1. Lactic Acid Bacteria in Vaccine Development
PhD thesis by Jacob Glenting
Bioneer A/S
And
Biocentrum
Technical University of Denmark
June 2007

2. 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

3. 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

4. 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

5. 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

6. 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

7. 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

8. 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

9. 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

10. 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

11. Chapter 1
General Introduction
8

12. 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

13. 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

14. 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

15. 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

16. 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

17. 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

18. 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

19. 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

20. 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

21. 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

22. 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

23. 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

24. 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

25. BioMed Central
Page 1 of 12
(page number not for citation purposes)
Microbial Cell Factories
Open AccessReview
Live bacterial vaccines – a review and identification of potential
hazards
Ann Detmer*1 and Jacob Glenting2
Address: 1Danish Toxicology Centre, Hørsholm, Denmark and 2Bioneer A/S, Hørsholm, Denmark
Email: Ann Detmer* - ad@dhigroup.com; Jacob Glenting - jag@bioneer.dk
* Corresponding author
Abstract
The use of live bacteria to induce an immune response to itself or to a carried vaccine component
is an attractive vaccine strategy. Advantages of live bacterial vaccines include their mimicry of a
natural infection, intrinsic adjuvant properties and their possibility to be administered orally.
Derivatives of pathogenic and non-pathogenic food related bacteria are currently being evaluated
as 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 that
contain the gene cassette encoding the antigen. With the increased knowledge of mucosal
immunity and the availability of genetic tools for heterologous gene expression the concept of live
vaccine vehicles gains renewed interest. However, administration of live bacterial vaccines poses
some risks. In addition, vaccination using recombinant bacteria results in the release of live
recombinant organisms into nature. This places these vaccines in the debate on application of
genetically modified organisms. In this review we give an overview of live bacterial vaccines on the
market and describe the development of new live vaccines with a focus on attenuated bacteria and
food-related lactic acid bacteria. Furthermore, we outline the safety concerns and identify the
hazards associated with live bacterial vaccines and try to give some suggestions of what to consider
during their development.
Background
Live vaccines have played a critical role from the begin-
ning of vaccinology. Indeed, the very first vaccination
experiment in the Western world was Jenner's inoculation
of a boy with the milder cowpox virus to protect against
the deadly smallpox. Although effective the technology
has safety problems associated with the risk of reversion
to a virulent organism and the possibility of causing dis-
ease in immune compromised individuals. Within the last
20 years the concept of live vaccines gains renewed inter-
est due to our increased immunological understanding
and the availability of molecular techniques making the
construction of safer live vaccines possible. This opens for
the development of new live bacterial vaccines that can
avoid the downsides of parenterally administered vaccine
because it (i) mimics the route of entry of many patho-
gens and stimulate the mucosal immune response (ii) can
be 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 down
stream processing. Broadly, live bacterial vaccines can be
classified 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 2006
Accepted: 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.
22

26. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 2 of 12
(page number not for citation purposes)
Furthermore, live bacterial vaccines can be designed to
induce an immune response to itself or to a carried heter-
ologous antigen. A non-virulent or attenuated derivative
of the pathogen is used to induce a response to the bacte-
rium itself. When used as a vaccine vehicle the bacterium
expresses 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. Both
classes of bacteria deliver the vaccine component to the
immune system whereby immunization may benefit from
the 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 antigen
but may also be allergens or therapeutics. A recent devel-
opment is the use of invasive bacteria for the delivery of
plasmid DNA vaccines to mammalian cells obtaining in
vivo synthesis of the plasmid-encoded antigen. As such,
the applications of live bacterial vaccines are extensive
and 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 commercial
live vaccine. One typhoid vaccine (Ty21a) contains live
attenuated Salmonella typhi and is administered orally
either 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 is
against 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 live
bacteria) for travellers from non-endemic regions and a
one log higher dose for residents in endemic regions (5 ×
109 live bacteria). The very few examples of live bacterial
vaccines on the market may be due to lack of success in
clinical trials. However, we believe that the safety of these
vaccines is another issue. Indeed, prophylactic vaccines
are given to healthy people and despite excellent safety
record they remain targets of un-substantiated allegations
by anti vaccine movements. Furthermore, future live vac-
cines will most likely be either targeted mutagenised or
equipped with foreign antigens and therefore considered
recombinant. 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 in
particular depend on considerations of safety issues. We
believe 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 live
vaccines and their safety aspects.
Attenuated pathogens as vaccines and vaccine vehicles
Lindberg [2] has excellently reviewed the history of live
bacterial vaccines. The first use of a live bacterial vaccine
was 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 more
efficacious V. cholerae vaccine, however still parenteral.
The first live oral V. cholerae vaccine candidate did not
appear until the 1980s. Later the V. cholerae strain CVD
103 Hg-R has been found to be both safe and immuno-
genic after a single oral dose. In 1996 a bivalent vaccine
waspresented including two strains of V. cholerae called
CVD 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. One
proposed strain was made streptomycin-dependent, but
failed to be efficacious in freeze-dried formulation [5].
Furthermore, the strain was genetically unstable and
reverted to virulence. Another S. typhi strain (Ty21a) with
a defect galE gene, as well as other not defined mutations,
requires an external source of galactose. This strain was
extensively evaluated in several field trials and has shown
excellent safety record [6]. Later, other auxotrophic strains
unable to synthesise essential compounds like aromatic
amino acids were developed and tested on human volun-
teers with variable safety and immunogenicity results [7-
10]. Attenuated live vaccines to prevent shigellosis have
also been proposed. Both genetically engineered or
selected mutants of Shigella have been tried but showed
side effects in clinical trials and points to the need of addi-
tional attenuation without hampering immunogenicity
[11-13]. Kotloff et al attenuated the guanine auxotrophic
Shigella flexneri 2a further by deleting two genes encoding
enterotoxins [14]. In a phase 1 trial this strain with inacti-
vated enterotoxin genes was better tolerated but still
immunogenic compared to the guanine auxotrophic
strain that contain active entoroxins.
Recombinant Shigella has also been proposed as a vaccine
vehicle [15]. Pathogenic Shigella has a virulence plasmid
encoding proteins involved in thesecretion apparatus and
proteins necessary for the entry process into human cells.
This invasive capacity can be used to deliver plasmid DNA
vaccines into mammalian cells [16]. Here, the delivered
plasmid DNA encodes an antigen, which is expressed by
the protein synthesis apparatus of the infected cells.
Diaminopimelate Shigella auxotrophs undergo lysis
unless 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 vaccine
vehicles of various proteins and plasmid DNA (Table 1).
In conclusion, the mimicry of natural infection makes
attenuated bacteria effective. The ability to deliver vaccine
components of different origins like e.g., HIV [15,17,18]
or piece of parasitic DNA [19] or gamete specific antigen
23

27. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 3 of 12
(page number not for citation purposes)
[20] make attenuated bacteria a versatile vaccinology tool.
However, in spite of the efforts in constructing attenuated
pathogens for use as bacterial vaccine vehicles none of
them has reached the market yet.
Lactic acid bacteria as vaccine vehicles
The potential of using lactic acid bacteria (LAB) for the
delivery of vaccine components is less exploited than
attenuated pathogens. Due to their safe status and the
availability of genetic tools for recombinant gene expres-
sion LAB are attractive for use as vaccine vehicles. Further-
more, their non-pathogenic status circumvents the need
to construct attenuated mutants. However, LAB are non-
invasive and the vaccine delivery to antigen presenting
cells may be less effective than invasive bacteria. Still, anti-
gen specific immune responses have been obtained with
several 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 can
explain a possible way to at least elicit a ClassII MHC
receptor presentation of the antigen. Even though the
transit time of Lactococcus lactis through the intestine is
less than 24 h in mice [22], a potent immune response has
been obtained with several antigens including tetanus
toxin fragment C (TTFC). Surprisingly, a similar response
was induced using dead or alive Lactococcus suggesting
that in situ antigen synthesis is not essential [23]. A slightly
better result was in the same study obtained with L.
plantarum, but also here a similar response was induced
from living or UV-light inactivated cells.
Active vaccination using LAB
The prospect of using live LAB as vaccine carriers has been
reviewed [24,25]. The most frequently used model anti-
gen is TTFC in which good results have been obtained
both in intranasal and oral mice models using strains of L.
plantarum and L. lactis [23,26]. Grangette et al [27] tested
cytoplasmic expression of TTFC antigen in both L.
plantarum and L. lactis showing protective effect in an oral
mouse model. Shaw et al [28] tested both cytoplasmic and
surface associated expression of same TTFC antigen and
found that cytoplasmic expression was superior to surface
exposed TTFC in L. lactis. In contrast, Bermúdez-Humarán
et al [29] tested human papillomavirus type 16 E7 antigen
sorted 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

28. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 4 of 12
(page number not for citation purposes)
wall-anchored antigen to induce the most potent immune
response. The different outcome of these experiments may
be explained by different stability of surface exposed TTFC
and E7 antigen. Intracellular expression of a labile antigen
can protect it from proteolytic degradation and environ-
mental stress encountered at the mucosal surfaces.
Genetic modification of the LAB cell wall rendering the
strain 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 racemase
mutants were more immunogenic than their wild type
counterparts. One explanation could be that oral immu-
nization is very dependant on a sufficiently large dose of
the antigen [27].
The use of live LAB as carriers of DNA vaccines has until
now not been an option as they are non-invasive and
therefore inefficiently deliver the plasmid DNA to the
cytoplasma 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 cells
in vitro, but also in vivo using an oral guinea pig model. To
determine the tropism of recombinant invasive strains
Critchley-Thorne el al used a perfusion bath with murine
ileal tissue and tested an invasive E. coli vaccine candidate
[31]. Although change of tropism of a bacterial carrier
opens for targeted delivery it introduces new safety issues
that should be addressed by persistence and distribution
studies of the bacterial strain after vaccination.
Active vaccination using recombinant L. johnsonii to treat
allergy has been suggested [32]. IgE epitopes was fused to
proteinase PrtB and cell wall-anchored. Subcutaneous
and intranasal immunization of mice induced a systemic
IgG response against human IgE. As such, allergy-induc-
ing IgE may be cleared by IgG antibodies induced by the
recombinant L. johnsonii. However, it remains to be
proven if these antibodies are protective in human
patients.
In conclusion, LAB has been successfully used for active
vaccination of animals like rodents (Table 2). Whether
LAB will be effective as a mucosal vaccine in humans can
only be answered by clinical trials. Furthermore, as the
dose 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

29. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 5 of 12
(page number not for citation purposes)
responses in animals is high it is unknown if the necessary
dose for use in humans will be feasible and cost effective.
Passive immunization using LAB
Protection by preformed antibodies or antibody frag-
ments is called passive vaccination. The pioneer experi-
ments were based on injection of antisera produced by
immunized animals like horse or sheep to combat for
example rattlesnake venom. Recently, passive immunity
was delivered using lactobacilli that secretes single-chain
antibodies [33]. In a rat caries model, colonisation of the
mouth with a L. zeae expressing a single-chain antibody
fragment recognizing the adhesion molecule of Streptococ-
cus mutans decreased the number of S. mutans and reduced
the development of caries. Recombinant Streptococcus gor-
donii displaying a microbiocidal single-chain antibody
(H6) has been used to treat vaginal candidiasis in a rat
model [34]. Although passive immunity has limits in its
temporary nature, these results suggest that LAB elegantly
can be used for the delivery of neutralising antibodies at
mucosal sites.
Allergy vaccines using LAB expressing allergens
For 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 happen
when vaccinating early in life [35]. However, induction of
tolerance can have positive clinical implications when the
purpose is to treat allergy. In a mouse model the use of a
recombinant L. plantarum expressing the house dust mite
allergen Der p1 as a fusion protein in the cytoplasm inhib-
ited house dust mite-specific T-cell responses [36]. In this
study mice were sensitized by immunization with the
house dust mite peptide and then given either L.
plantarum expressing Der p1 or L. plantarum without Der
p1. Both strains inhibited IFN-γ production by T cells. But
the decrease in production of-5 was only seen for the L.
plantarum expressing the Der p1 peptide antigen. This
indicates that the lactobacilli strain expressing Der p1 can
suppress the cytokine milieu promoting the Th2 allergic
response. Another example of the strain specific effect of
LAB on induction and maintenance of oral tolerance has
been 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 gnotobiotic
mice than was L. johnsonii (NCC 533). The allergen can
also be co-administered instead of recombinant expressed
by the LAB. Mucosal co-application of L. plantarum or L.
lactis together with birch pollen allergen Bet v1 shifted the
immune response towards an anti-allergic Th1 response
both in sensitized and un-sensitized animals [38]. Recom-
binant strains expressing immune polarizing cytokines
like IL-10 have also been developed and in vivo effects in
both mice [39] and pigs [40] have been observed. More
knowledge on the mechanisms behind skewing the
immune response is however needed to select the proper
strain with anti allergic immune polarization. Further-
more, the immune regulatory effect of one strain of LAB
may differ in allergic and non-allergic individuals. A down
regulation in allergic persons and an immune stimulating
effect in normal persons was observed when using same
strain of LAB [41].
Immune stimulatory effects of LAB
Among LAB's effect on the immune system there is a strain
dependent induction of cytokines. Different LAB strains
induce distinct mucosal cytokine profiles in BALB/c mice
[42] pointing at the importance of using one strain for
immune induction and another for induction of tolerance
or a partial down regulation of the immune system. The
same authors [43] also indicate growth phase dependent
differences 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 the
complexity of these observations, a human study has
shown that non-specific immune modulation by a given
strain of L. rhamnosus (GG, ATCC 53103) differs in
healthy and allergic subjects. In healthy persons the strain
was immune stimulatory whereas in allergic persons it
down-regulated an inflammatory response [44]. Interac-
tions between different LAB strains can also interfere with
the 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 shown
to elicit divergent patterns of colonisation and immune
response in germfree mice. Further evidence for an
immune modulating effect is seen when either L. lactis or
L. plantarum was used in a mouse model of birch pollen
allergy [38]. In combination with birch pollen allergen
Bet v1 both strains skewed the immune response from
Th2 to Th1 in sensitised mice as indicated by the IFN-γ/IL-
5 ratio. The immune polarizing effect of LAB has also been
observed in humans. A clinical trial showed a strain
dependent immune modulation of two different LAB
strains when administered together with an oral S. typhi
vaccine (Ty21a) [47]. Here, thirty healthy volunteers were
randomised into three groups receiving L. rhamnosus GG,
L. lactis or placebo for 7 days. On days 1, 3 and 5 the
Ty21a vaccine was given orally. Analysis showed a higher
number of specific IgA-secreting cells in the group receiv-
ing L. rhamnosus GG and a higher CR3 receptor expression
on neutrophils in the group receiving L. lactis. A partial
down regulation of the immune system has also been
observed. Atopic children receiving 2 × 1010 L. rhamnosus
GG daily for 30 days enhanced their IL-10 production in
sera as well as in mitogen-induced peripheral blood
mononuclear cells [41].
26

30. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 6 of 12
(page number not for citation purposes)
It can be concluded that immune polarization towards
either a Th2 or a Th1 response can be obtained using dif-
ferent LAB. As such the intrinsic immune modulatory
capacity of the LAB must be evaluated and selected to fit
the purpose of vaccination.
Safety concerns of the bacterial vaccine strain
Several safety concerns of the bacterial vaccine strain have
been raised (Table 3). Before using pathogenic bacteria for
vaccination purposes, its pathogenicity must be weakened
via attenuation. Attenuation usually involves deletion of
essential virulence factors or mutation of genes encoding
metabolic enzymes whose function is essential for sur-
vival outside the laboratory. Inactivation of a metabolic
gene has the advantage that the bacteria still express viru-
lence determinants important to elicit a protective
immune response. Appropriate stable auxotrophic strains
are usually not able to replicate in the human body and
can safely be used even in immune compromised individ-
uals. Defined deletions of at least two metabolic essential
genes are usually used [2] and decrease the probability of
reversion to virulence. To reduce the risk of spreading for-
eign genetic material to the environment the antigen
encoding 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 loose
the antigen-encoding cassette. The use of antibiotic resist-
ance genes as marker genes in vaccines is not encouraged
as 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 been
published and should be used as soon as possible in the
developmental process of a vaccine [48-50].
Another concern using live bacterial vaccines is the onset
of autoimmune responses like arthritis especially in
patients with the HLA-B27 tissue type [51]. However, the
risk is certainly lower than after natural infection. The
occurrence of such side effects can best be followed by
post launch monitoring and must always be evaluated
against the health risks associated to the disease itself. A
theoretical side effect of vaccines is the possible induction
of autoimmune reactions. However, there is no recom-
mendation to avoid vaccination of people with an ongo-
ing autoimmune disease like rheumatoid arthritis or
systemic lupus erythematosus if vaccination otherwise is
motivated [52]. In contrast, immune-compromised hosts
can have difficulties in handling replicating live attenu-
ated vaccines and should therefore not be vaccinated with
such vaccines. However, new ways of further attenuating
bacteria like combining auxotrophy with deletions of vir-
ulence genes [14] may open for the use of live vaccines to
immune-compromised hosts. In addition, immune-com-
promised people close to hosts vaccinated with live atten-
uated vaccines should be aware of the risk of cross
contamination with the vaccine strain.
Table 3: Safety concerns of the vaccine strain
Systemic disturbance Systemic infection
Conversion from avirulent to virulent bacterium
Translocation to organs
Disturbance of digestive processes
Inhibition of bacterial production of nutrients
Immune system Absorption of allergens through the intestinal epithelium
Induction of tolerance to pathogen instead of immunity
Induction or potentiation of autoimmunity
Bacterial mimicry of self-antigen
Metabolites Production of harmful/undesired metabolites including enzymatic
activities
Breakdown of chemicals to toxic metabolites
Implications for natural flora in GI tract Permanent colonisation of cell substrate in the intestine
Gene/plasmid transfer to host's indigenous flora
"Competitive exclusion" of indigenous flora
Unintentional transferral of cell substrate Unintentional transfer to other individuals
Unintentional 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

31. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 7 of 12
(page number not for citation purposes)
Adverse examples of human live vaccine strains causing
death and illness among domesticated animals are rare
but existent. In Mongolia in autumn 1979 the H1N1
influenza A vaccine virus may have caused a severe influ-
enza epizootic among camels [53]. No examples of
human bacterial vaccines causing problems among ani-
mals have been found in the literature but the possibility
exists and has to be both tested and evaluated before
release of a live bacterial vaccine. In general the spread of
live bacterial vaccines to the environment is a concern.
However, attenuated human pathogens are usually not
adapted to live outside its host. Therefore survival in the
environment is usually short. Vaccines based on recom-
binantLAB may result in the release of these bacteria in
nature, 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. Releasing
gene-modified organisms into the environment can cause
debate 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 the
thyA gene with the expressioncassette for human IL-10. As
a consequence, the L. lactis mutant is dependant on thy-
midine or thymine for growth, which is present in low
amounts in nature and in the human body. Furthermore,
acquirement of an intact thyA gene would recombine the
transgene out of the genome, resulting in reversion to its
wild type state.
Safety concerns of the antigen encoding sequence
In live bacterial vaccines the antigen-encoding gene is
either 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 the
plasmid in the vaccinee must be evaluated. The use of a
prokaryote plasmid replication unit of narrow host range
can limit the horizontal plasmid transfer to other bacteria
present in the vaccinated individual and prevent unde-
sired persistence of the plasmid. In particular for plasmid
DNA vaccines a study should identify which cells take up
and/or express the DNA and what is the fate of the DNA
within those cells as well as for how long the DNA persists
in the cells [54]. Nasal administration of a naked DNA-
vaccine in mice led to some accumulation of plasmid
DNA in the brain [55] illustrating the diffusion of the
plasmid after immunization. The amount of accumulat-
ing plasmid that is acceptable outside the target cells
needs to be further clarified.
The recombinant plasmid harboured by bacterial vaccine
vehicles may integrate in the genome of the recipient and
potentially cause oncogenesesis. Concerns about the
potential oncogenicity of biological products like contin-
uous cell line products (CCL), DNA vaccines and gene
therapy products have been raised [54]. In CCLs foreign
DNA should be avoided in the final product and a limit
has been defined as for maximal residual amount per
human dose. In DNA vaccines DNA is obvious present but
insertion of DNA should be avoided. Finally in the gene
therapy product DNA is both present and inserted but
insertional oncogenesis should be avoided. Integration of
foreign DNA into the host genome is by definition inser-
tional mutagenesis and can induce oncogenesis. There are
three 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 insertional
deactivation of a host suppressor gene. The mechanism
behind DNA integration into the chromosome is either by
random integration, homologous recombination or retro-
viral insertion [56]. The most probable cause of unwanted
integration is by random integration which occurs at a fre-
quency of approximately 10-4 [54]. Unwanted integration
by homologous recombination and retroviral insertion
can be avoided by omission of sequences necessary for
insertion [57]. Analysing the antigen encoding unit car-
ried by the bacteria for human homologous sequences
and 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 retroviral
vaccination [59]. The report of adverse effects in a French
gene therapy study, where 2 out of 10 patients developed
leukaemia within 3 years of [60,61], illustrates occurrence
of 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 plasmid
Transfer of vector to indigenous flora
Open reading frames coding for injurious peptides (allergens)
Imprecise transcription and translation
Specifically for DNA-vaccines Persistence of DNA
Permanent expression of the foreign antigen
Formation of anti-DNA antibodies
Transformation event
Spread of antibiotic resistance genes
28

32. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 8 of 12
(page number not for citation purposes)
effect due to integration of foreign DNA into host genome
has been performed and was found to be less than 10-16 to
10-19 per DNA molecule [62]. This frequency must be put
in relation to the spontaneous mutation frequency which
has been estimated in humans to occur at the rate of 1 in
every 50 million nucleotides incorporated during DNA
replication. 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 by
PCR 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 gene
activity at sites remote from insertion [66]. Different ani-
mal trial has foreseen possible adverse effects like in the
following two examples. Foreign DNA ingested by mice
has been shown to be covalently linked to mouse DNA
[67]. Foreign DNA has also been shown in association
with chromosomes in fetuses born by mice fed orally with
bacteriophage M13 DNA [68]. There is however, no evi-
dence for a germ line transmission of ingested foreign
DNA [66]. The de novo methylation that frequently occurs
with integrated foreign DNA has been suggested as being
a natural defence mechanism [69].
In conclusion, integration of the plasmid harboured by
bacterial vaccine vehicles is a potential hazard. Integration
of gene therapy vectors has been observed, but omitting
sequences driving the insertion may limit the possibility
for integration of the plasmid carried by the bacterium.
Plasmids for heterologous gene expression are usually
preferred due to its multi copy nature and higher gene
dosage. However, placing the antigen encoding genes on
to the bacterial chromosome may limit the spread of the
genes. The route of administration of the vaccine may also
be important when evaluating hazards. As live bacterial
vaccines is fit for mucosal administration one must
remember that ingestion of foreign DNA does occur every
day with our food and is as such not new.
Peptides can be absorbed through the mucosa and some
may induce an allergic reaction. The existence of genes in
the bacterial vaccine coding for such potential allergens or
other injurious peptides can be checked beforehand
searching for homologies to known allergens, as the full
sequence of the bacteria and plasmid should be known.
Vaccination using live bacterial vaccines or exposure to
the natural infections can lead to the formation of auto
reactive antibodies, especially in people prone to autoim-
mune diseases. However, the half life of the induced auto
antibodies 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 and
vaccination [70,72-77] and much controversy in this mat-
ter is still existing. However, convincing data establishing
a link between vaccination and autoimmunity in man are
still not presented. In a mouse model a difference in clin-
ical outcome was observed in two different mouse strains
in relation with auto-antibodies induced by vaccination
with dendritic cells loaded with apoptotic thymocytes
[78]. In normal BALB/c mice the presence of post vaccina-
tion autoantibodies was not associated with any clinical
or histological sign of autoimmunity. However, in mice
prone to autoimmunity (NZBxNZW) F1 a severe pathol-
ogy attributed to autoimmunity was observed. This differ-
ence in outcome attributed to the difference in genotype
has also been observed in humans and it can be con-
cluded that susceptibility to autoimmunity is determined
more by genetic factors than by vaccine challenge despite
the formation of post vaccination auto-antibodies [77]. A
vaccination or treatment with adjuvant can also activate
regulatory 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 solution
for individuals with a genetic profile prone to autoimmu-
nity.
Conclusion
Both attenuated bacteria like salmonella and food related
lactic acid bacteria have been developed as live vaccines
suitable for oral administration. Today, live vaccines
based on attenuated S. typhi and V. cholerae are available.
The development of bacterial vaccine vehicles carrying a
heterologous gene or a DNA vaccine is more problematic
and none has yet reached the market. Several bacteria
have been suggested as vaccine vehicles and especially lac-
tic acid bacteria are promising. Their safe status and
immune modulating capacity have been tested using
diverse vaccine components like antigens from infectious
diseases, allergy promoting proteins and therapeutic anti-
bodies. However, considerable safety issues against live
vaccine vehicles can be raised. Their recombinant nature
calls for a bio containment strategy and auxotroph
mutants may be the answer. The bacterial host must be
fully sequenced and evaluated using bioinformatics tools
for the production of allergy inducing peptides. The anti-
gen encoding gene cassette must be sequenced and
homologies to self proteins or allergy inducing proteins
should be addressed. Especially bacteria carrying recom-
binant plasmids the probability of horizontal gene trans-
fer to other bacteria present should be avoided by using
host restricted replication units. Furthermore, the plas-
mids should be evaluated for sequences facilitating inte-
gration into the human genome.
Authors' contributions
The authors contributed equally to this work.
Acknowledgements
29

33. Microbial Cell Factories 2006, 5:23 http://www.microbialcellfactories.com/content/5/1/23
Page 9 of 12
(page number not for citation purposes)
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.
References
1. 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 Biol
Stand 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 oral
cholera vaccine in adult volunteers from the United States
and Peru. Infect Immun 1997, 65:3852-3856.
4. Taylor DN, Sanchez JL, Castro JM, Lebron C, Parrado CM, Johnson
DE, 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, in
United States military personnel stationed in Panama. Infect
Immun 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 of
lyophilization. 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 and
immunogenicity of delta aroC delta aroD and delta cya delta
crp Salmonella typhi strains in adult volunteers. Infect Immun
1992, 60:536-541.
8. Sztein MB, Wasserman SS, Tacket CO, Edelman R, Hone D, Lindberg
AA, Levine MM: Cytokine production patterns and lymphopro-
liferative responses in volunteers orally immunized with
attenuated vaccine strains of Salmonella typhi. J Infect Dis
1994, 170:1508-1517.
9. Edwards MF, Stocker BA: Construction of delta aroA his delta
pur 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 attenuated
Salmonella enterica serovar typhi oral live vector vaccine
CVD 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 virG
Shigella sonnei live, attenuated, oral vaccine strain WRSS1
in 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 2a
vaccine 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, Sztein
MB, Picking WD, Levine MM: Deletion in the Shigella entero-
toxin genes further attenuates Shigella flexneri 2a bearing
guanine auxotrophy in a phase 1 trial of CVD 1204 and CVD
1208. J Infect Dis 2004, 190:1745-1754.
15. Shata MT, Hone DM: Vaccination with a Shigella DNA vaccine
vector induces antigen-specific CD8(+) T cells and antiviral
protective immunity. J Virol 2001, 75:9665-9670.
16. Courvalin P, Goussard S, Grillot-Courvalin C: Gene transfer from
bacteria 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 DNA
vaccine carried by attenuated Shigella. Vaccine 2003,
21:644-648.
18. Vecino WH, Morin PM, Agha R, Jacobs WRJ, Fennelly GJ: Mucosal
DNA vaccination with highly attenuated Shigella is superior
to attenuated Salmonella and comparable to intramuscular
DNA vaccination for T cells against HIV. Immunol Lett 2002,
82:197-204.
19. Soussi N, Milon G, Colle JH, Mougneau E, Glaichenhaus N, Goossens
PL: Listeria monocytogenes as a short-lived delivery system
for the induction of type 1 cell-mediated immunity against
the p36/LACK antigen of Leishmania major. Infect Immun
2000, 68:1498-1506.
20. Srinivasan J, Tinge S, Wright R, Herr JC, Curtiss RIII: Oral immuni-
zation with attenuated Salmonella expressing human sperm
antigen induces antibodies in serum and the reproductive
tract. Biol Reprod 1995, 53:462-471.
21. Geoffroy MC, Guyard C, Quatannens B, Pavan S, Lange M, Mercenier
A: Use of green fluorescent protein to tag lactic acid bacte-
rium strains under development as live vaccine vectors. Appl
Environ Microbiol 2000, 66:383-391.
22. Gruzza M, Fons M, Ouriet MF, Duval-Iflah Y, Ducluzeau R: Study of
gene transfer in vitro and in the digestive tract of gnotobiotic
mice from Lactococcus lactis strains to various strains
belonging 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 recombinant
Lactococcus lactis. Nat Biotechnol 1997, 15:653-657.
24. Mercenier A, Muller-Alouf H, Grangette C: Lactic acid bacteria as
live 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, Turneer
M, 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. Vaccine
2002, 20:3304-3309.
27. Grangette C, Muller-Alouf H, Hols P, Goudercourt D, Delcour J,
Turneer M, Mercenier A: Enhanced mucosal delivery of antigen
with cell wall mutants of lactic acid bacteria. Infect Immun
2004, 72:2731-2737.
28. Shaw DM, Gaerthe B, Leer RJ, Van Der Stap JG, Smittenaar C, Heijne
DBG, Thole JE, Tielen FJ, Pouwels PH, Havenith CE: Engineering
the microflora to vaccinate the mucosa: serum immu-
noglobulin G responses and activated draining cervical
lymph nodes following mucosal application of tetanus toxin
fragment 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, Le
Loir 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 lactis
is able to invade small intestine of guinea pigs and deliver
DNA into mammalian epithelial cells. Microbes Infect 2005,
7:836-844.
31. Critchley-Thorne RJ, Stagg AJ, Vassaux G: Recombinant
Escherichia coli Expressing Invasin Targets the Peyer's
Patches: 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 Lactobacillus
johnsonii 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 Dalen
PJ, Pouwels PH, Leer RJ, Kelly CG, van Dollenweerd C, Ma JK, Ham-
marstrom L: In situ delivery of passive immunity by lactobacilli
producing 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, Houghten
RA, Hoffman S, Klinman DM: Induction of neonatal tolerance by
30