Modulation of the T Cell Response to β-Lactoglobulin by Conjugation

β-LG or the conjugates were stimulated with β-LG, and the in vivo T cell response ... suppression of the generation of T cell epitopes by conjugatio...
0 downloads 0 Views 171KB Size
Download PDF
168

Bioconjugate Chem. 2003, 14, 168−176

Modulation of the T Cell Response to β-Lactoglobulin by Conjugation with Carboxymethyl Dextran Kazuo Kobayashi, Tadashi Yoshida, Koji Takahashi, and Makoto Hattori* Department of Applied Biological Science, Faculty of Agriculture, Tokyo University of Agriculture and Technology, Tokyo, Japan. Received July 31, 2002; Revised Manuscript Received October 22, 2002

We have previously prepared β-lactoglobulin (β-LG)-carboxymethyl dextran (CMD) conjugates with water-soluble carbodiimide and achieved reduced immunogenicity of β-LG. In the present study, to elucidate the mechanism for the reduced immunogenicity of β-LG, we investigated changes in the T cell response to β-LG after conjugation with CMDs differing in molecular weight (about 40 and 162 kDa). Lymph node cells from BALB/c, C3H/He, and C57BL/6 mice that had been immunized with β-LG or the conjugates were stimulated with β-LG, and the in vivo T cell response was then evaluated by BrdU (5-bromo-2′-deoxyuridine) ELISA as the ex vivo proliferative response. T cells from the conjugate-immunized mice showed a lower proliferative response than those from the β-LG-immunized mice. T cell epitope scanning, using synthesized peptides, showed that the T cell epitope profiles of the conjugates were similar to those of β-LG, whereas the proliferative response to each epitope was reduced. These results indicate that the lower in vivo T cell response with the conjugates was not due to induction of conjugate-specific T cells, but due to a decrease in the number of β-LG-specific T cells. After the lymph node cells from β-LG-immunized mice had been stimulated with β-LG or the conjugates, the efficiency of the antigen presentation of the conjugate to β-LG-specific T cells was evaluated by BrdU ELISA as the in vitro proliferative response. The antigen presentation of β-LG to the T cells was reduced by conjugation with CMD. In addition, conjugation with CMD enhanced the resistance of β-LG to cathepsin B and cathepsin D, which suggest that conjugation with CMD inhibited the degradation of β-LG by proteases in APC and led to suppression of the generation of antigenic peptides including T cell epitopes from β-LG. It is therefore considered that the suppressive effect on the generation of T cell epitopes reduced the antigen presentation of the conjugates and that this reduction led to a decrease in the number of β-LG-specific T cells in vivo. As a result, the decreased help to B cells by T cells would have reduced the antibody response to β-LG. We conclude that suppression of the generation of T cell epitopes by conjugation with CMD is important to the mechanism for the reduced immunogenicity of β-LG.

INTRODUCTION

Allergic reactions to food and dietary components have shown a tendency to increase in advanced countries, and food allergies have become an important human health problem. It has been reported that 12.6% of children in Japan had experienced an allergic reaction within 1 h after ingesting food and that the most common sources of food allergens involved were eggs (52.3% of food allergy cases), milk (31.8%), and seafoods (10.6%) (1). Although the fundamental therapy for food allergy is a food elimination diet, a prolonged period of food elimination affects a child’s growth (1). In particular, milk is a basic food component that plays a predominant role in the first * Corresponding author. Address: Department of Applied Biological Science, Tokyo University of Agriculture and Technology, 3-5-8 Saiwai-cho, Fuchu-City, Tokyo 183-8509, Japan. Telephone: +81-42-367-5879. Fax: +81-42-360-8830. E-mail: [email protected]. 1 Abbreviations: APC, antigen-presenting cells; β-LG, β-lactoglobulin; BrdU, 5-bromo-2′-deoxyuridine; CD, circular dichroism; CMD, carboxymethyl dextran; EDC, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide; ELISA, enzyme-linked immunosorbent assay; mAb, monoclonal antibody; 2-ME, 2-mercaptoethanol; MHC, major histocompatibility complex; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline; PEG, poly(ethylene glycol); SDS-PAGE, sodium dodecyl sulfatepolyacrylamide gel electrophoresis.

years of life. The major allergen in cow’s milk is β-lactoglobulin (β-LG), and about 82% of milk allergy patients are sensitive to this protein (2). β-LG, a major whey protein, is a globular protein of MW 18 400 consisting of 162 amino acids and possesses two disulfide bonds and one free cysteine (3). Although the physiological function of β-LG still remains unclear, it is tentatively considered to be the binding and transportation of small hydrophobic ligands such as retinol and fatty acids, and the protein is categorized as a member of the lipocalin superfamily (4). β-LG has a β-barrel structure (5, 6) which is a common feature among the lipocalins. This kind of molecule has high allergenic potential, and several allergens of animal origin belong to the lipocalin superfamily (7, 8). In terms of food science, β-LG is considered to be a valuable protein because it has useful functional properties such as emulsifying, foaming and gelling properties (9-11), as well as a high content of essential amino acids (3). It is therefore desirable to develop new methods that would reduce the allergenicity of β-LG. Although attempts to reduce the allergenicity of proteins have been made by enzymatic digestion and denaturation (12-15), these methods destroy the physiological functions of the proteins and bring about problems with their taste. In contrast, protein conjugation is superior to other hypoallergenic methods in that it can simultaneously

10.1021/bc020050o CCC: $25.00 © 2003 American Chemical Society Published on Web 12/17/2002

T Cell Response Modulated by β-LG−CMD Conjugates

achieve improved functions (thermal stability, solubility, emulsifying ability, etc.) and low allergenicity and immunogenicity, while maintaining the physiological functions of proteins (16-20). Many attempts to reduce the antigenicity and immunogenicity of proteins by conjugation methods have been made. Poly(ethylene glycol) (PEG) conjugates of recombinant IL-2 have been reported to elicit a 100-1000-fold lower level of antigen-specific IgG antibody production than recombinant IL-2 (21). High lipophilization of β-LG has been shown to be effective in reducing the production of anti-β-LG IgG and IgE (22). Conjugates of monoclonal and polyclonal antibodies with oxidized dextrans of low molecular weight have been reported to have optimal immunoreactivity, in vivo pharmacokinetics, and tumor localization properties, as well as low immunogenicity in vivo (23). The conjugation of allergens with oligodeoxynucleotides having CpG sequences has been shown to suppress allergic responses (24, 25). However, the mechanism for the reduced antigenicity and immunogenicity of the proteins is still unclear despite such work. Our objective in this study was to create hypoallergenic neoglycoconjugates of β-LG with high thermal stability and improved functional properties, while maintaining retinol-binding activity, and to elucidate the mechanism involved. We have achieved, in our previous studies (16-18, 26, 27), reduced immunogenicity, high thermal stability, improved emulsifying ability, and retained the retinol-binding activity, while maintaining an almost nativelike conformation, by conjugating β-LG with carboxymethyl dextran (CMD). We have shown that the molecular weight of the polysaccharides used as modifiers and the saccharide content in the conjugates affected the immunogenicity of β-LG and that one mechanism involved in the reduced immunogenicity of β-LG by conjugation with CMD was responsible for masking B cell epitopes by the saccharide chain from the results of B cell epitope scanning of β-LG and the conjugates (26, 27). We investigated, in the present study, changes in the T cell response to β-LG after conjugation with CMD to further elucidate the mechanism. Antigens delivered from an extracellular source are processed by proteases within APC to create antigenic peptides including T cell epitopes and then presented on the APC surface by major histocompatibility complex (MHC) class II molecules for recognition by T cells (28). The T cells provide help to B cells through direct cell-cell interaction and secretion of cytokines that promote their proliferation and differentiation into antibody-producing cells (29). Thus, modulation of the T cell response by conjugation is considered to affect the immunogenicity of proteins. In particular, we focus on changes in the epitope profiles of β-LG after conjugation in elucidating the mechanism. Although the T cell epitope profiles of proteins, including β-LG (30-32), have been reported, there have been hardly any reports on the changes in epitope profiles of proteins after conjugation with saccharides. Epitope scanning of proteins after conjugation provides useful information such as the elimination of epitopes and generation of neo-epitopes. After preparing covalently bonded β-LG-CMD conjugates with CMDs of different molecular weight (Mr ) about 40 and 162 kDa), we investigate the in vivo T cell response, T cell epitope profiles, antigen presentation to β-LG-specific T cells, and the susceptibility to cathepsin B and cathepsin D. Our data show that suppression of the generation of T cell epitopes by conjugation with CMD was involved in the mechanism for the reduced immunogenicity of β-LG.

Bioconjugate Chem., Vol. 14, No. 1, 2003 169 MATERIALS AND METHODS

Materials. Dextran T40 (Mr ) about 40 kDa) and Dextran 162 (Mr ) about 162 kDa) were purchased from Amersham Pharmacia Biotech (Buckinghamshire, U.K.) and Sigma Chemical Co. (St. Louis, MO), respectively. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) hydrochloride was purchased from Dojindo (Kumamoto, Japan). Carboxymethylation of Dextran. Each dextran sample was carboxymethylated with monochloroacetic acid under alkaline conditions as previously described (16-18, 26). The carboxymethyl dextrans prepared from Dextran T40 and Dextran 162 are termed CMD 40 and CMD 162, respectively. The numbers of carboxyl groups attached to one molecule of dextran were determined by 1 H NMR (400 MHz JEOL AL-400, Japan) to be 39 and 162, respectively, for CMD 40 and CMD 162. Preparation and Purification of β-LG. Crude β-LG (genotype AA) was prepared from fresh milk of a Holstein cow according to the method of Armstrong et al. (33). Crude β-LG was purified with a DEAE-Sepharose Fast Flow column (3.0 ID × 40 cm; Amersham Pharmacia Biotech, Buckinghamshire, U.K.) by referring to the method described previously (16). The purity of β-LG was confirmed by polyacrylamide gel electrophoresis (PAGE), performed by the method of Davis (34). Preparation and Purification of the β-LG-CMD Conjugates. The β-LG-CMD conjugates were prepared by using EDC as previously described (27). In brief, β-LG (1 g) and each CMD (CMD 40, 8.48 g; CMD 162, 9.375 g) in the molar ratio of amino groups of β-LG:carboxyl groups of CMD ) 1:10 were dissolved in distilled water (125 mL), and the pH value of the solution was adjusted to 4.75 with 1 M HCl. An EDC solution (1.67 g/10 mL) in a molar ratio of amino groups of β-LG:EDC ) 1:10 was added over a 30 min period, the pH value being maintained at 4.75 by adding 1 M HCl. The reaction mixture was incubated at 25 °C for 3 h before a 2 M acetate buffer at pH 5.5 (10 mL) was gradually added over a period of 10 min to stop the reaction. After dialyzing against distilled water at 4 °C and lyophilizing, the crude β-LGCMD conjugates were obtained. These crude conjugates were purified by hydrophobic chromatography and anionexchange chromatography as described previously (27). The conjugates prepared from CMD 40 and CMD 162 are termed Conj. 40 and Conj. 162, respectively. The amount of protein in each conjugate was determined by measuring the absorbance at 280 nm. The amount of CMD bound to β-LG was determined by measuring the absorbance at 490 nm after coloring by the phenol-sulfuric acid method (35). Immunization. Female BALB/c, C3H/He, and C57BL/6 mice (Clea Japan, Tokyo, Japan) at 6 weeks of age (7-8 animals/group) were subcutaneously immunized in the hind footpads and base of the tail with β-LG or the conjugates (100 µg as protein) emulsified in Freund’s complete H37 Ra adjuvant (Difco Laboratories, Detroit, MI). After 7 days, the inguinal and popliteal lymph nodes were removed. T Cell Proliferation Assay and T Cell Epitope Scanning. The T cell proliferation assay and T cell epitope scanning were performed in 96-well culture plates (Beckton Dickinson, Franklin Lakes, NJ) in 200 µL/well of an RPMI1640 medium (Nissui Pharmaceutical, Tokyo, Japan) supplemented with 0.03% glutamine, 0.2% NaHCO3, 2-mercaptoethanol (2-ME; 50 µM), penicillin (100 U/mL), streptomycin (100 µg/mL), and 1% autologous normal mouse serum. For T cell epitope scanning,

170 Bioconjugate Chem., Vol. 14, No. 1, 2003

Kobayashi et al.

Table 1. Immunological Properties of the β-LG-CMD Conjugatesa immunogenicity (%)b (anti-β-LG antibody response) mice

Conj. 40

Conj. 162

BALB/c C3H/He C57BL/6

54.2 54.0 59.7

39.5 44.0 34.5

novel immunogenicity (anti-conjugate antibody response) Conj. 40

Conj. 162

not emerged not emerged not emerged

B cell epitope distribution Conj. 40

Conj. 162

not changed not changed not changed

antibody response to epitopes Conj. 40

Conj. 162

reduced (Conj. 40 < Conj. 162) reduced (Conj. 40 < Conj. 162) reduced (Conj. 40 < Conj. 162)

a This table summarized the data reported previously (27). bBALB/c, C3H/He, and C57BL/6 mice were immunized i.p. with β-LG, Conj. 40, or Conj. 162 emulsified in Freund’s complete adjuvant and boostered 14 days later with the proteins emulsified in Freund’s incomplete adjuvant. Seven days after secondary immunizaton, each antiserum was prepared. Immunogenicity was evaluated by ELISA using the antisera. Anti-β-LG antibody response in the antisera prepared from the β-LG-immunized mice was as 100%, and that from conjugateimmunized mice was relatively calculated.

a series of overlapping 15-mer peptides moving one amino acid residue at a time in accordance with the amino acid sequence of β-LG were synthesized with a five-in-one B cell and T cell epitope scanning kit (Chiron Mimotopes, Clayton, Victoria, Australia) as described previously (26). The concentration of the synthesized peptides was approximately 1 nmol/µL from the results of an amino acid analysis. The lymph node cells were suspended at 5 × 105 cells/well in the culture plates and then stimulated with protein (β-LG or the conjugates) at various concentrations, a synthesized peptide solution (10 µL) or PBS (blank). Cultures were set up in triplicate for stimulation with the proteins, while one well for each peptide was tested for the peptide series. After culturing in a humidified atmosphere of 5% CO2 in air for 3 days at 37 °C, the T cell proliferative response was measured with a BrdU proliferation kit (Roche Molecular Biochemicals, Basel, Switzerland). In short, the cells in the culture plates were pulsed with a 100 µM BrdU solution (20 µL/well) for 2 h under the same culture conditions. The culture plates were centrifuged at 1250 rpm for 10 min at 4 °C and the supernatant was removed, before the plates were dried for 1 h at 60 °C. A FixDenat solution (200 µL) was added to each well, and the plates were incubated for 1 h at 25 °C. The FixDenat solution was removed, and peroxidaselabeled anti-BrdU mAb (100 µL/well) that had been diluted 100-fold with the Antibody dilution solution was added, before the plates were incubated for 2 h at 25 °C. The peroxidase-labeled anti-BrdU mAb solution was removed, and each well was washed three times with PBS (200 µL each). A tetramethylbenzidine solution (100 µL/well) was added, and the plates were incubated for 5-10 min. After adding 1 M H2SO4 (25 µL/well) to stop the enzymatic reaction, the absorbance at 450 nm was measured with an MPR-A4i microplate reader (Tosoh, Tokyo, Japan). The following criteria were used for the peptides adopted as positive in determining the T cell epitopes: (1) those which showed response greater than the mean value plus three times the standard deviation of the absorbance to the peptide (PLAQGGGGGGGGGGG) in the absence of the β-LG sequence (31), (2) those which showed positive response to at least two consecutive overlapping peptides, and (3) those which showed reproducibility in two individual experiments (31). The common amino acid sequences among the peptides that fulfilled these criteria were identified as the epitopes according to the method of Gammon et al. (36). Digestion of the β-LG-CMD Conjugates with Cathepsin B and Cathepsin D. β-LG or each conjugate (0.1% (w/v) as the protein concentration) was dissolved in a 0.2 M citric acid/Na2HPO4 buffer (pH 5.0) containing EDTA (1 mM) and 2% (v/v) 2-ME, and the solution was incubated at 37 °C for 12 h. Cathepsin B (EC 3.4.22.1) or cathepsin D (EC 3.4.23.5) from bovine spleen (Sigma

Chemical Co., St. Louis, MO) was added to the solution (enzyme:substrate ) 1:10), and the mixture was incubated at 37 °C for various times. Digestion was stopped by adding the loading buffer for sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and by heating at 100 °C for 5 min. The digested sample was applied to SDS-PAGE (37), and the gel was stained with Coomassie Brilliant Blue R-250. After destaining, the digestibility of each sample was evaluated by densitometry. RESULTS AND DISCUSSION

To reduce the immunogenicity of β-LG by conjugation with polysaccharides and to elucidate the mechanism involved, we have previously prepared two β-LG-CMD conjugates (Conj. 40 and Conj. 162) by using EDC and investigated the immunological properties of each conjugate in BALB/c, C3H/He and C57BL/6 mice (27). We prepared conjugates having a similar saccharide content with CMDs of different molecular weight to evaluate the influence of the molecular weight of the polysaccharides on the immunological properties of the conjugates. The respective molar ratios of β-LG to CMD in the conjugates were 8:1 and 7:1 for Conj. 40 and Conj. 162. The immunological properties of the conjugates are shown in Table 1. In all the three strains of mice, the immunogenicity of β-LG (production of the β-LG-specific antibody) was reduced without inducing any novel immunogenicity (production of a conjugate-specific antibody) by conjugation with CMD. The reduced immunogenicity was more evident in the case of Conj. 162 than Conj. 40. To elucidate the mechanism for the reduced immunogenicity of β-LG, we investigated changes in the B cell epitope profiles of the conjugates by ELISA, using the overlapping 15-mer peptides synthesized on the basis of the amino acid sequence of β-LG (27). The linear B cell epitope profile of each conjugate was similar to that of β-LG in terms of the epitope distribution, whereas the antibody response to each epitope was reduced. The reducing effect was Conj. 40 < Conj. 162, so we concluded that masking of the B cell epitopes by the saccharide chain was responsible for reducing the immunogenicity of β-LG. In Vivo T Cell Response of the β-LG-CMD Conjugates. Since T cells provide help to B cells through direct cell-cell interaction and secretion of cytokines that promote their proliferation and differentiation into antibody-producing cells (29), T cells are considered to play an important role in antibody response. To elucidate the mechanism for the reduced immunogenicity of β-LG by conjugation with CMD, we investigated the in vivo T cell response of each conjugate. After BALB/c, C3H/He, and C57BL/6 mice had been immunized with β-LG or the conjugates, the lymph node cells from the mice were

T Cell Response Modulated by β-LG−CMD Conjugates

Figure 1. Proliferative response to β-LG of lymph node cells from mice immunized with β-LG or the β-LG-CMD conjugates. After immunizing BALB/c (a), C3H/He (b), and C57BL/6 (c) mice with β-LG (b), Conj. 40 (2), or Conj.162 (3), the lymph node cells from the mice were removed and stimulated with β-LG at various concentrations. The magnitude of the in vivo T cell response was evaluated as the ex vivo proliferative response by BrdU ELISA. Each value is expressed as the mean absorbance at 450 nm and standard deviation of triplicate cultures after subtracting the background values (stimulated with PBS). Significant differences (p < 0.05) between β-LG and each conjugate were determined by Student’s t-test and are indicated by asterisks.

removed and then stimulated with β-LG at various concentrations. The magnitude of the in vivo T cell response was evaluated as the ex vivo proliferative response by BrdU ELISA (Figure 1). In BALB/c mice (Figure 1a), T cells from the groups immunized with the conjugates showed lower response than those from β-LGimmunized mice, this being particularly evident with Conj. 162. In C3H/He (Figure 1b) and C57BL/6 mice (Figure 1c), although the T cell proliferative responses of the Conj. 40- and Conj. 162-immunized groups were similar, the proliferative response of T cells from the conjugate-immunized groups was lower than that from the β-LG-immunized group. This result indicates that the response of the T cells to β-LG in the conjugate-immunized mice was reduced by conjugation with CMD. Two possibilities are considered as the reason for this result. (1) A decrease in the number of β-LG-specific T cells: since the induction of β-LG-specific T cells was suppressed by conjugation with CMD, the number of β-LG-specific T cells might have been reduced. (2) Changed T cell epitopes by conjugation: McCool et al. (38) have found that lymph node cells from mice immunized with a protein-polysaccharide conjugate recognized additional peptides that were not recognized by lymph node cells from mice immunized with the original protein. In our study, T cells in the conjugate-immunized mice might have shown the reduced response to β-LG owing to the induction of conjugate-specific T cells which could not recognize native β-LG by the generation of the

Bioconjugate Chem., Vol. 14, No. 1, 2003 171

neo-epitopes that are not recognized by β-LG-specific T cells after conjugation with CMD. T Cell Epitope Profiles of the β-LG-CMD Conjugates. Since the proliferative response of T cells to whole proteins is polyclonal, we next investigated in detail the T cell epitope profiles of the conjugates. After the lymph node cells from BALB/c, C3H/He, and C57BL/6 mice immunized with β-LG or the conjugates had been stimulated with the overlapping peptides synthesized on the basis of the sequence of β-LG, the T cell proliferative response to the peptides was measured by BrdU ELISA. Figure 2 shows the T cell epitope profiles of β-LG and the β-LG-CMD conjugates. The horizontal axis indicates the N-terminal amino acid residue number of each 15mer peptide corresponding to the position in the β-LG sequence, and the vertical axis indicates the T cell proliferative response to each peptide. The T cell epitopes identified according to the method of Gammon et al. (36) are summarized in Figure 3, in which the horizontal axis indicates the sequence number in β-LG and the line thickness indicates the intensity of the response to each epitope. The T cells from BALB/c mice immunized with β-LG showed proliferative response to peptides 1-7, 14-15, 23-29, 35-40, 42-43, 54-55, 60-70, 73-79, 101-109, 111-112, 119-122, 127-133, 133-140, 143-144, and 147-148 (Figure 2a). The T cell epitopes of β-LG that had been recognized in BALB/c mice were determined to be 7Met-15Val, 15Val-28Asp, 29Ile-37Ala, 40Arg-49Thr, 43 Val-56Ile, 55Glu-68Gln, 70Lys-74Glu, 79Pro-87Leu, 109 Asn-115Gln, 112Glu-125Thr, 122Leu-133Leu, 133Leu141 Lys, 140Leu-147Ile, 144Pro-157Glu and 148Arg-161His, the dominant epitopes being 70Lys-74Glu, 109Asn-115Gln, and 140 Leu-147Ile (Figure 3). No changes in the epitope distribution was apparent for T cells from the conjugateimmunized BALB/c mice, but the proliferative response to the epitopes was lower throughout the entire amino acid sequence than that of the β-LG-immunized group (Figure 2a-c). The magnitude of the response to Conj. 162 was lower than that to Conj.40. The T cell epitopes of the conjugates recognized in BALB/c mice were determined to be 6Thr-15Val, 25Ala-38Pro, 28Asp-41Val, 70Lys75 Lys, 109Asn-115Gln, 140Leu-144Pro and 148Arg-161His for Conj. 40, and 6Thr-15Val, 28Asp-38Pro, 69Lys-75Lys, 109 Asn-115Gln, and 140Leu-146His for Conj. 162 (Figure 3). The results obtained with the C3H/He and C57BL/6 mice are shown in Figure 2d-i, and the T cell epitope profiles of β-LG and the conjugates were determined as shown in Figure 3. The T cell epitopes of β-LG recognized in the C3H/He mice were determined to be 91 Lys-103Leu, 99Tyr-111Ala, 102Tyr-115Gln, 125Thr-137Asp, 137 Asp-143Leu, and 148Arg-161His, the dominant epitope being 137Asp-143Leu (Figure 3). In the C3H/He mice immunized with the conjugates, the epitope distribution of both conjugates was similar to that of β-LG, whereas the proliferative response to the epitopes was lower throughout the entire amino acid sequence than that of the β-LG-immunized group (Figure 2d-f). The signal intensity to the epitopes in the Conj. 162-immunized group was weaker than that in the Conj. 40-immunized group. The T cell epitopes of each conjugate recognized in C3H/He mice were determined to be 91Lys-101Lys, 98 Asp-111Ala, 119Cys-132Ala, 137Asp-144Pro and 140Leu151 Phe for Conj. 40, and 90Asn-101Lys, 101Lys-114Glu, 131 Glu-144Pro, and 139Ala-147Ile for Conj. 162 (Figure 3). The T cell epitopes of β-LG recognized in C57BL/6 mice were determined to be 15Val-27Ser, 108Glu-121Cys, and 123Val-130Αsp, the dominant epitope being 123Val-

172 Bioconjugate Chem., Vol. 14, No. 1, 2003

Kobayashi et al.

Figure 2. Proliferation of the lymph node cells from mice immunized with β-LG or the β-LG-CMD conjugates to the overlapping 15-mer peptides covering the amino acid sequence of β-LG. After immunizing BALB/c (a-c), C3H/He (d-f), and C57BL/6 mice (g-i) with β-LG (a, d, g), Conj. 40 (b, e, h), or Conj. 162 (c, f, i), the lymph node cells from the mice were removed and stimulated with the overlapping synthesized peptides. The proliferative response of T cells to the peptides was measured by BrdU ELISA. The horizontal lines in the figures indicate cutoff values to judge whether the observed response was positive. Representative epitope profiles from two individual experiments are shown. 130 Asp (Figure 3). No changes in the epitope distribution were apparent for the T cells from the conjugate-immunized C57BL/6 mice, but the proliferative response to the epitopes was lower throughout the entire amino acid sequence than that of the β-LG-immunized group (Figure 2g-i). The signal intensity to the epitopes in the Conj. 162-immunized group was weaker than that in the Conj. 40-immunized group. The T cell epitopes of the conjugates recognized in C57BL/6 mice were determined to be 123 Val-130Asp for Conj. 40 and 118Val--30Asp and 121Cys134 Glu for Conj. 162 (Figure 3). Totsuka et al. (31) have clarified the T cell epitope profiles of β-LG in BALB/c, C3H/He and C57BL/6 mice by a conventional 3H-thymidine incorporation assay. Although some differences are apparent between our results and those of Totsuka et al. with respect to the signal intensity, the epitope distribution is similar. We have previously shown that the glycosylation sites in β-LG after conjugation with CMD of 10 kDa were 47Lys, 60 Lys, 101Lys, and 138Lys (26). Other investigators (39-41) have reported that the preferential lactosylation sites of β-LG by the Maillard reaction were 47Lys and/or 100 Lys. The conjugates prepared in this study tended to reduce the response to the epitope regions around these Lys residues. The human T cell epitopes of β-LG have recently been identified as 1Leu-21Ser, 14Lys-29Ile, 30 Ser-47Lys, 47Lys-67Ala, 77Lys-97Thr, 97Thr-117Leu, and 142 Ala-162Ile by using synthesized peptides (32). Since most of these T cell epitopes are located in the vicinity of the plausible glycosylation sites just mentioned, the conjugates prepared in this study would modulate the T cell response in humans. It has been reported that a protein-polysaccharide conjugate showed a different pattern of T cell reactivity

to epitopes derived from original protein (38); however, our results from the epitope scanning of Conj. 40 and Conj. 162 show that the T cell epitope profiles of the conjugates were similar to those of β-LG in all the strains of mice tested, which indicates that the location of epitopes recognized by T cells induced in vivo in the conjugate-immunized mice did not change by conjugation with CMD. In fact, when lymph node cells from the conjugate-immunized mice were stimulated with β-LG or the conjugates, the proliferative response to the conjugate was no higher than that to β-LG (data not shown). The signal intensity to the epitopes in the conjugate-immunized groups was lower than that in the β-LG-immunized group in the three strains of mice, indicating a decrease in the number of β-LG-specific T cells owing to the low induction of β-LG-specific T cells in vivo after conjugation with CMD. Therefore, the low in vivo T cell response in the conjugate-immunized groups, as shown in Figure 1, is considered to have been due to the lower number of β-LG-specific T cells induced in vivo by conjugation with CMD. Antigen Presentation of the β-LG-CMD Conjugates to β-LG-Specific T Cells. To elucidate the mechanism for the decrease in number of β-LG-specific T cells in vivo by conjugation with CMD, we investigated antigen presentation of the conjugates to β-LG-specific T cells. After the three strains of mice had been immunized with β-LG, the lymph node cells were removed and stimulated with β-LG or the conjugates at various concentrations. The efficiency of the antigen presentation was evaluated as the in vitro T cell proliferative response by BrdU ELISA (Figure 4). In the BALB/c (Figure 4a) and C3H/He mice (Figure 4b), the proliferative response of the β-LG-specific T cells to Conj. 40 was similar to that

T Cell Response Modulated by β-LG−CMD Conjugates

Bioconjugate Chem., Vol. 14, No. 1, 2003 173

Figure 3. T cell epitope profiles of β-LG and the β-LG-CMD conjugates. The common regions of at least two overlapping peptides, which showed a response greater than the mean value plus three times the standard deviation of the absorbance when the lymph node cells were stimulated with the peptide (PLAQGGGGGGGGGGG) in the absence of the β-LG sequence and which showed reproducibility in two individual experiments, were identified as epitopes according to the method of Gammon et al. (36).

to β-LG, whereas the response to Conj. 162 was lower than that to β-LG. In C57BL/6 mice (Figure 4c), Conj. 40 and Conj. 162 both showed lower T cell proliferative response than that to β-LG. CMD 40 and CMD 162 had no effect on the T cell proliferative response, even at a concentration of 10 µM/well (data not shown). This lower T cell proliferative response indicates that the antigen presentation of β-LG to the β-LG-specific T cells was suppressed by the conjugation with CMD. In particular, the proliferative response of β-LG-specific T cells to Conj. 162 from CMD of the higher molecular weight was lower than that to β-LG in the three strains of mice. Hence, conjugation with CMD of high molecular weight is considered to be effective for suppression of the antigen presentation of β-LG which would lead to a decrease in the number of β-LG specific T cells in vivo. Susceptibility of the β-LG-CMD Conjugates to Cathepsin B and Cathepsin D. Three possible mechanisms are considered plausible for suppressing the antigen presentation of β-LG to T cells by conjugation with CMD: (1) suppression of the generation of T cell epitopes, (2) inhibition of binding between MHC and the antigenic peptide, and (3) inhibition of MHC-peptide complex recognition by the T cell receptor. To examine the first possibility, we investigated the susceptibility of

the conjugates to cathepsin B and cathepsin D, which are regarded as representative enzymes involved in the generation of antigenic epitopes (28). The processing of many antigens requires reduction of the disulfide bonds (28, 42) as well as proteolysis in APC, so the conjugates were digested with the cathepsins in the presence of 2-ME (Figure 5). About 90% of β-LG was digested with cathepsin B (Figure 5a) and cathepsin D (Figure 5b) after 8 h, whereas no more than 10% of the conjugates was digested. In respect of the endosomal protease involved in antigen processing, it has been suggested that the processing of most antigens occurs through the sequential action of more than one enzyme, that endoproteases including cathepsin D may be important in revealing and releasing antigens, and that exopeptidases including cathepsin B may trim epitopes to their final size (28, 43). Hence, the conjugates were digested with a mixture (1:1) of cathepsin B and cathepsin D (Figure 5c). β-LG was completely digested after about 2 h, whereas about 70% of each conjugate remained intact even after 8 h of digestion. β-LG was not digested when incubated for 8 h in the absence of the cathepsins (data not shown). In all experiments, both conjugates showed much higher resistance to the cathepsins than β-LG. These results indicate that conjugation with CMD inhibited the deg-

174 Bioconjugate Chem., Vol. 14, No. 1, 2003

Figure 4. Proliferative response to β-LG and the β-LG-CMD conjugates of the lymph node cells from the mice immunized with β-LG. After immunizing BALB/c (a), C3H/He (b), and C57BL/6 (c) mice with β-LG, the lymph node cells from the mice were removed and stimulated with β-LG (b), Conj. 40 (2), or Conj.162 (3) at various concentrations. The T cell proliferative response to the proteins was measured by BrdU ELISA. Each value is expressed as the mean absorbance at 450 nm and standard deviation of triplicate cultures after subtracting the background values (stimulated with PBS). Significant differences (p < 0.05) between β-LG and each conjugate were determined by Student’s t-test and are indicated by asterisks.

radation of β-LG by proteases in APC and would have led to suppression of the generation of T cell epitopes from β-LG. It is considered that this suppressive effect on the generation of T cell epitopes by conjugation with CMD reduced the antigen presentation of the conjugates. This proposal is supported by the findings that the lysozyme stabilized by intramolecular cross-linking depressed T cell epitope generation by increasing the resistance to proteolysis (44) and that proteophosphoglycan, a major product secreted from Leishmania mexicana amastigotes, did not elicit a specific CD4+ T cell response due to no degradation occurring in macrophages (45). The reduced antigen presentation of the conjugates would bring about a decrease in the number of β-LGspecific T cells in vivo and, as a result, the decreased help to B cells by T cells would reduce the antibody response to β-LG. Consequently, we conclude that suppression of the generation of T cell epitopes by conjugation with CMD played an important role in the mechanism for reducing the immunogenicity of β-LG. As for the other possibilities, the inhibition of binding between MHC and the antigenic peptide must be taken into consideration, because the plausible glycosylation sites already mentioned are located within some of the T cell epitopes identified in this study. In fact, T cell epitopes with internal but not external glycosylation did not bind MHC class II molecules and had less T cell proliferative response (46). Concluding Remarks. We have shown in this study that the suppressive effect on the generation of T cell epitopes by conjugation with CMD played an important role in the mechanism for the reduced immunogenicity

Kobayashi et al.

Figure 5. Susceptibility of β-LG and the β-LG-CMD conjugates to cathepsin B and cathepsin D. β-LG (b), Conj. 40 (2), and Conj.162 (3) were digested with cathepsin B (a), cathepsin D (b), and a mixture (1:1) of cathepsin B and cathepsin D (c) for various times in the presence of 2-ME. The digested samples were applied to SDS-PAGE, and the digestibility of the samples was evaluated by densitometry. The intensity of the bands of each sample at 0 min is taken as 100%, and that of the samples at subsequent times is relative to the value at 0 min. All results are expressed relative to the intact protein (%).

of β-LG. Our work is valuable in that we could clarify the change in the T cell epitope profile of a protein after conjugation with polysaccharides differing in molecular weight to elucidate the mechanism for the reduced immunogenicity. Taken together with our previous results (26, 27), both masking of the B cell epitopes by the saccharide chain and suppression of the generation of T cell epitopes in APC are considered to be important for the reduced immunogenicity of protein by conjugation with a polysaccharide. Both factors would be involved in the induction of a low IgE response by conjugation methods, and further studies focusing on these aspects should be carried out. We have particular interest in the change in qualitative T cell response to proteins after conjugation or modification (47-51). With regard to the difference in the molecular weight of CMD used as a modifier, although not all experiments showed a distinct difference between Conj. 40 and Conj. 162, the effect of Conj. 40 did not exceed that of Conj. 162. Conjugation with a polysaccharide of high molecular weight is therefore considered to be effective for modulating the T cell response to proteins. The results of the present study are essential to develop the design of novel hypoallergenic foods.

T Cell Response Modulated by β-LG−CMD Conjugates ACKNOWLEDGMENT

This work was supported in part by grant from Asahi Breweries Foundation and Kyowa Hakko Kogyo Co. Ltd. LITERATURE CITED (1) Iikura, Y., Imai, Y., Imai, T., Akasawa, A., Fujita, K., Hoshiyama, K., Nakura, H., Kohno, Y., Koike, K., Okudaira, H., and Iwasaki, E. (1999) Frequency of immediate-type food allergy in children in Japan. Int. Arch. Allergy Immunol. 118, 251-252. (2) Spies, J. (1973) Milk allergy. J. Milk Food Technol. 36, 225231. (3) McKenzie, H. A. (1971) β-Lactoglobulins. In Milk Proteins: Chemistry and Molecular Biology (H. A. McKenzie, Ed.) pp 257-330, Academic Press, New York. (4) Sawyer, L., and Kontopidis, G. (2000) The core lipocalin, bovine β-lactoglobulin. Biochim. Biophys. Acta 1482, 136148. (5) Papiz, M. Z., Sawyer, L., Eliopoulos, E. E., North, A. C. T., Findlay, J. B. C., Sivaprasadrao, R., Jones, T. A., Newcomer, M. E., and Kraulis, P. J. (1986) The structure of β-lactoglobulin and its similarity to plasma retinol-binding protein. Nature 324, 383-385. (6) Brownlow, S., Cabral, J. H. M., Cooper, R., Flower, D. R., Yewdall, S. J., Polikarpov, I., North, A. C. T., and Sawyer, L. (1997) Bovine β-lactoglobulin at 1.8 Å resolution - still an enigmatic lipocalin. Structure 5, 481-495. (7) Virtanen, T., Zeiler, T., Rautiainen, J., and Ma¨ntyja¨rvi, R. (1999) Allergy to lipocalins: a consequence of misguided T-cell recognition of self-and nonself? Immunol. Today 20, 398-400. (8) Wal, J. M. (1998) Cow’s milk allergens. Allergy 53, 10131022. (9) Shimizu, M., Saito, M., and Yamauchi, K. (1985) Emulsifying and structural properties of β-lactoglobulin at different pHs. Agric. Biol. Chem. 49, 189-194. (10) Waniska, R. D., and Kinsella, J. E. (1988) Foaming and emulsifying properties of glycosylated β-lactoglobulin. Food Hydrocolloids 2, 439-449. (11) Foegeding, E. A., Kuhn, P. R., and Hardin, C. C. (1992) Specific divalent cation-induced changes during gelation of β-lactoglobulin. J. Agric. Food Chem. 40, 2092-2097. (12) Ishizaka, K., Okudaira, H., and King, T. (1975) Immunogenic properties of modified antigen E. II. Ability of urea-denatured antigen and a polypeptide chain to prime T cells specific for antigen C. J. Immunol. 114, 110-115. (13) Kurisaki, J., Nakamura, S., Kaminogawa, S., and Yamauchi, K. (1982) The antigenic properties of β-lactoglobulin examined with mouse IgE antibody. Agric. Biol. Chem. 46, 2069-2075. (14) Watanabe, M., Suzuki, T., Ikezawa, Z., and Arai, S. (1994) Controlled enzymatic treatment of wheat proteins for production of hypoallergenic flour. Biosci. Biotechnol. Biochem. 58, 388-390. (15) del Val, G., Yee, B. C., Lozano, R. M., Buchanan, B. B., Ermel, R. W., Lee, Y. M., and Frick, O. L. (1999) Thioredoxin treatment increases digestibility and lowers allergenicity of milk. J. Allergy Clin. Immunol. 103, 690-697. (16) Hattori, M., Nagasawa, K., Ametani, A., Kaminogawa, S., and Takahashi, K. (1994) Functional changes in β-lactoglobulin by conjugation with carboxymethyl dextran. J. Agric. Food Chem. 42, 2120-2125. (17) Nagasawa, K., Takahashi, K., and Hattori, M. (1996) Improved emulsifying properties of β-lactoglobulin by conjugating with carboxymethyl dextran. Food Hydrocolloids 10, 63-67. (18) Nagasawa, K., Ohgata, K., Takahashi, K., and Hattori, M. (1996) Role of the polysaccharide content and net charge on the emulsifying properties of β- lactoglobulin-carboxymethyl dextran conjugates. J. Agric. Food Chem. 44, 2538-2543. (19) Hattori, M., Numamoto, K., Kobayashi, K., and Takahashi, K. (2000) Functional changes in β-lactoglobulin by conjugation with cationic saccharides. J. Agric. Food Chem. 48, 20502056. (20) Babiker, E. E., Azakami, H., Matsudomi, N., Iwata, H., Ogawa, T., Bando, N., and Kato, A. (1998) Effect of polysac-

Bioconjugate Chem., Vol. 14, No. 1, 2003 175 charide conjugation or transglutaminase treatment on the allergenicity and functional properties of soy protein. J. Agric. Food Chem. 46, 866-871. (21) Katre, N. (1990) Immunogenicity of recombinant IL-2 modified by covalent attachment of polyethylene glycol. J. Immunol. 144, 209-213. (22) Akita, E. M., and Nakai, S. (1990) Lipophilization of β-lactoglobulin: Effect on allergenicity and digestibility. J. Food Sci. 55, 718-723. (23) Fagnani, R., Hagan, M. S., and Bartholomew, R. (1990) Reduction of immunogenicity by covalent modification of murine and rabbit immunoglobulins with oxidized dextrans of low molecular weight. Cancer Res. 50, 3638-3645. (24) Shirota, H., Sano, K., Kikuchi, T., Tamura, G., and Shirato, K. (2000) Regulation of murine airway eosinophilia and Th2 cells by antigen-conjugated CpG oligodeoxynucleotides as a novel antigen-specific immunomodulator. J. Immunol. 164, 5575-5582. (25) Tighe, H., Takabayashi, K., Schwartz, D., Nest, G. V., Tuck, S., Eiden, J. J., Kagey-Sobotka, A., Creticos, P. S., Lichtenstein, L. M., Spiegelberg, H. L., and Raz, E. (2000) Conjugation of immunostimulatory DNA to the short ragweed allergen Amb a 1 enhances its immunogenicity and reduces its allergenicity. J. Allergy Clin. Immunol. 106, 124-134. (26) Hattori, M., Nagasawa, K., Ohgata, K., Sone, N., Fukuda, A., Matsuda, H., and Takahashi, K. (2000) Reduced immunogenicity of β-lactoglobulin by conjugation with carboxymethyl dextran. Bioconjugate Chem. 11, 84-93. (27) Kobayashi, K., Hirano, A., Ohta, A., Yoshida, T., Takahashi, K., and Hattori, M. (2001) Reduced immunogenicity of β-lactoglobulin by conjugation with carboxymethyl dextran differing in molecular weight. J. Agric. Food Chem. 49, 823831. (28) Fineschi, B., and Miller, J. (1997) Endosomal proteases and antigen processing. Trends Biochem. Sci. 22, 377-382. (29) Kaminogawa, S. (1996) Food Allergy, oral tolerance and immunomodulation - Their molecular and cellular mechanisms. Biosci. Biotechnol. Biochem. 60, 1749-1756. (30) Tsuji, N. M., Kurisaki, J., Mizumachi, K., and Kaminogawa, S. (1993) Localization of T-cell determinants on bovine β-lactoglobulin. Immunol. Lett. 37, 215-221. (31) Totsuka, M., Ametani, A., and Kaminogawa, S. (1997) Fine mapping of T-cell determinants of bovine β-lactoglobulin. Cytotechnology 25, 101-113. (32) Inoue, R., Matsushita, S., Kaneko, H., Shinoda, S., Sakaguchi, H., Nishimura, Y., and Kondo, N. (2001) Identification of β-lactoglobulin-derived peptides and class II HLA molecules recognized by T cells from patients with milk allergy. Clin. Exp. Allergy 31, 1126-1134. (33) Armstrong, J. M., McKenzie, H. A., and Sawyer, W. H. (1967) On the fractionation of β-lactoglobulin and R-lactalbumin. Biochim. Biophys. Acta 147, 60-72. (34) Davis, B. J. (1964) DISC ELECTROPHORESIS - II: Method and application to human serum proteins. Ann. N.Y. Acad. Sci. 121, 404-427. (35) Dubois, M., Gilles, K. A., Hamilton, J. L., Rebers, P. A., and Smith, F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356. (36) Gammon, G., Geysen, H. M., Apple, R. J., Pickett, E., Palmer, M., Ametani, A., and Sercarz, E. E. (1991) T cell determinant structure: cores and determinant envelopes in three mouse major histocompatibility complex haplotypes. J. Exp. Med. 173, 609-617. (37) Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. (38) McCool, T. L., Harding, C. V., Greenspan, N. S., and Schreiber, J. R. (1999) B- and T-cell immune response to pneumococcal conjugate vaccines: Divergence between carrier- and polysaccharide-specific immunogenicity. Infect. Immun. 67, 4862-4869. (39) Leonil, J., Molle, D., Fauquant, J., Maubois, J. L., Pearce, R. J., and Bouhallab, S. (1997) Characterization by ionization mass spectrometry of lactosyl β-lactoglobulin conjugates

176 Bioconjugate Chem., Vol. 14, No. 1, 2003 formed during heat treatment of milk and whey and identification of one lactose-binding site. J. Dairy Sci. 80, 22702281. (40) Fogliano, V., Monti, S. M., Visconti, A., Randazzo, G., Facchiano, A. M., Colonna, G., and Ritieni, A. (1998) Identification of a β-lactoglobulin lactosylation site. Biochim. Biophys. Acta 1388, 295-304. (41) Siciliano, R., Rega, B., Amoresano, A., and Pucci, P. (2000) Modern mass spectrometric methodologies in monitoring milk quality. Anal. Chem. 72, 408-415. (42) Collins, D. S., Unanue, E. R., and Harding, C. V. (1991) Reduction of disulfide bonds within lysosomes is a key step in antigen processing. J. Immunol. 147, 4054-4059. (43) Riese, R. J., and Chapman, H. A. (2000) Cathepsins and compartmentalization in antigen presentation. Curr. Opin. Immunol. 12, 107-113. (44) So, T., Ito, H.-O., Koga, T., Watanabe, S., Ueda, T., and Imoto, T. (1997) Depression of T-cell epitope generation by stabilizing hen lysozyme. J. Biol. Chem. 272, 32136-32140. (45) Aebischer, T., Harbecke, D., and Ilg, T. (1999) Proteophosphoglycan, a major secreted product of intracellular Leishmania mexicana amastigotes, is a poor B-cell antigen and does not elicit a specific conventional CD4+ T-cell response. Infect. Immun. 67, 5379-5385. (46) Mouritsen, S., Meldal, M., Christiansen-Brams, I., Elsner, H., and Werdelin, O. (1994) Attachment of oligosaccharides to peptide antigen profoundly affects binding to major his-

Kobayashi et al. tocompatibility complex class II molecules and peptide immunogenicity. Eur. J. Immunol. 24, 1066-1072. (47) Yang, X., Gieni, R. S., Mosmann, T. R., and HayGlass, K. T. (1993) Chemically modified antigen preferentially elicits induction of Th1-like cytokine synthesis patterns in vivo. J. Exp. Med. 178, 349-353. (48) Kohno, K., Ohtsuki, T., Suemoto, Y., Inoue, T., Taniguchi, Y., Usui, M., Ikeda, M., and Kurimoto, M. (1996) Regulation of cytokine production by sugi allergen-pullulan conjugate. Cell. Immunol. 168, 211-219. (49) Singh, N., Bhatia, S., Abraham, R., Basu, S. K., George, A., Bal, V., and Rath, S. (1998) Modulation of T cell cytokine profiles and peptide-MHC complex availability in vivo by delivery to scavenger receptors via antigen maleylation. J. Immunol. 160, 4869-4880. (50) Korematsu, S., Tanaka, Y., Hosoi, S., Koyanagi, S., Yokota, T., Mikami, B., and Minato, N. (2000) C8/119S mutation of major mite allergen Derf-2 leads to degenerate secondary structure and molecular polymerization and induces potent and exclusive Th1 cell differentiation. J. Immunol. 165, 2895-2902. (51) So, T., Ito, H.-O., Hirata, M., Ueda, T., and Imoto, T. (2001) Contribution of conformational stability of hen lysozyme to induction of type 2 T-helper immune responses. Immunology 104, 259-268.

BC020050O