| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of General and Experimental Pathology, University of Vienna, 1090 Vienna, Austria; and the
Institute of Immunology and Allergology, University of Bern, 3010 Bern, Switzerland
1Correspondence: Department of General and Experimental Pathology, University of Vienna, AKH, EBO-3Q, Waehringer Guertel 18–20, A-1090 Vienna, Austria. E-mail: erika.jensen-jarolim{at}akh-wien.ac.at
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Bet v 1 • Bet mim 1 • Prof mim 1 • type I allergy • bacteriophage • mimotope • oral immunization • digestion • food allergen • BALB/c mouse model
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. However, it has
been reported that patients also suffer from allergies to plant-derived
food without serological or apparent clinical pollen allergy
(3)
. Several possibilities for sensitization to food
allergens have been postulated. Food allergens, in contrast to most
food antigens, must be relatively resistant to gastric and eventually
pancreatic digestion in order to withstand the oral route and be
presented to the immune system in the gut (4
, 5)
.
Recently, an animal model for oral sensitization to food proteins has
been established (6)
. Furthermore, molecular mimicry of
allergens by gastrointestinal pathogens such as viruses might be
responsible for enhanced allergic responses to a respective allergen
(7
, 8)
. This is paralleled by the observation that
allergic asthma is aggravated by viral infections of the upper
respiratory tract (9
, 10)
.
Filamentous phages are virion particles that are able to infect
F-pili positive Escherichia coli and are frequently used to
display random peptides either on the virion coat protein pIII
(11
, 12)
or in multicopy fused to the coat protein pVIII
(13
, 14)
. Such phage peptide libraries display a
repertoire of a complexity in the order of 108 or
more possible structures. The random peptides are, depending on the
library, frequently composed of 6 to 20 amino acid residues, and have
been shown to mimic natural epitopes or ligands of monoclonal
antibodies (for review, see ref 15
). In previous
experiments, we have generated a mimotope of Bet v 1, the major birch
pollen allergen (16)
. This mimotope was selected from a
constrained random peptide phage library by using BIP1, a monoclonal
anti-Bet v 1 Ab2 (14
, 17)
, and is
designated Bet mim 1.
Depending on the epitope specificity, murine immunoglobulin G
(IgG)2
antibodies to Bet v 1 can
enhance or inhibit IgE binding to Bet v 1 in vitro
(18)
. The BIP1 monoclonal antibody (mAb) was described to
significantly enhance IgE binding to Bet v 1 (18
, 19)
. We
demonstrated recently that oral administration of Bet mim 1 induced an
anti-Bet v 1-specific IgG response in BALB/c mice (16)
.
Based on these observations, we tested the influence of an oral
immunization with Bet mim 1 on the subsequent development of type I
allergy to birch pollen. Mice were fed with phages displaying Bet mim 1
or control mimotopes or with phosphate-buffered saline (PBS), and
subsequently sensitized to Bet v 1 adsorbed to
Al(OH)3, an adjuvant known to induce IgE-antibody
responses (20)
. Type I allergic reactions were evaluated
by immediate type skin testing of sensitized mice pretreated with
phages or PBS. As the phages were applied by the oral route, we
examined the grade of resistance to physiological gastrointestinal
digestion of natural allergen epitopes and compared them to
phage-displayed mimotopes. Since a BIP1 epitope is also present on the
Bet v 1-homologous allergen Mal d 1 of apple, this allergen was
selected as a model for a pollen-related allergen from plant food
(21
, 22)
.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
or circular (14)
nonapeptides fused to coat protein pVIII of the filamentous
bacteriophage fd. Libraries were kindly provided by IRBM (Istituto di
Ricerce Biologia Molecolare, Rome, Italy). Logarithmically growing E. coli XL-1 Blue were infected with phages after panning and grown in SB medium containing carbenicillin (50 µg/ml) and tetracycline (10 µg/ml), purchased from Fluka Chemie AG (Buchs, Switzerland), for 2 h at 37°C. After infection with helper phage VCSM13 (1012 pfu/ml), the culture was incubated for 2 h at 37°C. Kanamycin (70 µg/ml) was added and the culture further incubated at 37°C, overnight. Phages were harvested by precipitation on ice for 30 min with 4% (w/v) polyethylene glycol 8000 and 3% (w/v) NaCl. After centrifugation (1000xg, 30 min, 4°C), pellets were resuspended in PBS/0.15% casein and centrifuged. Supernatants were immediately used for biopanning experiments.
Monoclonal antibodies
mAb BIP1 was raised by immunizing BALB/c mice with extract from
birch pollen (Allergon AB, Engelholm, Sweden) as described earlier
(17)
. This antibody recognizes Bet v 1, the major birch
pollen allergen.
Generation of the mimotopes Bet mim 1 and Prof mim 1
Biopannings were essentially performed according to Barbas et
al. (23)
, described in detail earlier (16)
.
In short, enzyme-linked immunoassay plates (Nunc, Roskilde, Denmark)
were coated with 4 µg BIP1 mAb in 100 µl bicarbonate buffer, pH
8.5. Wells were treated with a blocking buffer containing 3% (w/v)
bovine serum albumin (BSA) in PBS. 1011
plaque-forming units of freshly prepared phages of the pooled circular
and linear libraries were incubated at 37°C for 2 h. Unbound
phages were removed by washing with Tris-buffered saline/0.5% Tween
20. Bound phages were eluted with 0.1 M HCl pH 2.2 and neutralized.
Eluents were amplified by infecting E. coli XL-1 Blue cells.
Three rounds of biopanning were performed and the phage ligands were
characterized by colony screening and DNA sequencing. The phage clone
selected by BIP1 expressed the mimotope CFPYCYPSESA (16)
,
which we now designate Bet mim 1. From a different set of biopanning
experiments, a mimotope of profilins, a different class of allergens,
for human IgE was characterized (CAAISGYPVC) (24)
. This
mimotope is now termed Prof mim 1 and used as a control in inhibition
and immunization experiments.
Extracts from birch pollen and recombinant allergens rBet v 1 and
rMal d 1
Birch pollen purchased from Allergon AB was extracted as
described earlier (25)
. Protein concentrations were
determined with the Bio-Rad Protein assay (Bio-Rad Laboratories,
Hercules, Calif.). Recombinant allergens rBet v 1 and rMal d 1
(21
, 22)
were from Biomay Biotechnik Produktions-und
Handels-GesmbH (Linz, Austria).
Intragastric and intraperitoneal immunization of BALB/c mice
BALB/c mice were immunized by intragastric gavages (200 µl per
gavage, 2 x 108 cfu/ml PBS) with phages
expressing the BIP1 mimotope Bet mim 1 (CFPYCYPSESA) (group A,
n=5). Group B (n=5) received phages expressing a
cross-reactive IgE mimotope for profilin, Prof mim 1 (CAISGGYPVC)
(24)
; group C (n=5) received the same amount of
PBS only. All groups were fed the respective solutions on days 0, 7,
14, and 35. On days 49 and 63, all mice were immunized
intraperitoneally (i.p.) with 1 µg rBet v 1 in 50 µl PBS and 100
µl Al(OH)3. All immunizations were performed by
Biovendor-Laboratory Medicine (Brno, Czech Republic).
Type I hypersensitivity reaction skin test
Skin tests were performed 7 days after the last immunization,
essentially as described previously (20)
. In short, 100
µl of 0.5% Evans Blue (Sigma, Dreisenhofen, Germany) was injected
intravenously into the tail vein of the mice. Subsequently, 30 µl of
rBet v 1 or recombinant profilin (r profilin) (2.5 µg/ml each) was
injected intradermally into the shaved abdominal skin. The mast cell
degranulation compound 48/80 (20 µg/ml; Sigma) was injected as
positive control and PBS as negative control. After 15–20 min, mice
were killed, skinned, and the diameter of the color reaction was
measured on the inside of the skin. The color intensity of the reaction
was determined by a hand-held densitometer (Vipdens, Brixen, Italy).
The skin response index was calculated from staining density x
diameter.
Gelelectrophoresis, immunodots, and immunoblots
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)/blotting
SDS-PAGE was performed according to the Laemmli protocols
(26)
. Electrophoresis was performed in a vertical slab gel
apparatus (Biometra, Biomedizinische Analytik GmbH, Göttingen,
Germany) and proteins were blotted to nitrocellulose with 0.4 µm pore
size (Schleicher & Schüll, Dassel, Germany).
Immunodots
Allergens rBet v 1 and rMal d 1 were dotted in duplicate onto
nitrocellulose, 0.5 µg in 1 µl/dot. Phages were applied at 0.5 x 104 cfu/µl per dot. Dot blots were air dried
and saturated with 50 mmol/l PBS (pH 7.5), 0.5% (w/v) BSA, and 0.5%
(v/v) Tween 20. Five micrograms per milliliter of mAb BIP1 was used per
blot strip and incubated at 4°C overnight. After washing with
PBS/Tween, bound mouse IgG was detected by
125I-labeled sheep anti-mouse Ig (Amersham Life
Science, Little Chalfont, England).
Inhibition of mAb binding by phage
mAb BIP1 was used directly or after preincubation with
107 cfu Bet mim 1-phages or Prof mim 1-phages/ml
at 4°C, overnight.
Detection of rabbit antibodies in immunoblot/dot
Rabbit anti-Bet v 1 Ab was diluted 1:1000 and applied to blots
overnight, 4°C. Bound rabbit Ig was detected by
125I-labeled anti-rabbit Ig (Amersham).
Detection of human IgE or IgG in immunoblot/dot
For detection of IgE or IgG, blots were incubated with a serum
pool of equal parts of 10 human sera with high levels of anti-Bet v 1
IgE (1:4 for IgE, 1:20 for IgG). After washing with PBS/Tween, bound
IgE was detected by 125I-labeled anti-human IgE
or IgG Ab (MALT Allergie System, IBL GmbH, Hamburg, Germany). Blots
were washed, dried and exposed to Kodak Biomax MS autoradiography films
at -70°C.
Digestion experiments
Recombinant allergens rBet v 1 and rMal d 1 (0.5 mg/ml), or
phage preparations (0.5x107 cfu/ml) were treated
with PBS alone or, for gastric digestion, with 0.87 g/l pepsin (Sigma)
in HCl pH 2.0, 37°C, for 30 s, 1 min, 5 min, 15 min, or 30 min,
at constant agitation (5)
. For additional pancreatic
digestion, samples were subsequently treated with 422 mg/l trypsin
(Sigma) in 76 mmol/l NaHCO3, pH 7.7, for 30 min
at 37°C at constant agitation. The reaction was stopped by cooling
samples on ice. Samples were stored at -20°C or used immediately for
SDS-PAGE or dots.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, lane 1). Preincubation with 107 cfu Bet mim
1-phages/ml resulted in a 90% inhibition (Fig. 1
, lane 2). Therefore,
phages displaying the peptide CFPYCYPSESA could dose dependently reduce
the binding of BIP1 to blotted Bet v 1 from whole birch pollen extract.
Preincubation with 107 cfu/ml phages expressing a
cross-reactive IgE epitope of profilin (Prof mim 1; CAAISGYPVC)
(24)
did not affect the binding (Fig. 1
, lane 3).
|
Oral administration of phages to BALB/c mice enhances allergic skin
responses
Phages displaying Bet mim 1 (2x108 cfu/ml,
representing ~1012 mimotopes per feeding) were
administered to BALB/c mice by intragastric gavages (group A) on days
0, 7, 14, and 35. Control mice were fed with the same amounts of phages
expressing Prof mim 1 (group B) or with PBS only (group C). After four
intragastric applications, all animals were subjected to dual i.p.
immunizations using rBet v 1 with Al(OH)3 as
adjuvant. Skin testing was performed with rBet v 1, r profilin, the
histamine releasing compound 48/80 as positive control, and PBS as
negative control. One of the five mice in group A was a nonresponder.
Representative skins are shown in Fig. 2
. Only animals pretreated with Bet mim 1-phages showed substantial
enhancement of Bet v 1-induced skin reactivity (Fig. 2A
) as
compared with the Prof mim 1 and PBS control groups (Fig. 2B, C
). Intragastric application of Bet mim 1-phages without
consecutive i.p. sensitization by rBet v 1 did not induce type I skin
hypersensitivities in mice (data not shown).
|
A skin response index was calculated from color intensity and diameter
of the skin reactions (Fig. 3
). The relative skin responses to rBet v 1 were increased markedly in
the group fed with Bet mim 1-phages (n=4) as compared to
group B (Prof mim 1; n=5) or to group C (PBS;
n=5).
|
Digestion experiments with rBet v 1, rMal d 1, and phages
displaying Bet mim 1
The stability of the mimotope in comparison with the natural
epitopes on Bet v 1 or Mal d 1 was tested in a gastric and pancreatic
digestion assay. Figure 4
demonstrates that the recombinant allergens Bet v 1 and Mal d 1 (Fig. 4
, lanes 1) are already greatly degraded after 30 s treatment
under physiological gastric conditions (Fig. 4
, lanes 2). Longer
gastric digestion (30 s to 30 min) or additional pancreatic digestion
(Fig. 4
, lanes 3) did not affect these results. In
SDS-PAGE/immunoblotting, no digested fragments were detected by BIP1, a
rabbit anti-Bet v 1 Ab, or human IgE from a patients'serum pool
composed of 10 sera with high anti-Bet v 1 IgE levels.
|
As the fragments might be too small to be apparent in electrophoresis
but could still harbor antigenic epitopes, we performed an immunodot
assay (Fig. 5
). The recombinant allergens rBet v 1 and rMal d 1 and Bet mim 1-phages
were subjected to gastric digestion (30 s to 30 min) or gastric (30
min) and pancreatic digestion (30 min) (Fig. 5)
. In the
immunodot, epitopes for BIP1 as well as for human IgE and IgG were no
more detectable on rBet v 1 and rMal d 1 after 30 s gastric
digestion (Fig. 5
, lanes 2). Using the rabbit anti-Bet v 1 Ab, some of
the antigenic epitopes on the digested allergens could still be
detected after gastric and pancreatic digestion, although substantially
reduced (Fig. 5
, lanes 3).
|
In contrast, the artificial epitope Bet mim 1 displayed by phage was
still apparent after 30 min gastric digestion (Fig. 5
, lanes 2) and in
low amounts after combined 30 min gastric and 30 min pancreatic
digestion (Fig. 5
, lanes 3). BIP1 bound to untreated and (although to a
lesser extent) digested Bet mim 1-phages, but not to control phages
expressing Prof mim 1; data not shown. The polyclonal rabbit anti-Bet v
1 Ab recognized phage-displayed Bet mim 1, and binding intensities
decreased only slightly during digestion. Undigested Bet mim 1 was also
recognized by human IgG, but not by human IgE.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Monoclonal antibodies were important
tools for detecting cross-reactive epitopes on these homologous
molecules (28)
. In addition, it has been recognized that
the binding of mAbs to Bet v 1 can in vitro enhance or
inhibit subsequent IgE binding to the allergen (18
, 19
, 29)
. Preincubation of Bet v 1 with the murine mAb BIP1
significantly enhanced the binding of human IgE to Bet v 1 (18
, 19)
. In addition, enhanced IgE binding has also been shown using
human IgG mAbs in place of BIP1 (29)
. This effect was
explained by a possible modification of the conformation of Bet v 1
through the binding of BIP1, thereby increasing the accessibility of
epitopes for IgE.
We recently generated a mimotope for BIP1, termed Bet mim 1, and have
shown that oral immunization of mice with Bet mim 1 results in the
formation of Bet v 1-specific IgG (16)
. In the present
study, we tested the biological effect of an oral immunization with Bet
mim 1. We have chosen the in vivo type I skin test as a
highly sensitive method to measure the allergen-specific IgE
production. Oral immunization with Bet mim 1-displaying phages
per se did not induce allergen-specific IgE-antibodies,
which was reflected in negative type I skin tests (data not shown).
BALB/c mice fed with Bet mim 1-displaying phages or controls were
subsequently also treated by systemic sensitization with
Al(OH)3-adsorbed rBet v 1. In the group of mice
pretreated with Bet mim 1-displaying phages, Bet v 1-specific skin
reactivity was substantially enhanced. This enhancing effect did not
depend on the phages used as carrier material and was specifically
related to the type of mimotope used for the feedings. This is
concluded from the fact that mice fed with phages displaying the Prof
mim 1 mimotope of a nonrelated allergen (profilin) did not differ in
their skin reactivity to control mice pretreated with PBS. In
accordance with the in vitro effects of mAbs discussed above
(18
, 19
, 30)
, we conclude from our in vivo
model that preexisting allergen-specific IgG antibodies may enhance IgE
binding to the respective allergen, resulting in augmented mediator
release, at least in mice. It is conceivable that epitopes or mimotopes
that induce BIP1-type IgG also enhance type I allergic reactions to Bet
v 1 and its homologues in allergic patients. This mechanism may account
for the majority of birch pollen allergic patients, as 96% show IgE
binding to Bet v 1 (25)
. The presence of Bet v 1 mimics in
food may thereby trigger enhanced symptoms of birch pollinosis.
Digestion assays suggested that resistance to gastric digestion is a
characteristic of food allergens (5)
. Here we demonstrated
that the naturally occurring allergens Bet v 1 (22)
and
its apple homologue, Mal d 1 (21)
, were degraded within
seconds in a gastric digestion simulation assay in vitro. In
contrast, typical food allergens resist degradation for 30 to 60 min
(5)
. Consequently, Bet v 1 and Mal d 1 did not fulfill the
characteristics postulated for food allergens, at least not at
physiological acidic conditions in the stomach. This explains why type
I reactions after ingestion of apples typically affect the oral mucosa,
but do not lead regularly to gastrointestinal allergic reactions
(30)
. It is possible that under hypoacidic conditions in
the stomach or during diseases with accelerated gastrointestinal
passage, proper physiological digestion may be hindered, leading to
larger leftovers of allergens. Moreover, it has been observed that
groups of adult food allergic patients often are dominated by females,
which could be explained by a physiologically higher pH of their
gastric juices as compared to men (31
, 32)
.
In contrast, phage-displayed mimotopes were resistant to digestion for extended periods of time, as shown for Bet mim 1, which resisted gastric treatment for at least 30 min. This indicates that phages can be used as appropriate vectors for presenting allergen epitopes to the gut-associated lymphoid tissue.
Taken together, these data strongly suggest that mimics of allergenic epitopes present in the diet and resistant to digestion may modulate allergic responses via the formation of allergen-specific IgG. In addition, conditions leading to changes in the physiological pH of the gastric juices (e.g., antacid medications) may facilitate this immunization pathway in atopic individuals. Therefore, sensitization via the oral route by dietary mimotopes may represent a mechanism for aggravating type I allergic reactions, probably leading to an earlier onset of symptoms even at lower allergen dosage.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication February 1, 1999. Accepted for publication April 24, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This 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]
|
|
|
|
 |
|
 I. Scholl, E. Untersmayr, N. Bakos, F. Roth-Walter, A. Gleiss, G. Boltz-Nitulescu, O. Scheiner, and E. Jensen-Jarolim Antiulcer drugs promote oral sensitization and hypersensitivity to hazelnut allergens in BALB/c mice and humans Am. J. Clinical Nutrition, January 1, 2005; 81(1): 154 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. GANGLBERGER, K. GRÜNBERGER, B. SPONER, C. RADAUER, H. BREITENEDER, G. BOLTZ-NITULESCU, O. SCHEINER, and E. JENSEN-JAROLIM Allergen mimotopes for 3-dimensional epitope search and induction of antibodies inhibiting human IgE FASEB J, November 1, 2000; 14(14): 2177 - 2184. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
