| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Pathophysiology, University of Vienna, Vienna 1090, Austria;
Centre of Applied Genetics, University of Agricultural Sciences, Vienna 1190, Austria; and
Large Scale Biology Corp., Vacaville, California 95688, USA
1Correspondence: Department of Pathophysiology, University of Vienna, AKH-EBO-3Q, Waehringer Guertel 18–20, A-1090 Vienna, Austria. E-mail: Heimo.Breiteneder{at}akh-wien.ac.at
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: plant expression system • tobacco mosaic virus • recombinant allergen • BALB/c • Th2 response
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. IgE-mediated type I allergy is a serious and
increasing health problem in the western world. Consequently, there is
a rising need for accurate diagnosis applying purified, well-defined,
and standardized allergenic molecules. Recombinant DNA techniques are
applied to produce such proteins with immunological characteristics
identical to the naturally occurring proteins (2
3
4)
.
Recombinant allergens significantly improve accuracy and reliability of
diagnosis of type I allergies (2
3
4)
. Bet v 1, the major
allergen of birch pollen, was the first plant allergen that was cloned,
sequenced (5)
, and expressed in Escherichia
coli (6
, 7)
. Bet v 1 was subsequently studied in
great detail regarding its molecular biology (8
, 9)
and
immunological in vitro (10
11
12)
and in
vivo characteristics (13
14
15)
. In addition, the
3-dimensional structure of Bet v 1 has been determined
(16)
. In industrialized countries 5 to 7% of the
population suffer from type I allergies to tree pollen
(17)
. More than 95% of tree pollinosis patients produce
IgE directed against Bet v 1. Therefore, Bet v 1 represents a model
allergen for the study of type I diseases and is the allergen of choice
to evaluate new expression systems.
So far, Bet v 1 has been expressed in E. coli, as no
post-translational modifications of the natural allergen could be
detected by mass spectrometry (18)
. The yield using the
E. coli expression system pMW175/BL21(DE3) was ~10 mg of
purified recombinant Bet v 1 per liter of bacterial culture
(7)
. To obtain pure E. coli recombinant (r) Bet
v 1 for in vitro use, several purification steps were
necessary including ion-exchange and hydrophobic interaction
chromatography as E. coli extracts contain considerable
amounts of lipopolysaccharides (7)
. Although bacterial
expression systems have the advantage of speed and abundant production,
they are limited in their ability to express properly folded soluble
proteins without the need of denaturation and renaturation.
In comparison to E. coli, yeast-based expression systems
offer the advantages of a eukaryotic folding machinery, the absence of
endotoxin, and higher yields of the recombinant protein. Expression in
the yeasts Pichia pastoris and Saccharomyces
cerevisiae were successfully applied for allergens from avocado
(Pers a 1, ref 19
), latex (Hev b 7, ref 20
),
olive pollen (Ole e 1, ref 21
), cockroach (Bla g 4, ref
22
), house dust mite (Der p 2, ref 23
), and
bermuda grass pollen (Cyn d 1, ref 24
) resulting in yields
of up to 50 mg per liter of culture. However, several researchers have
experienced difficulties expressing certain plant proteins in E.
coli or yeast, resulting in low or no yield or the inability of
the expressed allergen to bind IgE. These problems are caused by the
lack of a plant-specific protein folding machinery in bacteria and
yeasts and by differences in glycosylation and/or codon usage. To
address these points and to circumvent low yields or the toxicity of
E. coli extracts and the need for complex purification
procedures, plant allergens are ideally expressed in a plant-based
system.
In particular, expression of heterologous proteins in plants to produce
edible vaccines has generated great interest in recent years. The
hepatitis B surface antigen (25)
, a Norwalk virus coat
protein (26)
, the E. coli heat-labile
entertoxin B subunit (27)
, and the cholera-toxin B subunit
(28)
were stably produced in their native immunogenic
forms either in tobacco or potato plants transformed by the
Agrobacterium method. A disadvantage of this type of plant
expression was the low expression level of the recombinant heterologous
proteins. In contrast, plant viral vectors offer a rapid production
system for generating substantial amounts of heterologous proteins
without having to go through the process of establishing transgenic
plants.
An attractive candidate virus is the tobacco mosaic virus (TMV), a
well-characterized, single-stranded RNA virus that replicates
extrachromosomally and redirects protein synthesis of host cells to
express high levels of heterologous proteins throughout the plant
(29)
. TMV-based vectors have been successfully applied to
produce medically relevant peptides or proteins in tobacco such as
malaria epitopes (30)
or tumor-derived single chain
variable fragments of antibodies (31)
.
Here we present for the first time the TMV-directed expression of an allergen in the tobacco-related species Nicotiana benthamiana and characterization of the immune response against this plant-produced allergen in humans in vitro and in mice in vivo. Furthermore, we demonstrate that Bet v 1 present in crude extracts of N. benthamiana plants displays identical immunological parameters to those of purified rBet v 1 produced in E. coli.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) were
designed with a SalI restriction site at the 5'-end
and a PstI restriction site at the 3'-end in order to
allow subsequent cloning (sense strand primer:
5'-TAGTCGACATGGGTGTTTTCACTTACGA-3'; antisense strand
primer: 5'-AGACTGCAGGTTGTAGGCATCGGAGTGTGC-3'
(priming regions underlined, SalI and PstI
sites in italics). PCR was performed in a 100 µl volume containing
0.5 µg purified Bet v 1 DNA prepared from clone pMW175/Bet v 1
(7)
, 25 pmol each of sense and antisense primer, 2.0 mM
MgCl2, 0.2 mM of each dNTP (dATP, TTP, dCTP, and
dGTP), and 2.5 U AmpliTaq DNA Polymerase in PCR buffer II (Perkin Elmer
Cetus, Norwalk, Conn.). PCR amplification was carried out on a Perkin
Elmer thermal cycler (GenAmpPCR System 2400). The amplification
procedure included a denaturation step at 94°C for 3 min, 30 cycles
of 45 s strand separation at 94°C, 30 s annealing at 63°C
for 30 s, and 30 s extension at 72°C, followed by an
elongation step of 10 min at 72°C. The PCR product was purified by
agarose gel electrophoresis, digested with SalI and
PstI, and ligated into respective sites of the
p4GD-PL vector containing a hybrid fusion of TMV-U1 and tobacco mild
green mosaic virus (29)
. The ligated DNA was used for the
transformation of competent E. coli XL1-Blue cells
(Clontech, Palo Alto, Calif.).
In vitro transcription of capped recombinant viral
RNA
Plasmid p4GD-Bet v 1 was prepared for in vitro
transcription by linearization with SfiI. Linearized plasmid
DNA was purified over spin columns (QIAquick Spin, Qiagen, Calif.) and
made blunt-ended by the DNA polymerase I Klenow Fragment (New England
Biolabs Inc., Boston, Mass.) for 15 min. Capped infectious RNA was
generated from 1 µg plasmid template using T7 RNA polymerase (Promega
Corporation, Madison, Wis.). Transcription reactions contained 40 mM
HEPES-KOH (pH 8.0), 24 mM MgCl2, 2 mM spermidine,
10 mM dithiothreitol, 0.5 mM each of ATP, CTP, and UTP, 0.2 mM GTP, 0.5
mM cap analog G(5')ppp(5')G (New England Biolabs), 1 U/µl RNasin
(Promega), 1 µg linearized DNA, and 0.6 U/µl T7 RNA polymerase in a
50 µl volume. Reactions were incubated for 2 h at 37°C.
Synthesis of RNA was analyzed by gel electrophoresis.
Inoculation of N. benthamiana plants
Carborundum-dusted leaves of N. benthamiana were
manually inoculated with 10 µl of the transcription reaction
containing 2 µg of in vitro transcribed viral RNA.
Immediately after inoculation, plants were rinsed with water and placed
in growth chambers. Plants were grown under 16 h of daylight at
22°C and observed daily for signs of infection such as necrotic local
lesions or systemic vein yellowing and variable leaf mottling. These
experiments were performed following the recommendations given by the
European Commission for GLP regarding genetically modified organisms.
Reverse transcriptase (RT)-PCR analysis of transfected plants
Two weeks after inoculation, total RNA was isolated by grinding
newly formed upper leaves with visual viral symptoms in liquid nitrogen
to powder and extracting with phenol/chloroform (TRI-Reagent, Molecular
Research Center, Cincinnati, Ohio). Reverse transcription was
performed on 1 µg total RNA using the GeneAmp RNA PCR kit (Perkin
Elmer Cetus). The cDNA synthesis was primed with a TMV-specific
antisense strand primer downstream the insertion site
(5'-TTTTTCCTTTTTTGTTTTCCG-3'). For RT-PCR, the same primer and a primer
upstream of the insertion site (5'-GATGATGATTCGGAGGCTACT-3') were used.
For control experiments, total RNA from uninfected N.
benthamiana and from plants infected with viral RNA that did not
contain the Bet v 1 coding sequence was subjected to RT-PCR. Sequence
analysis of the RT-PCR product was performed using the Thermosequenase
Fluorescent Labeled Primer Cycle Sequencing kit (Amersham Life Science
Ltd, London, U.K.) and the LI-COR DNA sequencer model 4000L (LICOR,
Lincoln, Nebraska).
Protein extractions
Leaves, stems, and roots of infected N. benthamiana
plants were harvested 21 days postinoculation. For Western blot
analysis, plant materials were frozen in liquid nitrogen and ground to
a fine powder in a Waring Blendor. The powder was then stirred
overnight at 4°C in a 10-fold excess (w/v) of 10 mM potassium
phosphate buffer at pH 8.0 containing 2 mM EDTA, 10 mM
diethyldithiocarbamate, 2% (w/v) polyvinylpolypyrrolidone, and 1 mM
PMSF. The supernatants obtained by centrifugation at 40,000
g and 4°C for 1 h were filtered, dialyzed against 10
mM potassium phosphate buffer, pH 8.0, and stored at -20°C until
use. Ten grams of birch pollen (Allergon AB, Välinge, Sweden)
were extracted in the same way, except that the extract was dialyzed
against phosphate-buffered saline (PBS), pH 7.0. Pollen-derived natural
(n) Bet v 1 was purified by affinity chromatography with an immobilized
anti-Bet v 1 monoclonal antibody, followed by reversed-phase
high-performance liquid chromatography (6)
.
Sera from atopic and nonatopic individuals and monoclonal
antibodies
Serum or plasma samples from five individual birch
pollen-allergic patients, a standard serum pool of eight birch
pollen-allergic patients, and a pool of 10 nonatopic volunteer donors
were used in this study. Atopy was determined by a clinical history of
allergic symptoms to birch pollen, by positive skin prick test
responses to birch pollen extract, and by positive radioallergosorbent
test (RAST) (classes 3.5–5.0) to birch pollen. Nonatopic subjects were
healthy individuals without any allergic history, negative RAST
results, and negative skin prick test responses to birch pollen
allergens. A murine monoclonal anti-Bet v 1 antibody (BIP1, ref
17
) was raised against natural Bet v 1. Hybridoma
supernatant was stored at -20°C and used in a 1:10 dilution of TBS
[50 mM Tris/150 mM NaCl/0.5% Tween 20/0.5% bovine serum albumin
(BSA); pH 7.4] for immunoblots.
Protein quantification
Protein concentrations were determined by the Bradford method
according to the Bio-Rad-Microassay Procedure (Bio-Rad, Hercules,
Calif.). A standard curve was produced with known concentrations of BSA
(Sigma-Aldrich, St. Louis, Mo.). Ultraviolet absorption at 280 nm was
used to estimate sample protein concentration. Protein quantification
of tobacco rBet v 1 was performed by densitometric analysis using image
analysis software (E.A.S.Y., Enhanced Analysis System, Herolab,
Wiesloch, Germany) to calculate band intensities and comparing to a
standard dilution series of E. coli rBet v 1.
Immunoblot studies
Protein extracts from birch pollen, purified rBet v 1
produced in E. coli (7)
, extracts from N.
benthamiana plants expressing Bet v 1, and extracts from control
plants infected with the viral vector alone were separated by sodium
dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) in 12%
polyacrylamide gels and transferred to nitrocellulose membranes by
electroblotting according to standard protocols (32)
.
Protein extracts were adjusted to contain 1 µg Bet v 1 per lane.
Blots were then incubated with sera (diluted 1:5 in blocking buffer),
plasma (diluted 1:2) from birch pollen-allergic patients, or BIP1, a
mouse monoclonal anti-Bet v 1 IgG antibody (17)
. Bound IgE
was detected by 125I-labeled rabbit anti-human
IgE (RAST RIA, Pharmacia, Uppsala, Sweden), diluted 1:10, and
visualized on Biomax MS film (Kodak, Rochester, N.Y.). Alternatively,
to detect BIP1, alkaline phosphatase-conjugated rabbit anti-mouse IgG
(Jackson Immuno Lab. Inc., West Grove, Pa.), diluted 1:1000, was
used and developed with the BCIP (5-bromo-4-chloro-3-indolyl
phosphate)/NBTC (4-nitroblue tetrazolium chloride) reagent solutions
(Boehringer Mannheim, Mannheim, Germany).
ELISA and ELISA cross-inhibition experiments
Microtiter plates (Nunc-Immuno Plate, Nalge Nunc International,
Roskilde, Denmark) were coated with 5 µg/ml of the monoclonal
antibody BIP1 (17)
as catching antibody in PBS. After
blocking with PBS, 0.05% Tween 20, 1% BSA, plates were incubated with
increasing concentrations of purified nBet v 1, E. coli rBet
v 1 or rBet v 1 from crude N. benthamiana leaf extract
ranging from 0.125 µg/ml to 2.0 µg/ml. Crude leaf extract from
N. benthamiana infected with the viral vector not harboring
the Bet v 1 coding sequence was applied as control in the same serial
dilution as Bet v 1 containing N. benthamiana leaf extract.
After washing five times with PBS and 0.05% Tween 20, rabbit anti-Bet
v 1 IgG (polyclonal serum, diluted 1:10,000 in PBS/0.05% Tween
20/0.5% BSA) was applied to the microtiter plates and incubated for
3 h at room temperature. Washing was followed by addition of
peroxidase-conjugated mouse anti-rabbit IgG (Jackson Immuno Lab.
Inc., 1: 2000 PBS 0.05% Tween 20/0.5% BSA). Color development was
performed using 0.1% ABTS (azinoethylbenzothiozoline sulfonic acid
substrate, Sigma-Aldrich) and the optical density was measured at 405
nm (490 nm as reference wavelength) after 30 min using an ELISA reader
(Dynatech MR 7000, Alexandria, Va.). Adequate controls were performed
for each incubation step by replacing the respective ligand by PBS
(data not shown).
For cross-inhibition experiments, the Bet v 1-specific serum pool was diluted 1:10 in Tris-buffered saline (TBS), 0.05% Tween 20, 1% BSA, and preincubated (overnight at 4°C) with 40.0, 20.0, 10.0, 5.0, 2.5, 1.25, and 0 µg/ml of purified nBet v 1, protein extract from N. benthamiana plants expressing Bet v 1, or BSA. Microtiter plates were coated with 5.0 µg/ml BIP1, followed by 2 µg/ml of purified nBet v 1 or rBet v 1 in a crude N. benthamiana leaf extract. Afterward, preincubated sera were applied onto the BIP1/nBet v 1 and BIP1/tobacco rBet v 1-coated plates and incubated overnight at 4°C. After washing, the plates were incubated with alkaline phosphatase-conjugated mouse anti-human IgE antibody (PharMingen, San Diego, Calif.) and developed with the substrate p-nitrophenyl phosphate. The color reaction was measured as described above, except that the reference wavelength was at 550 nm. Controls with TBS and a normal human serum pool assured the specification of the IgE detection system (data not shown).
Sensitization of BALB/c mice
Seven-week-old female BALB/c mice were purchased from Charles
River (Sulzfeld, Germany). Mice (n=5/group) were either
injected with 1 µg purified rBet v 1 produced in E. coli,
crude N. benthamiana leaf extract containing 1 µg Bet v 1,
or crude birch pollen extract containing 1 µg Bet v 1. The antigens
adsorbed to Al(OH)3 were intraperitoneally
administered in 150 µl volumes. Control mice were injected with 150
µl crude leaf extract from N. benthamiana infected with
the viral vector not harboring the Bet v 1 coding sequence.
Immunizations were performed three times, on days 0, 14, and 28. Six
days after the last immunizations blood samples were taken from the
tail vein, serum prepared and stored at -20°C until analyzed.
Cutaneous type I hypersensitivity reaction in BALB/c mice
Seven days after the last immunization (i.e., on day 35 of the
immunization protocol), intradermal skin tests were performed as
described previously (14)
. One hundred microliters of
0.5% Evans blue (Sigma-Aldrich) were injected intravenously (i.v.)
into the tail vein of the mice. Subsequently, 30 µl of birch pollen
extract (2.5 µg/ml) and E. coli rBet v 1 (2.5 µg/ml)
were injected intradermally into the shaved abdominal skin. The mast
cell degranulation compound 48/80 (20 µg/ml, Sigma) was used as a
positive control. PBS was injected as a negative control. After 20 min,
the mice were killed and the color intensity of the reaction was
compared with the individual positive control on the inverted skin.
Determination of allergen-specific antibody levels in sera of
BALB/c mice by ELISA
Allergen-specific antibody levels in mouse sera were determined
by ELISA (14)
. In brief, microtiter plates were coated
with E. coli rBet v 1 (2 µg/ml). Serum samples diluted
1/1000 for IgG1, 1/500 for IgG2a, and 1/10 for IgE antibodies were
applied in 100 µl volumes. After washing, anti-mouse IgG1, IgG2a, and
IgE antibodies (1 µg/ml, PharMingen) were applied overnight at 4°C.
Detection of bound antibodies was performed with peroxidase-conjugated
mouse anti-rat IgG antibodies (1/1000, Jackson Immuno Lab. Inc.). ABTS
(Sigma-Aldrich) was used for color development and the absorbance
was measured at 405 nm with a Dynatech microplate reader. The antibody
levels were expressed in optical density units. The baseline levels
derived from the preimmune sera were individually subtracted from the
respective immune sera. Statistical analyses were performed with the
Mann Whitney U test.
Determination of allergen-specific IgE antibody levels by passive
cutaneous anaphylaxis (PCA)
Passive cutaneous anaphylactic reactions were performed in
Wistar rats. Pooled sera from E. coli rBet v 1 or tobacco
rBet v 1 immunized BALB/c mice were diluted 1:4 and a 1:20 in PBS and
intradermally injected in a 50 µl volume on both sides of the back of
the rats. Preimmune sera, in a 1:4 dilution, were injected as negative
controls. After 24 h, 1 mg of birch pollen extract in 1 ml 0.5%
Evans blue was i.v. injected into each animal. The mast cell
degranulation compound 48/80 was intradermally injected as a positive
control (50 µl of 1 mg/ml). The reaction was read 25 min after the
antigen challenge. The diameter of the reaction was measured and the
area and the color intensity was evaluated on the inside of the skin.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Expression of Bet v 1 in N. benthamiana plants
Mechanical inoculation of N. benthamiana plants with
in vitro transcripts derived from p4GD-Bet v 1 DNA resulted
in systemic infection 7 days postinoculation visible as mild leaf
deformations, some variable leaf mottling, and growth retardation.
RT-PCR was performed with RNA isolated from newly formed upper leaves 2
wk postinoculation. Direct sequencing of the RT-PCR product confirmed
the correct sequence of the Bet v 1 mRNA transcribed from the viral
subgenomic promoter in systemically infected leaves. Twenty-one days
postinoculation, leaf, stem, and root material was harvested, weighed,
and extracted. Ten microliters of the supernatant from infected leaves
were separated by SDS-PAGE and stained by Coomassie brilliant blue. As
Bet v 1 and the abundant viral coat protein migrated at approximately
the same molecular mass range (17.5 kDa), a differentiation between the
two proteins was not possible.
Subsequently, expression of tobacco rBet v 1 was analyzed by Western
blotting using the anti-Bet v 1 monoclonal antibody BIP1 (Fig. 1
). Plant tissue was homogenized in a 10-fold excess of buffer (w/v). Ten
microliters of the supernatant resulting from the centrifugation were
loaded per lane corresponding to ~1 mg fresh tissue. A dilution
series of E. coli rBet v 1 (0.25 µg, 0.1 µg, 0.05 µg,
0.01 µg) was loaded on the same gel to estimate the amounts of rBet v
1 expressed in the different tissues. BIP1 detected tobacco rBet v 1 in
protein extracts of upper leaves (Fig. 1
, lane 5), lower leaves (the
sites of infection; Fig. 1
, lane 6), stems (Fig. 1
, lane 7), and roots
(Fig. 1
, lane 8) 21 days postinoculation. In contrast, no BIP1 binding
could be observed to proteins in extracts from N.
benthamiana plants infected with viral RNA without the Bet v 1
insertion (Fig. 1
, lane 9). Tobacco rBet v 1 amounts present in the
different plant tissues were visualized by the intensity of the BIP1
binding and calculated by densitometric comparison to the BIP1 binding
intensity to the E. coli rBet v 1 dilution series (Fig. 1
,
lanes 1–4). Tobacco rBet v 1 present in 10 µl supernatant extracted
from the various tissues was calculated at 250 ng in upper leaves, in
lower leaves at 50 ng, in stems at 200 ng, and in roots at 75 ng.
Consequently, Bet v 1 was calculated to be present at 250 µg/g fresh
upper leaves (0.025%). Total soluble protein extracted from leaves was
determined by the method according to Bradford to be 10 mg/g fresh
leaf. Therefore, rBet v 1 was accumulated to 2.525% of the soluble
protein in leaves.
|
Immunoblots experiments
Immunoblot analyses of birch pollen extract (Fig. 2
, lanes 1), purified E. coli rBet v 1 (Fig. 2
, lanes 2),
protein extract of N. benthamiana leaves expressing Bet v 1
(Fig. 2
, lanes 3) or from control plants infected with virus not
harboring the Bet v 1 coding sequence (Fig. 2
, lanes 4) were performed
with individual sera from birch pollen-allergic patients and the
standard serum pool. All individual sera as well as the standard serum
pool displayed IgE binding to nBet v 1 present in birch pollen, to
purified E. coli rBet v 1, and to Bet v 1 expressed in
N. benthamiana (Fig. 2
, lanes 1–3). No IgE binding could be
detected to leaf protein extracts from control plants (Fig. 2
, lanes
4). The serum pool of nonatopic donors did not show IgE binding to any
proteins. Similarly, the buffer controls performed for an identical set
of protein blots without the addition of sera remained negative (data
not shown).
|
ELISA and ELISA cross-inhibition experiments
To compare the binding capacity to the monoclonal anti-nBet v 1
antibody BIP1 serial dilutions of nBet v 1, tobacco rBet v 1 and
E. coli rBet v 1 were tested in ELISA experiments. For each
antigen concentration, four determinations were performed. Mean values
are given in Fig. 3
. Detection was carried out using polyclonal rabbit anti-Bet v 1 IgG and
peroxidase-conjugated mouse anti-rabbit IgG antibodies. At all
evaluated protein concentrations, a comparable binding capacity was
observed for each preparation of Bet v 1 in a dose-dependent manner
(Fig. 3)
. No specific binding was observed in the controls (data not
shown).
|
Using the standard serum pool of birch pollen-allergic patients,
cross-inhibition was determined by ELISA experiments. Natural Bet v 1
and tobacco rBet v 1 had the same inhibitory effect on IgE binding to
nBet v 1 (Fig. 4A
) and to tobacco rBet v 1 (Fig. 4B
),
respectively, demonstrating that the IgE binding epitopes of the two
proteins are equivalent. Amounts as low as 2.1 µg/ml for nBet v 1 and
2.3 µg/ml for tobacco rBet v 1 resulted in 50% inhibition of IgE
binding nBet v 1 (Fig. 4A
), whereas 50% inhibition of IgE
binding to tobacco rBet v 1 was achieved by a concentration of 2.5
µg/ml tobacco rBet v 1 or 3.1 µg/ml nBet v 1. Both proteins were
able to fully inhibit the IgE binding to solid-phase bound nBet v 1 or
tobacco rBet v 1. No inhibition of patients’ IgE binding to nBet v 1
or tobacco rBet v 1 was observed when sera were preincubated with BSA
(Fig. 4A, B
).
|
Cutaneous type I hypersensitivity reaction in BALB/c mice and PCA
BALB/c mice immunized with purified rBet v 1 produced in E.
coli [Fig. 5
; rBet v 1/Al(OH)3], crude N.
benthamina leaf extract containing Bet v 1 [Fig. 5
; tBet v
1/Al(OH)3], or crude birch pollen extract [Fig. 5
; BP/Al(OH)3] displayed equally strong
immediate type skin reactions 20 min after intradermal injection of
E. coli rBet v 1 (Fig. 5
, lower right quadrants) or birch
pollen extract (Fig. 5
, lower left quadrants) in comparison to the
positive control, the mast cell degranulation compound 48/80 (Fig. 5
,
upper left quadrants). No allergen-specific skin reactions were
elicited in the control mice immunized with crude leaf extract from
N. benthamiana plants infected with the viral vector without
the Bet v 1 coding sequence [Fig. 5
;
N.b./Al(OH)3]. Injections with PBS
alone did not elicit a skin reaction in any of the animals (Fig. 5
,
upper right quadrants). To confirm that the type I skin reactions were
due to allergen-specific IgE antibodies and not to allergen-specific
IgG1, PCA reactions were performed in Wistar rats using the immune sera
of the BALB/c mice. The size of the reaction elicited by injection of
tobacco rBet v 1 immune sera (1:4 dilution) into the rats was 89% of
that elicited by the E. coli rBet v 1 immune sera. The color
intensity of both reactions was comparable. The 1:20 dilution of both
mice immune serum pools induced only weak to negative reactions. The
PCA results underline that the type I skin tests were mediated by
comparable allergen-specific IgE antibody levels.
|
Allergen-specific antibody levels in sera of BALB/c mice
In accordance with the type I skin tests and PCA,
allergen-specific IgE antibodies, determined by ELISA, were equally
high in mice immunized with purified E. coli rBet v 1
(OD405=0.6±0.16) and nonpurified tobacco rBet v
1 (OD405=0.6±0.17). IgG1 antibody levels were
slightly higher in the E. coli rBet v 1-immunized mice
(OD405=2.3±0.4) than in tobacco rBet v
1-immunized mice (1.01±0.13), but the difference was not statistically
significant. IgG2a antibody levels were low in either of the groups
(E. coli rBet v 1: OD405=0.06±0.08,
tobacco rBet v 1: OD405=0.01±0.17). Thus,
immunization with E. coli rBet v 1 or tobacco rBet v 1 led
to a typical Th2-like immune response in vivo and in
vitro. The immune response after the immunization with BP extract
also displayed a Th2 profile (IgE:
OD405=0.29±0.15; IgG1:
OD405=1.75±0.01; IgG2a:
OD405=0.04±0.1). Immunization with extracts form
N. benthamina plants not expressing Bet v 1 did not result
in allergen-specific antibody responses as already observed in the type
I skin tests in vivo. Baseline OD405
units of preimmune sera from all animals were 0.156 ± 0.012 for
IgG1, 0.246 ± 0.029 for IgG2a, and 0.166 ± 0.014 for IgE.
The values reported above are those from the immune sera after
subtraction of the baseline values.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There is a growing need for the experimental, diagnostic, and clinical
use of pure allergens (2
3
4)
. Based on the detailed
information available on the birch pollen allergen Bet v 1
(5
6
7
8
9
10
11
12
13
14
15
16)
, this protein was chosen as a model allergen to
assess the possibilities of transient expression in plants infected by
recombinant viral vectors. As isolation from natural sources is
exceedingly time and cost intensive, allergens are conventionally
produced as recombinant proteins in E. coli. By comparison,
few examples for allergen expression in yeast have been published
(19
20
21
22
23
24)
. Recombinant production of properly folded
proteins in plants represents a major advantage over the E.
coli system, where recombinant proteins often have to be refolded
by denaturation/renaturation. In many cases only a small fraction of
the protein adopts the correct conformation. It is generally assumed
that the expression of plant allergens in a homologous system poses
fewer problems regarding the correct folding and solubility of
recombinant proteins. In addition, plant-expressed recombinant proteins
do not require extensive purification from bacterial toxins, making
production less time and cost intensive.
It has been demonstrated that transgenic plants that express foreign
proteins with pharmaceutical value represent a feasible alternative to
fermentation-based production systems.
Agrobacterium-mediated transformation has been successfully
applied for production of several subunit vaccines
(25
26
27
28)
. Mice fed with extracts or edible tissue of these
plants developed both serum and secretory antibodies specific to the
recombinant antigen (25
26
27
28)
. However, to obtain
genetically engineered plants after Agrobacterium-mediated
transformation is a very time-consuming process and the quantity of the
heterologous protein expressed does not always reach satisfactory
levels.
In contrast to conventional plant transgenic approaches, a second
strategy for the transient expression of foreign proteins in plants has
been developed using viral vectors. Biologically active
-trichosanthin and epitopes with immunoprotective properties have
been produced in tobacco plants transfected by recombinant TMV
(30
, 33
, 34)
. Very recently, the production of a
tumor-specific vaccine in N. benthamina plants has been
reported using a TMV-based expression system (31)
. The
expression and application of a single-chain variable region of an Ig
of a B cell lymphoma clearly demonstrated the speed and the efficiency
of this system (31)
. We have applied for the first time
plant virus-directed transient expression to produce an allergen in the
tobacco-related species N. benthamiana. It was our aim to
establish and evaluate the suitability of such a plant-based expression
system for the production of plant-derived allergens as an alternative
to E. coli or yeast-based production systems. Therefore, it
was required that the plant-expressed heterologous allergen displayed
immunological characteristics equivalent to the natural counterpart.
We used a TMV vector to transiently express the allergen Bet v 1 by
placing its coding sequence under the control of the viral coat protein
subgenomic promoter (29)
. We were able to demonstrate the
expression of Bet v 1 in leaves, stems, and roots of infected N.
benthamiana plants (Fig. 1)
. In leaves, Bet v 1 was accumulated to
2.525% of the total soluble protein, which is slightly higher as
reported for the TMV-directed expression of -trichosanthin
(33)
. These findings show that this system is capable of
producing high amounts of recombinant protein, exceeding those of
transgenic plants. Levels of 0.23% were reported for expression of the
Norwalk virus capsid protein in transgenic tobacco leaves and 0.3% for
the cholera toxin B subunit in potato microtubers (26
, 28)
. Bet v 1 expression in N. benthamiana by far
exceeds the 0.01% yield found for the hepatitis B virus surface
antigen or for E. coli LT-B in transgenic tobacco or potato
(25
, 27)
.
Binding of IgE antibodies to Bet v 1 is strictly related to the
molecule’s conformation. Slight changes in the conformation of Bet v 1
significantly lower its capacity to bind IgE (35)
. More
severe alterations of the molecule’s structure completely abolish its
IgE binding capability (35)
. We used several individual
sera and a standard serum pool of birch pollen-allergic patients’ sera
to cover the spectrum of IgE antibodies induced in atopics as a
response to the sensitization against Bet v 1. The IgE binding
capacities of Bet v 1 from birch pollen, rBet v 1 from E.
coli, and rBet v 1 from N. benthamiana were equivalent
(Fig. 2)
. No detectable cross-reactive protein was observed in the 17
kDa range in the extracts of N. benthamiana control plants
infected with the viral vector without the Bet v 1 coding sequence
(Fig. 2
, lanes 4). Cross-reactive allergens in the higher molecular
mass ranges as observed in the immunoblots have been detected in a wide
range of plant species and are assumed to be homologous conserved
proteins. Their identity is still under investigation.
In addition to the immunoblot studies, we have used ELISA and ELISA
inhibition assays to evaluate the immunological characteristics of Bet
v 1 produced in N. benthamiana plants. Comparable binding
properties to the anti-nBet v 1 monoclonal antibody BIP1 were observed
for natural Bet v 1 purified from birch pollen, Bet v 1 expressed in
N. benthamiana, and Bet v 1 produced in E. coli.
This result confirms that the BIP1 epitope is well preserved during the
expression of Bet v 1 in N. benthamiana (Fig. 3)
. Moreover,
patients’ sera preadsorbed with either nBet v 1 or tobacco rBet v 1
lost their IgE reactivity to nBet v 1 (Fig. 4A
) as well as
to tobacco rBet v 1 (Fig. 4B
) in a dose-dependent and
comparable manner. From these results, we conclude that the IgE binding
properties of the tobacco Bet v 1 are equivalent to Bet v 1 present in
birch pollen. These cross-inhibition experiments further indicate that
the conformation of Bet v 1 produced in N. benthamiana is
equivalent to the Bet v 1 from birch pollen, as even small changes in
the overall 3-dimensional structure abolish IgE binding reactivity
(35
, 36)
.
We had previously established a mouse model for type I allergy to birch
pollen. The immunization protocols used in this model led to increased
IgE/IgG1 vs. low IgG2a antibody levels and positive skin tests in
vivo, thus representing an immune response comparable to that of
human type I allergy (14
, 37
, 38)
. Consequently, this
mouse model is a useful tool to evaluate and characterize new systems
for recombinant allergen production. Previous studies have shown that
no significant immunological differences between purified E.
coli rBet v 1 and natural Bet v 1 from birch pollen exist
(6
, 15
, 38)
. We have applied this mouse model to compare
immune responses after immunization with standard purified rBet v 1
produced in E. coli to Bet v 1 produced in N.
benthamiana plants. Notably, the application of Bet v 1 in a total
plant protein extract (i.e., nonpurified) induced immune responses that
were equal in terms of antibody production profile as well as skin test
reactivity to those after immunization with purified E. coli
rBet v 1. We have further demonstrated that no antibody responses or
positive skin reactivity to the allergen were induced after
immunization with the crude extract from N. benthamiana
plants infected with the TMV vector not carrying the Bet v 1 coding
sequence.
Low toxicity, solubility, and production of high amounts of correctly
folded recombinant proteins at low cost are prerequisites for
developing new vaccination strategies. Vaccines against several
infectious diseases have already been produced in transgenic plants
(25
26
27
28)
. In particular, production of oral vaccines in
edible plant tissues expressing subunit vaccines has shown the
feasibility of this concept (25
26
27
28)
. Only recently the
concept of oral tolerance or oral vaccination has also gained interest
for the treatment of allergic diseases (39
40
41)
. The
required doses of recombinant allergens for oral applications are
substantial and therefore expensive. Using the TMV-directed transient
expression system, the above-mentioned requirements can be met. In any
case, this expression system, which offers a low cost means of
producing allergens in plants, will support a broad range of
applications in the field of molecular allergology.
|
ACKNOWLEDGMENTS |
|---|
Received for publication August 20, 1999.
Revision received December 13, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-tichosanthin in transfected plants by an RNA viral vector. Proc. Natl. Acad. Sci. USA 90,427-430This article has been cited by other articles:
|
|
 |
|
  I. Scholl, N. Kalkura, Y. Shedziankova, A. Bergmann, P. Verdino, R. Knittelfelder, T. Kopp, B. Hantusch, C. Betzel, K. Dierks, et al. Dimerization of the Major Birch Pollen Allergen Bet v 1 Is Important for its In Vivo IgE-Cross-Linking Potential in Mice J. Immunol., November 15, 2005; 175(10): 6645 - 6650. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP |
), E.
coli rBet v 1 (
), and rBet v 1 from protein extracts of
N. benthamiana leaves infected with the viral construct
(
) were compared in an ELISA assay for their binding capacity to the
mouse monoclonal anti-nBet v 1 antibody BIP1. Protein extracts from
N. benthamiana leaves infected with the virus without
the Bet v 1 coding sequence (x) were used as controls. Bet v 1 bound to
BIP1 was detected with a polyclonal rabbit anti-nBet v 1 antiserum.
