Hyposensitization therapy for atopic diseases has been conducted for decades but suffered from many problems including anaphylactic reactions. We previously developed a mutant protein of the major mite allergen Derf-2, C8/119S, which showed reduced binding to IgE. The C8/119S mutant was shown to exhibit more efficient hyposensitizing effect than Derf-2 in the animal model of allergic bronchial asthma. In the present study, we indicate that C8/119S exhibits markedly augmented immunogenicity for the proliferation of Derf-2-specific human T cells and T cell clones irrespective of the epitope specificity as compared with Derf-2. Furthermore, C8/119S has induced potent and almost exclusive differentiation of Th1 cells from the peripheral blood of atopic patients in vitro. Neither Ag dosage effect nor absence of B cell-mediated Ag presentation could fully account for these effects. C8/119S has been indicated to lose the characteristic β-barrel structure as judged by circular dichroism spectroscopic analysis and to polymerize solubly in physiological condition. Heating of Derf-2 also caused less stable molecular aggregation, but it hardly affected the secondary structure and failed to induce such a polarity toward the Th1 cell differentiation. These results have indicated that the degenerate secondary structure of C8/119S leading to stable molecular polymerization is primarily responsible for the marked increase in T cell-immunogenicity and the induction of exclusive Th1 cell differentiation in atopic patients. It has been suggested strongly that the recombinant C8/119S protein can provide an effective Ag with the least risk of anaphylaxis for allergen immunotherapy against house dust mite in human.
For the control and possible eradication of intractable atopic diseases against unavoidable allergens, hyposensitizing therapy, or allergen immunotherapy, has been the major realistic means (1, 2, 3). However, the therapy has suffered from a number of obstacles (4, 5). Although complications derived from the impurity of allergens have been overcome by the production of purified recombinant allergens, provocation of possibly life-threatening anaphylaxic reactions remains to be a serious adverse effect. In contrast, immunological mechanisms for the hyposensitization per se have been an issue of much arguments for decades, and a number of mechanisms have been proposed, including generation of blocking Ab, suppressor cells, induction of specific anergy, and clonal deletion (2, 3, 6, 7). Recently, shift of the balance between Th1 and Th2 cell responses has been stressed as the major underlying cause for atopic diseases, in which the Th2 response producing IL-4 predominates over Th1 response producing IFN-γ resulting in the production of excess IgE Ab (8, 9, 10). A number of different factors are shown to affect the balance of Th1 and Th2 responses, including the routes of Ag exposure (11), Ag concentration (12, 13), forms of Ags (14), affinity of Ags for the Ag receptors (15), types of APCs (13, 16), and genetic factors (17). It was reported that successful hyposensitization indeed reversed such a shifted balance of T cell response for the specific allergens (18).
Derf-2 is one of the major allergens of house dust mite responsible for the heavy atopic diseases, including bronchial asthma in childhood (19, 20). Unlike another major allergen Derf-1, Derf-2 is heat-stable and resists against drying in the sun. Nuclear magnetic resonance spectroscopic analysis has indicated that Derf-2 is a soluble and monomeric protein of a single-domain of Ig fold with β-sheet structure (21). We have previously reported a C8/119S mutant protein of Derf-2 with the cysteine residues at positions 8 and 119 being replaced by serines to abrogate an intramolecular disulfide bond (22). C8/119S shows much reduced binding to the Derf-2-specific IgE in atopic individuals, and is thus the least likely to provoke anaphylactic reactions in vivo (23). In the animal models of allergic asthma, it was shown that nasal administration of C8/119S induced significant hyposensitizing effect in the Derf-2-sensitized animals, leading to the reduced bronchial constriction upon Derf-2 challenge, whereas wild-type Derf-2 did so only marginally if any (24). Although these results have suggested that the recombinant C8/119S can be a promising Ag for the immunotherapy of human allergic diseases for mites, immunological mechanisms for the effective induction of hyposensitization by C8/119S remain to be verified. In the present study, we indicate that C8/119S exhibits markedly augmented immunogenicity for the Derf-2-specific T cells as compared with the native Derf-2. We further show that, unlike wild-type Derf-2, C8/119S induces the potent and almost exclusive differentiation of Th1-type cells from the peripheral blood of atopic individuals in vitro. Neither Ag dosage effect nor absence of B cell-mediated Ag presentation could fully account for the unusual polarity for Th1 cell differentiation by C8/119S. In contrast, structural analyses have revealed that disruption of an intramolecular disulfide bond between C8 and C119 results in the total abrogation of β-sheet structure and in the stable molecular polymerization. We suggest that such structural alteration resulting from the mutation is primarily responsible for the markedly polarized T1 cell differentiation.
Materials and Methods
Recombinant Derf-2 and mutant C8/119S proteins were produced in Escherichia coli, extracted with 6 M urea, renatured by the dialysis against 20 mM Tris-HCl (pH 9.0), and purified on anion exchange column chromatography as described previously (20). Correct disulfide bond formation of the purified Derf-2 was confirmed by peptide mapping.
Cell cultures and Th cell clones
Two atopic individuals, SK and SH, who showed typical allergic rhinitis and dermatitis, positive skin test, and radioallergosolvent test to Derf-2, were examined in the present study. PBMC (3) were cultured at 3 × 106 cells/well in 24-well culture plates in Yssel’s modified Iscove’s medium supplemented with 0.25% human serum albumin and antibiotics in the presence of Derf-2 (1.0 μg/ml). Sixteen days later, viable cells were recovered by Ficoll-Hypaque density gradient centrifugation, and cultured at 5 × 105 cells/well in Yssel’s medium supplemented with 10% pooled human AB serum in the presence of Derf-2 (1.0 μg/ml) and 3000 rad-irradiated autologous PBMC (1.5 × 106 cells/well) as APC unless otherwise specified. To obtain Th clones, antigenic stimulation as above was repeated two to four cycles and limiting dilution was performed in the presence of Derf-2 (1.0 μg/ml), IL-2 (20 U/ml), IL-4 (5 U/ml), and irradiated autologous PBMC. All T cell clones used in the present study were CD3+CD4+CD8−, and their specific response to Derf-2 was suggested to be restricted to HLA DQ. The clones were propagated in the culture in the presence of Derf-2, IL-2, IL-4, and irradiated autologous PBMC. Epitope mapping of the in vitro immunized T cells as well as independent Th clones was done by using 23 sets of overlapping synthetic 15-residue peptides derived from Derf-2, kindly provided by Dr. S. Ikeda (Research Laboratories of Torii, Osaka, Japan).
B cells were purified from PBMC with Dynabeads M-450 CD19 (Dynal, Oslo, Norway), and the beads were detached immediately by DETACHaBEAD (Dynal). The purity of B cell fraction was >99.5%, whereas B cell-depleted fraction contained <0.5% B cells as judged by flow cytometric analysis.
Sixteen days after the last stimulation with Ags, when MHC class II Ag expression on the T cells became negligible, T cells or Th clones were assessed for the Ag-specific proliferative response. T cells (5 × 104 cells/well) and irradiated APC (1.5 × 105 cells/well) were incubated in 96-well round-bottom plates in triplicate in the presence of varying concentrations of Ags without exogenous cytokines for 2 days. The cultures were pulsed with [3H]TdR (2 μCi/well) for the last 12 h, harvested, and radioactivity was measured by a scintillation counter.
T cells or Th clones were cultured as above for 2 days, and the culture supernatants were harvested. IL-4 and IFN-γ in the supernatants were determined by ELISA kit (Amersham Life Science, Buckinghamshire, U.K.). The binding of allergens to the specific IgE in the serum was determined by sandwich ELISA.
Derf-2 and C8/119S proteins were run on either 15% SDS-PAGE or modified Davis’ native PAGE, and stained with Coomassie brilliant blue.
Circular dichroism (CD)3 spectroscopy
Purified protein samples of Derf-2 and C8/119S were suspended in PBS at a final concentration of 60 μM. UV-CD spectra of the samples were measured in a circular quartz cuvette of 0.02 cm path length using a spectropolarimeter (Japan Spectroscopic, Tokyo, Japan). CD spectra were recorded from 250 to 190 nm at a digital resolution of 0.1 nm with scan speed of 10 nm/min. Three scans were signal-averaged, and the data sets were concatenated.
Light scatter particle sizing
Protein samples of Derf-2 and C8/119S were suspended in PBS at a final concentration of 600 μM and subjected to the submicron particle sizer systems (Nicomp, Santa Barbara, CA). Three hundred microliters of the samples were analyzed at a wave length of 514.5 nm at 20°C with a channel width of 6.0 ms, and volume weighted particle size distribution was calculated using a Nicomp distribution analysis program.
C8/119S mutant of Derf-2 stimulates the Derf-2-specific T cell clones by far more potently than wild-type Derf-2
As illustrated in Fig. 1,A, Derf-2 has three intramolecular disulfide bonds at C8–119, C21–27, and C73–78, among which C8–119 bond appears to be crucial to link the N and C termini and form an Ig fold. We first compared the immunogenicity for T cells between C8/119S and wild-type Derf-2 using three independent CD4+ Th clones established from the two atopic patients (SK and SH). Epitope mapping analysis using synthetic peptides has indicated that the Th clones recognized distinct epitopes of Derf-2, residues 14–23 (KKVMVDGCH) for Cl.SK:5:3G, 48–57 (KTAKIEIKAS) for Cl.SK:4:3:Dr, and 90–100 (YTWNVPKI) for Cl.SH:3:10C. Neither C8 nor C119 was included in the epitopes of any Th clones (Fig. 1,A). As shown in Fig. 1, B–D, C8/119S exhibited by far more potent activity than Derf-2 to stimulate the proliferation of all three Th clones in terms of both minimal required Ag concentrations and maximal proliferative response. Thus, 10–1000 times less concentrations of C8/119S could induce proliferative response for the Th clones than Def-2, and the maximal response reached nearly twice of that by Derf-2. C8/119S also induced the production of both IL-4 and IFN-γ much more strongly than Derf-2 with relative ratios of the two representative cytokines of Th2 and Th1 types, respectively, being largely unchanged (Fig. 1, B–D). Although not shown, neither Th clone produced detectable levels of IFN-γ or IL-4 in the absence of specific Ag. Because the augmented immunogenicity of C8/119S was observed similarly in all the independent Th clones with distinct specificity, the effect was unlikely to be due to the generation of neo-epitopes or cryptic epitopes by the mutation.
C8/119S mutant induces the potent and exclusive differentiation of Th 1-type cells from the peripheral blood of atopic patients in vitro
We intended to confirm the augmented immunogenicity of C8/119S using the primary T cells of the two atopic individuals. Fresh T cells from patient SK were stimulated in vitro with either 1 μg/ml Derf-2 or C8/119S twice for a month, and then challenged with either Ag at varying concentrations in the presence of irradiated autologous peripheral PBMC as APCs. As shown in Fig. 2,A, C8/119S induced by far more potent proliferation of the T cells than Derf-2 irrespective of the Ag forms used for the in vitro immunization, confirming the results for the Th clones. Although not shown, essentially the same results were obtained by using the T cells from another patient. We then examined whether the in vitro immunization with C8/119S affected the differentiation profiles of Th1 and Th2 cell types. PBMC from the two atopic patients were stimulated similarly with either Derf-2 or C8/119S twice and then challenged with the optimal dose of C8/119S (1 μg/ml) to assess the cytokine production. As shown in Fig. 2,B, T cells from both atopic individuals that had been stimulated with Derf-2 produced both IL-4 and IFN-γ. In quite a contrast, those that had been stimulated with C8/119S produced little or only marginal IL-4 and by far more IFN- γ following the challenge, resulting in the drastic shift of IL-4/IFN-γ ratio (0.428–0.013 in SK and 0.21–0.0025 in SH). Comparative epitope mapping analysis between the T cells of SK stimulated with Derf-2 and C8/119S revealed essentially similar patterns (Fig. 2 C), again eliminating the possibility for generation of dominant cryptic epitope in C8/119S.
In a given allergen, it has been indicated that the Ag concentrations affect the differentiation profiles of Th1- and Th2-type cells (12). Therefore, we comparatively examined the effect of concentrations of Derf-2 and C8/119S during the in vitro immunization. The primary T cells of patient SK were stimulated with varying concentrations of Derf-2 or C8/119S twice for a month, and then challenged with the optimal dose of C8/119S (1 μg/ml) to assess the cytokine production. As shown in Fig. 3, IFN-γ production reached maximally at 10 μg/ml of Derf-2 and gradually decreased with decreasing the concentrations of Derf-2. Inversely, least IL-4 production was induced by 10 μg/ml of Derf-2, whereas IL-4 production increased with decreasing the concentrations of Derf-2 with 10 ng/ml of Derf-2 inducing maximal IL-4 production following the challenge. The results confirmed the previous report using soluble Ags (12). At 100 μg/ml of Derf-2, the generation of specific T cells capable of producing either cytokine was severely depressed, suggesting that high-dose tolerance occurred. In contrast, in vitro immunization with C8/119S induced T cells producing much greater amounts of IFN-γ and negligible IL-4 at the entire range of concentrations examined from 100 μg/ml to 10 ng/ml. It was particularly noted that high-dose tolerance was not evident even at 100 μg/ml of C8/119S. The results have indicated that C8/119S induces the potent and almost exclusive differentiation of Th1-type cells from the primary T cell population of atopic individuals, regardless the Ag concentrations.
Absence of B cell-mediated Ag presentation is not the major factor for augmented T cell immunogenicity and selective induction of Th1 cell differentiation by C8/119S
Types of APC have been reported also to affect the Th1 and Th2 profiles during the T cell differentiation (12, 15). Because C8/119S almost totally fails to bind to Derf-2-specific Ab (Refs. 20, 22 , see also Fig. 5), it may be expected that C8/119S is unable to be presented by the Derf-2-specific B cells in the PBMC of sensitized patients, thereby affecting the T cell differentiation profile. To examine the possibility, a Derf-2-specific Th clone was stimulated with Derf-2 or C8/119S in the presence of irradiated total PBMC, B cell-depleted PBMC, or purified B cells from the autologous donor as APC. As shown in Fig. 4 A, Derf-2 induced significant proliferative response in the presence of purified B cells as well, although the response was somewhat less than that in the presence of total PBMC. The reduced efficiency may be due to the relatively higher radiosensitivity of B cell-mediated APC function (25). In contrast, C8/119S totally failed to stimulate the Th clone in the presence of purified B cells as APC, indicating that C8/119S is indeed unable to be presented by the Derf-2-specific B cells. However, even when B cell-depleted PBMC were employed as APC, C8/119S still induced much greater proliferative response than Derf-2. The results have indicated that the absence of B cell-mediated Ag presentation of C8/119S is not a major factor for its augmented immunogenicity for T cell proliferation.
We then addressed whether the B cell-mediated Ag presentation of Derf-2 affected the profile of cytokine production. As shown in Fig. 4,B, the T cells that had been stimulated twice with Derf-2 in the presence of B cell-depleted PBMC produced comparable level of IL-4 to those that had been stimulated with total PBMC, although IFN-γ production was significantly increased. Thus, IL-4/IFN-γ ratio significantly decreased from 0.7 to 0.21 by excluding B cells from APC. When stimulated with Derf-2 in the presence of purified B cells, IL-4/IFN-γ ratio was comparable or even greater (0.79), although the APC activity of irradiated B cells was significantly lower again. In contrast, in the same experiment cytokine production by T cells of the same patient (SK) that had been stimulated with C8/119S alone was by far markedly polarized toward Th1 pattern, the IL-4/IFN-γ ratio being 0.01 (Fig. 4,B). By using the B cell-depleted PBMC as APC, essentially the identical results were obtained, the IL-4/IFN-γ ratio being 0.007 (Fig. 4,B). Con- sistently, two independent Th clones established by the repeated stimulation with C8/119S alone almost exclusively produced IFN-γ with negligible IL-4 (Fig. 4,B) in contrast to the Th clones induced with Derf-2 (see Fig. 1). Thus, it has been suggested strongly that the absence of B cell-mediated Ag presentation per se does not fully, if any, account for the exclusive Th 1 cell differentiation by C8/119S either.
C8/119S mutation results in the stable molecular polymerization as a consequence of the degenerate secondary structure: structural implication for the augmented T cell-immunogenicity and selective Th1 cell induction
Results so far suggested that the altered immunogenicity of C8/119S might be attributable to its intrinsic molecular features distinct from Derf-2. The C8/119S mutation disrupts an intramolecular disulfide bond of Derf-2, which is expected to be critical for the Ig fold formation of the protein (see Fig. 1,A). Therefore, we first analyzed both Derf-2 and C8/119S by the PAGE. In the nondenaturing PAGE, C8/119S migrated much slower than Derf-2, whereas both proteins exactly comigrated at the expected molecular size of 15 kDa in the SDS-PAGE (Fig. 5,A). The results have suggested that C8/119S molecules may be polymerized into larger molecular masses in physiological condition. To estimate the extent of polymerization, the light scatter particle sizing analysis of C8/119S was performed in comparison with the native and heat-aggregated Derf-2. As shown in Fig. 5,B, the vast majority (>90%) of C8/119S exhibited a peak at the position of about five times larger mean diameter than that of Derf-2 (3.9 nm vs 20 nm). The results have confirmed that C8/119S is indeed polymerized solubly in physiological solution, in which around 125 molecules are estimated to polymerize on average. Heat treatment of Derf-2 at 100°C for 10 min also resulted in less yet significant molecular aggregation (Fig. 5,B). However, heat aggregation was apparently unstable, in that it was totally dissociated even in the nondenaturing PAGE (Fig. 5,A). CD spectrum analysis has revealed that β-strands of Derf-2 are almost totally abrogated by the C8/119S mutation, whereas heat treatment hardly affected them (Fig. 5,C). Reflecting this, C8/119S almost totally lost the ability to bind Derf-2 specific IgE as reported previously, whereas heated Derf-2 retained it (Fig. 5 D).
Aggregation of proteins is well known to result in the increase in immunogenicity (26, 27). Therefore, we compared the immunogenicity of C8/119S with heat aggregated Derf-2. As shown in Fig. 6,A, both C8/119S and heated Derf-2 induced much greater proliferative response of Derf-2-specific T cell clone than Derf-2. We then compared the differentiation of Th1- and Th2-type cells in the primary T cells from an atopic patient with C8/119S and heat-aggregated Derf-2. The PBMC were stimulated twice with the optimal concentration of wild-type Derf-2, C8/119S, or heat-treated Derf-2, and then challenged with C8/119S (1 μg/ml) to assess the production of IFN-γ and IL-4. Conforming to the previous results, T cells immunized with C8/119S almost exclusively produced IFN-γ with negligible IL-4, whereas those stimulated with Derf-2 produced both cytokines comparably (Fig. 6,B). In contrast, T cells immunized with heated Derf-2 were found to produce both IL-4 and IFN-γ with the comparable ratio to those stimulated with Derf-2, although the amounts tended to be less (Fig. 6 B). These results have indicated that the heat aggregation of Derf-2 with marginal degeneration in the structure results in the augmented immunogenicity for the proliferation of primed T cells but is not sufficient to induce the polarized Th1 cell differentiation. Thus, it is suggested that the degenerate secondary structure resulting in much more stable and extensive polymerization may be responsible for the exclusive induction of Th1 cell differentiation by C8/119S mutant.
It has been indicated that a number of factors affect the T cell-immunogenicity and profiles of Th1 and Th2 cell differentiation, including routes of Ag exposure (11), Ag doses (12, 13), Ag forms (14), and types of APC involved (13, 16). In the present study, we have indicated that a recombinant C8/119S mutant of Derf-2 allergen induces by far more potent proliferation than wild-type Derf-2 for the primed T cells as well as the established Th clones in vitro. Markedly augmented immunogenicity of C8/119S was observed for all the independent Derf-2-specific Th clones with distinct epitope specificity. Thus, the augmented immunogenicity of C8/119S was unlikely to be due to the generation of any new or cryptic dominant epitopes by the mutation. C8/119S hardly affected the profiles of IL-4 and IFN-γ production in the established Derf-2-specific Th clones. However, in the primary T cells from atopic individuals, C8/119S almost exclusively induced the Th1 type cells, whereas the optimal dose of native Derf-2 induced the generation of Th cells producing both IL-4 and IFN-γ. We managed to obtain two independent Th clones by the repetitive stimulation solely with C8/119S, and both of them indeed produced abundant IFN-γ with marginal IL-4 (Th1 type; Fig. 4,B). In contrast, a number of Th clones established with Derf-2 from the same atopic individual consistently produced significant IL-4 with variable levels of IFN-γ (our unpublished data, see also Fig. 1). Thus, C8/119S not only exhibits markedly augmented immunogenicity for the primed T cells but also strongly affects the Th1/Th2 differentiation profile in the primary T cells of atopic patients inducing a strong polarity toward Th 1 cells.
The profiles of Th1- and Th2-type cell differentiation was reported to depend on the concentrations of soluble Ags during the immunization, in which the higher Ag doses inducing greater proliferative response tended to favor the Th1-type response, whereas the lower doses the Th2 response (13). The rule apparently held true for the wild-type Derf-2, which is a soluble and monomeric protein (Fig. 3,A). However, C8/119S mutant induced the differentiation of T cells producing abundant IFN-γ with little IL-4 at the entire range of concentrations examined from 100 μg/ml to 10 ng/ml (Fig. 3,B). Notably, the induction of high-dose tolerance detected by 100 μg/ml of Derf-2 was not observed by the same concentration of C8/119S. Thus, the exclusive induction of Th1 cell differentiation was unlikely to reflect the shift of Ag dosage effect on the Th1/Th2 differentiation profile. By using a series of random mutation of an antigenic peptide derived from a house dust mite allergen Der pII, it was reported that the affinity of peptides for the corresponding T cells significantly affects the profiles of IFN-γ and IL-4 production (28). However, we failed to detect significant contribution of the mutated residues of Derf-2 to the activation of T cells by the epitope mapping analysis (Fig. 2 C).
It has been suggested also that types of APC can affect the profiles of Th1 and Th2 cell differentiation, in that the specific B cells tend to induce Th2 rather than Th1-type cells (15). For instance, it was reported that the denatured form of recombinant bee venom PlA2 allergen, which failed to bind to the specific Ab, tended to favor the induction of Th 1 cells in vitro, and that the effect was attributed to the absence of B cell-mediated Ag presentation (29). Because B cells specific for the Ags are expected to have much higher affinity for the Ag than other APCs, the preferential generation of Th2 cells with limited concentrations of Derf-2 may be ascribed partly to the difference in the APC. Present results have indicated that C8/119S, which fails to bind to the Derf-2 specific Ab, is indeed unable to stimulate the proliferation of specific T cells when purified B cells were used as APC. However, wild-type Derf-2 in the presence of B cell-depleted APC still induced Th cells producing comparable level of IL 4 with those stimulated in the presence of total PBMC as APC, whereas C8/119S induced exclusive IFN-γ production even in the presence of B cell-depleted APC. It was thus unlikely that the absence of B cell-mediated Ag presentation per se was the major factor for the selective Th 1 cell induction by C8/119S.
These results have implied collectively that the unique immunogenic features of C8/119S might be ascribed to its intrinsic molecular nature per se. Nondenaturing PAGE and light scattering particle sizing analysis have indicated that, unlike Derf-2, C8/119S is polymerized solubly in physiological condition, consisting of around 100 molecules. The polymerized form of C8/119S appeared to be more extensive and stable than the molecular aggregation of Derf-2 by heating, in that the former was retained in the nondenaturing PAGE, whereas the latter completely dissociated. We have recently succeeded in obtaining x-ray crystallography and confirmed that Derf-2 shows typical β-barrel structure consisting of three parallel β-strands on the one side and four parallel β-strands on the other twisted by around 30 degree (Y. Tanaka et al., unpublished observation). Present CD spectrum analysis, on the other hand, revealed that the β-strands were almost totally abrogated in C8/119S, whereas they were hardly affected by heating. Reflecting this, C8/119S almost totally lost the binding capacity to the Derf-2 specific IgE, whereas heat-aggregated Derf-2 retained it, consistent with the known heat stability of Derf-2 antigenicity. It is speculated that C8/119S mutation leads to the formation of random-coil structure by losing all β-strands. Such conformational change may result in the exposure of hydrophobic amino acid side chains representing nearly 40% of total residues on the molecular surface, which are otherwise buried inside the molecules leading to the quite stable molecular polymerization via abundant hydrophobic interactions.
Present results have indicated that, although heat-aggregated Derf-2 induces augmented proliferation of established Th clones comparably with C8/119S, it fails to induce the Th1 cell polarity. Thus, it may be suggested that the more extensive molecular polymelization in terms of stability and/or size of molecular masses is required for the polarized Th1 cell induction than for the augmented stimulation of T cell proliferation. It was reported previously that OVA chemically cross-linked by glutaraldehyde into very large masses consisting of nearly 1000 molecules on average inhibited the induction as well as on-going IgE production (30, 31), and preferentially elicited the induction of Th1 type cells in vivo (14). At present, exact mechanisms for the selective differentiation of Th1 cells by C8/119S remain to be further investigated, but a few possibilities may be considered. First, such stably polymerized Ags may generate antigenic peptides at by far higher local density in class II MHC on the surface of APC than monomeric Ags during the intracellular endosomal processing. Difference in the local density of antigenic peptide occupancy among MHC class II molecules is suggested to influence the destiny of differentiation of primed T cells toward Th1 and Th2 subsets (32, 33). Alternatively, stably polymerized proteins may exhibit additional “adjuvant” effect to activate APC, such as macrophages and dendritic cells. Lack of high-dose tolerance by C8/119S might support the possibility. Most recently, it has been indicated that a single transcription factor (T bet) is sufficient to dictate Th cell differentiation into Th1 type (34). It is tempting to speculate that scavenger receptors for the stably polymerized protein masses might be involved in providing signals for the APC to selectively induce Th1 cell differentiation (35), possibly via activation of such a transcriptional factor.
In either case, the distinct immunogenic features of C8/119S mutant, that is, loss of epitope(s) for the specific IgE Ab, markedly augmented immunogenicity for the primed T cells, and exclusive induction of Th1 cell differentiation, altogether make it a desirable and safer Ag for hyposensitization therapy. In fact, it was shown that repeated nasal administration of small doses of C8/119S in the mice sensitized with Derf-2 induced much more significant prevention from the bronchial construction upon challenge with Derf-2 than wild-type Derf-2 (23). The similar effects were confirmed also in monkeys (T. Yokoyama et al., unpublished observation). Although it remains to be examined systemically whether the present results obtained in two atopic patients hold true generally, they raise a good possibility for the clinical application of the C8/119S mutant allergen for desensitization therapy. Present results further imply that point mutation of other clinically important allergens critically affecting the secondary structure may provide an useful approach to develop safer and effective hyposensitizing Ags for the human atopic diseases.
We thank Dr. Ikeda for providing us synthetic peptides of Derf-2.
The work was supported by grants from the Ministry of Education, and Science, Japan.
Abbreviation used in the paper: CD, circular dichroism.