Plant-derived
Vaccines
As
a pioneer on the field of "edible vaccines" C. Arntzen has explored
using plants to express
antigens for oral delivery since 1992. The topic was reviewed several
times by
Mason and
Arntzen (1995), Mason et al. (1998a), Fischer et al. (1999), Ma (2000),
Mason et
al. (2002).
Growing
vaccines in fields has striking advantages in comparison to
conventional
fermenters. Purification for pathogenic contaminants are unnecessary
with plant
vaccines. Food vaccines are like subunit preparations in that they are
engineered to contain single antigens but do not bear genes that would
enable
whole pathogens to form (Demotz et al 2001). Additionally
bacteria undergo numerous cycles of genetic changes during the
fermentation
process (Shuler and Kargi 2001). Also plant genes go through recurrent
multiplication but are exposed to far less pressure than microbes are
in
fermenters and are therefore extremely stable.
Availability
of edible vaccines will provide a painless approach to deliver large
numbers of
vaccine antigens for human immunization by use of the common mucosal
immune
system (CMIS). The mucosal surface, which includes the oral,
respiratory,
synovial, gut, urinary and reproductive epithelium, is one of the first
important interfaces between pathogens and the host, and as such is
critical in
prevention of infectious disease (Kaufmann 2001,
Iijima et al. 2001, Eriksson and Holmgren 2002).
Mucosal
administration of vaccines allows an induction of appropriate immune
responses
to microbial antigens in systemic sites and peripheral blood as well as
in most
external mucosal surfaces. The
subunit vaccination includes formulations comprising protein antigens
with
adjuvants improving immunogenicity. Such vaccines have to be capable to
stimulate and activate T cell responses (Kaufmann 2001).
As various reports have demonstrated the efficacy of oral vaccination
(Lee and
Chen 1994, Ryan et al. 1997, Harokopakis et al. 1998, Jertborn et al.
2001,
Ruiz-Bustos 2000, Doherty et al. 2002)
through the mucosal immunization it
will be possible in the next years to prevent various infections.
By
consumption of plant-based vaccines it
is already possible to achieve immunogenicity against various viral and
bacterial diseases. Many studies were focused on viral pathogens. Recently,
immunogenicity against a hepatitis antigen was expressed in mice fed
with
transgenic potatoes developed by the group around Arntzen and Mason (Thanavala
et al. 1995, Richter et al. 2000, Kong et
al. 2001).
Successful
experiments with plant derived vaccines have also been carried out with
Norwalk
virus (Mason
et al. 1996, Tacket et al. 2000), rabies
(McGarvey et al. 1995, Yusibov
et al. 2002), FMDV (Wigdorovitz et al. 1999),
Dus
Santos et al. 2002), Respiratory Syncytial Virus (Sandhu et al. 2000,
Belanger
et al. 2000) and rotavirus (Yu et al. 2001).
However
also bacterial pathogens were targeted. Immunogenicity was expressed in
humans
by consumption of transgenic potatoes containing enterotoxin subunit B
from E.
coli (Mason et al.
1998b). Immunogenicity
in humans by a recombinant antigen delivered in transgenic potatoes was
demonstrated by Tacket et al. (1998). Potato
was also used to achieve protective efficacy
towards
cholera and enterotoxic E. coli by Yu
et al (2001) and Lauterslager et al. (2001),
that in principle is also possible in corn (Streatfield et al. 2000).
By plant transformation technique already first successful steps have
been
undertaken, but there are still a couple of milestones to be completed
(Bonetta
2002).
Besides
the unappetizing aspect of eating raw potatoes there are
further limitations in using the nuclear transformed potato
as a vaccine
delivery system. The technology is limited by very low expression
levels of
nuclear transformants and in most cases, immunogenic doses cannot be
delivered
in reasonably small portions. Until now only three plant derived
vaccines have
reached clinical tests, with most of them failing due to low antibody
response.
According to Mason et al. (2002) in order to address these problems we
markedly
need improvements in plant expression technology and inducible
expression of
transgenes.
One
questionable alternative to plant transformation for vaccine synthesis
was
plant-viral infection so far (Wigdorovitz et al. 1999, Belanger et al.
2000,
Yusibov et al. 2002). However this
transient system suffers from instability and frequent gene loss (Mason
et al.
2002). And above all there is still the
serious
problem of transgene spread. Similarly all the other approaches were carried out by nuclear
transformation bearing a high risk
of transgene propagation (Losey et al. 1999, Dale 1999). Outcrossing
via the pollen
mediated gene flow needs
to be substantially controlled and reduced (Jank and Gaugitsch 2001, Wenzel
2002).
In
principle there are only two reasonable ways to keep the transgenes
away from
neighbour-plants:
The
one way consists in the usage of plants with male sterility as
characterised in
potato (Lössl et al. 2000).
Use of plants that are infertile and clonally propagated could
facilitate
management of quality control and production of vaccines.
The other way to prevent transgene transmission is the integration of
transgenes
into the chloroplast genome.
Valuable
improvements can be provided to the current state of edible vaccination
by
application of the transplastome technology (Svab et al. 1990, Koop et
al. 1996,
Ruf et al. 2002):
There
are several advantages to synthesize vaccines in the plant cytoplasmic
genome
pledging for plastid transformation: First of all it is save to the
environment.
In contrast to nuclear transformants by cytoplasmic transformation
spread of
transgenic pollen to the environment can be reduced drastically, as
pollen
rarely contains plastids. By maternal inheritance of the plastid
(Corriveau and
Coleman 1988) pollen-mediated gene flow is reduced to a very small
probability.
Thus a much higher security is ensured compared to antigens transformed
to the
cell nucleus.
Additionally, in comparison to risky virus infection of crops needed
for food
supply cytoplasmic transformation technology does not utilize plant
pathogens.
Chloroplast
transformation allows multiple vaccinations per plant. By translation
of
polycistronic mRNAs the expression of entire biosynthetic pathways from
operons
is feasible.We have provided evidence that by cytoplasm genome
transformation it
is possible to express several genes with a single transformation
vector (Lössl
et al. 2003a). Thus a pyramidization of various subunit-antigens can be
realized
(Sette et al. 2001, Skeiky et al. 2002, Yu et al. 2001).
In
contrast to conventional plant transformation methods plastid
transformation is
a very precise genetic engineering technology. Site directed insertion
of
designed transgenes into the plastome allows to predict transformation
results
more precise than for nucleus transformations (Svab et al. 1990, Maliga
2001).
Transgene
integration into the plastome takes place via homologous recombination.
Thus, in
contrast to nuclear transformation we can exclude position effects.
Epigenetic
effects in transplastomic plastids are absent, co-suppression or gene
silencing
do not occur (Koop et al. 1996, Heifetz 2000), methylation in the
plastid genome
is rarely found (Ngernprasirtsiri et al. 1988a,b).
Additionally cytoplasmic encoded transgenes confer high levels of
transgene
expression and foreign protein accumulation which is due to polyploidy
of the
plastid genetic system. Usually a high stability of foreign peptides is
obtained with
an accumulation of novel protein of up to 40% of total soluble protein (Staub
et al. 2000, De Cosa et al. 2001).
Finally
through the availability of promoter sets and expression cassettes for
tissue-specific or inducible expression (Caddick et al. 1998, Roslan et
al.
2001) a tuning of the expression level of plastid encoded subunit
antigens is
practicable.
In
a first approach solanaceous plants with antigens that could elicit
immune
response are transformed. Practical considerations carried out by Herrera-Estrella
et al. (1999) and Stoger
et al. (2002) were taken into account to
choose this
plant family as target.
New vaccines are in focus according to their importance and feasibility
of
application. For further reasonable progress efforts should be
concentrated on
fruits that are inexpensive to produce, grown in (sub-)tropic regions,
eatable
raw, quick to transform, regenerate and easy to evaluate.
As most proteins are degraded when heated, food crops with an edible
raw fruit
are desirable for antigen subunit expression. Up till now only few
species meet
all the criteria required for production of an edible vaccine. Subspecies
of the solanaceae family can be propagated without high costs
vegetatively or by seeds and are ready for evaluation in clinical
trials within
shorter time than others.
The tomato fruit was already used
successfully as
target for oral immunization of mice expressing respiratory syncytial
virus (Belanger
2000, Sandhu et al. 2000), rabies virus (McGarvey
1995) and in another case for expression of acetylcholinesterase
for enzyme therapy (Mor et al. 2001).
However these
approaches were associated with the problems mentioned above.
A
plastid transformation system has recently been developed to transform
tomato
plastids, with the possibility of expressing high amounts of foreign
proteins
(Ruf et al. 2001). Being very similar in transformation and
regeneration to
tobacco (Svab et al. 1990, Koop et al. 1996) this system paves the way
to
efficient production of edible vaccines in tomato.
Synthesis
of vaccines in transformed chloroplasts is expected to confer an
evident
increase of expression level with a high accumulation of foreign
protein (Staub
et al. 2000, De Cosa et al. 2001). Protein amount is a critical issue as
protein content in the tomato fruit is relatively low and immunogenic
doses need
to be delivered in reasonably small portions. Therefore
it is necessary to increase the transgene expression by usage of
special
expression cassettes with variant promoters and 5´ untranslated
regions (UTRs).
To
the moment there is not much knowledge about post-translational
modifications (especially
glycosylation), acetylations disulfide bonds and correct folding of the
foreign
peptides when introduced to the plant cell. Only for individual
antigens some
information is available (Daniell et al. 2001). Further interactions
with plant
cell metabolism are not yet studied intensively and need to be
evaluated case
dependently.