Vaccine 23 (2005) 2042–2048

Review

Magnifection—a new platform for expressing recombinant

vaccines in plants
Y. Gleba∗ , V. Klimyuk, S. Marillonnet
Icon Genetics, Biozentrum Halle, Weinbergweg 22, D-06120 Halle (Saale), Germany

Available online 13 January 2005

Abstract

Today, plant biotechnology relies on two processes for delivery and expression of heterologous genes in plants: stable genetic transformation
and transient infection with viral vectors. Although much faster, the transient route until recently was limited because of virus’ low infectivity
and its inability to carry average-size or larger transgenes. A recently developed new generation transfection technology overcomes these
limitations by relying on Agrobacterium as an infective systemic agent that delivers viral replicons. This improved process is being used to
simultaneously start transient gene amplification and high-level expression in all mature leaves of a plant, and such a transfection can be done
on an industrial scale. This eclectic technology, called ‘magnifection’, combines advantages of three biological systems: vector efficiency
and efficient systemic DNA delivery of Agrobacterium, speed and expression level/yield of a plant RNA virus, as well as posttranslational
capabilities and low production costs of a plant. The proposed process allows for industrial production that does not require genetic modification
of plants, that is much faster than previous methods, and that is biologically safe. Numerous applications in the area of vaccine manufacturing
are being discussed.
© 2005 Elsevier Ltd. All rights reserved.

Keywords: Viral vector; Plants; Transient expression

Contents

1. Existing technologies for protein production in plants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2043

1.1. First generation production platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2043
1.2. Nuclear transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2043
1.3. Plastid transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2043
1.4. Transient expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2043
1.5. Viral transfection as an industrial platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2044
1.6. Two vector engineering strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2044

2. ‘Magnifection’—a new generation transfection technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2044

2.1. Very fast production and R&D process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2045
2.2. Very high expression levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2045
2.3. Inexpensive process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2046
2.4. Versatility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2046
2.5. Biosafety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2047

∗ Corresponding author. Tel.: +49 345 5559 887; fax: +49 345 5559 884.
E-mail address: gleba@icongenetics.de (Y. Gleba).

0264-410X/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.vaccine.2005.01.006
 Y. Gleba et al. / Vaccine 23 (2005) 2042–2048 2043

2.6. Remaining commercial risks and concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2047

2.7. Potential limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2047

3. Potential applications in the area of vaccines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2047

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2048

1. Existing technologies for protein production in plants have been shown to provide the necessary machinery
plants for expression of fully functional antibodies such as full size
immunoglobulins. Although the proteins expressed this way
1.1. First generation production platforms are in most cases fully functional, the protein yields that can
be obtained from nuclear transgenics are, with very few ex-
The search for inexpensive and efficient expression sys- ceptions, fairly low, in the range of 0.1–0.5% of total soluble
tems for production of recombinant proteins has turned the protein. The other most obvious shortcoming of the tech-
attention of researchers and entrepreneurs to plants as the po- nology is the long R&D duration time, usually 18 or more
tentially cheapest production hosts. The potential of ‘molec- months, required to generate the first 100 mg of protein sam-
ular farming’, the production of recombinant pharmaceutical ple [1–3].
proteins using plants as bioreactors, is being illustrated by
a number of recombinant protein products that are currently 1.3. Plastid transformation
undergoing clinical trials. All these product candidates have
been obtained from plants using expression processes de- The gene encoding the protein of interest can also be
veloped during late 1970s to early 1980s, and all currently integrated on a plastid chromosome. Plastids are plant or-
available ‘first generation’ expression methods suffer from ganelles of symbiotic origin and their genetic machinery
various limitations such as the long time frame necessary closely resembles that of bacteria. In a mature leaf cell,
for stable transformation, the low yield obtained with stable there are up to 100 plastids (chloroplasts), each containing
or transient systems, biosafety concerns around open field up to 100 copies of plastid circular DNA; because of this
cultivation of transgenic crops expressing therapeutic pro- high copy number, expression driven by a plastid gene can
teins, and the inability of transient systems to be scaled up. be as high as 35% of total soluble protein (the gene en-
These available production platforms fall into several cate- coding the large subunit of ribulose-1,5-bisphosphate car-
gories: nuclear transformation, plastid transformation, tran- boxylase/oxygenase, RUBISCO). Similarly, expression lev-
sient expression based on Agrobacterium-mediated delivery, els from transgenes incorporated in the plastid chromosome
and transient expression mediated by plant viral vectors. can reach 25% of total soluble protein (reviewed in [4]), such
high yield being probably the most attractive feature of the
1.2. Nuclear transformation technology. The expressed protein is however confined to the
plastid, an organelle that does not provide many desired post-
Most research in the area of plant-based production of translational modifications such as glycosylation and, like in
pharmaceutical proteins has been based on expression of bacteria, some proteins form inclusion bodies that have to be
transgenes stably integrated on a plant chromosome. The re- refolded during the purification process. Like nuclear trans-
sultant stably transformed transgenic plants are then used as formation, plastid transformation is also a slow process, and
production hosts. In many cases, the transgene of interest is unable to support rapid production of sub-gram quantities
is controlled by a strong constitutive promoter such as the of recombinant protein within 12–18 months.
35S CaMV promoter, and the primary product is in the leaf
biomass which has to be processed promptly after harvest- 1.4. Transient expression
ing for purification of the recombinant protein. In recently
developed approaches, expression of the gene of interest is The following two processes are being routinely used to-
controlled by a seed-specific promoter, and the protein ac- day: transient expression of genes delivered by Agrobac-
cumulates in mature seeds. The obvious benefit of the lat- terium and viral vector-based transient amplification in
ter technology is that, by storing the harvested seed, the up- plants. In both cases, expression relies on delivery of trans-
stream and downstream parts of the operation can be discon- genes by a vector (bacterium or virus), and there is no need
nected. The proteins can be targeted to different subcellular for stable transgene integration [5]. An obvious advantage
compartments, such as the cytoplasm, the nucleus, the plas- of such systems is speed (3–4 days necessary for full ex-
tids, the vacuole, or they can be secreted into the apoplast or pression in leaves infiltrated with agrobacteria, and 10–14
retained in the endoplasmic reticulum, thus providing vari- days for maximum expression from viral vectors). Leaves
ous choices for posttranslational modifications. Importantly, infiltrated by agrobacteria usually express low amounts of
2044 Y. Gleba et al. / Vaccine 23 (2005) 2042–2048
protein, similar to the expression levels seen in stably trans-
formed tissues under control of the same promoter (in a ma-
jority of cases, the 35S CaMV promoter is being used). It
is generally believed that Agrobacterium-based infiltration
cannot be used on an industrial scale and is limited to early
R&D stages; that perception stems from the low yields ob-
tainable using agroinfiltration and from technical difficul-
ties expected in designing the equipment for industrial-scale
throughput.
1.5. Viral transfection as an industrial platform
Virus-based technology, on the other hand, is scalable and
is being used on an industrial scale by at least one company,
Large Scale Biology Corp., Vacaville, CA. There has been
impressive technical progress over the last few years in the
development of production processes and hosts for expres-
sion using viral vectors (for the latest reviews, see, for exam-
ple [1,6–9]). These studies have changed our understanding
of both the advantages and the limitations of the viral vectors
currently in use.
One of the conclusions made by some researchers was
that in its present form, viral vectors suffer from obvious and
inherent shortcomings and that future progress may require
more radical approaches, in particular, the design of viral vec-
tors that are not simple carbon copies of a wild-type virus car-
rying a heterologous coding sequence, but are more integrated
with a transgenic plant host that has been pre-engineered to
provide different functions, including some functions earlier
controlled by the viral vector, thus allowing for a more effi-
cient, controlled and safe process.
1.6. Two vector engineering strategies
Among the functions that a wild-type virus is capable
of performing are: initial host infection, nucleic acid am-
plification/replication, protein translation, assembly of ma-
ture virions, cell-to-cell spread, long-distance spread, repro-
gramming of the host biosynthetic processes, suppression of
silencing, etc. Not all of these functions are necessarily re-
quired in a vector, whereas some other new functions, such as
high-level expression of the heterologous gene, are required.
There are two ideologically different approaches that could
be used to develop an ideal viral vector/host process.
Historically, the first one was to design a vector that is as
capable of infecting a plant as a wild-type virus, but is in addi-
tion engineered to carry and express a heterologous sequence
coding for a gene of interest (a ‘full virus’ vector strategy).
The progress, in this case, was the result of extensive studies
that focused on engineering a vector that was essentially a
fully functional virus, that, despite carrying and expressing a
heterologous sequence, has retained infectivity, stability and
systemic virulence in its host.
The other, later trend, is reflecting an ideology that ad-
mits inherent limitations of the viral process, and attempts to
‘deconstruct’ the virus by eliminating the functions that are
limiting (for example, too species-specific, or rate-limiting)
or undesired (such as the ability to create functional infectious
viral particles) and rebuilding the process by either delegating
the missing functions to the host (that has been genetically
modified to provide those functions in trans) or replacing
them with analogous functions that are not derived from a
virus (a ‘deconstructed virus’ vector strategy). In more ad-
vanced forms, these expression systems integrate elements
of the viral machinery, such as RNA/DNA amplification and
cell-to-cell movement, along with non-viral processes, such
as for example, amplicon formation/systemic delivery via
agrodelivery or activation from a chromosomal DNA, substi-
tution of systemic movement of the viral vector by intracellu-
lar generation of amplicons from a chromosomally encoded
pro-replicon or pro-virus [10].
2. ‘Magnifection’—a new generation transfection
technology
Icon Genetics, a German plant biotechnology company,
has developed a new generation expression platform that fully
utilizes advantages of a ‘deconstructed virus’ strategy and
effectively addresses most of the shortcomings of today’s
technologies [11–13].
The process which we term ‘magnifection’, is a simple
and indefinitely scalable protocol for heterologous protein
expression in plants, which is devoid of stable genetic trans-
formation of a plant, but instead relies on transient amplifi-
cation of viral vectors delivered to multiple areas of a plant
body (systemic delivery) by Agrobacterium (Figs. 1–3). Such
a process is in essence an infiltration of whole mature plants
with a diluted suspension of agrobacteria carrying T-DNAs
encoding RNA replicons. In this process, the bacteria as-
sume the (formerly viral) functions of primary infection and
systemic movement, whereas the viral vector provides for
cell-to-cell (short distance) spread, amplification and high-
level expression. A few adult tobacco plants are sufficient
for early construct optimisation and fast production of mil-
ligram or gram quantities of recombinant protein for pre-
clinical or clinical evaluation. The scale-up (industrial) ver-
sion is essentially the same, and requires a simple apparatus
for high-throughput vacuum-infiltration of whole plants. This
eclectic technology combines advantages of three biological
systems: vector efficiency and systemic delivery capabili-
ties of an Agrobacterium, speed and expression level/yield
of a virus, and posttranslational capabilities and low cost of a
plant. The magnifection platform effectively addresses most
of the major shortcomings of earlier plant-based technolo-
gies by providing the overall best combination of the fol-
lowing features: high expression level, high relative yield,
low up- and downstream costs, very fast and low cost R&D;
and low biosafety concerns. These features differentiate the
magnifection expression platform from the first generation
production platforms in plants and are detailed in the follow-
ing paragraphs.
 Y. Gleba et al. / Vaccine 23 (2005) 2042–2048 2045

Fig. 1. Time course of transgene (GFP) expression in magnifected leaves of Nicotiana benthamiana. Coomassie-stained polyaccrylamide protein gel showing:
M, protein standards in kDa; 1,14: crude extract from uninfected leaf tissue; 2–12: same, from infected leaf tissue 2–12 days after infiltration; 13, GFP standard;
the arrow indicates the GFP.

2.1. Very fast production and R&D process
The newly introduced industrial steps (material infiltra-
tion and short-term incubation) take less than 10 days of time
and require only relatively straightforward manipulations, all
with high throughput potential. The speed of scale up that the
technology offers is perhaps the most attractive element of
such industrial process. The research and development ver-
sion of the process is equally very fast, the rate-limiting steps
being vector engineering time (done in bacteria) and the in-
cubation time (a few days) necessary for viral amplification.
Milligram and gram quantities of recombinant protein are
available in 3–4 weeks, and the scale up to 100 kg is possible
within less then one year, thus the platform provides the high-
est possible speed of R&D in the industry. Such timelines are
unprecedented for traditional GMO-based plant biotechnol-
ogy.
2.2. Very high expression levels
Magnifection provides up to 5 g of recombinant protein
per kilogram of fresh leaf biomass; with the recombinant
protein relative yield being up to 80% of the total soluble
protein. These yields reflect the fact that, as a result of ampli-
fication by the viral vector, most protein in the system is the
protein of interest, thus, the platform addresses the volume
and cost of both upstream as well as downstream process
components. The yields are 10–100-fold higher, and relative

Fig. 2. Leaf of edible plant (red beet) expressing GFP, 7 days after infiltration; the picture on the right is photographed under UV light.
2046 Y. Gleba et al. / Vaccine 23 (2005) 2042–2048
Fig. 3. Magnifected Nicotiana benthamiana plants expressing recombinant protein (GFP), 7 days after infiltration, UV light.
yields are 10-fold or more higher than with other existing
plant-based alternatives. In addition, these parameters po-
sition the magnifection platform to out-perform traditional
fermentation and bioreactor processes in the production of
many important product classes.
2.3. Inexpensive process
The process is inexpensive, because plant leaf material
and the minimal amount of required bacterial biomass can
both be produced very cheaply. The high level of gene ex-
pression and the high relative protein yield achievable in
our amplification system also minimizes the cost of down-
stream purification. Earlier claimed advantages of making
recombinant proteins in seed, that allow producers to cou-
ple seasonal biomass production with the continuous down-
stream processing by storing seed until the processing can
be accommodated, are easily achieved with the magnifec-
tion platform. Although the protein is made in leaves, ac-
tivation of protein production is controlled by infiltration,
so the plants can be grown in batches and subject to pro-
cessing after the amplification in the batch has been ini-
tiated. Such a process allows the manufacturer to effec-
tively and seamlessly couple the front-end and downstream
operations.
We estimate that the cost of production using magnifection
in a greenhouse operation, assuming a yield of 0.8 g recombi-
nant protein per kilogram fresh biomass, will be in the range
of less than $1 per gram of crude protein, and the cost of goods
of purified protein produced under GMP-compliant condi-
tions will be less than $50 per gram. Based on yields obtained
with existing magnICON technology, downstream purifica-
tion costs are expected to be lower than with any other plant-
based, or established non-secreting, production platform.
2.4. Versatility
Magnifection has been shown to work with dozens of pro-
teins tested so far, including physiologically neutral, but also
highly toxic proteins and enzymes, single chain antibodies,
antigens, cytokines, hormones, etc. (Table 1). Tobacco is our
production host of choice (non-food, non-feed), but the tech-
nology works with many other, in particular edible, species.
Icon Genetics is not currently producing monoclonal antibod-
ies or proteins from very long genes (over 2.3 kb), but expect
to initiate such programs during the next 12 months. Repre-
sentative yields of protein obtained from fresh leaf biomass
include 1.2 g/kg for somatotropin, 5.1 g/kg for interferon alfa,
5.4 g/kg of several bacterial antigens, and 0.4–1.5 g/kg for
various single chain antibodies.
 Y. Gleba et al. / Vaccine 23 (2005) 2042–2048
Table 1
Plateform validation: variety of genes expressed using magnifection
• Interferons, growth hormones, colony stimulating factors (6* )
• Mabs, single chain antibodies, antibody fusions (15)
• Bacterial antigens, viral antigens, antigen mutants and fusions,
adjuvants, epitope carriers (16)
• Enzymes and enzyme inhibitors of animal, plant, microbial or
viral origins (6)
• Other (8)
• Total: over 50 proteins of human, animal, plant, bacterial, viral
origin
∗ Number of genes/constructs tested.
2.5. Biosafety
Because of very high expression levels, production of most
pharmaceutical and veterinary proteins can be performed
entirely in a contained facility, thus minimizing biosafety-
related risk. The whole process can be performed follow-
ing a straightforward protocol, similar to existing industrial
microbial technologies; it requires, in addition to well es-
tablished industrial upstream (greenhouse plant cultivation)
and downstream (protein extraction and purification) com-
ponents, a contained technology block that includes a small
bacterial fermentation apparatus, an apparatus for vacuum-
infiltration of batches of plants, and a chamber/greenhouse
for subsequent short-term incubation. Such a block would of
course require certain safety ‘locks’ to prevent the release
of agrobacteria into the open environment and to protect the
operating personnel.
As a result of high expression levels, the whole process
can be fully contained in a greenhouse: a 1 ha greenhouse is
capable of producing up to 500 kg of recombinant protein per
year. Comparable amounts made in transgenic corn (epicyte
data) or transfected tobacco utilizing first-generation viral
vectors (LSBC) would require a 1000-fold larger space. The
vectors used are based on our ‘deconstructed virus’ strategy
(they are lacking essential viral functions), thus no wild type
virus can be generated as a result of an uncontrolled recom-
bination event. In addition, the use of tobacco, a non-food,
non-feed plant, further mitigates the biosafety concerns. Icon
has developed innovative technology for the control of gene
flow from transgenic plants intended for use in agricultural
applications. This additional technology is also available, if
needed, for use with the magnifection platform.
2.6. Remaining commercial risks and concerns
Although the proposed commercial platform is new and its
potential was so far exemplified with only a modest number
of cases (Table 1), the remaining process risk is, in our opin-
ion, very low. Plants have been used to express a great variety
of antibodies and antigens, and, in a number of cases, the re-
search addressed and confirmed overall functionality (in case
of vaccines, immunogenicity) of the molecules produced in
plants (see reviews [1,5]). Many of these experiments were
based on the use of plant viral vectors, and the results re-
2047
ported are fully predictive of what should be expected when
using magnifection since the ‘full viral’ vector strategy and
‘magnifection’, are essentially the same processes from the
end-result point of view (amount, quality of the recombinant
protein and other metabolites in a plant cell). In addition, as
mentioned above, the magnifection process itself has been
validated through expression of a number of antigens of mi-
crobial and viral origin and a number of antibodies.
2.7. Potential limitations
There are potentially some areas in which the posttransla-
tional modifications, especially plant-specific glycosylation
pattern, of plant-made recombinant proteins, will render the
material non-functional, in case of a vaccine, poorly immuno-
genic. We have no evidence for this being the case so far, but
the data are very limited, and general conclusions can hardly
be drawn; any final answer in each particular case will require
the evaluation of the molecule of interest. It is worthwhile to
remember, however, that alternative expression platforms all
face similar limitations; in particular, microbial cells lack gly-
cosylation, and patterns of sugars added to proteins in yeast
or insect cells, or even avian cells, are all different from gly-
cosylation provided by human cells.
Like with any other expression platform, an economically
acceptable expression level of a gene of interest may not al-
ways be possible in a magnifected plant. Our exploration has
identified a couple of such difficult cases, one of them being
the hepatitis B surface antigen, which, when expressed at high
levels in a plant, is extremely toxic to the host. However, the
history of production host development in microbial biotech-
nology tells us that the yields seen in early experiments can
usually be dramatically improved through manipulation of
the gene, the vector and the production host.
A more difficult and challenging remaining task is adapta-
tion of the magnifection process for expression of oligomulti-
meric proteins such as IgG antibodies, since the viral vectors
have to be manipulated to express two or more polypeptides
in equimolar amounts. Our work in this area has allowed us to
identify several effective strategies, but the expression levels
achieved so far need further improvement.
3. Potential applications in the area of vaccines
The production process described lends itself to many ap-
plications in the area of vaccine research and production,
primarily because of speed of R&D, fast scale up and manu-
facturing process, and low production costs [14,15].
Rapid commercial production of specific recombinant
vaccine types in situations with time constraints could be an
especially promising application of the plant-based process
described here. Since the industrial cycle (from engineered
bacterial vector to pure protein) takes less than three weeks,
a rapid manufacturing of recombinant vaccine in response
to a pandemic, such as a flu, could perhaps be addressed
2048 Y. Gleba et al. / Vaccine 23 (2005) 2042–2048
better by magnifection then by the current industrial process.
We are not there yet, as more studies are required to test and
perhaps improve the efficacy of plant-expressed recombinant
vaccines such as influenza haemagglutinins, but the data ac-
cumulated thus far with other plant-derived vaccines are very
encouraging [1,2].
Because the production process in question is significantly
less expensive, compared to many other platforms, we expect
that it may become a source of affordable vaccines for de-
veloping countries. The calculated cost of goods for several
recombinant proteins in our system is expected to be below
$0.01 per one 0.1 mg vaccine dose, i.e. it will probably be
less than the cost of packaging. A capital cost of a GMP-
compliant facility built in the USA and capable of producing
100 kg purified recombinant protein per year is expected to
be less than $15 million, making such an upfront investment
a more affordable, and a less risky, proposition, as compared
to microbial or animal cell-based fermentation. The plant-
based production offers more flexibility: it can be used to
produce product batches of any size, as there are no technical
or economic constrains such as those imposed by the size
of microbial fermentors. For cost reasons, many biodefence
vaccines can also probably be produced inexpensively and
accumulated in required volumes and stored.
Production of animal vaccines, too, is limited primarily
by the manufacturing cost. Plant-based production is an ob-
vious alternative when traditional vaccine processes are not
possible, for example in cases when the virus kills the host
(flu vaccine made in hen eggs).
Oral, aerosol, nasal immunisation, transcutaneous skin
patches all show promise, and in many cases efficient proto-
cols will require higher doses, thus the manufacturing costs
will dictate the choice of the production platform, making the
magnifection process an obvious alternative.
Production of individualized medicines is a new and ex-
citing area of medical science, and at least one company,
Large Scale Biology Corp., Vacaville, CA, has conducted
successful clinical (phase 1) trials with individualized scFv
vaccines produced in plants and used in the treatment of non-
Hodgkin’s lymphoma [16,17]. The individualized single-
chain Fv in these experiments were obtained using plant viral
vectors, and the procedure involved engineering of a DNA
copy of the vector, in vitro transcription to generate an infec-
tious RNA, infecting a primary plant, and subsequently using
the viral particles collected from the primary host to infect
hundreds of plants. The magnifection process is more effec-
tive and fast, as there is no need for in vitro transcription and
the transfection process can be done immediately on many
plants in one step, thus significantly (by 10–14 days) reduc-
ing the time from DNA to patient, a very important parameter
in this case.
References
[1] Ma JK, Drake PM, Christou P. The production of recombinant phar-
maceutical proteins in plants. Nat Rev Genet 2003;4(10):794–805.
[2] Warzecha H, Mason HS. Benefits and risks of antibody and vaccine
production in transgenic plants. J Plant Physiol 2003;160(7):755–64.
[3] Nikolov ZL, Woodard SL. Downstream processing of recombi-
nant proteins from transgenic feedstock. Curr Opin Biotechnol
2004;15(5):479–86.
[4] Maliga P, Graham I. Molecular farming and metabolic engineering
promise a new generation of high-tech crops. Curr Opin Plant Biol
2004;7(2):149–51.
[5] Fischer R, Stoger E, Schillberg S, Christou P, Twyman RM. Plant-
based production of biopharmaceuticals. Curr Opin Plant Biol
2004;7(2):152–8.
[6] Awram P, Gardner RC, Forster RL, Bellamy AR. The potential of
plant viral vectors and transgenic plants for subunit vaccine produc-
tion. Adv Virus Res 2002;58:81–124.
[7] Fitzmaurice WP, Holzberg S, Lindbo JA, Padgett HS, Palmer KE,
Woffe GM, et al. Epigenetic modification of plants with systemic
RNA viruses. Omics 2002;6(2):137–51.
[8] Pogue GP, Lindbo JA, Garger SJ, Fitzmaurice WP. Making an ally
from an enemy: plant virology and the new agriculture. Annu Rev
Phytopathol 2002;40:45–74.
[9] Porta C, Lomonossoff GP. Viruses as vectors for the expres-
sion of foreign sequences in plants. Biotechnol Genet Eng Rev
2002;19:245–91.
[10] Gleba Y, Marillonnet S, Klimyuk V. Engineering viral expression
vectors for plants: the ‘full virus’ and the ‘deconstructed virus’ strate-
gies. Curr Opin Plant Biol 2004;7(2):182–8.
[11] Marillonnet S, Giritch A, Gils M, Kandzia R, Klimyuk V, Gleba Y.
In planta engineering of viral RNA replicons: efficient assembly by
recombination of DNA modules delivered by Agrobacterium. Proc
Natl Acad Sci USA 2004;101(18):6852–7.
[12] Marillonnet S, Thoeringer C, Kandzia R, Klimyuk V, Gleba Y. Effi-
cient transfection of a plant with viral RNA replicons using synthetic
DNA templates 2005, submitted for publication.
[13] Marillonnet S, Thoeringer C, Kandzia R, Klimyuk V, Marillonnet
S. Large-scale gene amplification and expression in plants using
Agrobacterium-mediated systemic delivery of viral replicons 2005,
submitted for publication.
[14] Poland GA, Murray D, Bonilla-Guerrero R. New vaccine develop-
ment. Bmj 2002;324(7349):1315–9.
[15] Abstracts of the Fourth World Congress on Vaccines and Imunisa-
tion. Tsukuba Science City, Tokyo, Japan, 2004.
[16] McCormick AA, Kumagai MH, Hanley K, Turpen TH, Hakim I,
Grill LK, et al. Rapid production of specific vaccines for lymphoma
by expression of the tumor-derived single-chain Fv epitopes in to-
bacco plants. Proc Natl Acad Sci USA 1999;96(2):703–8.
[17] McCormick AA, Reinl SJ, Cameron TI, Vojdani F, Fronefield M,
Levy R, et al. Individualized human scFv vaccines produced in
plants: humoral anti-idiotype responses in vaccinated mice con-
firm relevance to the tumor Ig. J Immunol Methods 2003;278
(1–2):95–104.