine expressing HIV gp120Gregers J Gram1, Anders Fomsgaard1, Mette Thorn1, Søren M Madsen2 and Jacob Glenting*2
Address: 1Department of Virology, State Serum Institute, Artillerivej 5, DK-2300 Copenhagen, Denmark and 2Vaccine Technology, Bioneer A/S, Kogle Alle 2, DK-2970Hørsholm, Denmark

Email: Gregers J Gram - [email protected]; Anders Fomsgaard - [email protected]; Mette Thorn - [email protected]; Søren M Madsen - [email protected]; Jacob Glenting* - [email protected]


AbstractFor reasons of efficiency Escherichia coli is used today as the microbial factory for production ofplasmid DNA vaccines. To avoid hazardous antibiotic resistance genes and endotoxins fromplasmid systems used nowadays, we have developed a system based on the food-grade Lactococcuslactis and a plasmid without antibiotic resistance genes. We compared the L. lactis system to atraditional one in E. coli using identical vaccine constructs encoding the gp120 of HIV-1.Transfection studies showed comparable gp120 expression levels using both vector systems.Intramuscular immunization of mice with L. lactis vectors developed comparable gp120 antibodytiters as mice receiving E. coli vectors. In contrast, the induction of the cytolytic response was lowerusing the L. lactis vector. Inclusion of CpG motifs in the plasmids increased T-cell activation morewhen the E. coli rather than the L. lactis vector was used. This could be due to the different DNAcontent of the vector backbones. Interestingly, stimulation of splenocytes showed higher adjuvanteffect of the L. lactis plasmid. The study suggests the developed L. lactis plasmid system as newalternative DNA vaccine system with improved safety features. The different immune inducingproperties using similar gene expression units, but different vector backbones and production hostsgive information of the adjuvant role of the silent plasmid backbone. The results also show thatcorrelation between the in vitro adjuvanticity of plasmid DNA and its capacity to induce cellular andhumoral immune responses in mice is not straight forward.

BackgroundGenetic immunization or DNA vaccination has initiated anew era of vaccine research. The technology involves theinoculation of plasmid DNA into a living host to elicit animmune response to a protein encoded on the plasmid[1]. The potential advantages of DNA vaccines include theinduction of cellular and humoral immune responses,flexible genetic design, lack of infection risk, stability of

reagents, and the relatively low cost of production in amicrobial host. These advantages are being exploited forinfections like HIV where traditional vaccines have provedunsuccessful [2-5]. The advantages of DNA vaccines havelead to extensive research primarily focused on theimmune responses induced against a variety of antigensbut less on the tools required for the microbial productionof plasmids. Plasmid DNA used for DNA vaccinations

Published: 29 January 2007

Genetic Vaccines and Therapy 2007, 5:3 doi:10.1186/1479-0556-5-3

Received: 21 October 2006Accepted: 29 January 2007

This article is available from: http://www.gvt-journal.com/content/5/1/3

© 2007 Gram et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


40

http://www.gvt-journal.com/content/5/1/3




Genetic Vaccines and Therapy 2007, 5:3 http://www.gvt-journal.com/content/5/1/3

have two general features reflecting its dual functionality:the unit responsible for propagation in microbial cellscalled the plasmid backbone, and the unit that expressesthe vaccine gene in the transfected tissue. For reasons ofefficiency, Escherichia coli, with its concomitant benefitsand drawbacks, has been the production host of choice.The benefits include a high DNA yield and well-estab-lished procedures for down-stream processing of the plas-mid [6]. However, as a gram-negative bacterium, E. colicontains highly immunogenic endotoxins, or lipopolysac-charides (LPS), in its outer membrane. Because of the netnegative charge of both LPS and DNA, these moleculesmay be co-purified by the ion exchange principle used inthe purification of plasmid DNA [7], although commer-cial kits do exist that can exclude LPS. On the other hand,the use of gram-positive hosts, none of which produceLPS, eliminate this dependency on the absolute efficiencyof LPS-removing kits. In addition, the E. coli vaccine plas-mids are typically maintained during growth by plasmid-encoded antibiotic resistance and addition of antibioticsto the growth medium. Thus, antibiotics may be contam-inants in the purified DNA vaccines with the potential ofinducing allergic responses in disposed individuals [8].Furthermore, there is much scientific and regulatory focuson the use of antibiotic resistance genes. The concern isthat the plasmid may transform the patient's microfloraand spread resistance genes [9]. We have developed a hostvector system for the production of plasmid DNA basedon Lactococcus lactis with improved safety properties con-cerning the microbial production host and the geneticcontent of the plasmid [10]. As host L. lactis is attractivefor production of vaccines due to its food grade status. Thedeveloped plasmid is free of antibiotic resistance genesand is based entirely on L. lactis genes; the minimal thetatype plasmid replicon [11] and the hom-thrB operon [12]encoding the homoserine dehydrogenase and homoser-ine kinase catalysing two steps in the biosynthesis of thre-onine. The use of this auxotrophic marker is based oncomplementation of a L. lactis host strain containing aninternal deletion in the hom-thrB operon on the chromo-some. Thus plasmid complementation relieves the strainsrequirement for threonine. Other food grade geneticmarkers have been developed in L. lactis to meet the highsafety demands of optimised and recombinant starter cul-tures for use in the manufacture of dairy products. Anexample is the use of suppressible pyrimidine auxotrophswhere suppressor tRNA is encoded by the plasmid andallows read through of the pyrF gene containing an ambermutation [13]. A selection marker based on lactose utili-zation has also been developed [14]. Both food-gradegenetic markers allow selection in milk containing bothlactose and low amount of pyrimidines. However, appli-cation of non-antibiotic-based genetic markers in L. lactisfor plasmid DNA production is new. We have investigatedthe use of the new L. lactis-based plasmid as a potential

HIV-1 DNA vaccine by including a eukaryote expressionunit encoding the gp120 surface molecule from the pri-mary CCR-5-trophic HIV-1BX08 [5,15]. Although nakedDNA vaccines may be very helpful in priming thehumoral and cellular immune system for subsequentboosting with more immunogenic agents like recom-binant virus or proteins [3,15], the relatively low potencyof the DNA vaccines themselves is problematic. The lackof immune stimulatory endotoxins or immune stimula-tory sequences (ISS) like CpG motifs may influence thepotency of the DNA vaccine preparation. Indeed, DNAitself acts as adjuvant [16-19], which depends on the con-tent of ISS CpG motifs and the dose given. In this respectit may be of importance that L. lactis genes are relativelyAT rich compared to the more GC rich E. coli genes andthus may contain fewer putative ISS.

The goal of this study was to develop an alternative hostvector system based on L. lactis and compare it to a tradi-tional one using same vaccine cassette. The developed sys-tem shows potential as a new plasmid production system,but gives also information about the adjuvant role of thesilent plasmid backbone in DNA vaccination.

MethodsConstruction of E. coli and L. lactis based expression vectorsThe E. coli expression plasmid WRG7079 (PowderJectInc., Madison, WI, USA) contains the kanamycin resist-ance gene, a ColE1-based origin of replication and aeukaryotic expression unit harboring the cytomegalo virus(CMV) promoter and enhancer region, Intron A, the tissueplasminogen activator (tPA) secretion signal, a polylinker,and the polyadenylation signal from bovine growth hor-mone (BGHpA). The synthetic gp120 gene with humancodons from the primary, CCR5-tropic, clade B HIV-1BX08has been cloned into WRG7079 [5] and is named pEC120(6.2 kb). A variant of pEC120 containing ISS CpG motifswas constructed. An 80 bp CpG cassette(ATCGACTCTCGAGCGTTCTATCGACTCTCGAGCGTTCTCACTCTCGAGCGTTCTCGCTAGACGTTAGCGT-TCAACGTTGA) containing 12 CpGs (bold) includinghuman and mouse ISS motifs [20-23], was introducedbetween the stop codon of the gp120 gene and theBGHpA. This plasmid was named pEC120CpG. Theeukaryotic expression units from pEC120 andpEC120CpG were cloned into the L. lactis pJAG5 cloningvector [10]. PJAG5 contains the hom-thrB auxotrophmarker, the minimal theta type replicon, and a polylinkerregion for cloning (Fig. 1). The threonine auxotrophic L.lactis strain, MG1614ΔhomthrB, was used for cloning andplasmid production [12]. Selection of primary transform-ants and growth under selective conditions was done indefined medium lacking threonine [10]. The HIV-1BX08gp120 expression unit was obtained from pEC120 using


41


PCR and primers homologous to the 5' end of the CMVpromoter (5'-CGGGGTACCCTTGTGCAATGTAACATCA-GAG-3') and to the 3' end of the BGHpA (5'-CGGGGTAC-CCTGTGGATAACCGTATTACCG-3'). Similarly, weisolated the gp120 expression cassette including the 80 bpCpG motif using PCR, the same primers and pEC120CpGas template DNA. The primers introduced terminal KpnIrestriction sequences (underlined). The gp120 expressionunits were ligated to KpnI-treated pJAG5 and electropo-rated into competent L. lactis MG1614ΔhomthrB. PCRand DNA sequencing confirmed the correct structure ofthe chimeric pJAG5 vectors. The resulting 7.8 kb vaccineplasmids were named pLL120 and pLL120CpG.

Plasmid preparation for DNA vaccine productionThe E. coli based plasmids were produced and purifiedusing Qiagen Midiprep according to the manufacturer(Qiagen, Hilden, DE). The L. lactis-based plasmids werepurified similarly but with modifications as follows. A100 ml (OD600 = 3) culture was harvested and washed in20 ml TE buffer containing 7% sucrose (STE). Cells wereresuspended in 4 ml STE containing 20 mg/ml lysozyme(Sigma) and 0.1 mg/ml RNase H (Sigma) and incubatedat 37°C for 60 min. Four ml NaOH-SDS buffer (#P2 from

Qiagen) was added and incubated for 5 min at room tem-perature. Four ml of ice cold potassium acetate buffer(#P3 from Qiagen) was added and mixed before incu-bated 30 min on ice and centrifuged at 20,000 g at 4°C for1 h. The clear supernatant was applied to midi cartridgesand thereafter handled as suggested (Qiagen). The qualityof the PBS dissolved DNA was analysed by agarose gelelectrophoresis and evaluation of the A260/A280 ratio.

In vitro expression of gp120BX08 proteinTo examine eukaryotic expression the syn.gp120BX08 gene,the human HEK293 kidney fibroblast cell line (ATCC,Rockville, MD) was transfected with pLL120,pLL120CpG, pEC120, and pEC120CpG, respectively,using Effectene transfection kit (Qiagen). Radioimmuno-precipitation (RIPA) of 35S-met and 35S-cys labelled pro-tein using gp120 specific antibodies was done asdescribed [5].

DNA vaccination6–7 weeks-old BALB/c mice were purchased from Bom-holdtgaard (Taconic, Ry, DK) and kept in groups of fiveper cage with food and water ad libitum and artificial light12 h per day. The acclimatization period before immuni-

The genetic anatomy of vaccine plasmidsFigure 1The genetic anatomy of vaccine plasmids. Vaccine plasmids from L. lactis (pLL120) and E. coli (pEC120) with CpG motifs. The plasmids contain a similar expression cassette which carries the CMV promoter and the gp120 gene fused to the signal sequence tPA, and a transcription terminator (BGHpA). Intron A is included for efficient splicing. The pLL120 vector backbone contains the minimal theta type plasmid replicon encoding the repB protein and the hom-thrB genetic marker encoding two threonine biosynthetic enzymes homoserine dehydrogenase and homoserine kinase. The pEC120 vector backbone contains the colE1 origin of replication and the kanamycin resistance (KanR) genetic marker.

CMV

IntronA

gp120

BghpA

repB

hom thrB

tPA

+/- ISS

pLL120

pLL120CpG

colE1

KanR

colE1

KanR

CMV

IntronA

gp120

BghpA

tPA

+/- ISS

pEC120

pEC120CpG


42


zation was 5 days. Mice were immunized intramuscularly(i.m.) in Tibialis anterior with 50 μl of 2 mg/ml plasmid inendotoxin-free PBS buffer using pLL120, pLL120CpG,pEC120, and pEC120CpG, respectively. The four vaccineconstructs were tested on four groups of 10 mice each (n= 10). A group of 5 mice received PBS buffer alone andserved as a negative control. Mice were vaccinated at week0, 9 and 15. Blood samples were drawn one day prior tothe first immunization as a pre-immune control serum(day 0), and thereafter every third week. The experimentwas terminated at week 19.

Serological assaysMouse anti HIV-1 gp120 antibodies were measured byindirect ELISA. Wells of polystyrene plates (Maxisorb,Nunc, Roskilde, DK) were coated for 2 days at room tem-perature with HIV-1 IIIB recombinant gp120 (Intracel) at0.2 μg/100 μl of carbonate buffer, pH 9.6. Before use theplates were blocked for 1 h at room temperature with 150μl/well of washing buffer (PBS, 0.5 M NaCl, 1% Triton-X-100) plus 2% BSA and 2% skim milk powder. After 3 × 1min washings, mouse plasma was added at 100 μl/welldiluted in blocking buffer and the ELISA plates were incu-bated for 90 min at room temperature using a microtiterplate shaker. The standard curve was made from a pool ofplasma consisting of BX08 gp120 positive mice sera [5].Plates were again washed 5 × 1 min and incubated 1 h atroom temperature with 100 μl/well of HRP-conjugatedrabbit anti-mouse IgG (DAKO-Cytomation, Glostrup,DK) diluted 1:1000 in blocking buffer. Colour was devel-oped with 100 μl/well of peroxidase enzyme substrateconsisting of 4 mg of o-phenylenediamine in 11 ml waterplus 4 μl hydrogen peroxide (30%, w/w). The reactionwas terminated after 30 min by 150 μl/well of 1 M H2SO4.The optical density (OD) of wells was measured at 492nm using a microplate photometer. Anti-HIV-gp120 IgGtiters were expressed as the reciprocal of the plasma dilu-tion resulting in an OD492 nm value of 0.500.

IgG1 and IgG2a were analysed by capture gp120 sandwichELISA [24,25]. Briefly, Maxisorb 96-well plates (NUNC)were pre-coated overnight with PBS containing 1 μg/mlsheep-anti-gp120 (Aalto, Dublin, IR), washed 5 times inPBS containing 0.05% Tween-20 and incubated for 2 h atroom temperature with 100 μl 1% Triton X-100 treatedcell-free culture supernatant from 5.25.EGFP.Luc.M7 cells(donated by Ned Landau, Salk Institute, CA, USA) whichwere infected chronically with HIV-1BX08. A dilution seriesof plasma samples from each mouse was incubated for 2h at room temperature. Wells were washed five times inPBS-Tween-20 and incubated with HRP-conjugated Ratanti-mouse IgG1 or -IgG2a antibodies (PharmingenBrøndby, DK) both diluted 1:1000 in dilution buffer (0.5M NaCl, 3 mM KCl, 15 mM KH2PO4, 6 mMNa2HPO4:2H2O, 2% Triton-X 100, 1% w/v BSA, 0.1% w/

v phenol-red) and incubated for 1 h at room temperature.Wells were washed five times as above and once using dis-tilled water before colorimetric development using 100 μlo-phenylenediamine substrate solution (Kem-En-Tec,Taastrup, DK). The reaction was stopped with 150 μl 1 MH2SO4 and absorbance measured at 490 nm.

Intracellular IFN-γ cytokine analysis by flow-cytometry (IC-FACS)Mouse spleens were removed aseptically at week 19 andgently homogenized to single cell suspension, washed 3times in cold RPMI1640 (Invitrogen, Taastrup, DK) sup-plemented with 10% fetal calf serum (FCS) (Invitrogen)and resuspended to a final concentration of 5 × 107 cell/ml. Spleens from individual mice were homogenised andtransferred to a round-bottom 96-well plate (2 × 105 cellsper well). Wells, containing a total volume of 200 μlRPMI1640 with 10% FCS, were incubated for 6 h at 37°Cwith 50 U/ml of IL-2 (Roche, Hvidovre, DK), monensin(3 μM) (Sigma, Brøndby, DK) with and without a 15-merpeptide derived from the V3-loop of HIV-1BX08 containinga conserved murine H-2Dd restricted CTL epitope (IGP-GRAFYTT). Wells were washed and cells re-suspended in50 μl medium containing fluorescent antibodies for therelevant surface markers, CD4-FITC, CD8-PECy7, andCD44-APC (Pharmingen) and incubated for 20 min inthe dark at 4°C. Then 100 μl monensin (3 μM) in PBS wasadded before centrifugation at 400 × g for 5 min. Cellswere resuspended in 100 μl monensin (3 μM) in PBS and100 μl 2% paraformaldehyde (Merck, Damstadt, DE) andincubated for 30 min in the dark at 4°C. Cells were pel-leted and resuspended in 100 μl PBS solution containing3 μM monensin and 1% BSA. Cells were pelleted andresuspended in 200 μl PBS containing 0.5 % Saponin(Sigma) and incubated in the dark for 10 min at roomtemperature. Cells were pelleted and incubated for 20 minat 4 °C with 50 μl 0.5 % saponin solution containing PEconjugated anti-IFN-γ antibodies (Pharmingen) to stainfor intracellular IFN-γ. Cells were washed twice with 100μl 0.5 % saponin/PBS solution before addition of a fixingsolution containing 2% paraformaldehyde. Fixed cellswere analysed on a BD LSR-II flowcytometer (BectonDickenson) using the BD FACS Diva software.

Cytotoxic T lymphocyte assayFor assay of cytotoxic T lymphocytes (CTL) the spleno-cytes were incubated 5 days with mitomycin-C treated (50μg/ml for 1 h) and CTL epitope peptide loaded mouseP815 (H-2Dd) stimulator cells at a ratio of 10:1 inmedium supplemented with 5 × 10-5 M β-mercaptoetha-nol. For assay of cytolytic CTL response to HIV-1BX08,P815 target cells were pulsed with 20 μg/ml of the HIV-1BX08 V3 peptide SIHIGPGRAFYTTGD containing the H-2Dd restricted CTL epitopeCorbet, 2. After stimulationsplenocytes were washed three times with RPMI1640 sup-


43


plemented with 10% FCS and resuspended to a final con-centration of 5 × 106 cells/ml. Cell suspension (100 μl)was added in triplicate to U-bottom 96-well microtiterplates and a standard 4 h 51Cr-release assay performed asdescribed [4].

In vitro splenocyt activation by plasmid DNASplenocytes were prepared from spleens of non-immu-nized 6–12 weeks old female BALB/c mice. Cells were cul-tured in 96-well U-bottom plates at 37°C and 5% CO2and maintained in RPMI1640 supplemented with 10%FCS, 1% penicillin/streptomycin, and 0.1% mercaptoeth-anol. Splenocytes (3 × 105 cells/well) were treated withRPMI1640 medium, plasmid (3 ug/well) or plasmid DNAtreated with DNase I (Biolab, Risskov, DK) using 1 UDNase per 1 ug DNA for 1 hr at 37°C followed by inacti-vation at 75°C for 10 min. The synthesis of IL-6 and IFN-γ in supernatants of cells cultured for 0, 1, 2, and 3 dayswas measured by sandwich ELISA kits (R&D Systems,Minneapolis, MN).

ResultsVectors and immune stimulatory CpG motifsThe characteristics of the L. lactis (pLL120 andpLL120CpG) and E. coli (pEC120 and pEC120CpG)based vectors are shown in Fig. 1. Sequence analysisshowed that the L. lactis vector backbone had a lower GCcontent (37%) than the E. coli vector (48%) and con-tained two copies of a known immune stimulatory mouseCpG motifs (GACGTT), while the E. coli vector backboneonly contained one CpG motif. Even though the syntheticgp120 gene is very GC rich (56%) because of the human-ized codons, it did not contain classical ISS motifs, but didharbour three human CpG motifs (two copies of GTCGTTand one copy of CTCGAG). The added CpG minigenecontained 12 CpG dinucleotides giving rise to six consen-sus ISS CpG motifs (5'-two purines-CpG-two pyrimi-dines-3') including two copies of the mouse ISS CpGmotifs (GACGTT and AACGTT) [22]. Thus, when insertedthe CpG minigene increased the numbers of mouse ISSmotifs by two. However, unidentified mouse ISS orimmune inhibitory CpG motifs may be present in theplasmids.

Eukaryotic gp120BX08 expression using a L. lactis based vectorTo evaluate the influence of the plasmid backbone on theexpression levels of the gp120 gene, human kidney cellswere transiently transfected. The expression of thesyn.gp120BX08 gene using the L. lactis or E. coli-based vec-tors was examined by RIPA of 35S-labelled supernatantproteins from transfected HEK293 cells and purified IgGfrom a pool of HIV sero-positive patients. 35S-labelledsupernatant from untransfected cells served as a negativecontrol and was also precipitated but did not produce a

protein with a molecular weight of 120 kDa (Fig. 2). Cellstransfected with the any of the four vaccine plasmids pro-duced a 120 kDa product (Fig. 2). The expression levels ofgp120 were comparable for pLL120 and pEC120 as wellas for pLL120CpG and pEC120CpG. For both types ofvectors the amounts of gp120 expressed seemed some-what lower when they harboured the CpG cassette. The L.lactis based vectors drive therefore antigen synthesis asefficiently as the E. coli based vectors. Furthermore, theproduced gp120 was immunogenic as antibodies fromHIV-1 sero positive patients recognize the in vitro pro-duced HIV protein (Fig. 2).

Antibody response after genetic immunizationWe examined the humoral responses in the DNA vacci-nated mice using ELISA (Fig. 3). The four different plas-mids (Fig. 1) were each administered to four groups of 10mice, respectively. Mice immunized i.m. with the pLL120and pLL120CpG at weeks 0, 9, and 15 induced specificIgG antibodies to the gp120 protein with the same kinet-ics and maximum titers as the mice receiving the E. colibased plasmids. The first immunization induced high spe-cific antibodies in all mice. The titers obtained were simi-lar to those obtained in previous mouse DNA vaccinationexperiments using the syn.gp120BX08 in WRG7079 [5].The second immunization at week nine boosted the anti-body titers about one hundred times in all groups. Furtherboosting at week 15 had less effect. The titers obtained ateach time point were indistinguishable from thoseobtained with the corresponding vectors containing theputative ISS CpG cassette. The negative control groupreceiving buffer alone showed no anti-gp120 antibody tit-ers above the background (data not shown). Thus, thekinetics and the magnitude of the anti-gp120 antibodyinduction was similar using the L. lactis-based pJAG5 andthe E. coli-based plasmid backbones.

The antibody subclass profile of the induced response wasanalysed. Here gp120 specific IgG1 and IgG2a were deter-mined by ELISA to be used as a surrogate marker for theTh1/Th2 balance (Fig. 4). Mice immunized with the L. lac-tis based plasmids induced an IgG1/IgG2a ratio indistin-guishable from that induced by the E. coli based pEC120and pEC120CpG. Furthermore, the IgG1/IgG2a ratiosobtained from the data from similar vectors with andwithout the CpG cassette were statistically indistinguisha-ble (two sample t-test, P > 0.05). All IgG1 and IgG2a ratioswere above 1 and below 1.5, which is indicative of amixed Th1:Th2 response with a slight Th2 predominance(Fig. 4).

CD8+ T lymphocyte immune responses after genetic immunizationWe examined the cellular immune response in mice aftergenetic immunizations with syn.gp120BX08 by cytolytic


44


CTL assay and staining for intracellular induction of IFN-γ (IC-FACS) (Figs. 5 and 6). A strong cellular immuneresponse was induced by the four vectors that were signif-icantly higher than both the background (unloaded targetcells) and the flat response obtained from unimmunizedmice (data not shown). This confirms the in vivo expres-sion and immunogenicity of gp120BX08 expressed fromboth L. lactis and E. coli vectors. After five days of epitopestimulation a somewhat higher cytolytic CTL responsewas obtained with the E. coli-based plasmid as comparedto the L. lactis-based plasmid (Fig. 5). Addition of the ISSCpG cassette increased this specific CTL response for bothvectors albeit more pronounced for the E. coli based vector(Fig. 5).

The cellular response was also analysed by IC-FACS stain-ing for direct epitope-peptide-induced IFN-γ productionin CD8+ T-cells (Fig. 6). The analysis confirmed specificcellular immunity induced by the vaccine gene in the fourvectors. This IFN-γ induction was, however, lower for theL. lactis based vectors than for the corresponding E. coli-based vectors.

In vitro stimulation of splenocytes by plasmid DNATo evaluate the immune stimulatory properties of the vac-cine plasmids splenocytes were co-incubated with plas-mid DNA and the induction of pro-inflammatory

cytokines were analysed. Interestingly, the L. lactis-basedvectors induced a higher synthesis of both IFN-γ and IL-6than the corresponding E. coli vectors (Fig. 7). Further-more, addition of the CpG minigene to the L. lactis vectorcaused a significantly higher induction of both IFN-γ andIL-6 compared to the pLL120 without the CpG minigene.To analyse if this stimulatory effect is caused by the plas-mid DNA or by contaminants in the DNA preparation theDNA was enzymatically degraded. DNase treatment of theplasmid solutions abolished the pro-inflammatory effect(Fig. 7). Therefore, the immune stimulatory induction bythe L. lactis vaccine vectors is directly linked to the plasmidDNA and not to contaminants that may be present in theDNA preparations.

DiscussionPlasmid DNA vaccines are the next generation of vaccines.As yet, research has mostly been focused on building func-tional vaccines by increasing the antigenicity of theencoded gene or by new plasmid delivery methods. There-fore, focus on the plasmid backbone and on the microbialproduction host has been limited. The key driver of thisstudy was to develop an alternative plasmid backbone inDNA vaccines and a new microbial production host andcompare it to a traditional used E. coli based DNA vaccine.This study represents to our knowledge the first alternativeto E. coli based plasmid DNA vaccines and provides new

Expression analysis of gp120Figure 2Expression analysis of gp120. In vitro expression of gp120 protein from L. lactis (pLL120) and E. coli (pEC120) plasmids with and without CpG motifs. RIPA of culture supernatants of 35S-cys/met-pulsed HEK-293 cells transfected with the vaccine plas-mids. Precipitation is done using pooled serum from HIV positive patients and the immunoprecipitates were analysed by SDS-PAGE. Positive control plasmid (Pos), untransfected cells (Neg), molecular marker (Marker). The gp120 is indicated.

gp120

220 kDa

97 kDa

66 kDa

pLL120pE

C120

pLL120CpG

pEC120C

pG

Neg

Pos

Pos

Marker

gp120

220 kDa

97 kDa

66 kDa

pLL120pE

C120

pLL120CpG

pEC120C

pG

Neg

Pos

Pos

Marker


45


information on the role of the plasmid backbone in DNAvaccines.

We examined the use of a novel L. lactis vector as the plas-mid backbone in DNA vaccines. The main advantages ofthis vector and its production strain are avoidance of anti-biotic resistance genes and antibiotic contaminants, lack

of LPS and endotoxins, and improved safety as the L. lactissystem may be regarded as food grade. In the presence ofthe thr-carrying plasmid backbone, the auxotrophic hoststrain is able to grow in a chemical defined medium in theabsence of threonine. The system is therefore compatiblewith use in a growth medium free of animal componentsthat may contain viruses and prions. Specifically, we con-structed L. lactis-based expression vectors with and with-out an ISS CpG unit containing the HIV-1BX08 gp120vaccine gene with humanized codons and developed asuitable process for purification of plasmid DNA from L.lactis. Furthermore, we demonstrated eukaryotic expres-sion in vitro, and DNA immunizations of mice and com-pared the resulting humoral and cellular immuneresponses with those obtained from induction with tradi-tional E. coli-based vectors [5].

Plasmid DNA purification from Gram-positive bacteria iscomplicated due to the presence of a thick peptidoglycancell wall. Purification of L. lactis plasmid DNA has reliedon modified alkaline procedures [26] including the use oftoxic substances like phenol, chloroform and ethidiumbromide. The method presented here convenientlyemploys the commercial ion exchange cartridges with afew modifications including an initial treatment using acell wall degrading enzyme. The developed plasmid prep-aration contained only low amounts of genomic DNAand RNA and was proven useful in routine analysis ofplasmids, transformation, in vitro transfection of human

Antibody subtypes following immunizationFigure 4Antibody subtypes following immunization. Ratio IgG1/IgG2a anti-gp120 antibodies from samples of week 18. Antibody subtypes of individual mice was determined and a mean of each group (n = 10) is shown with error-bars repre-senting 2 × SEM.

0,0

0,5

1,0

1,5

2,0

Rat

i oIg

G1/ Ig

G2a

pLL120 pLL120CpG pEC120 pEC120CpG0,0

0,5

1,0

1,5

2,0

Rat

i oIg

G1/ Ig

G2a

pLL120 pLL120CpG pEC120 pEC120CpG

Antibody response after DNA vaccinationFigure 3Antibody response after DNA vaccination. Mouse IgG anti-gp120 antibody titers at week 0, 3, 6, 9, 15, and 18. DNA immunizations were at week 0, 9, and 15. The antibody titer of individual mice was determined and presented as the mean of the group of 10 mice (n = 10). Error-bars represent +/- two times Standard Error of Mean (2 × SEM = SD/√n). White bars:pLL120CpG, cross-sectioned bars: pLL120, grey bars: pEC120, dotted bars: pEC120CpG.

week 0 week 3 week 6 week 9 week 15 week 19

103

104

105

106

107

108

IgG

anti- g

p12

0tite

rs(A

.U.)

week 0 week 3 week 6 week 9 week 15 week 19

IgG

anti- g

p12

0tite

rs(A

.U.)

102


46


293 cells, and in vivo DNA immunization. Although L. lac-tis is attractive in terms of safety the efficiency of the sys-tem is a drawback. Due to the medium copy number ofthe pJAG5 plasmid the yield during a fermentation proc-ess is 5 fold lower than that of high copy number plasmidsfrom E. coli (data not shown). However, higher copynumber plasmids were also tested in L. lactis but showedlower segregational stability or a less favourable distribu-tion between supercoiled and non-supercoiled plasmidforms (data not shown).

The transfection of human HEK293 cells demonstratedHIV-1BX08 gp120 expression using the L. lactis pJAG5 back-

bone as delivery vehicle. The difference in nucleotide con-tent and/or difference in methylation pattern in L. lactisDNA could potentially influence the expression and/orthe immune response. However, the intensity in RIPA ofgp120 produced from cells transfected with pLL120 andpLL120CpG was similar to that of cells transfected with E.coli vectors. This illustrates that the backbone vector didnot affect the level of expression, which may not be sur-prising since both vectors contain the same gp120 expres-sion unit. The reasons for a somewhat lower in vitroexpression from both vectors when the ISS CpG unit isincluded are not known but could be partly due to DNAmethylation and gene silencing [27,28].

Cellular immune response after DNA immunizationFigure 5Cellular immune response after DNA immunization. Peptide-specific CTL responses were determined by 51Cr-release assay at titrated levels of effector to target cell (E:T) ratios. Splenocytes or effector cells were pre-stimulated for five days in vitro with V3-loop peptide. Assay is done on individual mice and presented as a mean of each group +/- 2 × SEM, n = 10. Signif-

icant difference between pEC120CpG and pLL120CpG is indicated with asterisks ( ) (two sample t-test, P < 0.05).

0

10

20

30

40

50

60

100:1 50:1 25:1 13:1 6:1

E:T ratio

% s

pecific

lysis

pEC120CpG

pLL120

pEC120

0

10

20

30

40

50

60

100:1 50:1 25:1 13:1 6:1

E:T ratio

% s

pecific

lysis

pEC120CpG

pLL120CpG

pEC120

**

*

*

*

pLL120


47


The usefulness of pJAG5 to carry the eukaryotic gp120expression unit and induce an antibody response to theencoded gp120 was evaluated by naked DNA immuniza-tion. Mice serum showed IgG antibodies specific forgp120 and the induction followed the same kinetics andpotency as that obtained using the corresponding E. coliplasmids. Moreover the IgG1/IgG2a ratios obtained withthe L. lactis and the E. coli-based vectors were not statisti-cally different in this mouse model. Thus, the inducedTh1/Th2 balance determined by the IgG1/IgG2a surrogatemarker was similar for the L. lactis and E. coli based-vector,respectively. Interestingly, both antibody titers and IgG1/IgG2a ratios were largely unaffected by the introducedCpG minigene supposed to act as a Th1 adjuvant via theToll-like receptor 9 [29-33]. It cannot be ruled out that theuse of a different and less Th2-prone mouse strain orhigher doses of plasmid DNA could give minor differ-ences. Some investigators have also observed minimaleffect on the humoral immune response induced by DNAvaccine plasmids with different numbers of CpG motifspresent [34,35]. Thus, as opposed to the establishedimmune stimulatory effect of synthetic and single-stranded oligodeoxynucleotides, the effect of plasmidDNA is more controversial.

In contrast to the identical humoral response induced bythe four plasmids, CD8+ T-cells were activated less by theL. lactis-based vectors than by the E. coli-based vectors asmeasured by both IC-FACS and CTL activity. In previous

Non-specific immune stimulatory effect of vaccine plasmidsFigure 7Non-specific immune stimulatory effect of vaccine plasmids. Splenocyte supernatants were collected before stimula-tion (day 0) and three days post stimulation with the plasmids (day 3). The secretion of IL-6 and IFN-γ was analysed by ELISA. Cross-sectioned bars: pLL120CpG, white bars: pLL120, grey bars: pEC120, black bars: pEC120CpG, and bars with vertical lines: culture medium. Symbols: +: DNase treatment of plasmid preparation. Error bars represent SD of two independent experiments.

0

100

200

300

400

500

600

700

800

900

Day 0 Day 3

pg I

L-6/m

l

0

500

1000

1500

2000

2500

3000

3500

4000

Day 0 Day 3

+ + + + + + + + + + + ++ + + +0

-

+ + + + + + + + + ++ + + +

pg I

FN

g/m

l

Peptide-specific IFN-γ inductionFigure 6Peptide-specific IFN-γ induction. CD8+ splenocytes of immunised mice was determined by flow-cytometry of sam-ples taken at week 18. Error-bars represent +/- 2 × SEM.

Asteriks ( ) indicate significant differences (two sample t-test, P < 0.05) between pEC120 and pLL120 and between pEC120CpG and pLL120CpG.

0,0

1,0

2,0

3,0

4,0


% I

FN

G+

CD

8+

cells

0,0

1,0

2,0

3,0

4,0


% I

FN

G+

CD

8+

cells

0,0

1,0

2,0

3,0

4,0


% I

FN

G+

CD

8+

cells

**


48


studies of synthetic HIV-1 envelope genes from strainsBX08 and MN no correlation was found between the CTLresponse and the expression levels, secreted or mem-brane-bound antigens, mode of DNA delivery, T helpercell response, or antibody response [5,36,37]. This maynot be surprising since the mechanisms and constraintsinvolved are quite different. The similar humoral butlower cellular immune response from the L. lactis-basedvector could be due to a lower adjuvant effect on CD8+ Tcells exerted by differences in ISS CpG motifs since theintroduction of the CpG unit showed a tendency toincrease the CTL activities within each group of mice (Fig.5). Various CpG ISS's, their numbers, location, or theirspacing or the sequences flanking them may stimulate theimmunological active cells and their cytokine productionsdifferently, and in this way differentially influence theresulting immune responses [35]. Both the presence ofunknown CTL immune-neutralising sequences in theplasmid backbone [17,38] or lack of contaminating LPScan cause the lower CTL induction observed using the L.lactis based DNA vaccine. To explain this we evaluated theimmune activating capacity of the plasmids on spleno-cytes. Surprisingly, the L. lactis plasmids showed higher invitro immune stimulatory properties than the tested E. coliplasmids. To test if this pro-inflammatory effect was dueto contaminating components like LPS from E. coli or tei-choic acid from L. lactis, the plasmid was enzymaticallydegraded. The results indicate that the inflammatory stim-ulation of the splenocytes was due to the plasmid DNAitself. As such the in vitro stimulation assay cannot explainthe lower cytotoxic T-cell response induced by the two L.lactis vectors. Furthermore, the higher adjuvant effect ofthe L. lactis vectors in vitro does not correlate with the DNAimmunizations in vivo where the two systems induce sim-ilar antibody responses. This emphasizes the importanceof animal studies. Because of the complex nature of theadjuvant effect of plasmid DNA, we measured the result-ing effect of in vivo DNA vaccination using large amountsof DNA as routinely used in i.m. immunization withoutany pre-treatment or added adjuvant to allow for maximalimmune interference of the plasmids. The observed differ-ence in the cellular but not humoral immune inductionby the L. lactis and E. coli-based vectors may lead to newways to adjuvant DNA vaccines. Here, the L. lactis vectorcan be a tool to analyse DNA sequences that can be addedfor maximum induction of cellular immune responses.The development of an improved L. lactis based plasmidthat can induce a stronger cellular immune response ishowever essential for this to be a true alternative to E. colibased DNA vaccines.

ConclusionBy this study we suggest an alternative host vector systemfor use as plasmid DNA vaccines. The L. lactis system isbased on an endotoxin-free and safe organism and the

plasmid backbone is free of antibiotic resistance genes.The L. lactis based DNA vaccine vector expressed HIV-1BX08 gp120 vaccine component similarly to E. coli vectorsand induced similar humoral immune responses as corre-sponding E. coli based plasmids. However, compared to E.coli plasmids the L. lactis based DNA vaccine systeminduced a lower cellular immune. This is not due to lackof immune stimulatory effect as the L. lactis based plasmidshowed high pro inflammatory effect on splenocytes.Although the developed system is less potent compared tothe tested E. coli system it may be a valuable tool to iden-tify new CTL adjuvant components.

Competing interestsThe author(s) declare that they have no competing inter-ests.

Authors' contributionsGJG carried out the immunological studies together withJG. MT made the in vitro splenocyte assay testing theimmune stimulatory role of the plasmids. AFO supervisedthe immunological studies, the design of the vaccine plas-mids and helped to draft the manuscript. SMM gave help-ful input to the genetic design of the plasmids. JG madethe DNA vaccine constructions, vaccine production anddrafted the manuscript.

AcknowledgementsWe acknowledge Irene Jensen, Ulla Poulsen, Pernille Smith, and Anne Cath-rine Steenbjerg for excellent technical assistance. We thank Bjarne Albre-chtsen for critical reading of the manuscript. This work was financed by the Danish Ministry of Science and Innovation.

References1. Gurunathan S, Klinman DM, Seder RA: DNA vaccines: immunol-

ogy, application, and optimization. Ann Rev Immunol 2000,18:927-974.

2. Fomsgaard A: HIV-1 DNA vaccines. Immunol Lett 1999,65:127-131.

3. Hanke T, Samuel RV, Blanchard TJ, Neumann VC, Allen TM, BoysonJE, Sharpe SA, Cook N, Smith GL, Watkins DI, Cranage MP,McMichael AJ: Effective induction of simian immunodeficiencyvirus-specific cytotoxic T lymphocytes in macaques by usinga multiepitope gene and DNA prime-modified vaccinia virusAnkara boost vaccination regimen. J Virol 1999, 73:7524-32.

4. Robinson HL: New hope for an AIDS vaccine. Nat Rev Immunol2002, 2:239-250.

5. Corbet S, Vinner L, Hougaard DM, Bryder K, Nielsen HV, Nielsen C,Fomsgaard A: Construction, biological activity, and immuno-genicity of synthetic envelope DNA vaccines based on a pri-mary, CCR5-tropic, early HIV type 1 isolate (BX08) withhuman codons. AIDS Res Hum Retroviruses 2000, 16:1997-2008.

6. Ferreira GN, Monteiro GA, Prazeres DM, Cabral JM: Downstreamprocessing of plasmid DNA for gene therapy and DNA vac-cine applications. Trends Biotechnol 2000, 18:380-388.

7. Petsch D, Anspach FB: Endotoxin removal from protein solu-tions. J Biotechnol 2000, 76:97-119.

8. Azevedo V, Levitus G, Miyoshi A, Candido AL, Goes AM, Oliveira SC:Main features of DNA-based immunization vectors. Braz JMed Biol Res 1999, 32:147-153.

9. Davison J: Genetic exchange between bacteria in the environ-ment. Plasmid 1999, 42:73-91.

10. Glenting J, Madsen SM, Vrang A, Fomsgaard A, Israelsen H: A plas-mid selection system in Lactococcus lactis and its use for


49




























BioMedcentral

gene expression in L. lactis and human kidney fibroblasts.Appl and Environ Microbiol 2002, 68:5051-5056.

11. Pedersen ML, Arnved KR, Johansen E: Genetic analysis of theminimal replicon of the Lactococcus lactis subsp. lactis bio-var diacetylactis citrate plasmid. Mol Gen Genet 1994,244:374-82.

12. Madsen SM, Albrechtsen B, Hansen EB, Israelsen H: Cloning andtranscriptional analysis of two threonine biosynthetic genesfrom Lactococcus lactis MG1614. J Bacteriol 1996,178:3689-3694.

13. Sørensen KI, Larsen R, Kibenich A, Junge MP, Johansen E: A food-grade cloning system for industrial strains of Lactococcuslactis. Appl Environ Microbiol 2000, 66:1253-1258.

14. Platteeuw C, Alen-Boerrigter I, van Schalkwijk, de Vos WM: Food-grade cloning and expression system for Lactococcus lactis.Appl Environ Microbiol 1996, 62:1008-1013.

15. Vinner L, Wee EG, Patel S, Corbet S, Gao GP, Nielsen C, Wilson JM,Ertl HC, Hanke T, Fomsgaard A: Immunogenicity in Mamu-A*01rhesus macaques of a CCR5-tropic human immunodefi-ciency virus type 1 envelope from the primary isolate (Bx08)after synthetic DNA prime and recombinant adenovirus 5boost. J Gen Virol 2003, 84:203-13.

16. Davis HL: CpG motifs for optimization of DNA vaccines. DevBiol 2000, 104:165-169.

17. Sato Y, Roman M, Tighe H, Lee D, Corr M, Nguyen MD, SilvermanGJ, Lotz M, Carson DA, Raz E: Immunostimulatory DNAsequences necessary for effective intradermal gene immuni-zation. Science 1996, 273:352-354.

18. Pisetsky DS, Reich C, Crowley SD, Halpern MD: Immunologicalproperties of bacterial DNA. Ann N Y Acad Sci 1995,772:152-163.

19. Yamamoto S, Yamamoto T, Shimada S, Kuramoto E, Yano O, KataokaT, Tokunaga T: DNA from bacteria, but not from vertebrates,induces interferons, activates natural killer cells and inhibitstumor growth. Microbiol Immunol 1992, 36:983-997.

20. Krieg AM, Yi AK, Matson S, Waldschmidt TJ, Bishop GA, Teasdale R,Koretzky GA, Klinman DM: CpG motifs in bacterial DNA trig-ger direct B-cell activation. Nature 1995, 74:546-549.

21. Klinman DM, Yi AK, Beaucage SL, Conover J, Krieg AM: CpG motifspresent in bacteria DNA rapidly induce lymphocytes tosecrete interleukin 6, interleukin 12, and interferon gamma.Proc Natl Acad Sci USA 1996, 93:2879-83.

22. Sonehara K, Saito H, Kuramoto E, Yamamoto S, Yamamoto T, Toku-naga T: Hexamer palindromic oligonucleotides with 5'-CG-3'motif(s) induce production of interferon. J Interferon CytokineRes 1996, 16:799-803.

23. Rodriguez EG, Vazquez DM, Herrera AM, Duarte CA: Enhancedcell-mediated IFN-gamma-secreting activity against theHIV-1IIIB V3 peptide of the TAB9 multiepitope after DNAvaccine backbone engineering. Biochem Biophys Res Commun2003, 308:713-8.

24. Gram GJ, Bolmstedt A, Schønning K, Biller M, Hansen JES, OlofssonS: Detection of orientation-specific anti-gp120 antibodies bya new N-glycanase protection assay. APMIS 2002, 110:123-31.

25. Moore JP, Jarrett RF: Sensitive ELISA for the gp120 and gp160surface glycoproteins of HIV-1. AIDS Res Hum Retroviruses 1988,4:369-379.

26. O'Sullivan DJ, Klaenhammer TR: Rapid mini-prep isolation ofhigh-quality plasmid DNA from Lactococcus and Lactobacil-lus spp. Appl Environ Microbiol 1993, 59:2730-2733.

27. Emerman M, Temin HM: Genes with promoters in retrovirusvectors can be independently suppressed by an epigeneticmechanism. Cell 1984, 39:449-67.

28. Gram GJ, Nielsen SD, Hansen JE: Spontaneous silencing ofhumanized green fluorescent protein (hGFP) gene expres-sion from a retroviral vector by DNA methylation. J Hema-tother 1998, 7:333-41.

29. Brazolot MCL, Weeratna R, Krieg AM, Siegrist CA, Davis HL: CpGDNA can induce strong Th1 humoral and cell-mediatedimmune responses against hepatitis B surface antigen inyoung mice. Proc Natl Acad Sci USA 1998, 95:15553-15558.

30. Kadowaki N, Ho S, Antonenko S, de Waal MR, Kastelein RA, BazanF, Liu YJ: Subsets of human dendritic cell precursors expressdifferent toll-like receptors and respond to different micro-bial antigens. J Exp Med 2001, 194:863-870.

31. Klinman DM, Yamshchikov G, Ishigatsubo Y: Contribution of CpGmotifs to the immunogenicity of DNA vaccines. J Immunol1997, 158:3635-3639.

32. Klinman DM, Barnhart KM, Conover J: CpG motifs as immuneadjuvants. Vaccine 1999, 17:19-25.

33. Krieg AM, Wu T, Weeratna R, Efler SM, Love-Homan L, Yang L, YiAK, Short D, Davis HL: Sequence motifs in adenoviral DNAblock immune activation by stimulatory CpG motifs. ProcNatl Acad Sci USA 1998, 95:12631-12636.

34. Biswas S, Ashok MS, Srinivasan VA, Ragarajan PN: Evaluation of theprotective efficacy of rabies DNA vaccine in mice using anintracerebral challenge model. Curr Sci 1999, 76:1012-1016.

35. Zelenay S, Elias F, Fló J: Immunostimulatory effects of plasmidDNA and synthetic oligonucleotides. Eur J Immunol 2003,33:1382-1392.

36. Vinner L, Nielsen HV, Bryder K, Corbet S, Nielsen C, Fomsgaard A:Gene gun DNA vaccination with Rev-independent syntheticHIV-1 gp160 envelope gene using mammalian codons. Vac-cine 1999, 17:2166-75.

37. Shiver JW, Davies ME, Perry HC, Freed DC, Liu MA: Humoral andcellular immunity elicited by HIV-1 vaccination. J Pharm Sci1996, 85:1317-1324.

38. Krieg AM: CpG motifs in bacterial DNA and their immuneeffects. Annu Rev Immunol 2002, 20:709-760.


50































































Chapter 3

Cell surface display of Bet v 1 on immunomodulatory lactobacilli: potential oral delivery vehicle for treatment of birch pollen allergy

51

Cell surface display of Bet v 1 on immunomodulatory lactobacilli: potential oral delivery vehicle for treatment of birch pollen allergy

Jacob Glenting1§, Simon Skjøde Jensen1, Jens Brimnes2, Mercedes Ferreras2, Søren

Michael Madsen1, Peter Ravn1

1 Bioneer A/S 2 ALK-Abello §Corresponding author

Email addresses:

JG: [email protected]

SSJ: [email protected]

JB: [email protected]

MF: [email protected]

SMM: [email protected]

PRA: [email protected]

52

mailto:[email protected]






Abstract

Background

Lactobacilli are attractive organisms for use as delivery vehicles of mucosal vaccines.

Some strains of lactobacilli can modulate the response of immune competent cells like

dendritic cells lining the gastrointestinal tract. The benefits of co-delivery of adjuvant

in proximity to the vaccine component make bacterial surface display an obvious

strategy for antigen presentation. However, surface display of proteins can lower the

access to vaccine epitopes by steric hindrance and change the surface architecture of

lactobacilli and thereby affect their immune modulating activity. Here, we tested

different lactobacil cell wall anchoring motifs and compared their capacity to display

two passenger proteins: an enzyme and an allergen. We also tested if these non-native

surface structures changed the immunomodulatory effect of lactobacilli.

Results

Using a plasmid based expression system different anchor signals were compared.

The surface display system was based on secretion using a sec-dependent signal

peptide and an LPXTG anchoring signal for the sortase-catalyzed linkage of the

recombinant protein to the cell surface. Different anchor signals from L. plantarum

WCFS1 was selected based on their consensus sequence. To avoid that the protein is

embedded in the cell wall and not exposed properly different length of spacers

between the LPXTG motif and the passenger protein were compared. Display of a

nuclease enzyme showed that only an anchor with a 450 aa long spacer region

supported immobilization of an active nuclease. This protein anchor was used for

surface display of the major birch pollen ellergen Bet v 1. The lactobacil produced

and chimeric Bet v 1 was as immunereactive as the control Bet v 1 with native aa

composition as tested by inhibition assay and sera from birch pollen allergic patients.

The immunomodulatory effect of two Lactobacillus strains was evaluated. The

strains showed divergent polarizing effect during co-incubation with murine bone

marrow derived dendritic cells. Surfaced display of Bet v 1 did not alter the

IL10/IL12p70 cytokine response from the dendritic cells. However addition of a

chemical cross linking agent showed that disruption of the molecular patterns on the

bacterial surface alters the immunomodulatory effect of lactobacilli.

53

Conclusions

Protein anchors with a long spacer can support the display of active enzymes and

immunological active allergen when expressed on the surface of lactobacilli.

Although surface molecules are key factors in the activation of dendritic cells, cell

wall display of Bet v1 using a LPXTG sorting signal and a protein spacer did not alter

the adjuvant properties of two strains of lactobacilli.

Background The use of food related lactic acid bacteria (LAB) as mucosal vaccine vectors is

becoming more important as reports on successful delivery of therapeutics, allergens

and antigens using recombinant LABs are increasing. LAB are attractive as live

mucosal vaccines because they: (i) are considered a GRAS-organism (generally

regarded as safe), (ii), are able to survive through the intestinal tract [Tuohy et al.,

2007], (iii), show mucosal adhesive phenotype [Tallon et al., 2007], and (iv) posses

immune modulating capacity [Christensen et al., 2002, Maassen et al., 2000, Lan et

al., 2005, Mohamadzadeh et al., 2005]. Studies have evaluated the use of

recombinant LABs as carriers of antigens [Hannify et al., 2007, Oliveira et al., 2006,

Ramasamy et al., 2006, Bermudez-Humaran et al., 2005, Pusch et al., 2005, Perez et

al., 2005, Cortes-Perez et al., 2005, Grangette et al., 2001], allergens [Hazebrouck et

al., 2007, Daniel et al., 2006, Adel-Patient et al., 2005], cytokines [Wu et al., 2006,

Bermudez-Humaran et al., 2003, Steidler et al., 2000], and antibody fragments

[Marcotte et al., 2006, Krüger et al., 2002]. In some delivery systems, the expressed

protein is secreted from the bacterium; in others it is accumulated intracellularly to

protect the passenger protein. However, the advantage of co-presenting adjuvant and

antigen in close proximity makes bacterial surface display particular interesting for

mucosal delivery of vaccines

Lactobacilli are a natural constituent of the human gut flora [Reuter et al., 2001] and

ascribed several beneficial effects including modulation of mucosal immune

responses [Perdigon et al., 2002, Vitini et al., 2000]. Protein components that supports

survival and colonisation of the GI-tract has been described [Bron et al., 2004,

Boekhorst et al., 2006]. Amongst these are adhesins that upon contact can bind the

intestinal mucosal surfaces [Buck et al., 2005, Boekhorst et al., 2006 Ramiah et al.,

54

2007]. The capacity of lactobacilli to survive and adapt to the gut-milieu, adhere to

mucosal surfaces, and stimulate the immune system, makes them suitable as vaccine

delivery vehicles at mucosal surfaces. Protein display on the surface of lactobacilli

involves secretion to the extra cellular milieu and covalent attachment of the

passenger protein to the bacterial surface [reviewed by Hansson et al., 2001].

Typically a C-terminal sorting signal and a spacer peptide to the passenger protein are

used. The sorting signal includes the LPXTG cleavage domain, followed by multiple

hydrophobic amino acids, and a tail of six or seven positively charged residues at the

extreme C-terminus. Upon secretion through the membrane the hydrophobic and

charged residues retains the protein in the membrane and the LPXTG site is exposed

for the surface located enzyme, sortase, which catalyzes cleavage and covalent

immobilization of the protein to the cell surface. Such protein targeting system using

the Lactococcus lactis based Usp45 signal peptide and Streptococcus pyogenes M6

cell wall anchor motif was shown to efficiently display a reporter protein on the

surface of lactobacilli [Dieye et al., 2001]. In Lactobacillus casei the poly-gamma-

glutamic acid synthetase complex A from Bacillus subtilis was used as an anchoring

motif for surface display of the spike protein of Corona virus [Lee et al., 2006] and

the E7 protein from human papilloma virus [Poo et al., 2006].

Despite the successful development of surface display in lactobacilli, there may be

issues relating to the quality, authenticity, and accessibility of the immobilized

passenger protein. Indeed, surface-anchored enzymes like ß-galactosidase and ß -

lactamase show reduced catalytic activity compared to their free forms [Van der Vaart

et al., 1997, Strauss et al., 1996]. An important strategy of designing surface display

systems is therefore to ensure both immobilization and accessibility of the

immunogen. In this study we compared variants of two native protein anchors for

surface display on lactobacilli and evaluated their efficiency in relation to surface

anchoring, display, and effect on the adjuvant properties of the host.

55

Figure 1 - The genetic components of the surface display systemThe expression cassette of the plasmid used for surface display of nuclease. The nucB is in translational fusion with the secretion signal sequence (ss) and the anchor signal. The sortase cleavage site LPXTG is boxed and depicted in the C-terminal of the anchor. The horizontal axis shows length of the anchor measured in amino acids. Note that pUP540 is without anchor and results in complete secretion of NucB to the culture medium.

Results and Discussion

Cell surface display of recombinant proteins on lactobacilli

Gene expression tools for use in microorganisms usually vary in efficiency. Surface

anchoring of therapeutic components to lactobacilli should ensure stable

immobilization of an active protein component. We therefore tested variants of two

different anchor signals and compared them to a similar secreted protein without

anchor. The Staphylococcus aureus nuclease (NucB) was used as an export specific

reporter for secretion and cell wall anchoring [Poquet et al., 1998] and combined with

a constitutive promoter, isolated from a glycolytic housekeeping gene of L. plantarum

299v, and a signal sequence from an extracellular factor from same strain.. Anchor

sequences were identified in the published genome sequence of L. plantarum WCFS1

where 25 open reading frames with LPXTG motives have been identified

[Kleerebezem et al., 2003]. Anchor signals were isolated from lp_0197 and lp_2486

based on their high consensus LPXTG motive. Furthermore, lp_0197 displays

interesting repeat structures in the spacer region, and lp_2486 have a high CAI

(Codon Adaptation Index) indicating that lp_2486 are highly translated, and hence it

could be speculated that the anchor signal is efficient too. To ensure efficient surface

display and accessibility two variants of both anchors were cloned with 54 and 199 aa

(lp_0197) or 49 and 459 aa spacer regions (lp_2486), respectively. In total four

variants of NucB anchoring were tested (Figure 1). Analysis of colonies of L.

plantarum 299v transformed with the nucB-plasmids showed translocation of

nuclease using all five plasmid constructs and a chromogenic overlay assay. The

nuclease activities in cell cultures were influenced on type of anchor and length of

56

Figure 2 - Quantitive measurement of NucB surface display using different protein anchors Amount of nuclease in cell cultures (white bars) and attached to the surface of bacteria (Cross sectioned bars). Data are mean of three independent experiments +/- SEM.

spacer (Figure 2). Highest expression was obtained using pSM2044 and pSM2045 (~

9 mg/L), which is comparable to that of UP540 cultures secreting anchorless NucB.

This is in accordance with the fact that similar promoters and signal sequences are

used. However, following a thorough wash in PBS buffer only the anchor from

pSM2045 with the long spacer showed nuclease activity on the surface (Figure 2).

The lack of nuclease activities using the other anchors suggests inefficient surface

display of nuclease or that the enzyme activity is inhibited when nuclease is located in

proximity to the cell surface. Expression and NucB targeting using pSM2045 was

evaluated by detecting the presence of nuclease in different culture fractions.

Surprisingly, the enzyme assays showed low amounts of nuclease in the supernatant

of PSM2045, whereas nuclease was absent in the supernatant from the strain

harbouring the fully secreted variant of NucB (Figure 3).

Figure 3 - Evaluation of supernatant, wash and cellular fractions using the pPSM2045 surface display system Enzyme activity assays on L. plantarum 299v expressing secreted and surface anchored nucB. Samples: SN: supernatant, CUL: culture, CW: Cell wall anchored, W1-W3: presence of nucB in consecutive wash fractions, Sum: SN+CW+W1+W2+W3. Data are mean of three independent experiments +/- SEM.

57

However, the anchorless nuclease was found in the cell fraction as a free form and

was easily released from the cell by washing in PBS with neutral pH. This could be

due to non-covalent ionic or hydrophobic interactions between the nuclease and the

bacterial cell surface. Indeed, the pH in an outgrown culture of L. plantarum is below

the pI of NucB (pI=5) resulting in a net positive charge of the protein with affinity for

the negative charge of lipoteichoic acid on the bacterial cell surface. In contrast, after

PBS wash showed PSM2045 cell surface anchored nuclease representing 35% of the

total nuclease activity in the culture. Although the anchor signals were selected based

on consensus LPXTG motifs and supported efficient production of NucB their

capacities to stably immobilize a passenger protein was not optimal. Even the most

promising candidate (pSM2045) resulted in liberation of nuclease from the cell

surface. Whether this phenomenon is instability, protease activity, or simply

illustrates a turn over of surface proteins is unknown. The better anchor signal from

this comparison was pSM2045, which resulted in 2 mg/L immobilized nuclease

equivalent to approximately 5×104 molecules per cell.

Figure 4 – IgE inhibition assayBinding of patients pooled serum IgE to biotinylated rBet v 1 inhibited by control rBet v 1 (■), Bet v 1 expressed by L. rhamnosus UP616 (∆) and L. acidophilus X37 (□). The inhibition level is represented as a function of the concentration of the competitor protein.

Immunological activity of surface displayed Bet v 1

Model proteins like nuclease are quantative indicators of protein targeting in

lactobacilli. However, surface display of vaccine components require preserved

immunereactivity. Surface display relies on the secretion and targeting machinery of

58

the host. The expression of non-native genes in lactobacilli can therefore affect the

protein authenticity and hence the immunereactivity. Furthermore, LPXTG-mediated

anchorage fixates the C-terminal of the passenger protein, which may disrupt the

native folding and limit exposure of important vaccine epitopes. We therefore

analysed the immunological properties of Bet v 1 produced as a chimera with the

protein anchor of pSM2045 in two species: Lactobacillus acidophilus X37 and

Lactobacillus rhamnosus UP616. Western blot of total cell protein lysate showed

expression of Bet v 1 as a fusion protein (data not shown). The capacity of the

chimeric Bet v 1 to bind IgE from sera of birch pollen allergic patients was analysed

(Figure 4). Bet v 1 from both strains of Lactobacillus can fully block the IgE binding

of a standard rBet v 1.2801, which displays similar IgE epitopes as native Bet v 1.

This suggests that the Bet v 1 anchor fusion produced in both of the tested lactobacilli

display all of the IgE epitopes present in Bet v 1. In addition, Bet v 1 from L.

rhamnosus UP616 competes with standard rBet v 1 for IgE at lower concentrations

than that of the rBet v 1 from L. acidophilus X37. Therefore L. rhamnosus UP616

produces higher amounts of accessible Bet v 1 than L. acidophilus X37, which also is

reflected on the band intensities on the western blot (data not shown).

Immunomodulatory effect of lactobacilli expressing Bet v 1

Since the surface displayed Bet v 1 showed immunereactivity using both strains we

tested whether they also showed retained intrinsic immune stimulatory effect. One

strategy to counteract the allergic disease eliciting Th2-like responses is to skew the

immune response towards Th1 or T-regulatory responses and thereby inhibit the IgE

mediated allergic response [Valenta et al., 2002]. Lactobacilli can activate dendritic

cells (DC) and facilitate polarization of naive DCs towards Th1 or T-regulatory

responses [Mohamadzadeh et al., 2005, Foligne et al., 2007 a/b]. The mechanisms

behind these effects are largely unknown, but molecular structures of the bacterial cell

wall and microbial nucleotides and believed to be involved, possibly through

activation of Toll-like receptors on DCs [Matsuguchi et al., 2003, Rachmilewitz et al.,

2004]. Surface display of Bet v 1 alters the cell wall architecture of the bacterial host

strain. Therefore we compared the immunomodulatory effect on murine bone marrow

derived dendritic cells of both wild type and recombinant lactobacilli expressing Bet v

1. Two lactobacilli were selected on the basis of their divergent immune polarizing

59

Figure 5 – Induction of cytokine production in dendritic cells by co-incubation with lactobacilliCytokines IL10 and IL12p70 were analyzed by ELISA in supernatants collected from 8-day cultures of murine bone marrow-derived DC that have been cultured for additional 15 h with different concentrations of bacteria (10, 100 µg/mL). White bars: IL12p70, cross sectioned bars: IL10, WT: wild type bacteria, Betv1: bacteria expressing surface anchored Bet v 1, GLU: glutaraldehyde treated bacteria. Data are mean +/-SEM from three independent co-incubations of DC and bacteria.

effect. L. acidophilus X37 is a strong IL12 inducer with concomitant low IL10

induction, where L. rhamnosus UP616 show a more balanced IL12 and IL10

induction. For both strains the cytokine induction was dependent on bacterial cell

concentration as observed previously [Christensen et al., 2002, Zeuthen et al., 2006]

(Figure 5). Expression of Bet v 1 on the cell surface did not alter the level or balance

between IL10 and the biologically active IL12p70 cytokines from the dendritic cells.

However, addition of glutaraldehyde to the bacteria prior to co-incubation with the

dendritic cells reduced the immune polarizing effect of both strains. This may not be

surprising as glutaraldehyd is a chemical cross linker that forms a Schiff-base (-H=N-

) with the amino groups of proteins. Hence, the surface pattern on the bacteria is

changed, which might alter the conformation of potential TLR agonists, thereby

reducing the adjuvant effect of the lactobacilli. Indeed, the DC data suggest that

preservation of the protein surface architecture is essential for the maintenance of

especially the IL12 inducing effect of L. acidophilus X37 (Figure 5). Although the

level of Bet v 1 display using the selected protein anchor was higher in L. rhamnosus

UP616 than L. acidophilus X37, the latter shows more promise as a Th1 promoting

carrier of immune reactive Bet v 1, whereas L. rhamnosus UP616 might promote

induction of tolerance towards the allergen.

60

Conclusions The purpose of surface display of vaccine components on LAB is co-delivery of the

immunogen together with immune activating signals. We analysed different

Lactobacillus protein anchoring signals for covalent linkage of a reporter protein and

the Bet v 1 allergen. A consensus LPXTG domain together with a spacer of 450 aa

could present active enzyme and Bet v 1 with similar accessible epitopes as that of

free Bet v 1. The immunemodulatory effect of L. rhamnosus and L. acidophilus with

surface displayed Bet v 1 was retained and comparable to their wild type counterparts,

and therefore might represent a promising oral vaccine delivery vehicle for treatment

of birch pollen allergy.

Methods Bacterial strains and growth conditions.

The bacterial strains are listed in Table 1. L. plantarum was grown at 37º C on Man-

Rogosa-Sharp (MRS) medium (Oxoid, Hampshire, United Kingdom) without aeration

and supplemented 5 µg/mL chloramphenicol (Cam) (Merck, Darmstadt, Germany)

when appropriate. L. lactis was grown at 30°C in M17 medium (Oxoid) supplemented

with 0,5% glucose (GM17). When appropriate, 5 μg/mL Cam was added. Escherichia

coli DH10B (Invitrogen, Carlsbad, Ca) was used as an intermediate cloning host and

added 10 μg/mL Cam, when appropriate. Glutaraldehyd (GLA) treatment of bacteria

was done as follows. Bacteria were harvested by centrifugation and re-suspended in

M9 buffer (0.6% Na2HPO4, 0.3% KH2PO4, 0.5% NaCl, 0.025% MgSO4) and adjusted

to a cell density of 1010 cells per mL. A fixed amount of L. plantarum cells (1010) was

incubated with 0.2% GLA (Sigma-Aldrich, St. Lois MO) for 50 min at room

temperature. Subsequent to the GLA treatment, the cell mixture was subjected to

centrifugation to give a cell fraction and a supernatant. The isolated cells were

thoroughly washed twice in M9 buffer.

DNA isolation and manipulation.

Chromosomal [Johansen & Kibenich, 1992] and plasmid DNA [O´sullivan &

Klaenhammer, 1993] from L. lactis and L. plantarum were prepared as described. E.

coli plasmids were isolated using a plasmid extraction kit from Genomed (Bad

Oeynhausen, Germany) as recommended. L. lactis and L. plantarum was made

61

electro-competent and transformed as described [Holo & Nes et al., 1989], while

competent E. coli DH10B cells were purchased (Invitrogen, Groningen, The

Netherlands) and transformed as recommended. DNA restriction and modification

enzymes, New England Biolabs (Beverly, MA) were used as recommended. DNA

was sequenced with a Thermo Sequenase fluorescent-labelled primer cycle

sequencing kit (Amersham Pharmacia, Uppsala, Sweden), Cy5-labeled primers, and

an ALFexpress DNA sequencer (Amersham Pharmacia).

Construction and tests of cell surface display system

To identify efficient cell wall anchoring signals a test system based on secretion and

surface display of the plasmid borne S. aureus nuclease was developed. The nuc gene

was PCR amplified using primers fnuc0 and rnuc and inserted into pCR2.1-TOPO.

Subsequently, the nucB fragment was isolated on an EcoRV-NheI fragment and

inserted into pNZ8048 digested with BsaAI and XbaI resulting in pTSA24. The

resulting plasmid, pTSA24, contains a multiple cloning region upstream of the nucB

gene allowing insertion of promoters and signal peptides. A strong consensus

promoter located upstream of the central regulator controlling the expression of

glycolytic enzymes in L. plantarum 299v was amplified by PCR using primers P1-

EcoRV and P2-XhoI and then inserted into pCR2.1-TOPO. The promoter was

subsequently isolated on a EcoRI fragment and inserted into the unique EcoRI present

in pTSA24 resulting in pUP411. The desired insertion relative to the nuc gene was

determined by DNA sequencing. Subsequently, a signal peptide which is homologous

to the extracellular protein lp_3067 from L. plantarum WCSF1 was PCR amplified

using L. plantarum 299v as template and the two primers CAD65239fwd and

CAD65239rev. The PCR fragment encoding the signal peptide contained was

subsequently inserted in-frame with the nucB gene present in pUP411 using the

unique BamHI site. The desired orientation of the signal peptide relative to the nuc

gene was determined by DNA sequencing. The plasmid was named pUP537.

Transformation of pUP537 into L. plantarum 299v resulted in strain UPO540

Candidate anchor signals were selected from L. plantarum WCFS1 using the

published sequence and LPXTG consensus sequence to identify surface displayed

proteins. Two anchored proteins, lp_0197 and lp_2486 were selected among 25

62

Table 1. Strains, plasmids, and primers used in this study and their relevant characteristics

Material Relevant feature or sequence Source Strains L. plantarum WCFS1 Isolated from human [Kleerebezem et al.,

2003] L. plantarum 299v Isolated from human [Adlerberth et al., 1996]

UP540 Harbours expression vector for NucB secretion This study PSM2042 Harbours anchoring plasmid pPSM2042 This study PSM2043 Harbours anchoring plasmid pPSM2043 This study PSM2044 Harbours anchoring plasmid pPSM2044 This study PSM2045 Harbours anchoring plasmid pPSM2045 This study

L.acidophilus X37 Isolated from human [Zeuthen et al., 2006] L.rhamnosus UP616 Isolated from human Own isolate E. coli DH10B Intermediate cloning host Invitrogen L. lactis MG1363 Intermediate cloning host [Gasson et al., 1983] Plasmids

pCR2.1 E. coli cloning vector Invitrogen pNZ8048 Lactobacillus expression vector [de Ruyter et al., 1996] pUP540 Expression vector for NucB secretion This study pPSM2042 nucB, Anchor from lp_0197, 54 aa spacer This study pPSM2043 nucB, Anchor from lp_0197,199 aa spacer This study pPSM2044 nucB, Anchor from lp_2486, 49 aa spacer This study pPSM2045 nucB, Anchor from lp_2486, 459 aa spacer This study pPSM719 pPSM2045, mutated BamH1 site This study pPSM719-

Betv1 Betv1, Anchor from lp_2486, 459 aa spacer This study

Primers Pnuc-BbsI/NheI-lp0197kort

5´-GGAGCGAAGA CAACGCTGAT TCAGGTCAAG CTAGCGTTGT CACAACGAAA ACTGAGACAG CTAAGTTAGT CAAGC-3´

Fnuc0 5´GATATCTGCAGCTCGAGCCCGGGAATTCATATGGATCCTCACAAACAGATAACGGCG 3

Rnuc 5´-GCTAGCTAAAAATTATAAAAGTGCCACTAGCAGCAGTG 3’ P1-EcoRV 5´- GATATCTTTAACGCGCCATTTGG 3 P2-XhoI 5´-CTCGAGCTCTTACCACAACATGAATTATAAC 3’ CAD65239fwd 5´-GGATCCACAAGGAGGAAATTCTAGTATGAAGTCAATGTTAGG 3' CAD65239rev 5´-GGATCCATTTTCTGCAGCATTGGC 3 Plp0197-Cterm-BclI/BbsI

5´-CGATATAGCG TTGTCTTCTG ATCATTAATT AAATCGCCGT TGACGACTAA CAACTGTTAG C-3´

Plp0197lang-Nterm-NheI

5´-CGATGCTAGC CAAGTGCTTG TGTTGAATGA GCAAAGCGC-3´

Plp2486lang-Nterm-NheI

5´-CGATGCTAGC GATTCTGATG GTAATGTGAC GGCTGCTGG-3´

Plp2486kort-Nterm-NheI

5´-CGATGCTAGC ACGGCACCAG CAACCACGGT TTCTGATG-3´

Plp2486-Cterm-BclI

5´-CGATTGATCA TTATGCTTCA TGCTTCCGAC GAGAAACGC-3´

BclI-Betv1-N 5´-CCGATTTGAT CAGGTGTGTT TAATTATGAG ACTGAGACC-3´ NheI-Betv1-Cny 5´-CGATGCTAGC GTTGTAGGCA TCGGAGTGTG CC-3´

proteins containing the LPXTG motif. Both proteins are relatively long and anchors

are in complete agreement with the LPXTG consensus from lactobacilli [Kleerebezem

et al., 2003]. Two variants of both anchors were cloned with 54 or 199 aa spacer

regions (lp_0197) or 49 or 459 aa spacer regions (lp_2486), respectively. Anchors

were isolated from L. plantarum WCFS1 genomic DNA using PCR and primers as

stated in table 1. pUP540 was digested with Bbs1, which cuts in the 3´end of nucB.

Here was the 320 bp PCR fragment containing the lp_0197 (54 aa spacer) anchor

signal with terminal sites BbsI inserted. L. lactis was used as an intermediate cloning

host resulting in pPSM2042. Using pPSM2042 the anchor signal was replaced by the

730 bp lp_0197 (199 aa spacer) using restriction sites BclI-NheI sites of both

fragment and plasmid resulting in pPSM2043. Likewise were the two anchors (281

and 1512 bp) from lp_2486 inserted into BclI-NheI treated pPSM2042. This resulted

63

in pPSM2044 and pPSM2045. L. lactis transformants containing the surface display

vectors were verified by sequencing. Plasmid DNA of pPSM2042, pPSM2043,

pPSM2044, pPSM2045, and pUP540 was made from L. lactis and electroporated into

competent L. plantarum 299v and plated on to MRS containing 5 µg/mL Cam.

Nuclease overlay assay was used to identify expression of active nuclease in colonies

and was done as described [Ravn et al., 2000]. The level nuclease expression was

evaluated by nuclease assay as described [Ravn et al., 2003] on cell cultures, washed

cells using PBS, and culture supernatants harvested by 4000 RPM in 15 minutes.

Cell surface display of Bet v 1

pPSM2045, which contains DNA encoding the signal peptide of apf located between

two BamHI sites, was digested with BamHI and ligated to a BglII-BamHI fragment,

containing the same signal peptide. This fragment was generated using PCR with

primers CAD65239fwd and CAD65239rev and 299v template DNA . The resulting

plasmid, pPSM719, is identical to pPSM2045 except for a single-base substitution

upstream of the signal peptide eliminating the upstream BamHI site (Table 1).

pPSM719 was treated with BamHI and NheI. The gene encoding Bet v 1 was obtained

from ALK-Abello and cloned using PCR and primers BclI-Betv1-N and NheI-Betv1-

Cny introducing terminal BclI and NheI restriction enzyme sites. The gene encoding

Bet v 1 was ligated to linearized pPSM719 and transformed into L. lactis MG1363.

Transformants were checked by colony PCR using Bet v 1 specific primers. The

pPSM719-Betv1 containing Bet v 1 was sequenced and transformed into

electrocompetent L. acidophilus and L. rhamnosus. Cam resistant clones was checked

for Bet v1 expression by western blotting using whole cell that have been boiled (95°

C 10 min) in 10 % SDS prior to SDS-PAGE. Bet v 1 specific serum was a kind gift

from ALK-Abello.

Specific human serum IgE inhibition assay

Recombinant Bet v 1 anchored to the cell surface of lactobacilli was extracted as

follows. A 10 mL culture was grown overnight, harvested, and washed in PBS buffer.

The cells were resuspended in 500 µL TE buffer supplemented with 20 mg/mL

lysozyme (Sigma) and incubated 1 h at 37° C. Surface extracted protein were isolted

from cell by centrifugation at 4000 RPM 25 min. A pool of equal volumes of sera

from seven birch pollen allergic patients was used for specific serum IgE inhibition

64

assays. All patients had a clinical history of birch pollinosis and were

radioallergosorbent test class 2 or more. rBet v 1 was biotinylated at a molar ratio of

1:5 (rBet v 1:biotin). The rBet v 1 shows similar IgE epitopes as native Bet v 1. The

inhibition assay was performed on ADVIA Centaur System (Bayer, Kgs. Lyngby,

Denmark) as follows: A serum sample (25 µl) was incubated with paramagnetic beads

(solid phase) coated with a monoclonal mouse antihuman IgE Ab (ALK-Abello´,

Hørsholm, Denmark), washed, resuspended, and incubated with a mixture of

biotinylated rBet v 1 and inhibitor (nonbiotinylated Bet v 1 extracted from

lactobacilli) in dilution series. The amount of biotinylated rBet v 1 bound to the serum

IgE on the solid phase was estimated from the measured relative light units (RLUs)

after incubation with acridinium-ester labelled streptavidin. The degree of inhibition

was calculated as the ratio between the RLUs obtained using buffer and mutant as

inhibitor.

In vitro stimulation of Dendritic cells

Bone marrow cells were isolated and cultured as described [Lutz et al.,1999] with

modifications. Femora and tibiae from two female C57BL/6 mice, 8–12 wk (DTU,

Denmark), were removed and stripped of muscles and tendons. After soaking the

bones in 70% ethanol for 2 min and rinsing in PBS, both ends were cut with scissors

and the marrow was flushed with PBS using a 27-gauge needle. Cell clusters were

dissociated by repeated pipetting. The resulting cell suspension was centrifuged for 10

min at 300 × g and washed once in PBS. Cells were resuspended in RPMI 1640

(Sigma-Aldrich, St. Louis, MO) supplemented with 4 mM L-glutamine, 100 U/ml

penicillin, 100 μg/ml streptomycin, 50 μM 2-ME, 10% (v/v) heat-inactivated FBS

(Atlanta Biologicals, Norcross, GA), and 15 ng/ml murine GM-CSF. GM-CSF was

added as 5–10% (v/v) culture supernatant harvested from a GM-CSF-producing cell

line. The GM-CSF produced was quantified using a specific ELISA kit (BD

PharMingen, San Diego, CA). To enrich for DC, 10 ml of cell suspension containing

3×106 cells was seeded per 100-mm bacteriological petri dishes (day 0) and incubated

for 8 days at 37°C in an atmosphere of 5% CO2. An additional 10 ml of freshly

prepared medium was added to each plate on day 3. On day 6, 9 ml from each plate

was centrifuged for 5 min at 300 × g, and the resultant cell pellet was resuspended in

10 ml of fresh medium, and the suspension was returned to the dish. On day 8, cells

were used to evaluate the effects of lactobacilli on cytokine release. Nonadherent cells

65

were gently pipetted from petri dishes containing 8-day old DC-enriched cultures. The

collected cells were centrifuged for 5 min at 300 × g and resuspended in medium

supplemented with 10 ng/ml GM-CSF. Cells were seeded in 48-well tissue culture

plates at 0,5 × 106/500 μl/well, and then to each well was added (100 μl/well) either

bacteria with and without Bet v 1 (10–100 μg/ml), and GLU treated bacteria (10–100

μg/ml). After a stimulation period of 15 h at 37°C in 5% CO2, culture supernatant was

collected and stored at -80°C until cytokine analysis. IL-12(p70) and IL-10 were

analyzed using commercially available ELISA kits (R&D,Minneapolis) according to

the manufacturer’s instructions.

Authors' contributions

JAG drafted the manuscript and supervised the experiments. SSJ and JAG made the

dendritic cell assay. JB and MF tested the immunereactivity of surface displayed Bet

v 1. SMM gave helpful input to the genetic constructions and edited the manuscript.

PRA and JAG made the genetic construction and characterizations of the surface

display system.

Acknowledgements This work was partially financed by the Danish Ministry of Science and Innovation.

References


Bermudez-Humaran LG, Langella P, Cortes-Perez NG, Gruss A, Tamez-Guerra RS, Oliveira SC et al.: Intranasal immunization with recombinant Lactococcus lactis secreting murine interleukin-12 enhances antigen-specific Th1 cytokine production. Infect Immun 2003, 71: 1887-1896.

Bermudez-Humaran LG, Cortes-Perez NG, Lefevre F, Guimaraes V, Rabot S, cocer-Gonzalez JM et al.: A novel mucosal vaccine based on live Lactococci expressing E7 antigen and IL-12 induces systemic and mucosal immune responses and protects mice against human papillomavirus type 16-induced tumors. J Immunol 2005, 175: 7297-7302.

Boekhorst J, Wels M, Kleerebezem M, Siezen RJ: The predicted secretome of Lactobacillus plantarum WCFS1 sheds light on interactions with its environment. Microbiology 2006, 152: 3175-3183.


Buck BL, Altermann E, Svingerud T, Klaenhammer TR: Functional analysis of putative adhesion factors in Lactobacillus acidophilus NCFM. Appl Environ Microbiol 2005, 71: 8344-8351.

66

Christensen HR, Frokiaer H, Pestka JJ: Lactobacilli differentially modulate expression of cytokines and maturation surface markers in murine dendritic cells. J Immunol 2002, 168: 171-178.

Cortes-Perez NG, Azevedo V, cocer-Gonzalez JM, Rodriguez-Padilla C, Tamez-Guerra RS, Corthier G et al.: Cell-surface display of E7 antigen from human papillomavirus type-16 in Lactococcus lactis and in Lactobacillus plantarum using a new cell-wall anchor from lactobacilli. J Drug Target 2005, 13: 89-98.

Daniel C, Repa A, Wild C, Pollak A, Pot B, Breiteneder H et al.: Modulation of allergic immune responses by mucosal application of recombinant lactic acid bacteria producing the major birch pollen allergen Bet v 1. Allergy 2006, 61: 812-819.


del-Patient K, Ah-Leung S, Creminon C, Nouaille S, Chatel JM, Langella P et al.: Oral administration of recombinant Lactococcus lactis expressing bovine beta-lactoglobulin partially prevents mice from sensitization. Clin Exp Allergy 2005, 35: 539-546.

Dieye Y, Usai S, Clier F, Gruss A, Piard JC: Design of a protein-targeting system for lactic acid bacteria. J Bacteriol 2001, 183: 4157-4166.

Foligne B, Zoumpopoulou G, Dewulf J, Ben YA, Chareyre F, Sirard JC et al.: A key role of dendritic cells in probiotic functionality. PLoS ONE 2007, 2: e313.

Foligne B, Nutten S, Grangette C, Dennin V, Goudercourt D, Poiret S et al.: Correlation between in vitro and in vivo immunomodulatory properties of lactic acid bacteria. World J Gastroenterol 2007, 13: 236-243.

Gasson MJ: Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 1983, 154: 1-9.

Grangette C, Muller-Alouf H, Goudercourt D, Geoffroy MC, Turneer M, Mercenier A: Mucosal immune responses and protection against tetanus toxin after intranasal immunization with recombinant Lactobacillus plantarum. Infect Immun 2001, 69: 1547-1553.

Hanniffy SB, Carter AT, Hitchin E, Wells JM: Mucosal delivery of a pneumococcal vaccine using Lactococcus lactis affords protection against respiratory infection. J Infect Dis 2007, 195: 185-193.

Hansson M, Samuelson P, Gunneriusson E, Stahl S: Surface display on gram positive bacteria. Comb Chem High Throughput Screen 2001, 4: 171-184.

Hazebrouck S, Pothelune L, Azevedo V, Corthier G, Wal JM, Langella P: Efficient production and secretion of bovine beta-lactoglobulin by Lactobacillus casei. Microb Cell Fact 2007, 6: 12.

Holo H, Nes IF: High-Frequency Transformation, by Electroporation, of Lactococcus lactis subsp. cremoris Grown with Glycine in Osmotically Stabilized Media. Appl Environ Microbiol 1989, 55: 3119-3123.

Johansen E, Kibenich A: Isolation and characterization of IS1165, an insertion sequence of Leuconostoc mesenteroides subsp. cremoris and other lactic acid bacteria. Plasmid 1992, 27: 200-206.

Kleerebezem M, Boekhorst J, van KR, Molenaar D, Kuipers OP, Leer R et al.: Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 2003, 100: 1990-1995.

Kruger C, Hu Y, Pan Q, Marcotte H, Hultberg A, Delwar D et al.: In situ delivery of passive immunity by lactobacilli producing single-chain antibodies. Nat Biotechnol 2002, 20: 702-706.

Lan JG, Cruickshank SM, Singh JC, Farrar M, Lodge JP, Felsburg PJ et al.: Different cytokine response of primary colonic epithelial cells to commensal bacteria. World J Gastroenterol 2005, 11: 3375-3384.

Lee JS, Poo H, Han DP, Hong SP, Kim K, Cho MW et al.: Mucosal immunization with surface-displayed severe acute respiratory syndrome coronavirus spike protein on Lactobacillus casei induces neutralizing antibodies in mice. J Virol 2006, 80: 4079-4087.

Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F, Romani N et al.: An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods 1999, 223: 77-92.

Maassen CB, van Holten-Neelen C, Balk F, den Bak-Glashouwer MJ, Leer RJ, Laman JD et al.: Strain-dependent induction of cytokine profiles in the gut by orally administered Lactobacillus strains. Vaccine 2000, 18: 2613-2623.

67

Marcotte H, Koll-Klais P, Hultberg A, Zhao Y, Gmur R, Mandar R et al.: Expression of single-chain antibody against RgpA protease of Porphyromonas gingivalis in Lactobacillus. J Appl Microbiol 2006, 100: 256-263.

Matsuguchi T, Takagi A, Matsuzaki T, Nagaoka M, Ishikawa K, Yokokura T et al.: Lipoteichoic acids from Lactobacillus strains elicit strong tumor necrosis factor alpha-inducing activities in macrophages through Toll-like receptor 2. Clin Diagn Lab Immunol 2003, 10: 259-266.

Mohamadzadeh M, Olson S, Kalina WV, Ruthel G, Demmin GL, Warfield KL et al.: Lactobacilli activate human dendritic cells that skew T cells toward T helper 1 polarization. Proc Natl Acad Sci U S A 2005, 102: 2880-2885.

O'Sullivan DJ, Klaenhammer TR: Rapid Mini-Prep Isolation of High-Quality Plasmid DNA from Lactococcus and Lactobacillus spp. Appl Environ Microbiol 1993, 59: 2730-2733.

Oliveira ML, Areas AP, Campos IB, Monedero V, Perez-Martinez G, Miyaji EN et al.: Induction of systemic and mucosal immune response and decrease in Streptococcus pneumoniae colonization by nasal inoculation of mice with recombinant lactic acid bacteria expressing pneumococcal surface antigen A. Microbes Infect 2006, 8: 1016-1024.

Perdigon G, Maldonado GC, Valdez JC, Medici M: Interaction of lactic acid bacteria with the gut immune system. Eur J Clin Nutr 2002, 56 Suppl 4: S21-S26.

Perez CA, Eichwald C, Burrone O, Mendoza D: Rotavirus vp7 antigen produced by Lactococcus lactis induces neutralizing antibodies in mice. J Appl Microbiol 2005, 99: 1158-1164.

Poo H, Pyo HM, Lee TY, Yoon SW, Lee JS, Kim CJ et al.: Oral administration of human papillomavirus type 16 E7 displayed on Lactobacillus casei induces E7-specific antitumor effects in C57/BL6 mice. Int J Cancer 2006, 119: 1702-1709.

Poquet I, Ehrlich SD, Gruss A: An export-specific reporter designed for gram-positive bacteria: application to Lactococcus lactis. J Bacteriol 1998, 180: 1904-1912.

Pusch O, Boden D, Hannify S, Lee F, Tucker LD, Boyd MR et al.: Bioengineering lactic acid bacteria to secrete the HIV-1 virucide cyanovirin. J Acquir Immune Defic Syndr 2005, 40: 512-520.

Rachmilewitz D, Katakura K, Karmeli F, Hayashi T, Reinus C, Rudensky B et al.: Toll-like receptor 9 signaling mediates the anti-inflammatory effects of probiotics in murine experimental colitis. Gastroenterology 2004, 126: 520-528.

Ramasamy R, Yasawardena S, Zomer A, Venema G, Kok J, Leenhouts K: Immunogenicity of a malaria parasite antigen displayed by Lactococcus lactis in oral immunisations. Vaccine 2006, 24: 3900-3908.

Ramiah K, van Reenen CA, Dicks LM: Expression of the mucus adhesion genes Mub and MapA, adhesion-like factor EF-Tu and bacteriocin gene plaA of Lactobacillus plantarum 423, monitored with real-time PCR. Int J Food Microbiol 2007, 116: 405-409.

Ravn P, Arnau J, Madsen SM, Vrang A, Israelsen H: The development of TnNuc and its use for the isolation of novel secretion signals in Lactococcus lactis. Gene 2000, 242: 347-356.

Ravn P, Arnau J, Madsen SM, Vrang A, Israelsen H: Optimization of signal peptide SP310 for heterologous protein production in Lactococcus lactis. Microbiology 2003, 149: 2193-2201.

Reuter G: The Lactobacillus and Bifidobacterium microflora of the human intestine: composition and succession. Curr Issues Intest Microbiol 2001, 2: 43-53.

Steidler L, Hans W, Schotte L, Neirynck S, Obermeier F, Falk W et al.: Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 2000, 289: 1352-1355.

Strauss A, Gotz F: In vivo immobilization of enzymatically active polypeptides on the cell surface of Staphylococcus carnosus. Mol Microbiol 1996, 21: 491-500.

Tallon R, Arias S, Bressollier P, Urdaci MC: Strain- and matrix-dependent adhesion of Lactobacillus plantarum is mediated by proteinaceous bacterial compounds. J Appl Microbiol 2007, 102: 442-451.

Tuohy KM, Pinart-Gilberga M, Jones M, Hoyles L, McCartney AL, Gibson GR: Survivability of a probiotic Lactobacillus casei in the gastrointestinal tract of healthy human volunteers and its impact on the faecal microflora. J Appl Microbiol 2007, 102: 1026-1032.

Valenta R: The future of antigen-specific immunotherapy of allergy. Nat Rev Immunol 2002, 2: 446-453.

Van D, V, te BR, Chapman JW, Toschka HY, Klis FM, Verrips CT: Comparison of cell wall proteins of Saccharomyces cerevisiae as anchors for cell surface expression of heterologous proteins. Appl Environ Microbiol 1997, 63: 615-620.

68

Vitini E, Alvarez S, Medina M, Medici M, de Budeguer MV, Perdigon G: Gut mucosal immunostimulation by lactic acid bacteria. Biocell 2000, 24: 223-232.

Wu C, Yang G, Bermudez-Humaran LG, Pang Q, Zeng Y, Wang J et al.: Immunomodulatory effects of IL-12 secreted by Lactococcus lactis on Th1/Th2 balance in ovalbumin (OVA)-induced asthma model mice. Int Immunopharmacol 2006, 6: 610-615.

Zeuthen LH, Christensen HR, Frokiaer H: Lactic acid bacteria inducing a weak interleukin-12 and tumor necrosis factor alpha response in human dendritic cells inhibit strongly stimulating lactic acid bacteria but act synergistically with gram-negative bacteria. Clin Vaccine Immunol 2006, 13: 365-375.

69

Chapter 4

DNA inversion Controls Expression of a Mannose Specific Adhesin from Lactobacillus plantarum

70

DNA inversion Controls Expression of a Mannose Specific Adhesin from Lactobacillus plantarum

Jacob Glenting1§, Søren M. Madsen1, Kim Holmstrøm1, Peter Ravn1, Bjørn Holst1

1Bioneer A/S, DK-2970 Hørsholm, Denmark

§Corresponding author

Email addresses:

JG: [email protected]

SMM: [email protected]

KH: [email protected]

PR: [email protected]

BH: [email protected]

71






Abstract

Background

Lactobacillus plantarum is a natural inhabitant of the human gastrointestinal tract and

is ascribed health benefiting effects. The bacterium’s adherence to the mucosal

surfaces is most likely a prerequisite for exertion of these effects. As mucosal surfaces

are covered with glycoproteins the molecular mechanisms behind the sugar binding

capacity of L. plantarum was investigated.

Results

A mannose specific adhesin Msa has been identified in L. plantarum WCFS1. L.

plantarum 299v was tested for mannose adhesion and binds with higher affinity than

strain WCFS1. The molecular mechanism behind this adhesion was identified as a

protein highly homologous to Msa. An msa knock out mutant strain lost its affinity

for mannose and adhesion to Caco-2 cells. The expression of msa in strain 299v and

WCFS1 was compared and showed that the weaker mannose affinity of WCFS1 was

due to lower msa expression. However, msa was expressed with phase variation in

WCFS1 resulting in a heterogenic culture of bacteria with high and low affine binding

to mannose. The variable expression of msa is controlled by inversion of a 104 bp

region upstream of msa. This invertible element is flanked by 14 bp perfect inverted

repeats. Transcriptional analysis showed that msa expression is increased in the ON

orientation of the invertible element and arrested in the opposite OFF orientation. The

msa promoter was mapped and the invertible DNA element was analysed for potential

RNA secondary structures. The OFF orientation results in formation of an mRNA

hairpin structure in the 5´ end of the mRNA, which may cause arrest of transcription.

This hairpin is not as energetically favourable in the ON orientation.

Conclusions

Two L. plantarum strains was analysed for expression of Msa. Both strains express a

highly homologous Msa. However, L. plantarum 299v showed constitutive msa

expression, whereas the msa was phase variable expressed in WCFS1. Phase variation

was controlled by inversion of the DNA element upstream msa most likely being

catalyzed by recombinase-like proteins. These findings illustrate that although genes

72

encoding adhesins are conserved amongst different Lactobacillus species the presence

of recombinase-like proteins can change the expression pattern significantly.

Background Lactobacilli are widely distributed in nature and are found in the gastrointestinal (GI)

and urogenitary tract of both humans and animals. Some strains of lactobacilli are

considered beneficial for humans and are attributed a major role in the positive health

effects of probiotics (reviewed by [Rolfe 2000]). The health claims associated with

intake of lactobacilli include maintenance of a balanced immune response in the

mucosal associated lymphatic tissue (MALT) [Perdigon et al., 1999], support of the

GI endothelial barrier by up-regulating host-production of mucin [Mack et al., 2003],

and inhibition of the growth of pathogenic bacteria by competing for space and

nutrients [Bernet et al., 1994] or releasing antimicrobial compounds like bacteriocins

[Gotteland et al., 2006]. The molecular mechanisms behind these effects of

lactobacilli are not covered in detail. However, a prerequisite of lactobacilli to

modulate host responses and maintaining a healthy gut flora is most likely adherence

to the mucosal surfaces of the gastrointestinal tract. The molecular factors responsible

for adhesion and colonisation of these organisms have therefore been investigated

using tissue samples, cell lines and components of the extracellular matrix. The

ligands of these adhesins have been identified as sugar components [Adlerberth et al.,

1996]. And ECM proteins like fibronectin [Hynonen et al., 2002], mucin [Granato et

al., 2004, Miyoshi et al., 2006], and collagen [Sillanpaa et al., 2000]. These so-called

adhesion factors of lactobacilli are a chemical diverse group and include both protein

and non-protein components of the bacterial cell surface. Recently, a mucus adhesion

promoting protein MapA was identified in L. reuteri as a 26 kDa cell surface protein

that binds receptor-like molecules on Caco-2 cells [Miyoshi et al., 2006]. Non-

proteianous cell surface molecules like lipoteichoic acid of L. johnsonii have been

shown to participate in the adhesion to intestinal cells [Granato et al., 1999]. In this

species unexpected surface proteins has been identified as adhesion factors. The

elongation factor EF-TU, normally involved in protein synthesis and without apparent

signal peptide or cell wall anchoring motive, was identified as a cell surface protein

mediating adhesion to intestinal cells [Granato et al., 2004]. Surprisingly was also the

73

GroEL heat shock protein found on the surface of the same species and identified as

an adhesion factor [Bergonzelli et al., 2006]. The dual role of these house-keeping

intracellular proteins illustrates the importance of experimental work and that

functional analysis of genes requires more than DNA homology searches.

One of the most abundant species of lactobacilli that colonize the human GI tract is

Lactobacillus plantarum [Ahrne et al., 1998]. Several adhesion mechanisms have

been described in L. plantarum. The L. plantarum strain 299v is an efficient coloniser

of the human GI tract and adhere to colon epithelial HT29 cells in vitro [Adlerberth et

al., 1996]. The adhesion mechanism was shown to involve a mannose specific

component and was inhibited by protease treatment of the bacteria and treatment with

per-iodate that oxidises sugar hydroxyl groups of the HT29 cells. The protein(s) and

gene(s) responsible for this mannose adhesive phenotype were however not identified.

Recently, the complete genome sequence of L. plantarum WCFS1 was published

[Kleerebezem et al., 2003]. Through homology gene analysis at least 12 cell surface

proteins was predicted to be adhesion factors [Boekhorst et al., 2006]. Amongst these

were mucus and fibronectin binding proteins, and also a 1000 amino acid mannose

specific adhesin (Msa) encoded by the lp_1229 gene [Pretzer et al., 2005]. Although

the mannose adhesive phenotype of L. plantarum strain WCFS1 and 299v seems

identical it is unknown if the same molecular components are involved.

The changing habitat of lactobacilli from food products to the GI-tract requires

activation of essential proteins that support colonisation and survival in the new

environment. Cloning systems that capture promoters from GI-tract induced genes has

been developed. The in vivo expression technology (IVET) was applied to L.

plantarum. Here, a genome library of 1-2 kb fragments was cloned on to a plasmid in

fusion with a gene encoding resolvase. DNA fragments with promoters, which are

activated in the GI-tract will express the resolvase that in turn excise the genome-

placed-gene encoding resistance to erythromycin [Bron et al., 2004]. Metabolic and

stress related genes were, as expected, induced. But also a bacteriocin gene was up-

regulated. However, four cell surface proteins were also induced illustrating that L.

plantarum change its surface architecture through passage in the GI-tract of mice.

Lactobacilli can also regulate surface proteins through other mechanisms than

74

regulation of promoter activities. The S-protein is a major surface protein of some

Lactobacilli that crystallises to form a two-dimensional layer on the bacterial cell

surface. Several functions have been ascribed to S-layers including a protective

covering layer, a unit for anchoring surface proteins, and as an adhesin [Toba et al.,

1995, Antikainen et al., 2002, vall-Jaaskelainen et al., 2003]. The gene encoding the

S-layer protein is in L. acidophilus regulated by chromosomal rearrangements where

the 6 kb silent gene is inverted and placed under the S-promoter [Boot et al., 1996].

Genome rearrangements and inversions as mechanisms for controlling expression of

especially surface proteins have been described for several other bacteria. A well-

known example is type 1 fimbriae of E. coli, which shows a multiphasic alteration in

its expression profile [reviewed by Henderson et al., 1999]. Fimbrial proteins are

adhesins and play an important role in the mannose-specific adhesion to host-surfaces.

Phase variable expression of fimbriae-like proteins (MR/P) was also observed in the

uropathogenic Proteus mirabilis [Zhao et al., 1997]. Here a 252 bp DNA fragment

that carries the MR/P promoter is inverted by a recombinase capable of switching the

invertible element from either ON to OFF or OFF to ON. The number of identified

genes, which show phase variable expression, is continuously increasing. However,

the mechanisms are relatively conserved and usually involve recombinases that

targets inverted DNA repeats flanking the invertible DNA element.

In this study the molecular factors of the mannose binding phenotype in L. plantarum

299v was compared to that of L. plantarum WCFS1. Interestingly, the two strains are

genetically very similar but differ notably in the expression of the mannose specific

adhesin. The adhesin in L. plantarum WCFS1 expressed with phase variation,

whereas L. plantarum 299v show a continous high expression of the adhesion.

Results Identification of mannose specific adhesin Msa in L. plantarum strain 299v

In L. plantarum strain WCFS1 the msa gene (NCBI locus tag lp_1229) gene was

identified to encode a mannose-specific adhesin (Msa) [Pretzer et al., 2005]. We

analysed the yeast agglutination capacity of different strains of lactobacilli and found

that especially L. plantarum 299v scored high in an agglutination assay [Table 1]. To

determine if an Msa-like protein is present in strain 299v we screened for an msa

75

Figure 1 - Mannose specific adhesin of L. plantarum 299v

The domains of the Msa. SP: Signal peptide, R17: repeats of 17 aa, conA: concanavalin A like domain, R100: 100 aa repeats, Δ36: 36 aa insertion compared to Msa of strain WCFS1, LPXTG: protein anchordomain recognized by the surface associated sortase.

homologue. Sequence analysis showed an msa homologue of 115 kDa with a signal

peptide and LPXTG anchor domain (Fig. 1). The Msa homologue contains similar

functional domains as Msa from WCFS1 including a sugar binding domain with

homology to the lectin concanavalin A. Compared to Msa of strain WCFS1 the

homologue contains 11 scattered bp changes and a 108 bp insertion. This insertion

encodes 36 aa and is an almost perfect repeat of the upstream sequence indicating that

a duplication event may have taken place. The highly similar Msa from strain 299v

and WCFS1 suggest that the stronger adhesive phenotype of 299v is due to a higher

expression of msa. Therefore we investigated the intergenic sequence upstream of

msa (see Additional file 1). Interestingly, the 400 bp region upstream the translation

start site of msa is in the two strains almost identical with only one nucleotide change.

Table 1 - Yeast agglutination phenotype of selected bacteria

Yeast agglutination assay was done and scored as follows: +++ strong agglutination forming large visible aggregates, ++ agglutination forming visible aggregates, + agglutination only visible by microscopy, ÷ no agglutination. Assay was on three independent cultures of each strain.

Deletion of the mannose-specific adhesin in strain 299v

In L. plantarum WCFS1 the mannose-specific agglutination of yeast cells is not

visible without microscopic analysis. In contrast, L. plantarum strain 299v forms large

visible aggregates (Table 1). To determine if additional proteins are responsible for

the mannose-specific adhesion of strain 299v an Msa-mutant was constructed. By

homologous recombination a deletion strain was constructed. This ΔMsa strain,

named 299vΔMsa, contains an msa gene where the concanavalin A like region was

deleted (Fig. 1). Furthermore, the deletion introduced a frame shift inhibiting

translation of the distal part of msa. The resulting phenotype of the msa-

76

knockout strain was a complete loss of yeast aggregating ability (Table 1).

Figure 2 - Quantification of surface displayed Msa using ELISA

Amount of surface displayed Msa on different lactobacilli as a function of cell density. Error bars are +/-SEM based on three independent ELISA experiments

LC

But 299v ΔMsa also showed lack of the clumping phenotype normally seen for

lactobacilli cultures. Complementation of strain 299vΔMsa by episomal expression of

msa was tested to restore the mannose adhesive phenotype. However, introduction of

the msa gene on a high copy number plasmid (pMSAHC ) lead to cell death (data not

shown) most likely due to high and lethal msa expression levels as a result of

increased gene dosage. In contrast, transformation of 299v ΔMsa with msa placed on

a lower copy number plasmid (pMSALC ) restored the agglutinating phenotype of the

strain (Table 1). We also tested msa complementation in a wild type bacterial species

that lacks the ability to agglutinate yeast cells. Introduction of an msa expression

plasmid (pMSALL ) in L. lactis MG1363 resulted in a mannose specific adhesion

phenotype, whereas cells containing control plasmids without the msa gene did not

adhere to yeast cells (Table 1). These results show that the msa gene product is

responsible for the mannose-specific adhesion of L. plantarum strain 299v. The msa

knock-out and gene complementation indicate that msa alone encodes this phenotype

and that the mechanisms for secretion and anchoring of Msa seems conserved

amongst different lactic acid bacteria.

77

Figure 3 - Mannose affinity purification of the Con A domain

The Con A domain was expressed and secreted by L. lactis. Culture supernatant was applied to a mannan column and the fraction eluted with mannose is shown as titrated samples

Quantative measurement of cell surface located Msa

Although agglutination assays are indicative of the amounts of Msa on the bacterial

surface a more quantitative Msa measurement was established using ELISA. To

obtain antisera specific f or Msa we cloned the Con A like domain (Fig. 1) and

expressed it in L. lactis using a plasmid construct that secretes the product

(pAMJ1294). The lectin functions of the Con A like domain was used for purification.

Application of culture supernatants from L. lactis – pAMJ1294 to columns containing

mannan residues resulted in a clear fraction of Con A after wash and elution using

high concentrations of mannose (Fig 2). The predicted function of the Con A domain

was therefore indeed responsible for the mannose affinity of this protein. The purified

Con A domain was used to raise Msa specific antibodies in rabbits.

Quantative measurement of cell surface Msa was done using whole bacteria and

ELISA. As expected there was a clear correlation between the number of 299v cells

and the amount of Msa resulting in a maturation curve (Fig. 3). The Msa knock-out

strain (299v Δ Msa) had low interaction with Msa specific antibodies, whereas

complementation of this strain with pMSA (299v ΔMsa+pMSALC) restored the

surface display of Msa. The weak agglutinating phenotype of the WCFS1 strain

correlates with the ELISA that showed lower amounts of cell surface displayed Msa

on this bacterial strain compared to 299v (Fig. 3).

78

Figure 4 - Adhesion of lactobacilli to Caco-2 cells

Adhesion of L. plantarum 299v and the msa knock out derivative to Caco-2 cells. Adhesion is measured as number of adhered bacteria per cell. The mean of three independent wells are presented as the mean with +/- SEM

The role of Msa in adhesion to epithelial cells

The role of Msa in bacterial adherence to epithelial cells was analysed by co-

incubating bacteria and Caco-2 cells in vitro. Caco-2 cells express many of the

markers associated with normal small-intestine villus cells and are therefore often

used to study bacterial adherence. Indeed the strain 299v bound to Caco-2 cells after

wash (Fig. 4). In contrast was the binding capacity of 299v Δ Msa impaired (Fig. 4).

The complete inhibition of adhesion to Caco-2 cells by deletion of msa suggests a

central role of Msa in lactobacilli’s adhesion to eukaryotic cells.

Phase variation of the mannose binding capacity of lactobacilli

During standardisation of a high throughput yeast agglutination assay using bacterial

micro-cultures, strain WCFS1 showed a low frequency of cultures with a strong

agglutinating phenotype that was visible without microscopic analysis. This

phenotype (named Agg+ ) showed a similar agglutination phenotype as strain 299v,

but only represented 2% of an overnight grown culture (n=1000 independent cultures

tested). To evaluate if this Agg+ phenotype is locked and passed on during cell

division 103 colonies of a 30 generation culture were screened in a micro well

agglutination system. Here, 82% of the microcultures were switched back to the OFF

position. As such, the switching frequency of ON to OFF is higher than that of cells

switching from the OFF to ON phenotype. L. plantarum 299v was also tested for

79

variation in its mannose binding phenotype. However, 1000 independent micro

cultures showed identical agglutinating phenotype illustrating that ON and OFF

switching of msa expression is less frequent or absent in strain 299v compared to

WCFS1 under these growth conditions.

Figure 5 - The organisation of the intergenic region upstream msa

The 104 bp invertible fragment is located between to 14 bp inverted repeats IRL and IRR. The ON orientation leads to Agg+ phenotype of WCFS1 whereas the OFF orientation of the invertible fragment leads to the phenotype with low affine mannose binding. IRL and IRR are perfect inverted repeats where the sequence of the top strand of IRL is the same as the sequence of the bottom strand of IRR read from 5´ → 3´ direction.

Phase variable Msa expression is controlled by DNA inversion

The DNA sequence of the msa promoter region of a number of WCFS1 clones with

Agg+ phenotype was analysed and compared to the sequence of strain WCSF1 and

299v. Sequence analysis of the intergenic region of lp_1228 and lp_1229 (msa) of

WCFS1 and WCFS1- Agg+ showed an inversion of a 104 bp segment (Fig. 5,

additional file 1). The inverted segment is flanked by two perfect inverted repeats IRL

and IRR of 14 bp (Fig. 5). To analyse the effect of segment inversion on msa

expression a transcriptional analysis of msa was done. A 499 bp fragment covering

the distal end of lp_1228 and proximal end of lp_1229 (msa) was isolated from the

WCFS1 and WCFS1 Agg+ strains and cloned into the L. lactis pAK80 promoter probe

vector. The orientation corresponding to the msa OFF situation resulted in weak blue

colonies using pAK80 and L. lactis. In contrast, the ON orientation in the same msa

transcription direction gave strong blue colonies. The difference in LacLM activity on

agar plates was also seen in assays conducted in

Figure 6 - Test of the msa intergenic sequence in a promoter probe vector

The intergenic fragment was inserted upstream lacLM and tested for promoter activity. The two orientations representing the WSFS1 Agg+ situation (ON) and the WCFS1 (OFF) situation are shown. Promoter activities are given as ß-galactosidase activities as a mean of three independent experiments +/- SEM.

80

suspension where the ON orientation is two times as active as the OFF orientation

(Fig. 6). As such, the orientation of the 104 bp fragment between the inverted repeats

influences the expression of the downstream gene. Transcription of msa was also

analysed in the relevant lactobacilli (Fig. 7).

Figure 7 - Transcription of msa in different lactobacilli

Northern blot using a Con A specific probe. The Msa (~1000 aa) covers 3 kb of the transcript.

Northern blot using a concanavalin A specific probe showed high msa transcription

(~3,5 kb) in WCFS1-ON and 299v, whereas a weak transcript was detected in

WCFS1. The promoter probe and Northern blot demonstrate that inversion of the 114

bp affects transcription of msa, which explain the different phenotypes of WCFS1 and

WCFS1- Agg+ isolates. The size of the hybridisation product suggests that msa is

transcribed in a monocistronic fashion.

Mapping of the msa promoter

To investigate the mechanisms involved in the suppression or activation of msa

transcription the promoter was mapped by primer extension. Using a reverse primer

located in the 5´ end of the msa mRNA resulted in a 240 bp extension product with a

5´ mRNA sequence initiating 17 bp upstream IRL (Additional file 1). Because the

msa promoter is located outside the invertible DNA element a simple promoter flip-

flop mechanism can be ruled out. However, transcription may also be regulated by

secondary mRNA structures. Interestingly, two additional inverted repeats are formed

around the IRR in the OFF orientation of the inverted fragment (Fig 8).

81

Figure 8 - Formation of inverted repeats around IRR in the OFF orientation

Inversion of the 104 bp fragment in the OFF orientation leads to formation of two inverted repeats (striped arrows) that may form a mRNA stem loop structure. During inversion the inverted repeats are disrupted and the stem loop structure is less energetically favourable.

These two inverted repeats, located in the untranslated leader of the msa transcript,

may give rise to an mRNA hairpin structure causing arrest of transcription.

Furthermore, the hairpin structure may embed a putative ribosome binding site (Fig.

8). Formation of the hairpin structure is less energetically favourable and disrupted by

inversion and corresponds to the ON phenotype (Fig. 8). The formation of such a stem

loop structure in the untranslated leader of the msa transcript may explain the lack of

msa transcription in the phase-OFF state of strain WCFS1. Although, similar

expression inhibiting mRNA structures can be formed in strain 299v, msa is

efficiently transcribed and translated. The mechanism that escapes this regulation in

strain 299v is unknown but may be due to additional transcription factors. It may here

be important that 299v show one bp change in the region upstream msa. This bp

change is in fact located in the putative -35 region (Additional file 1 following the

references) and may indicate an alternate regulation of msa in strain 299v compared

to strain WCFS1.

82

Discussion The close interaction of lactobacilli and the human gastrointestinal lineage is most

likely fundamental for the health benefits of these bacteria. The human mucosal

surfaces are covered with mucus glyco-proteins and especially mannose residues are

important in bacterial adhesion [Vimal et al., 2000]. Sugar binding adhesins have

been therefore been a focus for researchers elucidating the host-microbe interactions

of lactobacilli. The mannose binding phenotype of lactobacilli has been described in

several strains of L. plantarum [Adlerberth et al., 1996, Pretzer et al., 2005, Sun et al.,

2007]. In this study a homologue to the mannose specific adhesin of WCFS1 was

identified in L. plantarum strain 299v. The two strains bind mannose by the same

molecular mechanisms. But the regulation of Msa expression is divergent in the two

lactobacilli. Strain WCFS1 display phase variation in Msa expression and the majority

of the culture is binding mannose with low affinity. However, 299v showed a

consistent strong mannose binding phenotype without any phase variation under the

tested laboratory conditions.

The Msa of WCFS1 was identified by in-silico phenotype-genotype mapping [Pretzer

et al., 2005]. We analysed the mannose adhesive phenotype of different lactobacilli

and observed that especially 299v was potent in a yeast agglutination assay. Although,

the monocistronic msa operon is highly similar in 299v and WCFS1 their expression

levels differ. Normally WCFS1 express low amount of Msa as observed by

transcriptional and protein analysis. This is surprising as Msa was shown to be

essential for bacterial adhesion to epithelial cell at least under laboratory conditions.

However, the surface architecture of lactobacilli may shift during adaptation to GI-

tract conditions [Bron et al., 2004a/b]. Other persistent colonisers of the human GI-

tract like Helicobacter pylori adapt to the environment in the GI-tract by genome

rearrangements turning on expression of a special set of proteins [Blaser et al., 2006].

Indeed, the mannose adhesive phenotype of WCFS1 showed phase variation at a low

frequency under laboratory growth conditions. The phase variable expression of Msa

is a result of inversion of a 104 bp DNA element positioned upstream msa. Most

likely is transcription arrested in the OFF orientation of the invertible fragment by

formation of a stem loop structure around the IRR. By inversion of the DNA fragment

the stem loop structure is disrupted allowing transcription from the promoter upstream

83

IRL. Most often phase variable gene expression is regulated by inversion of promoter

elements. In E. coli the expression of fimA is controlled by promoter inversion. Here

several protein factors including an integrase binds to recognition sites based on

inverted repeats on the target DNA [reviewed by Henderson et al., 1999]. The

identified inverted repeats in WCFS1 and 299v flanking the 104 bp invertible region

may be recognition sites of recombinases. However, sequence analysis did not show

IRL and IRR to be known recognition sequences of integrase or recombinase like

proteins.

The mechanism that suppresses msa transcription in WCFS1 is abolished in 299v.

Despite the presence of inverted repeats in msa upstream region in 299v we did not

observe any genome rearrangements and phase variations. The mechanisms behind is

unknown, but could be due to alternative regulatory transcription factors allowing

read through the stem loop structure. Interestingly, 299v contain a base pair change in

the putative -35 box of the msa promoter (Additional file 1), which may suggest

alternative transcription factors are used in 299v compared to WCFS1. The detailed

mechanism of the phase variable msa expression in WCFS1 is to be described.

However, the frequency of phase variation, the invertible DNA element between

perfect inverted repeats, and the formation of secondary RNA structures in the OFF

orientation is related to regulation of fimbrial expression in E. coli [reviewed by

Henderson et al., 1999] and P. mirabilis [Zhao et al., 1997]. The components that

activate the DNA inversion in L. plantarum WCFS1 are unknown. However,

recombinases are suggested to be involved in the inversion of the gene encoding the

S-layer protein in L. acidophilus [Boot et al., 1996]. Such recombinases may be

activated in the GI tract and in turn activate the msa expression. Interestingly, in L.

plantarum expression of the recombinase codV is induced during passage through the

GI-tract [Bron et al., 2004a/b]. Such enzyme activities may catalyze DNA inversion

of the upstream msa element in WCFS1.

The reasons behind DNA inversion as a regulation mechanism as opposed to

promoter activation or translational control are unknown. However, DNA inversion is

an all or nothing control system and expression of several operons and even gene

islands may be controlled by one single recombinase. The identification of DNA

84

inversion as another layer of gene expression control in lactobacilli is exiting and

underlines how important experimental work and in vivo models are.

Conclusions The mannose specific adhesive phenotype of two strains of lactobacilli was analysed.

The L. plantarum 299v adheres to mannose residues with higher affinity than L.

planatarum WCFS1. A mannose specific adhesin is responsible for this phenotype

and is also a key adhesion factor during binding to the epithelial lineage as observed

by in vitro adhesion assays. The Msa is expressed with phase variation in WCFS1

whereas expression in 299v is constitutive. Phase variable expression of msa was due

to DNA inversion of a 104 bp upstream DNA element. In the ON orientation a strong

msa transcription was observed, whereas the OFF orientation arrests transcription.

The arrest of transcription may be due to formation of RNA secondary structures.

Methods

Bacterial strains and growth conditions.

The bacterial strains are listed in Table 2. L. plantarum was grown at 37º C on Man-

Rogosa-Sharp (MRS) medium (Oxoid, Hampshire, United Kingdom) without aeration

and supplemented 3 µg/mL erythromycin (Merck, Darmstadt, Germany) when

appropriate. L. lactis was grown at 30°C in M17 medium (Oxoid) supplemented with

0,5% glucose (GM17). 5-Bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-Gal)

was used at a concentration of 160 µg/ml in agar plates for L. lactis. Escherichia coli

DH10B (Invitrogen, Carlsbad, Ca) was used as an intermediate cloning host and

added 250 μg/mL erythromycin when appropriate.

DNA isolation and manipulation.

Chromosomal [Johansen & Kibenich, 1992] and plasmid DNA [O´sullivan 1993a]

from L. lactis and L. plantarum were prepared as described. E. coli plasmids were

isolated using a plasmid extraction kit from Genomed (Bad Oeynhausen, Germany) as

recommended. L. lactis and L. plantarum was made electro-competent and

transformed as described [Holo & Nes, 1995], while competent E. coli DH10B cells

were purchased (Invitrogen, Groningen, The Netherlands) and transformed as

85

recommended. DNA restriction and modification enzymes, New England Biolabs

(Beverly, MA) were used as recommended. DNA was sequenced with a Thermo

Sequenase fluorescent-labelled primer cycle sequencing kit (Amersham Pharmacia,

Uppsala, Sweden), Cy5-labeled primers, and an ALFexpress DNA sequencer

(Amersham Pharmacia).

Yeast agglutination assay

Bacteria were tested for their ability to agglutinate yeast cells as described by

[Adlerberth et al., 1996], with minor modifications. Briefly, overnight grown cultures

of same cell density were washed in PBS pH 7.4 and resuspended in the same volume

of PBS. 50 µl bacteria solution was transferred to micro plates with flat bottom wells

already containing 50 µl bakers yeast suspended in PBS. After shaking on a micro

plate shaker for 2 min, agglutination was read by visual inspection of the wells.

Intense agglutination gives a clear solution with one big clump. Wells were also inspected

under the microscope after transferring the solution to glass slides.

Cloning of msa and construction of a deletion mutant in L. plantarum 299v.

Cloning of the open reading frame of the lp_1229 homologue in L. plantarum 299v

was done using PCR and primers P1 and P2 (Table 2), which introduces terminally

BamH1 restriction sites. Template DNA was purified genomic DNA from L.

plantarum 299v. PCR products were purified using the GFXTM

purification kit

(Amersham Bio-sciences) and cloned into the pCR2.1-TOPO vector (Invitrogen) as

recommended by the manufacturer. The resulting pAMJ2075 contains the full reading

frame of msa. The msa homologue was sequenced with appropriate located primers.

The upstream msa region was sequenced using primers P3 and P4 and genomic DNA

as template. The sequence was compared to L. plantarum WCFS1 using BLAST-N.

The msa was isolated on a BamH1 fragment from pAMJ2075 and inserted into similar

sites of the low copy number expression plasmids pTRKL2v1 resulting in pMSALC.

The msa gene including the promoter from pTRKL2v1 was isolated on a PstI-FspI

fragment and inserted into compatible PstI and BsaI sites of pNZ8048 resulting in the

high copy number plasmid pMSAHC. The two plasmids pMSALC and pMSAHC were

used for plasmid complementation studies.

86

Construction of an msa deletion mutant in L. plantarum 299v was done as follows.

Two sub-fragment of lp_1229 were PCR amplified for construction of an integration

vector containing an msa with an internal deletion. Using primer-sets P3 + P4 and P5

+ P6, respectively, the region upstream the ConA-like domain and the region

downstream the ConA-like domain were amplified by PCR. PCR fragments were

introduced complementary ends (Table 2, P4 and P5 underlined) to allow annealing of

the two PCR fragments. Another PCR with primers P3 and P6 and annealed PCR sub

fragments as template DNA resulted in an assembled PCR product, which combines

the 5´- and 3´-ends of lp_1229, without the Con A-like region. This PCR product was

cloned in pCR2.1_-TOPO, leading to plasmid pAMJ2052. Besides the flanking EcoRI

restriction sites present in pCR2.1_-TOPO, the assembled PCR fragment contains an

internal EcoRI restriction site close to the 5´-end. After digestion with EcoRI the

lp_1229 ΔConA fragment from pAMJ2052 was isolated and inserted into pTN1 [Neu

et al., 2003] in the corresponding EcoRI site. The cloning was done in L. lactis

MG1363 leading to plasmid, pAMJ1271. Primary transformants were verified using

PCR with primers pTN1frw and pTN1rev. The pTN1 vector replicates at 30 °C

whereas replication is arrested at 41 °C, allowing the use of the vector for single copy

integration in L. plantarum 299v. pAMJ1271 was electroporated into competent L.

87

plantarum 299v and cells were plated on MRS agar containing Ery and incubated for

48 hours at 30 °C. A colony was picked and cultured overnight at the permissive

temperature (30 °C) in MRS with Ery, then diluted 103 fold in fresh MRS without

Ery, and shifted to the non-permissive temperature (41 °C). Overnight cultures, grown

at 41 °C, were diluted 104, plated on MRS with Ery and incubated for 48 hours at

41°C to obtain single copy integrations of pAMJ1271 into the chromosome. Plasmid

integration was verified as Ery resistant clones and by PCR, using the primers P6 and

P7 (Table 2). Excision of the integrated plasmid by a single-crossover event was

induced by incubation in non-selective MRS broth at the permissive temperature (30

°C). Bacteria were spread on MRS (without Ery, 30 °C) and individual colonies were

inoculated as duplicates to micro wells (96-well-plates) containing 100 µL MRS with

and without Ery, respectively. After incubation at 30 °C cultures sensitive to Ery were

checked for their ability to agglutinate yeast cells. Ery sensitive clones represent

either msa deletion mutants or the wild-type strain. A deletions mutant, named

299vΔMsa, was identified by its lack of the agglutinating phenotype and confirmed

by sequencing the DNA region covering the deletion sites using primers P8 and P9.

Recombinant expression of the lp_1229 ConA domain and affinity purification

The Con A-like domain of lp_1229 was PCR amplified using primers P10 and P11

and L. plantarum 299v genomic DNA. The primers introduced terminal BglII and

SalI restriction sites allowing for cloning in the L. lactis expression vector

pKWY2589-773 (Table 2). Culture supernatants of an L. lactis-pAMJ1294 culture

was analysed by SDS-PAGE. Affinity purification was done by applying crude

culture supernatants to prepacked mannose columns (Amersham). Non-binding

proteins was removed by washing with PBS and mannose binding proteins was eluted

by addition of 20 mM mannose in PBS. 0,5 mL samples was collected and analysed

by 14% SDS-PAGE.

Quantification of surface displayed Msa using ELISA

Purified recombinant Con A-like domain was mixed with Freund’s incomplete

adjuvant and used for immunisation of a rabbits (6 doses of 0,15 mg protein) by

DakoCytomation A/S, Denmark. The Msa specific ELISA was done as follows.

Overnight grown cultures of same cell density were washed in PBS (pH 7.4 ) and

resuspended in 1/10 volume of PBS. Serial 2 folds dilutions were made in PBS. 50 µl

88

of each cell dilution were transferred in doublets into microtiter plates containing 50

µl PBS with 3 % skim milk and 1/1000 diluted anti-serum raised against the Con A-

like domain. Samples were incubated for 60 min. After 2 times washing the cells to

remove unbound antibodies, pellets were re-suspended in 100 µl PBS buffer

containing 3% skim milk and 1/4000 dilution of alkaline phosphatase conjugated anti-

rabit IgG according to the manufacture (KPL, Tåstrup, Denamrk). After incubation

for another 60 min, cell were washed 3 times in PBS and re-suspended in 100 µl

alkaline phosphatise substrate (KPL). Reaction were stopped by the addition of 100

µl 2,5% EDTA, typically after 10 min at room temperature. Cells were precipitated

and 100 µl supernatant were transferred to another microtiter plate. The absorbance at

595 nm was measured and plotted as a function of cell density

Adhesion of lactobacilli to Caco-2 cells.

Caco-2 cells were cultured in DMEM (Lonza, Vallenbæk Strand, Denmark)

supplemented with 10% fetal calf serum (Lonza), 1% penicillin/streptomycin (PS)

(Lonza), 1% glutamax (Lonza) and 1% non-essential amino acids (NEAA) (VWR,

Rødovre, Denmark), for 4-6 hour at 37°C until settled. Medium was changed and the

cells were further incubated. After two days ¾ of the medium were shifted with fresh

medium, and at 70% confluence cells were trypsinated by washing 2 times in PBS

without Ca2+ and Mg2+ and addition of 1 ml trypsin in 9 ml DMEM for 3-7 min until

cells were released. Caco-2-cells at a concentration of 4x105 cells per well were

transferred to NunclonTM 6-well Multidishes (NUNC, Roskilde, Denmark) for growth

prior to the adhesion assay. Cells were incubated 16 days with exchange of medium

every second day. Lactobacilli were grown for 2 days and diluted to an optical density

of 0.5 measured at 600 nm using conditioned medium corresponding to approximately

5x108 CFU/ml. Prior to the adhesion assay the Caco-2 cells were washed twice in 3

ml PBS and subsequently incubated at 37°C for 30 min with 750 µl PBS. An equal

volume of the bacterial suspension was hereafter added. For each strain 3 wells were

used. Adhesion was allowed to occur for 1 hour after which the non-adhesive bacteria

were removed by 4 washes in PBS. To visualize the bacteria each well was fixed for

5-10 min with methanol followed by staining with Giemsa (Sigma-Aldrich, Brøndby,

Denmark). For each well the bacteria were inspected visually under bright field

microscopy using a 100x objective. Ten randomly selected regions of each well were

acquired as digital images, and the bacterial count in each micrograph was manually

89

counted. For each well the average number of bacteria per micrograph was calculated.

The average number of bacteria from all three wells representing the same strain was

calculated by averaging the average numbers obtained from the individual wells.

Cloning of invertible fragments in promoter cloning vector pAK80 and –galactosidase assays

The 499 bp fragment (containing the 104 bp invertible element) upstream lp_1229

was PCR amplified using P12 and P13, which introduced terminal BamH1 restriction

sites. Genomic DNA from L. plantarum WCFS1 and WCFS1 agg+ was used as

template DNA. The amplified fragment was cloned in both orientations into the

promoter probe vector pAK80 [Israelsen et al., 1995] in fusion with a promoter less

lacLM gene. The ligation mixtures were transformed into L. lactis MG1363, plated on

GM17 Ery and clones containing the BamH1 insert were verified by PCR using

primers P12 and P13. The orientation of the insert was verified by sequencing using

primer P14. The resulting four plasmids were pAK80-ON (the WCFS1 agg+

orientation), pAK80-ON(-) (Reverse fragment orientation) and pAK80-OFF (the

WCFS1 orientation) and pAK80-OFF(-) (Reverse fragment orientation).

Transformants were tested on GM17 Ery supplemented with X-gal and β-

galactosidase activity was measured on exponentially growing cultures as described

[Israelsen et al., 1995].

Transcriptional analysis by northern blot and primer extension

Overnight grown cultures of L. plantarum were harvested and total RNA was

extracted with Pure RNA Isolation Kit (Roche, Hvidovre, Denmark) following the

instruction of the manufacturer. Total RNA was separated on a 1% (w/v) agarose gels.

Blotting, hybridisation and washing conditions were as previously described [Madsen

et al., 1996]. A 786 bp PCR fragment, obtained by P10 and P11, covering the Con A

part of lp_1229 was labelled with [ 32P]ATP and T4 polynucleotide kinase and used

as a probe in Northern hybridisation.

Primer P13 was used in a primer extension reaction to determine the start of

transcription of msa and examine the potential existence of a promoter within the

intergenic region between msa and lp_1228. P13 was end-labelled with [ 32P]ATP

90

and T4 polynucleotide kinase. The Primer Extension System-AMV Reverse

Transcriptase (Promega, Birkerød, Denmark) was used for the primer extension

reaction. The primer was annealed to total RNA by incubating at 52 °C for 20 min and

cooling at room temperature for 10 min. The extension reaction was conducted at 42

°C for 30 min. The extension product was separated on an 8 % polyacrylamide

denaturing gel and sized using a radiolabelled X174 DNA marker ladder. A sequence

reaction was done on plasmid pAK80-ON using primer P13 and the dideoxy method

as described by the manufacturer [Promega].

Authors' contributions

JG drafted the manuscript. BH and JG did the experiments. SMM gave valuable

suggestions through out the study and edited the manuscript. KH carried out the cell

adhesion assay and edited the manuscript. SA provided valuable know how on

lactobacilli and its adhesion factors. PR contributed to the transcriptional analysis.

Acknowledgements We acknowledge Annemette Jørgensen and Ulla Poulsen Bioneer A/S for excellent

technical assistance.

References


Ahrne S, Nobaek S, Jeppsson B, Adlerberth I, Wold AE, Molin G: The normal Lactobacillus flora of healthy human rectal and oral mucosa. J Appl Microbiol 1998, 85: 88-94.

Antikainen J, Anton L, Sillanpaa J, Korhonen TK: Domains in the S-layer protein CbsA of Lactobacillus crispatus involved in adherence to collagens, laminin and lipoteichoic acids and in self-assembly. Mol Microbiol 2002, 46: 381-394.

Bergonzelli GE, Granato D, Pridmore RD, Marvin-Guy LF, Donnicola D, Corthesy-Theulaz IE: GroEL of Lactobacillus johnsonii La1 (NCC 533) is cell surface associated: potential role in interactions with the host and the gastric pathogen Helicobacter pylori. Infect Immun 2006, 74: 425-434.

Bernet MF, Brassart D, Neeser JR, Servin AL: Lactobacillus acidophilus LA 1 binds to cultured human intestinal cell lines and inhibits cell attachment and cell invasion by enterovirulent bacteria. Gut 1994, 35: 483-489.

Blaser MJ: Microbes adapt to inner space. Nat Med 2006, 12: 994-996.

Boekhorst J, Wels M, Kleerebezem M, Siezen RJ: The predicted secretome of Lactobacillus plantarum WCFS1 sheds light on interactions with its environment. Microbiology 2006, 152: 3175-3183.

91

Boot HJ, Kolen CP, Pouwels PH: Interchange of the active and silent S-layer protein genes of Lactobacillus acidophilus by inversion of the chromosomal slp segment. Mol Microbiol 1996, 21: 799-809.


Bron PA, Grangette C, Mercenier A, de Vos WM, Kleerebezem M: Identification of Lactobacillus plantarum genes that are induced in the gastrointestinal tract of mice. J Bacteriol 2004, 186: 5721-5729.


Gasson MJ: Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 1983, 154: 1-9.

Gotteland M, Brunser O, Cruchet S: Systematic review: are probiotics useful in controlling gastric colonization by Helicobacter pylori? Aliment Pharmacol Ther 2006, 23: 1077-1086.

Granato D, Perotti F, Masserey I, Rouvet M, Golliard M, Servin A et al.: Cell surface-associated lipoteichoic acid acts as an adhesion factor for attachment of Lactobacillus johnsonii La1 to human enterocyte-like Caco-2 cells. Appl Environ Microbiol 1999, 65: 1071-1077.

Granato D, Bergonzelli GE, Pridmore RD, Marvin L, Rouvet M, Corthesy-Theulaz IE: Cell surface-associated elongation factor Tu mediates the attachment of Lactobacillus johnsonii NCC533 (La1) to human intestinal cells and mucins. Infect Immun 2004, 72: 2160-2169.

Henderson IR, Owen P, Nataro JP: Molecular switches--the ON and OFF of bacterial phase variation. Mol Microbiol 1999, 33: 919-932.

Holo H, Nes IF: Transformation of Lactococcus by electroporation. Methods Mol Biol 1995, 47: 195-199.

Hynonen U, Westerlund-Wikstrom B, Palva A, Korhonen TK: Identification by flagellum display of an epithelial cell- and fibronectin-binding function in the SlpA surface protein of Lactobacillus brevis. J Bacteriol 2002, 184: 3360-3367.

Israelsen H, Madsen SM, Vrang A, Hansen EB, Johansen E: Cloning and partial characterization of regulated promoters from Lactococcus lactis Tn917-lacZ integrants with the new promoter probe vector, pAK80. Appl Environ Microbiol 1995, 61: 2540-2547.

Johansen E, Kibenich A: Isolation and characterization of IS1165, an insertion sequence of Leuconostoc mesenteroides subsp. cremoris and other lactic acid bacteria. Plasmid 1992, 27: 200-206.

Kleerebezem M, Boekhorst J, van KR, Molenaar D, Kuipers OP, Leer R et al.: Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 2003, 100: 1990-1995.

Mack DR, Ahrne S, Hyde L, Wei S, Hollingsworth MA: Extracellular MUC3 mucin secretion follows adherence of Lactobacillus strains to intestinal epithelial cells in vitro. Gut 2003, 52: 827-833.

Madsen SM, Albrechtsen B, Hansen EB, Israelsen H: Cloning and transcriptional analysis of two threonine biosynthetic genes from Lactococcus lactis MG1614. J Bacteriol 1996, 178: 3689-3694.

Miyoshi Y, Okada S, Uchimura T, Satoh E: A mucus adhesion promoting protein, MapA, mediates the adhesion of Lactobacillus reuteri to Caco-2 human intestinal epithelial cells. Biosci Biotechnol Biochem 2006, 70: 1622-1628.

Neu T, Henrich B: New thermosensitive delivery vector and its use to enable nisin-controlled gene expression in Lactobacillus gasseri. Appl Environ Microbiol 2003, 69: 1377-1382.

O'Sullivan DJ, Klaenhammer TR: Rapid Mini-Prep Isolation of High-Quality Plasmid DNA from Lactococcus and Lactobacillus spp. Appl Environ Microbiol 1993a, 59: 2730-2733.

O'Sullivan DJ, Klaenhammer TR: High- and low-copy-number Lactococcus shuttle cloning vectors with features for clone screening. Gene 1993b, 137: 227-231.

Perdigon G, Vintini E, Alvarez S, Medina M, Medici M: Study of the possible mechanisms involved in the mucosal immune system activation by lactic acid bacteria. J Dairy Sci 1999, 82: 1108-1114.

Pretzer G, Snel J, Molenaar D, Wiersma A, Bron PA, Lambert J et al.: Biodiversity-based identification and functional characterization of the mannose-specific adhesin of Lactobacillus plantarum. J Bacteriol 2005, 187: 6128-6136.

92

Rolfe RD: The role of probiotic cultures in the control of gastrointestinal health. J Nutr 2000, 130: 396S-402S.

Sillanpaa J, Martinez B, Antikainen J, Toba T, Kalkkinen N, Tankka S et al.: Characterization of the collagen-binding S-layer protein CbsA of Lactobacillus crispatus. J Bacteriol 2000, 182: 6440-6450.

Sun J, Le GW, Shi YH, Su GW: Factors involved in binding of Lactobacillus plantarum Lp6 to rat small intestinal mucus. Lett Appl Microbiol 2007, 44: 79-85.

Toba T, Virkola R, Westerlund B, Bjorkman Y, Sillanpaa J, Vartio T et al.: A Collagen-Binding S-Layer Protein in Lactobacillus crispatus. Appl Environ Microbiol 1995, 61: 2467-2471.

vall-Jaaskelainen S, Lindholm A, Palva A: Surface display of the receptor-binding region of the Lactobacillus brevis S-layer protein in Lactococcus lactis provides nonadhesive lactococci with the ability to adhere to intestinal epithelial cells. Appl Environ Microbiol 2003, 69: 2230-2236.

Vimal DB, Khullar M, Gupta S, Ganguly NK: Intestinal mucins: the binding sites for Salmonella typhimurium. Mol Cell Biochem 2000, 204: 107-117.

Zhao H, Li X, Johnson DE, Blomfield I, Mobley HL: In vivo phase variation of MR/P fimbrial gene expression in Proteus mirabilis infecting the urinary tract. Mol Microbiol 1997, 23: 1009-1019.

Additional files

Additional file 1 – DNA sequence of the intergenic sequence upstream the msa gene

The translation start sites of lp_1228 and lp_1229 are indicated. The putative msa promoter with -10 and -35 box is shown. Transcription start site is indicated (*) and the one base pair change (C →T) in strain 299v is indicated.

93

Chapter 5

Recombinant Production of Immunological Active Peanut

Allergen Ara h 2 using Lactococcus lactis

94

Recombinant Production of Immunological Active Peanut Allergen Ara h 2 using Lactococcus lactis

Jacob Glenting1§, Lars K. Poulsen2, Kentaro Kato3, Søren M. Madsen1, Hanne

Frøkiær4, Camilla Wendt1, Helle W. Sørensen1

1 Bioneer A/S, DK-2970 Hørsholm, Denmark 2 Allergy Clinic 7751, National University Hospital, DK-2100 Copenhagen, Denmark 3 Department of Medical Biochemistry and Genetics, University of Copenhagen, DK-

2200 Copenhagen, Denmark 4 Biocentrum DTU, DK-2800 Kgs. Lyngby, Denmark

§Corresponding author

Email addresses: JG: [email protected]

LK: [email protected]

KK: [email protected]

SM: [email protected]

HF: [email protected]

CW: [email protected]

HS: [email protected]

95








Abstract

Background

Natural allergen sources can supply large quantities of authentic allergen mixtures for

use as immunotherapeutics. However, such extracts are complex, difficult to define,

vary from batch to batch, and are in some cases too immunological active. The use of

recombinant expression systems for allergen production can alleviate some of these

issues. Several allergens have been tested in high-level expression systems and in

most cases show immunereactivity comparable to their natural counterparts. The

gram positive lactic acid bacterium Lactococcus lactis is an attractive microorganism

for use in the production of protein therapeutics. L. lactis is considered food grade,

free of endotoxins, and is able to secrete the heterologous product together with few

other native proteins. Hypersensitivity to peanut represents a serious allergic problem.

Some of the major allergens in peanut have been described. However, for therapeutic

usage more information about the individual allergenic components is needed. In this

paper we report recombinant production of the Ara h 2 peanut allergen using L. lactis.

Results

A synthetic ara h 2 gene was cloned into an L. lactis expression plasmid containing

the P170 promoter and the SP310mut2 signal sequence. Flask cultures grown

overnight showed secretion of the 17 kDa Ara h 2 protein. A batch fermentation

resulted in 40 mg/L recombinant Ara h 2. Purification of Ara h 2 from the culture

supernatant was done by hydrophobic exclusion and size separation. Mass

spectrometry and N-terminal analysis showed a recombinant Ara h 2 of full length

and correctly processed by the signal peptidase. The immunological activity of

recombinant Ara h 2 was analysed by ELISA using antibodies specific for native Ara

h 2. The recombinant Ara h 2 showed comparable immunereactivity to that of native

Ara h 2.

Conclusions

Recombinant production of Ara h 2 using L. lactis can offer high yields of secreted,

full length and immunologically active allergen. The L. lactis expression system can

support recombinant allergen material for immunotherapy and component resolved

allergen diagnostics.

96

Background The objective of allergen immunotherapy is to counteract an already established

pathological immune response against the administered protein. The most frequently

used form in the clinic is specific immunotherapy, which involves repeated

subcutaneous injection of increasing doses of adjuvant-bound allergen extract [1].

Recently, needle free and mucosal vaccination such as sublingual administration has

been successfully exploited using allergens from house dust mite and cat dander [2]

and the grass allergen Phl p 5 [3,4]. Allergen immunotherapy relies on repeated

immunizations for a relative long period. The therapeutic strategy, particularly the

sublingual variant, requires therefore relatively large amounts of allergen and

demands high quality standards of the source of allergen. Most therapies use allergen

extracts from natural sources, which contain the native (iso)forms of the proteins.

Crude extracts prepared from natural sources can however be difficult to standardise

and contain difficult to define mixtures of allergens (reviewed by [5]). In addition to

the protein allergens, they also contain non-allergenic proteins and other substances.

Recombinant produced allergens may increase the safety of immunotherapy and

overcome some of the problems associated with natural allergen extracts [6]. The

most important allergens have been cloned and sequenced. The use of these genes for

recombinant allergen expression can facilitate i) high yield allergen production with

low biological or batch to batch variation ii) material for refined and component-

resolved allergy diagnosis iii) allergen preparations of defined purity and composition

iv) development of engineered hypoallergens that show reduced binding to IgE. The

drawbacks of recombinant production are associated with lack of product-authenticity

and that some therapies require multiple allergens, some of which are yet unknown.

High-level expression systems for production of allergens have been developed.

These are based on either bacteria or eukaryotes. The birch pollen Bet v 1 allergen has

been produced using the T7 based Escherichia coli system with a yield of 8-10 mg

purified allergen per litre culture [7]. Plants have also been tested as recombinant

allergen factories. The olive pollen allergen, Ole e 3 and Ole e 8, was produced in

Arabidopsis thaliana and showed similar biological activities as their natural

97

counterpart [8]. The choice of recombinant expression system for allergen production

is a balance between product yield, authenticity and immunereactivity, and cost

effectiveness. In most cases, the immunereactivity of recombinant allergens is

comparable with their natural counterparts (reviewed by [9]). Microbial based

expression systems are simple and cost effective. However, more complicated and

eukaryotic based expression systems are necessary where post translational

modifications like glycosylation play an essential role in the allergenicity of the

protein. An example is the Cit s 1 from oranges bearing a single N-glycan, which is

the target of the IgE response to this protein [10]. Recombinant systems with differing

post translational machineries may therefore produce allergens with same amino acid

composition, but with different allergenecity. Therefore has different expression

systems been compared. The dust mite allergen Lep d 2, causing disease amongst

farmers, was produced in E. coli and adherent cell cultures at yields of 1 mg/L and 4

mg/L, respectively [11]. Both types of recombinant Lep d 2 products showed

immune-reactivity similar to that of natural Lep d 2 when tested against patient sera.

Interestingly, this indicates that IgE epitopes are retained in the recombinant proteins,

each produced by two very different expression hosts.

Recombinant production of genetically engineered hypoallergens represents a step

towards improvement of immunotherapy. By site directed mutagenesis allergens with

epitopes that display reduced IgE binding, but retained T-cell epitopes, can be

developed [12]. The approach was first developed for dust mite allergen Der p 2

where engineered protein variants without disulfide bonds showed reduced IgE

binding [13]. A recombinant engineered hypoallergen variant of Bet v 1 was better

tolerated compared to its natural counterpart in a clinical trial with allergic patients

[14]. It is generally accepted that engineered hypoallergens opens for immunotherapy

with higher, but still safe, doses of allergen.

Peanut allergy is one of the most severe food allergies with a prevalence of 0,6% in

UK, USA and Australia [15-17]. Treatment relies today on strict avoidance of peanuts

in the diet and ready-to-access self injectable epinephrine. The major allergens in

peanut have been identified as Ara h 1-3 [18], where Ara h 2 is the most frequently

recognised [19]. More than 90% of the patients allergic to peanut have IgE specific to

Ara h 2 [20]. The gene encoding Ara h 2 has been used for plasmid DNA vaccination.

98

Oral administration of chitosan formulated plasmid DNA induced Ara h 2 expression

in the intestinal epithelium and reduced Ara h 2 specific IgE activity in a mouse

model [21]. Although direct inoculation of plasmid DNA into the patient is a simple

vaccine strategy, the technique still suffers from efficacy and safety issues [22].

Recombinant engineering and production of the Ara h 2 protein has also been tested.

A hypoallergen variant of Ara h 2 with altered IgE epitopes showed reduced IgE-

binding compared to the wild type Ara h 2 [23]. Expression of Ara h 2 in E. coli

showed similar conformational features of the heterologous and native Ara h 2

product [24]. The E. coli produced Ara h 2 was however accumulated intracellularly

and an affinity-tag was added to the N-terminal of the protein to facilitate purification.

The present study tested the use of the lactic acid bacteria Lactococcus lactis as

microbial production host of the major peanut allergen Ara h 2. L. lactis is a gram

positive bacterium with food grade status due to its long history of use in the

manufacturing of dairy products. We used a plasmid based and secretory expression

system [25] for production of recombinant Ara h 2 (rAra h 2) by 1 litre batch

fermentation and a simple purification protocol. The rAra h 2 was characterized by

mass spectrometry, N-terminal sequencing and immunological analysis.

MKFNKKRVAIATFIALIFVSFFTISSIQDAQAAERS Ara h 2

Signal peptidase

SP310mut2

Figure 1 - Anatomy of Ara h 2 expression vector pAMJ399-arah2L. lactis expression vector pAMJ399. The ara h 2 gene is inserted in fusion with the signal sequence (SP310mut2). The cleavage site of the signal peptidase and the N-terminal of Ara h 2 isindicated. P170: L. lactis P170 promoter, Terminator: transcriptional terminator, ermB: gene conferringerythromycin resistance, repD repE: L. lactis replication unit, p15A: E. coli replication region.

99

28 kDa

17 kDa

14 kDa

M CHW4 CHW9 Neg

Figure 2 - Expression analysisSDS-PAGE of culture supernatants of L. latis strain CHW4 and CHW9 harbouring the pAMJ399-arah2 and L. lactis strain MG1363 containing plasmid pAMJ399 as negative control (Neg). A clear band of17 kDa represents secreted rAra h 2.

Results

Expression and purification of rAra h 2 allergen using L. lactis in batch fermentation

To support a high yield expression of the 17 kDa Ara h 2 protein, the gene was

chemically synthesised and codon optimized to use most abundant tRNAs of L. lactis.

Potential translation inhibiting secondary RNA structures was also deleted from the

sequence. The 465 bp synthetic gene was inserted into the pAMJ399 expression

vector in translational fusion with the signal sequence (Fig. 1). The signal sequence

encodes a 32 amino acid signal peptide leading to protein secretion through the sec-

dependent pathway [26]. Upon secretion the signal peptide was cleaved off releasing

rAra h 2 with a synthetic N-terminal (AERS) extension to the extra cellular milieu

(Fig. 1). Expression of Ara h 2 was first tested in flask experiments using different

mutant strain backgrounds. The two strains tested were derivatives of L. lactis

MG1363, made by chemical mutagenesis, and identified as high secretors of

recombinant proteins (data not shown). Supernatants from overnight cultures of

CHW4 and CHW9, both harbouring pAMJ399-arah2, were analysed and a distinct

band of 17 kDa were be detected in the Ara h 2 transformed strains opposed to the

negative control strain harbouring pAMJ399 without the Arah2 gene inserted (Fig. 2).

Strain CHW9 seemed to produce more recombinant product than CHW4 and was

selected for further testing in 1 litre batch fermentation. Fermentation using CHW9

100

Figure 3 - Growth profile of Ara h 2 producing L. lactis CHW9 during fermentationCell densities plotted against time of fermentation. Samples are taken every hour as indicated by numbers. an

d synthetic medium gave a maximum cell density of OD600 =12 (Fig. 3). As the cell

density increased and reached transition to stationary phase the P170 promoter was

induced and a 17 kDa secreted product was accumulated in the culture supernatant

(Fig. 4). SDS-PAGE analysis of culture supernatant (sample 11) and a molecular

weight standard of known concentration was made to estimate the concentration of

rAra h 2 to 40 mg/L (data not shown). No intracellular accumulation of rAra h 2 was

detected (data not shown), which suggests that the secretion of rAra h 2 is not the

rate-limiting step in the protein production.

Figure 4 - Culture samples from fermentation Crude samples of supernatants corresponding to sample 1-11. Arrow indicates rAra h 2.

Purification of rAra h 2

Although rAra h 2 is secreted to the culture supernatant as the dominant protein in the

solution (Fig. 4) a simple purification procedure was established to exclude the native

protein secreted by L. lactis. Culture supernatant from the fermentation broth was

applied to a hydrophobic interaction column and fractions containing the 17 kDa rAra

h 2 band were collected (Fig. 5). To separate the 17 kDa protein from other proteins

still present after the hydrophobic exclusion the pooled fractions were purified by two

101

cycles of filtrations to eliminate proteins below 10 kDa and above 50 kDa. The two

purification steps resulted in a purified rAra h 2 product as analysed by SDS-PAGE

(Fig. 5).

40 41 42 43 44 45 46 47 48 M rAra h 2100

80

12

36

6

Characterization of rAra h 2 The use of non-native hosts for heterologous protein production can affect the product

authenticity due to degradation of the recombinant product, imprecise cleavage of the

signal peptide and non-native posttranslational modifications. Therefore, experiments

were set-up to determine if the rAra h 2 corresponded to the theoretical predictions.

The molecular weight of rAra h 2 was determined by MALDI-TOF mass

spectroscopy and showed a peak at m/z 18438,98 g/mol (Fig. 6), which corresponds to

the theoretical value of Ara h 2 being 18437,46 g/mol. No other dominant protein

forms were detected in the tested range of 10-30 kDa. To verify that the pre-protein

was precisely cleaved by the L. lactis signal peptidase, the rAra h 2 was N-terminally

sequenced. The first 20 amino acids were identified as

AERSRQQWELQGDRRCQSQL, which match the N-terminal of rAra h 2 and the in

silico prediction of the cleavage site (see additional file 1). In summary, L. lactis

secreted a full length rAra h 2, which corresponds to the theoretical molecular weight.

The rAra h 2 does however contain a four amino acid extension (AERS) originating

from the SP310mut2 signal peptide, which is not present in the native Ara h 2.

Figure 6 - MS-MALDI TOF analysis of rAra h 2Mass spectrum analysis of rAra h 2 showing the dominant peak at 18438,98 Da in the analysed areabetween 10-30 kDa.

102

Biological activity of heterologous Ara h 2

Bacteria lack the post translational machinery responsible for protein modifications

like glycosylation and phosphorylation. L. lactis may therefore produce an rAra h 2

with predicted amino acid composition, but with a different conformation than that of

native Ara h 2. Therefore the immunological equivalence of rAra h 2 and Ara h 2,

isolated from a natural source, was compared by ELISA using rabbit antisera raised

against purified, native Ara h 2. The rAra h 2 showed strong serum-reactivity

compared to the control nuclease protein produced by a similar L. lactis strain (Fig.

7). The immune reactivity was therefore related to rAra h 2 and not to other native

proteins produced by L. lactis and still present in the rAra h 2 solution. The cross-

reactivity of rAra h 2 with sera raised by native Ara h 2 was evidenced by perfect

parallelism with a peanut extract indicating that most if not all antibody epitopes of

native Ara h 2 is present in rAra h 2 (Fig. 7). This demonstrates the immunological

activity of the L. lactis produced rAra h 2.

Figure 7 – Immune reactivity of rAra h 2Sandwich ELISA using sera from rabbits immunized with purified native Ara h 2. Concentrations on the primary axisreefers to protein concentration of peanut extract, rAra h 2, and nuclease. The L. lactis produced rAra h 2 was run

in two-fold dilutions as was the negative protein control (an L. lactis produced nuclease). Curve Std. peanut: extract of Ara h 2 from natural source, curve rAra h 2: L. lactis produced rAra h 2, curve Sham: L. lactis produced nuclease. Error bars represent SEM of duplicate determinations.

Discussion Some of the problems associated with allergen immunotherapeutics based on natural

sources can be overcome using gene engineering and recombinant production of

103

allergens. The increasing number of sequenced genomes allows for uncomplicated

cloning of the allergen encoding cDNA and expression to be done in suitable

organisms. Synthetic genes encoding allergens can be synthesised at an affordable

prize and be optimized for the specific expression host to support high yield

production. Furthermore, in a recombinant production strategy it is relatively simple

to engineer hypoallergens with weaker IgE binding epitopes [12,23], which can lower

the side effects associated with immunotherapy. Natural allergen products can be

heterogeneous from batch to batch and contain undesired and unknown substances.

Recombinant production can support a standardised and defined allergen production.

However, recombinant production relies on the use of non-native hosts, which may

affect the conformational features of the allergen protein. An altered protein

conformation or posttranslational modifications like glycosylation or disulfide bond

formation can influence the immunereactivity.

In this study we tested L. lactis for production of a major peanut allergen. The

advantages of L. lactis include its food grade status, lack of endotoxins, and high

protein secretion capacities. Furthermore, the genetic tools for generation of allergen

variants and high trough put screening are well developed.

To ensure high level expression of rAra h 2 the gene was synthetically designed and

optimized for expression in L. lactis. Furthermore, a high copy number pAMJ399

expression vector was used to increase the gene dosage and maximise rAra h 2

production. Both CHW4 and CHW9 mutant strains efficiently secreted a rAra h 2 of

expected molecular mass. The tendency of higher rAra h 2 expression of CHW9

compared to CHW4 cannot be pinpointed to any known mutation as both strains are

from a library of chemically and randomly mutagenesised strains and selected as high

protein-secretors. The performance of L. lactis strain CHW9 with pAMJ399-arah2

was tested in batch fermentation with yield of biomass of OD600=12. The successful

production of 40 mg/L of recombinant Ara h 2 is to our knowledge the highest

reported yield in a microbial-based production of Ara h 2. The secretion of rAra h 2 to

a supernatant with few native proteins simplifies the downstream processing and

purification. Indeed, a two-step purification using hydrophic exclusion and size

separation resulted in a pure product as analysed by SDS-PAGE. The potential of

adding purification tags could be applied and may ease purification even more.

104

However, a protein tag adds an extra processing step of tag removal by protease

cleavage.

Allergens may be complex structures and need higher organisms than prokaryotes for

production of a proper folded and modified allergen. In addition can recombinant

production lead to product degradation due to proteolytic activity or premature

termination of translation. The protein was therefore further characterized. SDS-

PAGE analysis of culture supernatant showed only one band corresponding to the 17

kDa of rAra h 2 and suggests that no truncated forms of the protein were produced

(Fig. 2). Furthermore, mass spectrometry analysis of non-digested rAra h 2 gave a

molecular mass comparable to the predicted amino acid composition (Fig. 6). This is

also in accordance with native Ara h 2, which is without posttranslational

modifications like glycosylation and phosphorylation [27]. However, native Ara h 2

contain four conserved disulphide bridges [28], but the L. lactis produced rAra h 2 is

free of disulphide bridges as observed by comparing the migration length of reduced

and non-reduced rAra h 2 samples on SDS-PAGE (data not shown). Lack of

intramolecular disulphide formation during protein synthesis in L. lactis has been

reported [29], which is in accordance with the fact that no putative thiol-disulphide

oxidoreductases has been identified in L. lactis so far. However, spontaneous

formation of disulphide bonds may explain how L. lactis can produce biological

active proteins where S-S bridges are essential eg. the IL-12 cytokine [30]. The N-

terminal of rAra h 2 was also characterized. The native signal peptide of Ara h 2 was

interchanged by the L. lactis optimized signal peptide SP310mut2. This ensured

efficient secretion and correct processing of the signal peptide. Indeed, SP310mut2

was cleaved as predicted (see additional file 1) releasing a rAra h 2 with native amino

acid composition, but with a synthetic AERS N-terminal extension.

Although the capacity of L. lactis to secrete high levels of rAra h 2 is attractive the

biological activity of rAra h 2 is essential. To compare the immunoreactivity of the

rAra h 2 to that of natural Ara h 2 an ELISA was established. The L. lactis produced

rAra h 2 showed similar immune reactivity as that of natural Ara h 2. This indicates

that most of the antibody epitopes are retained in the recombinant Ara h 2. However,

to establish biological equivalence with native Ara h 2 a large panel of sera from

clinically allergic patients should be tested by radioallergosorbent test (RAST) and in

105

basophil histamine release experiments. Such clinical documentation was, however,

beyond the scope of the present study.

The use of rAra h 2 with wild type amino acid composition for immunotherapy may

be associated with side effects due to its high immunereactivity. By site directed

mutagenesis Ara h 2-hypoallergens with weaker IgE epitopes may be obtained. Such

hypoallergens with few amino acid changes will have close resemblance to rAra h 2

and should be as highly expressed as the wild type allergen in L. lactis. However, the

lack of disulphide bonds in rAra h 2 may be sufficient to lower its IgE binding

capacity as deletion of S-S bridges has been shown to disrupt IgE eitopes and lower

the immunereactivity of proteins [13]. If this is the case for the L. lactis produced

rAra h 2 is to be shown. The rAra h 2 may also be a valuable tool for component

resolved allergy diagnosis.

L. lactis has also been used for the in vivo delivery of therapeutic components.

Steidler et al. showed that L. lactis could deliver active IL-10 to the gastrointestinal

tract by orally feeding mice with recombinant bacteria [31], which recently has been

tested in a clinical study with positive results on patients suffering from Crohn´s

disease [32]. In situ delivery of allergy components has also been evaluated using live

recombinant lactic acid bacteria. Intranasal administration of birch pollen Bet v 1-

secreting-L. lactis reduced allergen specific IgE in a mouse model [33]. Recently,

immunomodulatory L. casei has been developed to secrete the important milk allergen

β-lactoblobulin [34], which currently is being tested as a therapeutic treatment in an

allergy mouse model. The rAra h 2 producing strain presented in this study could also

be used for in vivo allergen delivery, where the intrinsic adjuvant capacity of lactic

acid bacteria is exploited.

Conclusions Recombinant allergen production can alleviate some of the problems associated to

allergen products based on natural sources. Recombinant allergens represent an

important tool in component resolved allergy analysis, development of engineered

hypoallergens, and production of allergens with defined composition and purity.

106

Peanut allergy is an example where recombinant allergens can provide material for

component resolved analysis of the allergic response and guide the engineering of

safer hypoallergens. We showed that L. lactis can sustain high-level expression of 40

mg/L of full-length rAra h 2 and developed a simple purification protocol. The

produced rAra h 2 was biologically active and showed similar immunereactivity as

natural Ara h 2. The relatively simple production and purification process of rAra h 2

makes L. lactis an interesting expression system for the production of the major

allergens present in peanuts.

Methods Bacteria and media

L. lactis was grown at 30°C in M17 medium (Oxoid, Hampshire, United Kingdom)

supplemented with 0,5% glucose (GM17). When appropriate, 1 μg/mL erythromycin

(Merck, Darmstadt, Germany) was added. Fed batch fermentation was done using 1 L

L. lactis culture, which was cultivated for 25 h in LM3-30 [35] in 2 L fermentors at

30° C. 5 M KOH was automatically added to maintain pH at 6 and the agitation rate

was set at 300 rpm. Growth was monitored by measuring OD600.

Construction of rAra h 2 producing L. lactis

Plasmid DNA from L. lactis was prepared as described [36]. L. lactis was made

electro-competent and transformed as described [37]. DNA restriction and

modification enzymes, New England Biolabs (Beverly, MA) were used as

recommended. The gene encoding Ara h 2 was synthetically manufactured by Geneart

(Regensburg, Germany) using the Geneoptimizer program for optimal expression in

L. lactis. The 459 bp gene was added terminally Bgl II and Sal I restriction enzyme

recognition sites by the manufacturer for cloning in pAMJ399 in fusion to the signal

sequence SP310mut2 (Fig. 1). L. lactis clones with pAMJ399 harbouring the Ara h 2

gene in Bgl II and Sal I sites were verified using PCR and primers specific for the

synthetic Ara h 2 gene. A PCR positive clone was tested for expression of rAra h 2 by

overnight growth in GM17 supplemented with 1 µg/mL erythromycin. This plasmid

was named pAMJ399-arah2. Transformation of the mutant strains PSM565 and

DOL7 with pAMJ399-arah2 resulted in CHW4 and CHW9, respectively.

107

Expression and purification of rAra h 2

Samples were taken every hour from fermentation and divided into cell pellet and

supernatant by centrifugation at 8000 x g for 5 min and frozen at -20˚ C. Culture

supernatants were TCA precipitated and analysed. Cell pellet was lysed using glass

beads and cellular debris was removed by centrifugation at 15000 x g for 15 min at 4˚

C and the intracellular protein fraction was isolated as the soluble fraction. Intra- and

extracellular protein fractions were analysed using 14% SDS-PAGE. Culture

supernatant from the end of fermentation was adjusted with (NH4)2SO4 to 1M and

applied to a phenyl sepharose column (GE-Heathcare, Hillerød, Denmark ) and

washed in buffer A (Tris-Cl 50mM, (NH4)2SO4 1M, pH 8). Protein was eluted with

buffer B (Tris-Cl 50mM, pH 8) and collected automatically in 5-mL fractions using

Frac-100 (Pharmacia Biotech, Freiburg, Germany). Fractions were analysed by SDS-

PAGE. Selected fractions containing rAra h 2 were pooled and desalted by a PD-10

Sephadex column (Sigma-Aldrich, Brøndby, Denmark) and small and large protein

impurities were removed by Amicon ultra-4 columns (Millipore, Billerica, MA) with

cut off values of 10 kDa and 50 kDa.

MALDI-TOF analysis

Molecular weight of rAra h 2 was determined by matrix-assisted laser desorption

ionization time-of flight mass spectrometry (MALDI-TOF MS). 0,5 µL of purified

protein (0,5-1 ng) was applied on a tip and mixed with 1 µL of 1,3 mg/mL of HABA

(2-(4-Hydroxyphenylazo)benzoic acid from Sigma-Aldrich) dissolved in H2O/CH3CN

(1:1) solution. The mass spectra were obtained on a Voyager-DETM Pro instrument

(Applied Biosystems, Weiterstadt, Germany) operating at an accelerating voltage of

20 kV in the linear mode with the delayed extraction setting. Recorded data were

processed using Data ExplorerTM 4.0 software.

N-terminal sequencing

Purified rAra h2 was loaded onto a 14% SDS-PAGE gel. After separation the protein

was electro blotted onto a PVDF membrane at 175mA for 1 h in a Semi Dry Blotter II

(Kem-En-Tec, Tåstrup, Denmark) using a 10mM CAPS blotting solution containing

6% methanol. The membrane was stained with 0,1% Coomassie Brilliant blue in 1%

acetic acid and 60% methanol for 1 min and destained for 5 min in 40% methanol

108

until the background was light purple/blue. The membrane was rinsed in deionised

water and air-dried. The relevant protein band was excised and subjected to N-

terminal sequencing using Edman degradation on a Procise Protein Sequencer 494

(Applied Biosystems).

ELISA

A 96-well ELISA plate was coated overnight at 4° C with rabbit antiserum raised

against purified and native Ara h 2 (a kind gift from Dr. W-M Becker,

Forschungszentrum Borstel, Germany). Excess coating buffer was discarded and

blocking buffer (1 mg/mL gelatine in washing buffer) was added to the wells and the

plate was placed on a shaker at room temperature for 1 h. The plate were washed 3

times in washing buffer (NaCl 80,0 g/L, KCl 2,0 g/L, KH2PO4 2,0 g/L, Na2HPO4 14,4

g/L, Tween 206 g/L, pH 7,4). A 100 µL of positive control and 100 µL of rAra h2

were added to the wells and serial diluted and incubated 2 h at 37° C. Peanut extract

served as positive control and was obtained using unshelled peanuts (Brüder Kunz

GmbH), which were grounded in a mortar with a pestle and prepared as described

[38] and kept at -20˚ C. Negative control in ELISA was TE buffer and culture

supernatant of a L. lactis MG1363 strain expressing the Staphyloccous aureus

nuclease (21 kDa). After incubation the plate were washed three times in washing

buffer. Subsequently, 100 µL of biotinylated Ara h 2 specific antibody, diluted

1:2500, was added to each well and plates were incubated for 2 h at 37° C. Extravidin

diluted 1:5000 in washing buffer was applied to each well (100 µL) and left at room

temperature on a shaker for 30 min. Following 3 washes in washing buffer the

substrate (o-Phenylenediamine dihydrochloride tablets dissolved in water added

hydrogen peroxide) was added and the plate incubated 10-15 min. at room

temperature in the dark. The reaction was stopped with 150 µL stop solution. The

absorbance at 450 nm was measured.

Competing interests None

109

Authors' contributions JG designed and supervised the experiments. JG drafted the manuscript. HW made

most of the experiments and edited the manuscript. SM and CW gave helpful

contributions on expression and fermentation experiments. HF gave valuable know-

how on allergens. KK supervised the mass spectrometry analysis. LP supervised the

immunological studies of rAra h 2 and edited the manuscript.

Acknowledgements We would like to thank Ulla Poulsen and Anne Cathrine Steenbjerg Bioneer A/S for

excellent technical assistance. We would like to acknowledge Ole Cai Hansen

Bioneer A/S for assistance during the purification process of the allergen.

Development of the Ara h 2 assay was supported by the EU Commission

(FAREDAT, QLRT-2001-00301) and the work of Louise Bjerremann Jensen and

Mona H. Pedersen, Allergy Clinic at the National University Hospital, is gratefully

acknowledged. We thank Pia Wium. Novozymes A/S for the N-terminal sequencing

of allergens.

References

1. Norman PS: Immunotherapy: 1999-2004. J Allergy Clin Immunol 2004, 113: 1013-1023.

2. Wilson DR, Torres LI, Durham SR: Sublingual immunotherapy for allergic rhinitis. Cochrane Database Syst Rev

2003, CD002893.

3. Dahl R, Kapp A, Colombo G, de Monchy JG, Rak S, Emminger W et al.: Efficacy and safety of sublingual

immunotherapy with grass allergen tablets for seasonal allergic rhinoconjunctivitis. J Allergy Clin Immunol

2006, 118: 434-440.

4. Khinchi MS, Poulsen LK, Carat F, Andre C, Hansen AB, Malling HJ: Clinical efficacy of sublingual and

subcutaneous birch pollen allergen-specific immunotherapy: a randomized, placebo-controlled, double-blind,

double-dummy study. Allergy 2004, 59: 45-53.

5. Spangfort MD, Larsen JN: Standardization of allergen-specific immunotherapy vaccines. Immunol Allergy Clin

North Am 2006, 26: 191-1vi.

6. Valenta R, Niederberger V: Recombinant allergens for immunotherapy. J Allergy Clin Immunol 2007, 119: 826-

830.

110

7. Hoffmann-Sommergruber K, Susani M, Ferreira F, Jertschin P, Ahorn H, Steiner R et al.: High-level expression and

purification of the major birch pollen allergen, Bet v 1. Protein Expr Purif 1997, 9: 33-39.

8. Ledesma A, Moral V, Villalba M, Salinas J, Rodriguez R: Ca2+-binding allergens from olive pollen exhibit

biochemical and immunological activity when expressed in stable transgenic Arabidopsis. FEBS J 2006, 273:

4425-4434.

9. Chapman MD, Smith AM, Vailes LD, Arruda LK, Dhanaraj V, Pomes A: Recombinant allergens for diagnosis and

therapy of allergic disease. J Allergy Clin Immunol 2000, 106: 409-418.

10. Poltl G, Ahrazem O, Paschinger K, Ibanez MD, Salcedo G, Wilson IB: Molecular and immunological

characterization of the glycosylated orange allergen Cit s 1. Glycobiology 2007, 17: 220-230.

11. Olsson S, van Hage-Hamsten M, Whitley P, Johansson E, Hoffman DR, Gafvelin G et al.: Expression of two

isoforms of Lep d 2, the major allergen of Lepidoglyphus destructor, in both prokaryotic and eukaryotic

systems. Clin Exp Allergy 1998, 28: 984-991.

12. Vrtala S, Focke-Tejkl M, Swoboda I, Kraft D, Valenta R: Strategies for converting allergens into hypoallergenic

vaccine candidates. Methods 2004, 32: 313-320.

13. Smith AM, Chapman MD: Reduction in IgE binding to allergen variants generated by site-directed

mutagenesis: contribution of disulfide bonds to the antigenic structure of the major house dust mite allergen

Der p 2. Mol Immunol 1996, 33: 399-405.

14. van Hage-Hamsten M, Kronqvist M, Zetterstrom O, Johansson E, Niederberger V, Vrtala S et al.: Skin test

evaluation of genetically engineered hypoallergenic derivatives of the major birch pollen allergen, Bet v 1:

results obtained with a mix of two recombinant Bet v 1 fragments and recombinant Bet v 1 trimer in a Swedish

population before the birch pollen season. J Allergy Clin Immunol 1999, 104: 969-977.

15. Emmett SE, Angus FJ, Fry JS, Lee PN: Perceived prevalence of peanut allergy in Great Britain and its

association with other atopic conditions and with peanut allergy in other household members. Allergy 1999, 54:

380-385.

16. Sicherer SH, Munoz-Furlong A, Burks AW, Sampson HA: Prevalence of peanut and tree nut allergy in the US

determined by a random digit dial telephone survey. J Allergy Clin Immunol 1999, 103: 559-562.

17. Woods RK, Thien F, Raven J, Walters EH, Abramson M: Prevalence of food allergies in young adults and their

relationship to asthma, nasal allergies, and eczema. Ann Allergy Asthma Immunol 2002, 88: 183-189.

18. de Leon MP, Rolland JM, O'Hehir RE: The peanut allergy epidemic: allergen molecular characterisation and

prospects for specific therapy. Expert Rev Mol Med 2007, 9: 1-18.

19. Koppelman SJ, Wensing M, Ertmann M, Knulst AC, Knol EF: Relevance of Ara h1, Ara h2 and Ara h3 in peanut-

allergic patients, as determined by immunoglobulin E Western blotting, basophil-histamine release and

intracutaneous testing: Ara h2 is the most important peanut allergen. Clin Exp Allergy 2004, 34: 583-590.

20. Stanley JS, King N, Burks AW, Huang SK, Sampson H, Cockrell G et al.: Identification and mutational analysis of

the immunodominant IgE binding epitopes of the major peanut allergen Ara h 2. Arch Biochem Biophys 1997,

342: 244-253.

111

21. Roy K, Mao HQ, Huang SK, Leong KW: Oral gene delivery with chitosan--DNA nanoparticles generates

immunologic protection in a murine model of peanut allergy. Nat Med 1999, 5: 387-391.

22. Glenting J, Wessels S: Ensuring safety of DNA vaccines. Microb Cell Fact 2005, 4: 26.

23. King N, Helm R, Stanley JS, Vieths S, Luttkopf D, Hatahet L et al.: Allergenic characteristics of a modified peanut

allergen. Mol Nutr Food Res 2005, 49: 963-971.

24. Lehmann K, Hoffmann S, Neudecker P, Suhr M, Becker WM, Rosch P: High-yield expression in Escherichia coli,

purification, and characterization of properly folded major peanut allergen Ara h 2. Protein Expr Purif 2003,

31: 250-259.

25. Madsen SM, Arnau J, Vrang A, Givskov M, Israelsen H: Molecular characterization of the pH-inducible and

growth phase-dependent promoter P170 of Lactococcus lactis. Mol Microbiol 1999, 32: 75-87.

26. Ravn P, Arnau J, Madsen SM, Vrang A, Israelsen H: Optimization of signal peptide SP310 for heterologous

protein production in Lactococcus lactis. Microbiology 2003, 149: 2193-2201.

27. Lehmann K, Schweimer K, Reese G, Randow S, Suhr M, Becker WM et al.: Structure and stability of 2S albumin-

type peanut allergens: implications for the severity of peanut allergic reactions. Biochem J 2006, 395: 463-472.

28. Barre A, Borges JP, Culerrier R, Rouge P: Homology modelling of the major peanut allergen Ara h 2 and

surface mapping of IgE-binding epitopes. Immunol Lett 2005, 100: 153-158.

29. van de GM, van der Wal FJ, Kok J, Venema G: Lysozyme expression in Lactococcus lactis. Appl Microbiol

Biotechnol 1992, 37: 216-224.

30. Bermudez-Humaran LG, Langella P, Cortes-Perez NG, Gruss A, Tamez-Guerra RS, Oliveira SC et al.: Intranasal

immunization with recombinant Lactococcus lactis secreting murine interleukin-12 enhances antigen-specific

Th1 cytokine production. Infect Immun 2003, 71: 1887-1896.

31. Steidler L, Hans W, Schotte L, Neirynck S, Obermeier F, Falk W et al.: Treatment of murine colitis by Lactococcus

lactis secreting interleukin-10. Science 2000, 289: 1352-1355.

32. Braat H, Rottiers P, Hommes DW, Huyghebaert N, Remaut E, Remon JP et al.: A phase I trial with transgenic

bacteria expressing interleukin-10 in Crohn's disease. Clin Gastroenterol Hepatol 2006, 4: 754-759.

33. Repa A, Grangette C, Daniel C, Hochreiter R, Hoffmann-Sommergruber K, Thalhamer J et al.: Mucosal co-

application of lactic acid bacteria and allergen induces counter-regulatory immune responses in a murine

model of birch pollen allergy. Vaccine 2003, 22: 87-95.

34. Hazebrouck S, Pothelune L, Azevedo V, Corthier G, Wal JM, Langella P: Efficient production and secretion of

bovine beta-lactoglobulin by Lactobacillus casei. Microb Cell Fact 2007, 6: 12.

35. Glenting J, Madsen SM, Vrang A, Fomsgaard A, Israelsen H: A plasmid selection system in Lactococcus lactis and

its use for gene expression in L. lactis and human kidney fibroblasts. Appl Environ Microbiol 2002, 68: 5051-

5056.

36. O'Sullivan DJ, Klaenhammer TR: Rapid Mini-Prep Isolation of High-Quality Plasmid DNA from Lactococcus

and Lactobacillus spp. Appl Environ Microbiol 1993, 59: 2730-2733

37. Holo H, Nes IF: Transformation of Lactococcus by electroporation. Methods Mol Biol 1995, 47: 195-199.

38. Poulsen LK, Pedersen MH, Platzer M, Madsen N, Sten E, Bindslev-Jensen C et al.: Immunochemical and biological

quantification of peanut extract. Arb Paul Ehrlich Inst Bundesamt Sera Impfstoffe Frankf A M 2003, 97-105.

112

Additional files

Additional file 1 – Signal peptide prediction

The amino acid sequence of rAra h 2 is submitted to the prediction server http://www.cbs.dtu.dk/services/SignalP/ using the gram positive bacteria as settings. The C-score is the cleavage site score. For each position in the submitted sequence, a C-score is reported, which should only be significantly high at the cleavage site. Y-max is a derivative of the C-score combined with the S-score resulting in a better cleavage site prediction than the raw C-score alone. This is due to the fact that multiple high-peaking C-scores can be found in one sequence, where only one is the true cleavage site. The cleavage site is assigned from the Y-score where the slope of the S-score is steep and a significant C-score is found. The S-mean is the average of the S-score, ranging from the N-terminal amino acid to the amino acid assigned with the highest Y-max score, thus the S-mean score is calculated for the length of the predicted signal peptide.

113

Chapter 6

Conclusions and concluding remarks

114

Conclusions and concluding remarks

The thesis has investigated the use of LAB in vaccine development. The applications of LAB as

tools in vaccine strategies are several. Therefore the present PhD thesis also investigated these

versatile usages. The study brings several new findings and observations to the scientific

community:

• Published literature which discuss the safety aspect of vaccines, where researchers may find

some guides and counsel for the development of safer vaccines

• An antibiotic and endotoxin free plasmid DNA production system, which is the first

published alternative to E. coli based production systems of plasmid DNA vaccines

• Increased knowledge on protein anchors for vaccine display and their affect on the

immunomodulatory effect of the host strain

• Analysis of adhesins in lactobacilli and description of their regulation including

identification of a new regulatory mechanism in L. plantarum

• Development of a microbial production system for high level expression of a major peanut

allergen

Although the developed L. lactis plasmid DNA vaccine system was successful in induction of

immune responses in mice it will most likely not replace present E. coli systems. The efficiency in

terms of yield of E. coli is superior. However, if the L. lactis system was further developed using

novel high copy number plasmids the complete lack of endotoxins and antibiotic resistance genes

will be attractive. An important observation was the difference in induction of cellular immune

responses, to the encoded HIV-1 gp120, using the L. lactis and E. coli plasmids, respectively. As

the levels of gp120 expression from the two plasmids were comparable it may suggest that the L.

lactis DNA contain sequences that inhibit CTL responses. Therefore, the L. lactis system may also

be useful to elucidate the cellular and humoral activating signals of prokaryotic DNA.

115

For the live vaccine expressing Bet v 1 animal studies should be conducted to establish the usage of

the developed tools. Here the use of LAB strains that show different immune polarizing capacity

should be tested. Their in vitro stimulation of dendritic cells could then be compared with their

ability to redirect undesired immune allergy responses. Indeed, establishment of an in vivo

correlation of dendritic cell based assays is needed.

The mannose specific adhesion and its regulation is an interesting feature of the L. plantarum.

Several key issues need to be addressed. The details and molecular component involved in the

regulation should be analysed. Over expression of the codV may shed light on the involvement of

recombinases in the regulation. Specific mutagenesis of the inverted repeats could also pin point

how vital these are in the regulation and in formation of stem loop structures. Furthermore, the

genome of L. plantarum WCFS1 is published and could be analyzed of the described invertible

elements in other gene operons. The findings also illustrate that bacterial surface structures may

have a different composition depending on the environment and laboratory conditions may breed

non-adhesive lactobacilli that in the human body are strongly adhesive.

L. lactis was efficient in producing a peanut allergen. Although the immunereactivity was compared

to native allergen, the reactivity towards sera from patients hypersensitive to peanut should be

established. At the present time we are inviting collaborators to analyse the immunereactivity of L.

lactis produced Ara h 2 in more depth using patient sera.

116

Appendix

A plasmid selection system in Lactococcus lactis and its use for gene

expression in L. lactis and human kidney fibroblasts

117

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Oct. 2002, p. 5051–5056 Vol. 68, No. 100099-2240/02/$04.00�0 DOI: 10.1128/AEM.68.10.5051–5056.2002Copyright © 2002, American Society for Microbiology. All Rights Reserved.

A Plasmid Selection System in Lactococcus lactis and Its Use for GeneExpression in L. lactis and Human Kidney Fibroblasts

Jacob Glenting,1* Søren M. Madsen,1 Astrid Vrang,1 Anders Fomsgaard,2 and Hans Israelsen1

Department of Lactic Acid Bacteria, Biotechnological Institute, DK-2970 Hørsholm,1 and Department of Virology,Statens Serum Institut, DK-2300 Copenhagen,2 Denmark

Received 17 April 2002/Accepted 16 July 2002

We report the development of a nonantibiotic and nonpathogenic host-plasmid selection system based onlactococcal genes and threonine complementation. We constructed an auxotrophic Lactococcus lactisMG1363�thr strain which carries deletions in two genes encoding threonine biosynthetic enzymes. To achieveplasmid-borne complementation, we then constructed the minimal cloning vector, pJAG5, based on the genesencoding homoserine dehydrogenase-homoserine kinase (the hom-thrB operon) as a selective marker. Usingstrain MG1363�thr, selection and maintenance of cells carrying pJAG5 were obtained in threonine-freedefined media. Compared to the commonly used selection system based on erythromycin resistance, thedesigned complementation system offers a competitive and stable plasmid selection system for the productionof heterologous proteins in L. lactis. The potential of pJAG5 to deliver genes for expression in eukaryotes wasevaluated by insertion of a mammalian expression unit encoding a modified green fluorescent protein. Thesuccessful delivery and expression of genes in human kidney fibroblasts indicated the potential of the designednonantibiotic host-plasmid system for use in genetic immunization.

The continuous discovery of new vaccines and therapeuticsposes a challenge to systems available for heterologous proteinproduction. A suitable choice of host and production condi-tions is important for the manufacture of a pharmaceutical-grade product. We have recently developed a plasmid expres-sion system for use in the lactic acid bacterium Lactococcuslactis (28). In this paper, we present a nonantibiotic alternativefor selection of plasmid-harboring bacterial cells for use inheterologous gene expression.

Due to its food-grade status, L. lactis is attractive for bothproduction and live delivery of vaccines and therapeutics. Re-cently, several examples of nondairy applications of this bac-terium have been reported, and the development of plasmid-borne gene expression systems has made it possible to use L.lactis for the production of heterologous proteins (1, 5, 7, 11,20, 21). Furthermore, L. lactis has been placed in a new class oflive bacterial vaccine carriers derived from gram-positive, non-pathogenic, and noninvasive bacteria (reviewed in reference33). Recently, live delivery was accomplished by feeding miceimmunogen-synthesizing L. lactis organisms, thus obtainingmucosal delivery of the Helicobacter pylori urease subunit Bantigen (22) and of the pneumococcal type 3 capsular polysac-charide antigen (12). The delivery of therapeutics using liverecombinant L. lactis has been demonstrated by in situ secre-tion of interleukin-10 for treatment of colitis in mice (37).

Although L. lactis is generally regarded as safe, this statuscan be compromised by the introduction of foreign DNA nec-essary for the synthesis of recombinant proteins. Usually, high-copy-number plasmids are used for high-level expression ofrecombinant proteins (1, 5, 7, 11, 20, 21). A simple way to

prevent plasmid loss is to use plasmid-encoded antibiotic re-sistance markers and grow the bacteria in the presence ofantibiotics. The chief drawbacks of this approach are the po-tential loss of selective pressure as a result of antibiotic deg-radation (as in the case of �-lactamase) and contamination ofthe biomass or purified protein by antibiotics and resistancegenes, which is unacceptable from a medical point of view.

Alternative genetic markers for L. lactis have been devel-oped. Depending on the type of selection, they can be placedin two groups: resistance and complementation markers. Ex-amples of resistance markers that confer immunity to an addedagent, such as nisin (8) or the metal ions cadmium (Cd2�) (24)and copper (Cu2�) (23), have been designed for plasmid main-tenance. Although some strains of L. lactis are naturally resis-tant to nisin and metal ions, the dominant nature of resistancemarkers makes them versatile, as they can be used in differentlactococcal strains.

The use of auxotrophic markers is based on complementa-tion of a mutation or deletion in the host chromosome and istherefore strain specific. In L. lactis, the first example wasbased on complementation of a lacF mutant strain deficient inlactose utilization (26). In two other systems, auxotrophicmarkers complement purine- and pyrimidine-auxotrophicstrains using genes encoding nonsense tRNA suppressors (6,36). In these systems, expression of the plasmid-borne suppres-sor tRNA gene allows read-through of a nonsense mutation(s)in the genes encoding purine or pyrimidine biosynthetic en-zymes. Both systems permit selection in milk or other mediathat contain no, or small amounts of, purines or pyrimidines.

We previously constructed a threonine-auxotrophic deriva-tive of the L. lactis MG1614 strain containing an internal de-letion in the hom-thrB operon (27). The bicistronic hom-thrBoperon encodes the homoserine dehydrogenase and homo-serine kinase catalyzing two of the five steps converting aspar-

* Corresponding author. Mailing adress: Department of Lactic AcidBacteria, Biotechnological Institute, Kogle Alle 2, DK-2970 Hørsholm,Denmark. Phone: (45) 45 16 04 44. Fax: (45) 45 16 04 55. E-mail:[email protected].

5051118

tate to threonine. Thus, a hom-thrB strain cannot synthesizethreonine and therefore needs an external source of threonine.

In this paper, we report the construction of a plasmid selec-tion system in L. lactis based on complementation of a threo-nine-auxotrophic strain using the hom-thrB operon. The plas-mid is based only on a lactococcal replicon and the hom-thrBoperon. We demonstrate its use for heterologous gene expres-sion in L. lactis and for delivery of genes for in vitro expressionin mammalian cells.

MATERIALS AND METHODS

Strains and growth conditions. Strains and plasmids are listed in Table 1.Unless otherwise stated, Escherichia coli was grown at 37°C in Luria-Bertanimedium (35), while L. lactis was grown at 30°C in M17 medium (Oxoid, Basing-stoke, United Kingdom) supplemented with 0.5% glucose (GM17) or in syn-thetic LM3-30 medium (A. Vrang, unpublished data) containing 3% glucose andall amino acids except aspartic acid. When appropriate, either 100 �g of ampi-cillin/ml or 250 �g of erythromycin (Ery)/ml was added for E. coli and 1 �g ofEry/ml was added for L. lactis. 5-Bromo-4-chloro-3-indolyl-�-D-galactopyrano-side (X-Gal) was used at a concentration of 160 �g/ml in agar plates for L. lactis.One-liter L. lactis cultures were cultivated for 30 h in LM3-30 in 2-liter fermen-tors at 30°C; 5 M KOH was automatically added to maintain the pH at 6, and theagitation rate was set at 300 rpm. Growth was monitored by measuring theoptical density at 600 nm, and samples of supernatants were taken every hour.Complemented MG1363�thr prototrophs were selected on 1.5% agarose platescontaining LM3-30 without threonine using cells that had been washed in 0.9%NaCl to remove threonine.

DNA isolation and manipulation. Chromosomal (17) and plasmid (30) DNAsfrom L. lactis were prepared as described previously. E. coli plasmids wereisolated using a plasmid extraction kit from Genomed (Bad Oeynhausen, Ger-many) as recommended by the manufacturer. L. lactis was made electrocompe-tent and transformed as described previously (14), while competent E. coliDH10B cells were purchased (Invitrogen, Groningen, The Netherlands) andtransformed as recommended by the manufacturer. DNA restriction and mod-ification enzymes (New England Biolabs, Beverly, Mass.) were used as recom-mended by the manufacturer.

Plasmid DNA was sequenced with a Thermo Sequenase fluorescently labeled

primer cycle-sequencing kit (Amersham Pharmacia, Uppsala, Sweden), Cy5-labeled primers, and an ALFexpress DNA sequencer (Amersham Pharmacia).

Construction of the threonine-auxotrophic L. lactis MG1363�thr strain. Thethreonine-auxotrophic MG1363�thr strain was constructed as described previ-ously(27) using the integration vector pSMA507 (Table 1). This plasmid containsan Ery resistance marker, a plasmid origin of replication from E. coli, the lacLMgenes, and the L. lactis hom-thrB genes from MG1614 lacking the distal 310-bpend of hom and the proximal 57 bp of thrB (27). Plasmid pSMA507 was elec-troporated into L. lactis MG1363, and Ery-resistant transformants were isolatedand cultured in GM17 without Ery for 75 generations. Cells were plated ontoGM17 plates with X-Gal and screened for loss of the plasmid and concomitantloss of �-galactosidase activity. Chromosomal DNAs from white colonies andfrom an MG1363 wild-type colony were amplified to screen for deletions in thehom-thrB operon by PCR using the primers pThr-frw1 and pThr-rev1 (Table 2).PCR of strain MG1363 DNA gave rise to a 2,083-bp fragment, while DNA fromthe deletion mutant gave a 1,716-bp fragment.

TABLE 1. Strains and plasmids

Bacterium or plasmid Characteristic(s) Reference or source

L. lactisMG1363 10MG1614 10MG1614�thr �hom �thrB 27MG1363�thr �hom �thrB This work

E. coli DH10B Laboratory strain Invitrogen

PlasmidspGEM7Zf(�) lacZ bla Promega, Southampton, United KingdompCR2.1 lacZ bla kan InvitrogenpEGFP-N1 CMV promoter-EGFP-SV40 poly(A); kan Clontech, Palo Alto, Calif.pG3E pGEM-3Zf(�)::1.2 kb; ermL 32pSMA231 pGEM-7Zf(�)::1.8 kb; �hom-thrB 27pSMA507 �hom �thrB lacLM ermL 27pSMBI109 Citrate replicon-P170 promoter-SP310mut2; nucB ermL A. Vrang, unpublishedpAK80 Citrate replicon 16pJAG1 pCR2.1::0.6 kb; 5� end of hom This workpJAG2 pGEM-7Zf(�)::2.4 kb; hom-thrB This workpJAG3 pG3E::1.8 kb; citrate replicon This workpJAG4 pJAG3::2.4 kb; hom-thrB This workpJAG5 Citrate replicon; hom-thrB This workpJAG6 pJAG5::0.9 kb; P170 promoter-SP310mut2; nucB This workpJAG7 pCR2.1::1.6 kb; CMV promoter-EGFP-SV40 pA This workpJAG8 pJAG4::1.6 kb; CMV promoter-EGFP-SV40 pA This work

TABLE 2. Nucleotide sequences of PCR primers

Primer Sequencea

pThr-frw1 (XbaI) .......5�-TATCGTCTAGACTGATTAATCTGTCAGTAAAATAGAAG-3�

pThr-rev1 ....................5�-GCTACTTCTAAATTATTTGTC-3�pThr-rev2 (EcoRI).....5�-ATTAAGAATTCCCAACAGATGTGTAATTT

TATCAGATGAAAATGAATTAGCCAAAGTTCTTAAAATAGGAATACCCCC-3�

pCMV-frw (SpeI).......5�-TATCGACTAGTTAGTTATTAATAGTAATCAATTACGGGG-3�

pBGH-rev (SpeI) .......5�-TATCGACTAGTTGATGAGTTTGGACAAACCACAACT-3�

pNuc-frw (SacI) .........5�-TTCGCGAGCTCGAGGGGAAGTAATT-3�pNuc-rev (KpnI).........5�-TATCGGGTACCCGATCTAAAAATTATAAA

AGTGCCA-3�pM13-frw.....................5�-GTAAAACGACGGCCAGT-3�pM13-rev.....................5�-CAGGAAACAGCTATGAC-3�

a Enzyme recognition sequences are underlined, and the altered EcoRI site inpThr-rev2 is italicized.

5052 GLENTING ET AL. APPL. ENVIRON. MICROBIOL.

119

Cloning of the L. lactis hom-thrB operon. Part of the hom-thrB operon ofMG1614 was cloned in pSMA231, which contains 1,800 bp of the 3� end of thehom gene and the complete thrB gene (27). The 5� end of hom, including thepromoter region, was PCR amplified with the primers pThr-frw1 and pThr-rev2using MG1614 chromosomal DNA as a template. pThr-frw1 was designed tointroduce an XbaI recognition sequence, and pThr-rev2 was designed to annealto the region located 403 to 480 bp downstream of the start codon of hom. Thisregion includes two EcoRI sites located 408 and 471 bp downstream of the homstart codon. To remove the former EcoRI site, pThr-rev2 was further designedto introduce an A-to-T substitution at this position (Table 2). The resulting PCRproduct was cloned into pCR2.1, giving rise to pJAG1, which was sequencedusing the sequencing primers pM13-frw and pM13-rev. The 5� end of hom wasmoved from pJAG1 into pSMA231 using XbaI and EcoRI, which resulted inpJAG2, harboring the complete hom-thrB operon.

Construction of the pJAG4 E. coli-L. lactis shuttle vector containing the hom-thrB marker. Plasmid pG3E is derived from pGEM-3Zf(�) and confers Eryresistance in E. coli (32). A 1,757-bp EcoRI fragment from pAK80 containing thelactococcal minimal replicon from the citrate plasmid (31) was inserted into theEcoRI site of pG3E. The resulting pJAG3 conferred Ery resistance on trans-formed MG1363�thr cells. pJAG2 was digested with XbaI and HindIII, and thecomplete hom-thrB operon was isolated on a 2,401-bp fragment and inserted intosimilarly digested pJAG3 to give pJAG4.

Construction of the pJAG5 L. lactis cloning vector. HaeII digestion of pJAG4produced a 4,636-bp fragment containing the citrate replicon, the hom-thrBoperon, and the polylinker. To favor self-religation, the isolated fragments wereligated in a large volume. The resulting vector, pJAG5 (Fig. 1), showed comple-mentation of the MG1363�thr strain on LM3-30 plates without threonine.

Construction of the pJAG6 expression vector containing the Staphylococcusaureus nuclease gene. We have constructed plasmid pSMBI109, which contains agene cassette harboring the inducible P170 promoter (A. Vrang et al., unpub-lished), the optimized signal peptide SP3mut2 (34; P. Ravn, J. Arnau, S. M.Madsen, A. Vrang, H. Israelsen, unpublished data), and the S. aureus nucBnuclease (Nuc) gene (4). This gene cassette was PCR amplified using the pNuc-frw and pNuc-rev primers, which introduced terminal 5� SacI and 3� KpnIrecognition sequences. The resulting 914-bp fragment was digested with SacI andKpnI and ligated to the similarly digested pJAG5, resulting in pJAG6.

Construction of pJAG8 for eukaryotic expression of the green fluorescentprotein. The gene encoding the red-shifted variant of the green fluorescentprotein (EGFP) was isolated from pEGFP-N1 using PCR and the primerspCMV-frw and pBGH-rev. Both primers contained terminal SpeI sequences.The resulting �1,650-bp fragment included the cytomegalovirus (CMV) pro-moter and enhancer regions and the simian virus 40 (SV40) polyadenylationsignal sequence (SVpA). This fragment was cloned into pCR2.1, producingpJAG7. The EGFP expression cassette was isolated as an SpeI fragment frompJAG7 and ligated into XbaI-digested pJAG5 to produce pJAG8.

Plasmid stability. To test the segregational stability of pJAG5 during expo-nential growth, transformed MG1363�thr cells were grown for 100 generationsunder nonselective conditions. Every 10 generations, diluted samples were cul-tured on plates with and without threonine. During fermentation, segregationalstability was tested at the transition to stationary phase and again 15 h after theonset of stationary phase. Strains with pJAG6 were tested on LM3-30 mediumwith and without threonine, and strains with pSMBI109 were tested on GM17medium with and without Ery.

Nuclease activity determinations. Nuc activities were determined by incuba-tion of culture supernatants with sonicated salmon DNA as a substrate followedby precipitation in ice-cold perchloric acid and measurement of absorbance at260 nm (Vrang et al., unpublished). One unit of Nuc is defined as the amount ofnuclease that will produce 1 �mol of acid-soluble polynucleotide per min fromnative DNA.

In vitro expression of EGFP from pJAG8 in human kidney cells. Adherenthuman 293 kidney fibroblast cells were incubated in RPMI medium (Gibco, NewYork, N.Y.) supplemented with 10% fetal calf serum for 24 h at 37°C in thepresence of 5% CO2. The cells were transfected with pJAG8 using the Effectenetransfection kit (Qiagen, Hilden, Germany) as described by the manufacturer.Expression of EGFP was visualized by fluorescence microscopy at 488 nm 24 hafter transfection.

RESULTS

Construction of a threonine-auxotrophic L. lactis MG1363�thrstrain. To construct MG1363�thr, we used the integrationvector pSMA507, which harbors a truncated hom-thrB operonand lacLM and is unable to replicate in L. lactis. First, trans-formed cells viable on GM17 plates containing X-Gal and Eryshould have pSMA507 integrated into their chromosomes dueto homologous recombination into the hom-thrB genes andshould form blue colonies. Second, without antibiotic pressure,cells whose chromosomes would undergo a second homolo-gous recombination between neighboring sets of hom-thrBgenes should form white Ery-sensitive colonies. Following thisstrategy, genomic DNAs from 10 white colonies were analyzedby PCR, which for all isolates indicated the presence of aninternal hom-thrB deletion. Phenotypic tests on plates withoutthreonine confirmed the threonine auxotrophy of these iso-lates, one of which was named strain MG1363�thr.

Construction and testing of a hom-thrB-complementingshuttle vector, pJAG4. Details of plasmid construction aredescribed in Materials and Methods and Table 1. For replica-tion in L. lactis, the minimal replicon from the citrate plasmidof Lactococcus lactis subsp. lactis biovar diacetylactis wascloned into pG3E to give pJAG3, which conferred Ery resis-tance on transformed MG1363�thr cells. The assembled hom-thrB operon from pJAG2 was then added to give pJAG4.MG1363�thr transformed with pJAG4 carrying hom-thrB wasable to grow on medium without threonine, demonstratingcomplementation. To compare the transformation efficienciesof the hom-thrB and Ery resistance markers on pJAG4, trans-formed cells were spread on LM3-30 plates without threonineand LM3-30 plates with Ery, respectively. The experiment wasdone twice in triplicate and showed that the numbers of CFUwere similar (hom-thrB complementation, 106 � 9 CFU; Eryselection, 112 � 11 CFU) on the two types of plates, indicatingthat hom-thrB selection was as efficient as Ery selection.

Construction and testing of the minimal lactococcal comple-mentation vector, pJAG5. We aimed at constructing a minimalcomplementation vector based solely on lactococcal DNAwhich contained only complementing genes, an origin of rep-lication, and a versatile polylinker and no antibiotic resistancegene or nonlactococcal DNA, such as an E. coli origin ofreplication. Therefore, the citrate replicon, the polylinker,and hom-thrB were isolated from pJAG4 on a single frag-ment, which was religated and electroporated into strainMG1363�thr. Prototrophic colonies were isolated on LM3-30plates without threonine, and a subsequent plasmid analysisconfirmed the structure of the resulting cloning vector, pJAG5

FIG. 1. Minimal cloning vector pJAG5. The arrows indicate thedirection of transcription. repB encodes the replication protein of thecitrate replicon, and hom and thrB encode homoserine dehydrogenaseand homoserine kinase, respectively. Only unique restriction sites areindicated.

VOL. 68, 2002 PLASMID SELECTION SYSTEM IN LACTOCOCCUS LACTIS 5053

120

(Fig. 1). This plasmid was stably maintained under nonselec-tive conditions, since no plasmid loss was observed duringcultivation for 100 generations in LM3-30 medium.

Secretion of S. aureus nuclease in fermentations. The capac-ity of the designed complementation system for use in plasmid-borne gene expression was compared to that of the currentlyused selection system based on antibiotic resistance. As a re-porter gene, we used nucB fused to the optimized SP3mut2signal peptide sequence. The reporter gene was transcribedfrom the pH- and growth-phase-regulated P170 promoter (28).We analyzed the level of nuclease production using this re-porter gene in two vectors, both containing the theta-typecitrate replicon but harboring either the hom-thrB-comple-menting marker (pJAG6) or the Ery resistance marker(pSMBI109). These plasmids were tested in MG1363 andMG1363�thr under different growth conditions, and the levelsof Nuc activity were compared after 30 h of fermentation andat an optical density at 600 nm of approximately 6 (Table 3).First, the influence of a chromosomal hom-thrB deletion wasinvestigated by comparing the Nuc expression of pSMBI109-transformed MG1363 and MG1363�thr strains in the presenceof threonine. As the fermentation showed identical levels ofNuc activity in the two strains, the use of a hom-thrB deletionmutant did not affect the level of expression under nonselectiveconditions. Second, we compared pJAG6(hom�-thrB�) andpSMBI109(ermL�) in MG1363�thr under the same condi-tions. The results showed that the Nuc activities using theseplasmids in the auxotrophic strain were also identical. Third,we compared the two plasmid-host systems under selectiveconditions and found that the Nuc activity of the MG1363�thr/

pJAG6 system was slightly higher than that of the MG1363/pSMBI109 system (13.2 � 0.8 versus 10.5 � 0.5 U/ml). Thisresult indicated that endogenous synthesis of threonine did notaffect heterologous gene expression. However, the growth rateof MG1363�thr/pJAG6 was slightly (10%) impaired underthese conditions. Relief of the selective pressure by the pres-ence of threonine, however, did not lead to pJAG6-freeMG1363�thr cells, as the numbers of CFU on LM3-30 plateswith and without threonine were similar (data not shown).

The results show that the complementation system offers aneffective and stable plasmid selection system for production ofheterologous Nuc in L. lactis.

Expression of green fluorescent protein in human kidneyfibroblasts. To test whether pJAG5 was capable of transienttransfection of eukaryotic cells and gene delivery for in situexpression, we cloned the reporter cassette harboring theCMV promoter, the EGFP gene, and the SVpA polyadenyla-tion signal sequence into pJAG5, resulting in pJAG8. Intro-duction of pJAG8 into the nuclei of the cells was facilitated byusing lipid-formulated plasmid. It should be noted that pJAG8contains no eukaryotic selection marker or replication systemfor maintenance in fibroblast cells. The fibroblasts were visu-alized by phase-contrast microscopy, but only transfected cellsexpressing the cytoplasmic reporter EGFP were visible on epi-fluorescence micrographs (Fig. 2). The successful transfectionof eukaryotic cells and the detectable expression of EGFP (Fig.2) demonstrated the utility of the pJAG5 cloning vector forgene delivery and in situ expression in eukaryotic cells.

DISCUSSION

While the presence of antibiotic resistance genes on recom-binant plasmids allows for efficient selection and maintenancein transformed cells, their use is undesirable in medical bio-technology, as even trace amounts of �-lactams, such as pen-icillin, or macrolides, such as Ery, can cause anaphylactic shockin sensitized persons (9, 18). Furthermore, the continuedspread of antibiotics and their resistance genes, leading tomultiresistant pathogens, is a topic of public concern. Here, wedescribe the development of a nonantibiotic plasmid selectionsystem based on complementation for use in L. lactis. Thesystem includes (i) a minimal cloning vector (pJAG5) encodingthreonine biosynthetic enzymes and harboring a lactococcal

FIG. 2. Adherent human 293 kidney fibroblasts transfected with pJAG8. (A) Phase-contrast micrograph illuminated at 488 nm. (B) Epifluo-rescence micrograph of the same region. U indicates a nonexpressor cell, while T indicates a transfected cell expressing EGFP. Magnification,�400.

TABLE 3. Nuclease determinations during fermentation

Strain Plasmid Conditionsa Nuc (u/ml � SD)b

MG1363�thr pJAG6 �Thr 13.2 � 0.8pJAG6 �Thr 11.8 � 0.8pSMBI109 �Thr 10.9 � 0.8

MG1363 pSMBI109 �Thr 11.1 � 0.4pSMBI109 �Thr, Ery 10.5 � 0.5pSMBI109 �Thr 11.8 � 0.3

a �Thr, without threonine; �Thr, with threonine; Ery, with Ery.b Nuc activities are presented as the means of three Nuc activity determina-

tions.


121

theta-type replicon and a polylinker and (ii) a threonine-aux-otrophic L. lactis MG1363 host strain, permitting selection inthreonine-free growth media such as the synthetic LM3-30.Notably, pJAG5 is highly stable and can be maintained withoutselective pressure for 100 generations. This is most likely dueto the theta-type mode of replication, which gives structuraland segregational stability (19). Furthermore, L. lactis vectorsof the theta family have narrow host ranges (15, 25, 39), pre-venting or at least reducing horizontal plasmid transfer toother microorganisms.

Potential drawbacks of a selection system dependent oncomplementation of an auxotrophic strain are growth rate re-duction and decrease in heterologous protein productioncaused by the necessity of endogenous threonine synthesisunder selective conditions. However, both the growth profilesand the Nuc expression of the hom-thrB-complemented threo-nine-auxotrophic strain were similar in media with and withoutthreonine, indicating minimal pleiotropic effects. Furthermore,our designed antibiotic-free selection system is as effective asthe Ery-dependent system and can be employed when antibi-otic-free conditions are desired, for instance, during produc-tion of proteins or plasmid DNA for medical use or productionof recombinant L. lactis for use in live delivery of vaccines ortherapeutics.

The application of pJAG5 (Fig. 1) in eukaryotic gene ex-pression was analyzed using the green fluorescent protein (2).An optimized (3) and codon-modified (13) EGFP gene con-trolled by the constitutive CMV promoter (29) was used forhigh-level expression and brighter fluorescence. The smallgene size (714 bp) and the ease with which EGFP expressioncan be visualized (no substrate needs to be added) make itideal for studying plasmid-mediated gene transfer. As shown inFig. 2, we demonstrated EGFP expression in human kidneyfibroblasts using pJAG8 transfection. This shows that the min-imal pJAG5 vector can be used for delivery and in situ expres-sion of eukaryotic genes, which to our knowledge is the firstexample of an L. lactis-derived plasmid being successfully usedfor gene expression in eukaryotic cells. Since genetic immuni-zation (reviewed in references 38 and 40) involves the directdelivery of plasmid DNA into patients to elicit an immuneresponse toward the encoded antigen, the use of a vectorwithout an antibiotic resistance marker is obviously desirable.We are investigating the potential use of the pJAG5 vector ingene delivery for genetic immunization.

ACKNOWLEDGMENTS

We thank Lasse Vinner and Sylvie Corbet for helpful discussions.We acknowledge Anne Cathrine Steenbjerg, Annemette Jørgensen,and Birgit Johansen for excellent technical assistance and thank BjarneAlbrechtsen for critical reading of the manuscript.

REFERENCES

1. Bermudez-Humaran, L. G., P. Langella, A. Miyoshi, A. Gruss, R. T. Guerra,D. O.-L. Montes, and Y. Le Loir. 2002. Production of human papillomavirustype 16 E7 protein in Lactococcus lactis. Appl. Environ. Microbiol. 68:917–922.

2. Chalfie, M., Y. Tu, G. Euskirchen, W. W. Ward, and D. C. Prasher. 1994.Green fluorescent protein as a marker for gene expression. Science 263:802–805.

3. Cormack, B. P., R. H. Valdivia, and S. Falkow. 1996. FACS-optimizedmutants of the green fluorescent protein (GFP). Gene 173:33–38.

4. Davis, A., I. B. Moore, D. S. Parker, and H. Taniuchi. 1977. Nuclease B. Apossible precursor of nuclease A, an extracellular nuclease of Staphylococcusaureus. J. Biol. Chem. 252:6544–6553.

5. de Ruyter, P. G., O. P. Kuipers, and W. M. de Vos. 1996. Controlled geneexpression systems for Lactococcus lactis with the food-grade inducer nisin.Appl. Environ. Microbiol. 62:3662–3667.

6. Dickely, F., D. Nilsson, E. B. Hansen, and E. Johansen. 1995. Isolation ofLactococcus lactis nonsense suppressors and construction of a food-gradecloning vector. Mol. Microbiol. 15:839–847.

7. Djordjevic, G. M., and T. R. Klaenhammer. 1998. Inducible gene expressionsystems in Lactococcus lactis. Mol. Biotechnol. 9:127–139.

8. Froseth, B. R., and L. L. McKay. 1991. Molecular characterization of thenisin resistance region of Lactococcus lactis subsp. lactis biovar diacetylactisDRC3. Appl. Environ. Microbiol. 57:804–811.

9. Gallagher, D. M., and D. P. Sinn. 1983. Penicillin-induced anaphylaxis in apatient under hypotensive anaesthesia. Oral Surg. Oral Med. Oral Pathol.56:361–364.

10. Gasson, M. J. 1983. Plasmid complements of Streptococcus lactis NDO712and other lactic streptococci after protoplast-induced curing. J. Bacteriol.154:1–9.

11. Gil, M. T., I. Perez-Arellano, J. Buesa, and G. Perez-Martinez. 2001. Secre-tion of the rotavirus VP8� protein in Lactococcus lactis. FEMS Microbiol.Lett. 203:269–274.

12. Gilbert, C., K. Robinson, R. W. Le Page, and J. M. Wells. 2000. Heterolo-gous expression of an immunogenic pneumococcal type 3 capsular polysac-charide in Lactococcus lactis. Infect. Immun. 68:3251–3260.

13. Haas, J., E. C. Park, and B. Seed. 1996. Codon usage limitation in theexpression of HIV-1 envelope glycoprotein. Curr. Biol. 6:315–324.

14. Holo, H., and I. F. Nes. 1995. Transformation of Lactococcus by electropo-ration. Methods Mol. Biol. 47:195–199.

15. Horng, J. S., K. M. Polzin, and L. L. McKay. 1991. Replication and tem-perature-sensitive maintenance functions of lactose plasmid pSK11L fromLactococcus lactis subsp. cremoris. J. Bacteriol. 173:7573–7581.

16. Israelsen, H., S. M. Madsen, A. Vrang, E. B. Hansen, and E. Johansen. 1995.Cloning and partial characterization of regulated promoters from Lactococ-cus lactis Tn917-lacZ integrants with the new promoter probe vector, pAK80.Appl. Environ. Microbiol. 61:2540–2547.

17. Johansen, E., and A. Kibenich. 1992. Isolation and characterization ofIS1165, an insertion sequence of Leuconostoc mesenteroides subsp. cremorisand other lactic acid bacteria. Plasmid 27:200–206.

18. Jorro, G., C. Morales, J. V. Braso, and A. Pelaez. 1996. Anaphylaxis toerythromycin. Ann. Allergy Asthma Immunol. 77:456–458.

19. Kiewiet, R., S. Bron, K. de Jonge, G. Venema, and J. F. Seegers. 1993. Thetareplication of the lactococcal plasmid pWVO2. Mol. Microbiol. 10:319–327.

20. Kok, J. 1996. Inducible gene expression and environmentally regulated genesin lactic acid bacteria. Antonie Leeuwenhoek 70:129–145.

21. Kuipers, O. P., P. G. de Ruyter, M. Kleerebezem, and W. M. de Vos. 1997.Controlled overproduction of proteins by lactic acid bacteria. Trends Bio-technol. 15:135–140.

22. Lee, M. H., Y. Roussel, M. Wilks, and S. Tabaqchali. 2001. Expression ofHelicobacter pylori urease subunit B gene in Lactococcus lactis MG1363 andits use as a vaccine delivery system against H. pylori infection in mice.Vaccine 19:3927–3935.

23. Leelawatcharanas, V., L. G. Chia, P. Charoenchai, N. Kunajakr, C.-Q. Liu,and N. W. Dunn. 1997. Plasmid-encoded copper resistance in Lactococcuslactis. Biotechnol. Lett. 19:639–643.

24. Liu, C. Q., N. Khunajakr, L. G. Chia, Y. M. Deng, P. Charoenchai, and N. W.Dunn. 1997. Genetic analysis of regions involved in replication and cadmiumresistance of the plasmid pND302 from Lactococcus lactis. Plasmid 38:79–90.

25. Lucey, M., C. Daly, and G. Fitzgerald. 1993. Identification and sequenceanalysis of the replication region of the phage resistance plasmid pCI528from Lactococcus lactis subsp. cremoris UC503. FEMS Microbiol. Lett. 110:249–256.

26. MacCormick, C. A., H. G. Griffin, and M. J. Gasson. 1995. Construction ofa food-grade host/vector system for Lactococcus lactis based on the lactoseoperon. FEMS Microbiol. Lett. 127:105–109.

27. Madsen, S. M., B. Albrechtsen, E. B. Hansen, and H. Israelsen. 1996.Cloning and transcriptional analysis of two threonine biosynthetic genesfrom Lactococcus lactis MG1614. J. Bacteriol. 178:3689–3694.

28. Madsen, S. M., J. Arnau, A. Vrang, M. Givskov, and H. Israelsen. 1999.Molecular characterization of the pH-inducible and growth phase-depen-dent promoter P170 of Lactococcus lactis. Mol. Microbiol. 32:75–87.

29. Manthorpe, M., F. Cornefert-Jensen, J. Hartikka, J. Felgner, A. Rundell, M.Margalith, and V. Dwarki. 1993. Gene therapy by intramuscular injection ofplasmid DNA: studies on firefly luciferase gene expression in mice. Hum.Gene Ther. 4:419–431.

30. O’Sullivan, D. J., and T. R. Klaenhammer. 1993. Rapid mini-prep isolationof high-quality plasmid DNA from Lactococcus lactis and Lactobacillus spp.Appl. Environ. Microbiol. 59:2730–2733.

31. Pedersen, M. L., K. R. Arnved, and E. Johansen. 1994. Genetic analysis ofthe minimal replicon of the Lactococcus lactis subsp. lactis biovar diacety-lactis citrate plasmid. Mol. Gen. Genet. 244:374–382.

32. Petersen, A., J. Josephsen, and M. G. Johnsen. 1999. TPW22, a lactococcaltemperate phage with a site-specific integrase closely related to Streptococcusthermophilus phage integrases. J. Bacteriol. 181:7034–7042.

VOL. 68, 2002 PLASMID SELECTION SYSTEM IN LACTOCOCCUS LACTIS 5055

122

33. Pozzi, G., and J. M. Wells. 1997. Gram-positive bacteria, vaccine vehicles formucosal immunization. Springer-Verlag, Berlin, Germany.

34. Ravn, P., J. Arnau, S. M. Madsen, A. Vrang, and H. Israelsen. 2000. Thedevelopment of TnNuc and its use for the isolation of novel secretion signalsin Lactococcus lactis. Gene 242:347–356.

35. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

36. Sørensen, K. I., R. Larsen, A. Kibenich, M. P. Junge, and E. Johansen. 2000.A food-grade cloning system for industrial strains of Lactococcus lactis. Appl.Environ. Microbiol. 66:1253–1258.

37. Steidler, L., W. Hans, L. Schotte, S. Neirynck, F. Obermeier, W. Falk, W.Fiers, and E. Remaut. 2000. Treatment of murine colitis by Lactococcuslactis secreting interleukin-10. Science 289:1352–1355.

38. Tighe, H., M. Corr, M. Roman, and E. Raz. 1998. Gene vaccination: plasmidDNA is more than just a blueprint. Immunol. Today 19:89–97.

39. von Wright, A., S. Wessels, S. Tynkkynen, and M. Saarela. 1990. Isolation ofa replication region of a large lactococcal plasmid and use in cloning of anisin resistance determinant. Appl. Environ. Microbiol. 56:2029–2035.

40. Wolff, J. A., R. W. Malone, P. Williams, W. Chong, G. Acsadi, A. Jani, andP. L. Felgner. 1990. Direct gene transfer into mouse muscle in vivo. Science247:1465–1468.


123

phd thesis glenting j

Documents

management of bioneer

phd study

jacob glenting bioneer

simon jensen bioneer

lactococcus lactis

vaccine development

hiv gp12039 chapter

dna vaccine