WO2003015716A2

WO2003015716A2 - Immunoglobulin e vaccines and methods of use thereof

Info

Publication number: WO2003015716A2
Application number: PCT/US2002/026986
Authority: WO
Inventors: Swey-Shen Alex Chen; Yong-Min Yang; Theresa Barankiewicz; Zhong Chen
Original assignee: Ige Therapeutics, Inc.
Priority date: 2001-08-13
Filing date: 2002-08-08
Publication date: 2003-02-27
Also published as: WO2003015716A3; EP1497654A2; US7604955B2; AU2002326751A1; US20030073142A1; EP1497654A4

Abstract

The present invention concerns the prophylaxis or therapy of immunoglobulin E-mediated disease conditions. The present invention discloses methods for identifying peptides that induce a cytotoxic T-lymphocyte response against immunoglobulin E. The present invention further discloses compositions that are able to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, and methods for their use.

Description

IMMUNOGLOBULIN E VACCINES AND METHODS OF USE THEREOF

The present application claims benefit of priority to the following application, which is incorporated by reference in its entirety herein: U. S. Provisional Patent Application No. 60/312,120, entitled "Molecular Constructs of IgE Vaccines for Allergic Asthma and IgE- Mediated Allergic Diseases, and Methods of Use Thereof, filed on 13 August 2001.

TECHNICAL FIELD The present invention relates generally to the field of immunology and allergy, and more specifically concerns compositions useful in the prophylaxis or therapy of immunoglobulin E- mediated disease conditions, and methods for their use.

BACKGROUND Immunoglobulin E (IgE) plays a central role in atopic hypersensitivity conditions including allergic rhinitis, allergic asthma, food allergies, and contact allergies (atopic dermatitis) and non-atopic hypersensitivity conditions including anaphylaxis, urticaria, and hives (Galli and Lantz, 1999, Fundamental Immunology, ed. Paul, p. 1127). IgE is identified as the presumptive causative agent responsible for passive transfer of cutaneous reactivity to allergens. Serum levels of IgE were shown to correlate closely with cutaneous reactivity to common allergens, as well as with bronchial hyper-responsiveness (Sears et al, 1991, N Engl. J. Med., 325:1067). Syndromes of allergic rhinitis and asthma are attenuated when levels of circulating IgE are neutralized by antibodies (Chang, 2000, Nature Biotech, 18:157).

IgE-mediated allergic inflammation is believed to be initiated upon triggering of IgE- sensitized mast cells with allergens. Inflammatory mediators include antihistamine and leukotrienes. The most common therapeutic approach to allergic inflammation is to block target organs from inflammatory mediators secreted by mast cells, for example through the use of antihistamines, which are believed to block the binding of histamine to their HI receptors on vascular endothelial cells and vascular, respiratory, and gastrointestinal smooth muscle cells. Antihistamines are particularly helpful in allergic rhinitis, ocular allergies, urticaria, and atopic dermatitis. Newly developed non-sedating antihistamines are popular. However, antihistamines have been minimally beneficial in allergic asthma, and have little ability to affect the expression of chronic allergic inflammation in the airways of these patients. Several different leukotriene receptor antagonists are now available, and these agents show promise in reducing the signs and symptoms of chronic allergic asthma. Although these agents are generally preferred to antihistamines in this clinical setting, they only partially attenuate clinical evidence of the disease (Boushey, 2001, J. Allergy Clin. Immunol, 108:S77) probably due to the multiple entities of mast cell small molecules mediators as well as a host of cytokines such as interleukin 4 (IL-4), interleukin 5 (EL-5), and tumor necrosis factor alpha (TNF-alpha), and eotaxin and other chemokines secreted by mast cells (Willis-Karp, 1999, Annu. Rev. Immunol, 17:255).

As an alternative to the complexity of blocking the pharmacological effects of mast cell mediators, there have been various approaches to competitively inhibit IgE binding to and sensitization of its type I high affinity Fc receptor (FceRI). FceRI receptor antagonists can be for example, a long piece of IgE polypeptide of 72 amino acids in length derived from constant region heavy chain domains of CH2 to CH3 (Helm et al, 1988, Nature, 331:180). Moreover, there have been reports that short synthetic peptides from five to sixteen amino acid sequences from the CH4 domain interfere with direct IgE sensitization to FceRI on mast cells (Stanworth et al, U. S. Patent 5,955,076). Immunotherapy by allergen desensitization aims to shift the immune system response from developing Th2-type helper T-lymphocytes to Thl helper T-lymphocytes. The general paradigm is that interleukin-4, secreted by Th2, plays a central role in initiating and augmenting allergen-specific IgE production. Interleukin-4 in turn plays a key role in skewing dichotomous T-lymphocyte development into the Th2 subset, thus further augmenting IgE production. Th2 commitment, in turn is believed to lead to production of more LL-4 (Mosmann and Coffinan, 1989, Ann Rev. Immunol, 7:145). A source of endogenous IL-4 is also believed to be contributed by mast cells and basophils, by a unique subset of IL-4 secreting CD8+ T- lymphocytes, and by NK1.1+ CD4 T-lymphocytes. There have been numerous attempts to downregulate Th2 development by administering recombinant allergens to patients, but due to the multiple sources of IL-4 and the complex cytokine feedback circuits in dichotomous CD4 T- lymphocyte development, this approach tends to yield variable results (Kay and Lessof, 1992, Clin. Exp. Allergy, S3:l; Seymour et al, 1998, J. Exp. Med., 187:721).

Another approach to down-regulate the levels of circulating IgE resides is the administering of an anti-IgE antibody, that is to say, passive immunization against IgE. One such anti-IgE antibody recognizes a mast cell IgE binding site and thus blocks IgE sensitization to mast cells (Chang, 2000, Nature Biotech., 18:157). The partially humanized monoclonal antibody E25 (MAb-E25) (Chang, US Patent No. 5,428,133) is currently in a Phase m clinical trial. Passive immunization against anti-IgE results in the scavenging of circulating IgE, but would not be expected to substantially inhibit IgE production in atopic patients. Additionally, it has been shown that circulating IgE-anti-IgE complexes accumulate to a level ten- fold higher in passively immunized patients than that prior to treatment. This result is probably due to the increase in half-life of circulating IgE from about 2 days (circulating free IgE) to about 14 days (IgE-anti-IgE complexes), possibly because the humanized IgG used as the passive vaccine protects IgE from degradation. One concern from this is that IgE dissociated from the IgE-anti- IgE complex may sensitize mast cells. Another disadvantage to this passive vaccine is that a high dose of 300-500 mg per patient is routine. Benefits of anti-IgE treatment appear transiently during the period of intensive treatment, and the patient's condition significantly worsens two to three days after cession of injection of MAb-E25. Moreover, since MAb-E25 is not completely humanized (about 5% mouse sequence remains), neutralizing antibodies may be produced upon a second or subsequent treatments, thus further diminishing the efficacy of this approach (Corne et al, 1997, J. Clin. Inv., 99:879; Milgrom et al, 1999, N. Engl. J. Med., 341:1966). Furthermore, the paratope-specificity of MAb-E25 carries the additional risk of inducing anti-MAb-E25 antibodies, that is to say, anti-idiotypic antibodies (Nogel et al, 2000, J. Mol. Biol, 298:729; Jerne, 1993, Scan. J. Immunol, 38:1) bearing an internal image that mimics the original conformation of the FceRI binding site. Being bivalent, such antibodies can potentially crosslink FceRI and induce mast cell degranulation even in the absence of circulating IgE.

A long-term solution for IgE-mediated allergic diseases is urgently needed. One approach is to actively immunize against IgE. Existing approaches have focussed on using an entire IgE heavy chain domain as the immunogen (Hellman et al, U. S. Patent 5,653,980). Although conceptually attractive, these approaches may lead to severe or even life-threatening IgE-mediated anaphylaxis, since antibodies raised against multiple antigenic epitopes across intact or largely intact IgE domains may cause extensive IgE-mediated mast cells/basophil activation. The contact region between the IgE heavy chain CH2 and CH3 domains and the FceRI receptor spans 5600 Angstrøm according to the recent crystallographic structure obtained for the complex (Garman et al, 2000, Nature, 406:259). The safety of using MAb-E25 resides in its blocking of the binding to FceRI.

It is believed that conformation-sensitive B-lymphocyte epitopes are poorly represented by their corresponding synthetic, non-conformationally restricted peptides, even when these are coupled to a foreign carrier protein. There is at present no method for identifying and constructing IgE B-lymphocyte vaccines that correspond to constrained epitope such as those at mast cell binding sites. There is thus a great deal of unmet medical need to design other innovative means to reduce levels of IgE production.

In contrast, major histocompatibility class I (MHC class I)-restricted cytotoxic T- lymphocyte (CTL) epitopes are believed to be sequence-dependent. The present invention aims at developing a method of active immunization against IgE, using as immunogen a short stretch of sequence universally present on the IgE heavy chain constant region, independent of any conformation requirement. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 depicts a schematic of an example of a vaccine of the present invention, consisting of immunogenic peptides derived from immunoglobulin E (IgE) that induce a cytotoxic T- lymphocyte (CTL) response against IgE peptides (depicted by shaded triangles) naturally processed and presented by a major histocompatibility complex class I (MHC-I) protein

(depicted by shaded oval on cell surfaces). Such peptide-MHC-I complexes can be recognized by cytotoxic T-lymphocytes via T-cell receptors (TCR). For example, in IgE-producing cells, IgE peptides are naturally processed from the IgE heavy (H) chain, and presented on the MHC-I protein. Alternatively, an antigen-presenting cell (APC) can present naturally processed IgE peptides on the MHC-I protein. Such an immunized cytotoxic T-lymphocyte or "killer cell" recognizes and lyses cells presenting the IgE peptides on the MHC-I molecule.

FIG. 2 depicts the three differentiation stages of IgE-producing B-lymphocytes, which are examples of targets for IgE-specific cytotoxic T-lymphocytes. Pre-switched B-lymphocytes express IgE heavy chain germ line transcripts under the influence of allergenic stimulations and may express low levels of IgE heavy chain prior to gene rearrangement. IgE-switched B- lymphocytes exhibit a high density of surface IgE anchored to the cell membrane via IgE membrane exons 1 and 2. IgE-secreting plasma cells express the highest levels of secretory IgE and thus the highest levels of naturally processed and presented IgE on MHC-I molecules. All three B-lymphocytes stages can serve as targets for an IgE-specific CTL response. Shaded triangles represent IgE peptides naturally processed and presented on the cell surface by an MHC-I molecule.

FIG. 3 depicts results of a standard chromium release assay using effector (E) cells at the indicated ratios to target (T) cells. Values shown are means of triplicate cultures with error bars indicating the standard deviation. FIG. 3 A: Effector cells were cytotoxic T-lymphocytes obtained from BALB/c mice (expressing murine MHC-I haplotype H-2^d) immunized with one of the test peptides pi or p2, together with the co-stimulatory factor ovalbumin peptide (OVAp), in N-[l-(-2,3-dioleoyloxy)propyl]-NN,N-trimethylammonium methylsulfate (DOTAP) liposomes. Target cells were P815 murine mastocytoma cells (expressing murine MHC-I haplotype H-2 ) pre-incubated with the appropriate test peptide (pi or p2) and interferon-gamma. These results show that both IgE-derived test peptides pi and p2, delivered with a co-stimulatory peptide and with liposomes, were able to elicit a CTL response in BALB/c mice that was cytotoxic to target cells displaying the peptides. FIG. 3B: Effector cells were cytotoxic T-lymphocytes obtained from BALB/c mice immunized with test peptide p3, with or without OVAp, in the presence or absence of DOTAP liposomes. Target cells were P815 murine mastocytoma cells that had been contacted with the test peptide p3 and interferon-gamma. These results show that immunization with p3 in the presence of the co-stimulatory peptide OVAp was able to induce a CTL response in BALB/c mice that was cytotoxic to the p3-treated P815 target cells; this CTL response was augmented by delivery of the peptides in liposomes. Immunization with the test peptide p3 alone, or with the co-stimulatory peptide OVAp in liposomes, did not induce a substantial CTL- response in BALB/c mice against the target cells; delivery of p3 in liposomes also amplified the CTL response relative to that seen with p3 alone.

FIG. 4 depicts the aligned amino acid sequences and sequence identification numbers for the test peptides pi, p2, p3, and p4. Underlined in the sequence of test peptide p3 are the two residues between which the serine endoproteinase Glu-C cleaves (marked by an arrow).

FIG. 5 depicts results of a standard chromium release assay using effector (E) cells at the indicated ratios to target (T) cells. Values shown are means of triplicate cultures with error bars indicating the standard deviation. Effector cells were cytotoxic T-lymphocytes obtained from C57BL/6 mice (expressing murine MHC-I haplotype H-2^b) that had been immunized with either the test peptide p4 with OVAp, or with the test peptide p5 in liposomes. Target cells were murine EL-4 lymphoma cells (expressing murine MHC-I haplotype H-2^b) that had been contacted with the test peptide p4 and interferon-gamma. These results show that mice immunized with either of the test peptides p4 or p5 developed comparable levels of immunoglobulin E-specific CTL response. This showed that p5 had been efficiently internalized into antigen-presenting cells, naturally processed and presented through the exogenous major histocompatibility complex class I pathway to yield immunogenic p4-derived peptide-MHC-I molecule complexes.

FIG. 6 depicts results of a standard chromium release assay using effector (E) cells at the indicated ratios to target (T) cells. Values shown are means of triplicate cultures with error bars indicating the standard deviation. Effector cells were cytotoxic T-lymphocytes obtained from BALB/c mice (murine MHC-I haplotype H2^d) that had been immunized with 2 x 10⁴ syngeneic bone marrow-derived dendritic cells (DC) (treated as described above with p3 and OVAp), and then treated or not in vivo with the anti-murine CD40 antibody FGK-45. FIG. 6 A: Target cells were P815 murine mastocytoma cells (murine MHC-I haplotype H2^d) that had been contacted with the test peptide p3 and treated overnight with 50 units per milliliter interferon-gamma. FIG. 6B: Target cells were IgE-secreting 26.82 hybridomas, which endogenously synthesize and naturally process and present murine immunoglobulin E peptides associated with the major histocompatibility complex class I molecule H-2^d. These results demonstrate that the test peptide p3 served as an immunogenic peptide when presented by antigen-presenting cells (dendritic cells that recognize p3) and were able to elicit in the DC-immunized mice a CTL response to naturally processed and presented IgE peptides. Dendritic cells incubated with p3 and OVAp were sufficient for inducing this CTL response. The magnitude of this CTL response was not amplified by either: (a) in vivo treatment of the DC-immunized mice with the anti-CD40 antibody FGK-45, or (b) incubation of the dendritic cells with p3 and OVAp with DOTAP liposomes. FIG. 7 depicts results of a standard chromium release assay using effector (E) cells at the indicated ratios to target (T) cells. Values shown are means of triplicate cultures with error bars indicating the standard deviation. Effector cells were cytotoxic T-lymphocytes obtained from 129/SvEv mice (murine MHC-I haplotype H2^b) that either carried the wild-type IgE gene (IgE +/+) (FIG. 7C and FIG. 7D), or that were defective in the IgE gene (IgE -/-) (FIG. 7A and FIG. 7B). The mice were immunized as indicated with the test peptide p3 in the presence of OVAp, and then treated in vivo with the anti-CD40 antibody FGK-45 or with rat IgG2a, or were immunized with dendritic cells (treated with p3 and OVAp in the presence or absence of DOTAP liposomes), and then treated or not in vivo with the anti-CD40 antibody FGK-45. Target cells were murine immunoglobulin E-secreting B4 hybridoma cells (murine MHC-I haplotype H-2 ) (FIG. 7A and FIG. 7C), or murine EL-4 lymphoma cells (murine MHC-I haplotype H-2^b), which had been contacted with the test peptide p3 and treated with interferon-gamma (FIG. 7B), or 26.82 hybridomas (which produce endogenous murine IgE associated with the murine MHC-I haplotype H-2^d) (FIG. 7D). An IgE peptide-specific CTL response was observed in 129/SvEv, IgE-/- mice immunized with p3 with OVAp in liposomes, or with DC treated with p3 and OVAp. Cytotoxic T-lymphocytes from these IgE-/- mice were able to lyse target cells that naturally processed and presented IgE peptides bound to H-2^b, regardless of whether the IgE peptides were biosynthesized endogenously (as in the B4 hybridoma cells, FIG. 7A) or were introduced exogenously (as in the EL-4 lymphoma cells, FIG. 7B).

A similar IgE peptide-specific CTL response was observed in 129/SvEv, IgE+/+ mice immunized with p3 with OVAp in liposomes, or with DC treated with p3 and OVAp. This CTL response was also specific for IgE peptides bound to a specific MHC-I haplotype as the CTL were able to lyse B4 hybridoma cells (H-2^b haplotype) (FIG. 7C) but not 26.82 hybridoma cells (H-2^d haplotype) (FIG. 7D). These results show that: a) MHC-I-restriction of the CTL response is observed at both the inductive phase and effector phase, b) immunization methods using different modes of co-stimulation were sufficient to overcome tolerance to self-IgE peptides in the IgE +/+ mice, c) IgE test peptides, processed and presented by DC, are at least similar to endogenously produced, naturally processed and presented IgE peptides that are bound to MHC-I of IgE-secreting plasma cell, and d) the CTL response induced by DC immunization was not amplified by treatment with anti-CD40 antibody. FIG. 8 depicts the inhibition of immunoglobulin E production in vitro and in vivo by immunoglobulin E-specific cytotoxic T-lymphocytes. FIG. 8A and FIG. 8B depict results of a commercial double-sandwich ELISA assay for immunoglobulin E. In FIG. 8A, cytotoxic T- lymphocytes (at an effector-to-target cell ratio of 10 to 1) elicited by p3 immunization of H-2 haplotypic mice decreased IgE production by H-2 haplotypic IgE-secreting cells (26.82 hybridoma cells) but did not appreciably decrease IgE production by H-2 haplotypic IgE- secreting cells (B4 hybridoma cells). These results show the MHC-I molecule specificity of the CTL response induced by a single test peptide. In FIG. 8B, cytotoxic T-lymphocytes from the p3-immunized mice markedly inhibited IgE production by LPS and IL-4 stimulated B- lymphocyte cultures in vitro at effector-to-target cell ratios of 1:10 or 1:30, whereas cytotoxic T- lymphocytes obtained from the mice immunized with only OVAp as a control did not decrease IgE production. These results show the test peptide-induced CTL response can inhibit IgE production by cells that produce IgE in situ. FIG. 8C depicts results of passive cutaneous anaphylaxis (PC A) measurements in BALB/c mice that had been immunized with p3 or pi with OVAp in DOTAP liposomes, or with only OVAp in DOTAP liposomes, or with only saline. The mice were subsequently challenged with keyhole limpet hemocyanin (KLH), and the allergen-specific (KLH-specific) IgE in pooled blood was then determined by PCA titres. These results show that immunization with a test peptide, together with appropriate modes of co- stimulation, can substantially decrease allergen-induced IgE production in vivo.

FIG. 9 depicts results of fluorescence-activated cell sorting (FACS) experiments. Five test peptides (peptide numbers 539.11, 539.12, 539.8, 539.9, and 539.10, representing three of the four CH domains of human IgE) from the panel of 12 high-, moderate-, and low-affinity test peptides shown in Table I were tested in RMA-S-A2.1 cells (murine TAP-deficient RMA-S cells that express HLA-A2.1). A nonameric peptide derived from the human immunodeficiency virus (HJN), was tested as an example of a nonameric HLA-A2.1-specific peptide that is not derived from human IgE. RMA-S-A2.1 cells were incubated with the individual test peptides, in the presence or absence of human beta-2 microglobulin. Cells were incubated with fluorescein isothiocyanate (FITC)-labelled anti-HLA-A2.1 antibody. FACS analysis was performed on a FACScan, with about ten thousand events collected and analyzed. Increase of HLA-A2.1- specific fluorescence intensity by one log was observed in cultures incubated with any of the five human IgE test peptides tested (test peptides number 539.8, 539.9, 539.10, 539.11, and 539.12, shown respectively in FIG. 9A, FIG. 9C, FIG. 9D, FIG. 9E, and FIG 9F). In contrast, no appreciable increase of HLA-A2.1 -specific fluorescence intensity was detected in cultures incubated with medium alone (FIG. 91). These results show that the five IgE test peptides were able to phenotypically correct defective MHC-I molecule expression in RMA-S-A2.1 cells. The increase of HLA-A2.1 -specific fluorescence intensity is comparable among the five IgE test peptides, and between the five IgE test peptides and the HIV control peptide (HJN) (FIG. 9G). These results show that different CH domains of human IgE can provide a plurality of therapeutic targets for IgE-specific cytotoxic T-lymphocytes. The addition of exogenous human beta-2 microglobulin ("beta2-m") did not substantially enhance surface expression of HLA-A2.1 (FIG. 9A versus FIG 9B; FIG. 9G versus FIG. 9H), showing that these IgE peptides, which were predicted to bind strongly to HLA-A2.1, induced a high or possibly maximal expression of HLA-A2.1 at the cell surface.

FIG. 10 schematically depicts a B-lymphocyte from mice that are transgenic for a human MHC-I molecule and for human immunoglobulin E. Mice were engineered to transgenically express both human HLA-A2.1 and human immunoglobulin E (tg hulgE) (heavy chain constant region and membrane domain, fused into the murine VH gene). The resulting animals express human HLA-A2.1 at the cell surface, and produce the chimeric human murine immunoglobulin E (comprising human IgE heavy chain and the endogenous murine IgE light chain) as well as the endogenous murine IgE. FIG. 11 schematically depicts construction of an antigen-specific, mouse/human chimeric immunoglobulin gene. The murine IgE VH segment (mVH) was ligated via the Sail restriction site near the beginning of the human IgE CHI domain, followed by the remainder of the IgE heavy chain constant region domains (CH2, CH3, and CH4) and the membrane domains (Ml and M2). The resulting chimeric construct was cloned into the modified pUC19 vector, pHSE3', which contains the MHC-I H-2 promoter and immunoglobulin heavy chain enhancer region.

FIG. 12 A depicts the different human immunoglobulin E isoforms cDNA amplified by reverse transcriptase-polymerase chain reaction from total RNA isolated from U266 cells (lane 2), and cloned into pGEM vectors (lanes 3-7 and 10). Three IgE isoforms are predominantly assembled by U266 cells: the classical membrane long isoform (lane 3), the membrane short isoform (lanes 4, 5, and 7), and the membrane long isoform (lane 6). In addition, the main IgE isoform secreted by U266 cells was amplified by RT-PCR (lane 9) and cloned into pGEM (lane 10). FIG. 12B depicts the adeno viral IgE construct consisting of the adenoviral vector pAd-lacZ, into which was cloned the full-length cDNA of human IgE membrane long form (pAd-IgEml). FIG. 13 depicts results of a Western blot using goat anti-human IgE IgG as the primary antibody, rabbit anti-goat IgG antibody, conjugated with horseradish peroxidase (HRP) as the secondary antibody, and 3,3',5,5'-tetramethylbenzidine as a substrate. The full-length cDNA of human IgE membrane long form was cloned into the mammalian expression vector pShuttle and the resulting construct used to transfect human kidney 293 cells. Transient expression of the pSml3 insert (pSml3) with full membrane-encoded exon sequences was confirmed by the presence of an immunoreactive protein transcript with the predicted molecular weight of about 65 kDa, which was absent in the null vector (pShuttle) (FIG. 13A). The adenoviral IgE construct pAd-IgEml, which also contained the full-length cDNA of human IgE membrane long form, was amplified and subjected to SDS-PAGE and Western blot. The heavy chain gene product expressed using pAd-IgEml appeared as the same molecular weight but migrated as a broader band, which was absent in the null vector (pAd-lacZ) (FIG. 13B). FIG. 14 depicts two approaches to testing cytotoxic T-lymphocytes, from transgenic mice that express HLA-A2.1 and that have been immunized (for example, with a selected immunogenic IgE peptide), for the ability to lyse IgE producing cells and to inhibit IgE production. The two approaches use different target cell types. In the first approach (depicted in FIG.14A), the target cells are cells in which the MHC-I molecule is contacted with test peptide exogenously. As shown in FIG. 14 A, a specific example of such target cells are TAP-deficient RMA-S cells transgenically expressing HLA-A2.1. Another example of such target cells are the human cells T2. The target cells are contacted (pulsed) with IgE peptides that bind to and induce cell-surface expression of HLA-A2.1, allowing the cytotoxic T-lymphocytes to recognize and lyse these target cells. In the second approach (depicted in FIG.14B), the target cells are cells in which the MHC-I molecule presents endogenously produced IgE peptides. As shown in FIG. 14B, a specific example of such target cells are human U266 IgE myeloma cells, which have the HLA-A2.1 haplotype.

FIG. 15 A, FIG. 15B, FIG. 15C, and FIG. 15D depict alignment of the immunoglobulin E heavy chain (including complete CHI to CH4 constant region domains) amino acid sequences for human, sheep, rat, pig, mouse, horse, duck, dog, cow, chimpanzee, and cat. Shaded regions indicate regions of sequence homology.

FIG. 16 depicts a phylo genetic tree representing the inferred evolutionary divergence between the immunoglobulin E heavy chain polypeptides of human, sheep, rat, pig, mouse, horse, duckbilled platypus, dog, cow, chimpanzee, Trichosurus (brushtail possum), and cat, calculated using the progressive multiple sequence alignment program ClustalW and using the alignment parameters given. The numerical value on top of each line indicates the evolutionary distance for that respective species from the most recent common ancestor of IgE.

FIG. 17 A depicts the different starting regions and combinations of membrane exons observed in differentially spliced human immunoglobulin E isoforms. FIG. 17B lb depicts splicing maps diagramming the classical secreted IgE isoform (which lacks the membrane exons), as well as the large membrane isoform, the short membrane isoform, the long secreted isoform, the M2" membrane isoform, and the short secreted isoform. The fully assembled IgE products have differentially spliced frames, which can result in differing translated sequences due to shifting of the reading frame, even if the spliced messages overlap a large part of the exon sequence. IgE peptides generated by such different isoforms that differ in the amino acid sequences of membrane exons can provide additional therapeutic targets for cytotoxic T- lymphocytes.

FIG. 18 depicts regions of the modified GFP_UV vector used to express cytotoxic T- lymphocyte IgE peptides by a three-step PCR strategy. This GFPuv vector was modified by the additional of restriction enzyme sites into site 6 in loop 6 and site 7 in loop 7. FIG. 18A depicts site 6, which is located at Gin 157 in loop 6, and contains ligation sites for the restriction enzymes Bgl It, Kpn I and BssH I. FIG. 18B depicts site 7, which is located at Glu 172 in loop 7, and contains ligation sites for the restriction enzymes Nhe I, Xba I, and Pst I. Oligonucleotides specifying defined sequences were ligated into these sites. FIG 19 depicts restriction enzyme digests of two clones (clone number 6/4 and clone number 6/5) obtained by the above three-step PCR strategy. These digests verified that the modified vector contained all the required sites as designed. Correct size of fragments, as predicted from the vector depicted in FIG. 18A and FIG. 18B, was observed for both clones following digestion by either Sphl and Bgl II or by Sphl and BssH I. FIG. 20A depicts restriction digests of six clones (clones Ce3-BC, Ce3-DE, Ce3-FG,

Ce2-3, Ce4-dest, and Ce4-crea) containing oligonucleotides encoding IgE cytotoxic T- lymphocyte epitopes, cloned as inserts into site 6 of the GFP v vector. Lanes marked with "M" contain a DNA molecular weight ladder; lanes marked with "*" contain positive clones. The clones were expressed in JM109 cells and the cell lysates screened by Western blot, the results of which are depicted in FIG. 20B. Lane 1 contains lysates of the untransformed host E. coli strain JM109; lanes 2 and 12 contain lysates of cells transformed with GFPuv vector with the site 6 modification; lanes 3 and 4 contain lysates of cells transformed with the Ce3-BC clone; lanes 5 and 6 contain lysates of cells transformed with the Ce2-3 clone; lanes 7 and 8 contain lysates of cells transformed with the Ce4-crea clone; lanes 9 and 10 contain lysates of cells transformed with the Ce4-dest clone; lane 11 contains lysates of cells transformed with GFPuv vector; lane 13 contains lysates of cells transformed with the Ce3-FG clone; and lane 14 contains lysates of cells transformed with the Ce3-DE clone. FIG. 20C depicts the nucleotide sequence (underlined nucleotides) of the six IgE cytotoxic T-lymphocyte epitopes cloned into site 6: Ce3-BC (SEQ ID NO. 56), Ce3-DE (SEQ ID NO. 57), Ce3-FG (SEQ ID NO. 58), Ce2-3 (SEQ ID NO. 59), Ce4-dest (SEQ ID NO. 60), and Ce4-crea (SEQ ID NO. 61).

SUMMARY

FIG. 1 illustrates one aspect of the present invention, which enables one to identify IgE peptides that are naturally processed and presented by MHC-I molecules and can elicit cytotoxic T-lymphocytes that are able to lyse cells displaying such peptides. The present invention is believed to promote the afferent limb of the immune system by inducing a cytotoxic T lymphocyte response to naturally processed and presented IgE peptides, using any necessary and appropriate modes of T-lymphocyte co-stimulation. It is known that there are at least five differentially spliced IgE mRNAs, which encode different translated products in IgE-producing plasma cells or normal human peripheral blood lymphocytes (PBL) (Batista et al, 1995, J. Immunol, 154:209). FIG. 2 diagrammatically illustrates that IgE molecules with the correctly encoded sequences and proper polypeptide folding can be secreted or assembled on cell surface via the IgE membrane exons. Misfolded IgE heavy chains, or products encoded by alternatively spliced, incomplete messages, are processed or degraded via a proteolytic pathway in the endoplasmic reticulum (ER). Peptides may be further exported outside for processing by proteosomes and re- transported into the ER (Pamer & Cresswell, 1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). Homeostatic levels of these naturally processed IgE peptides are constantly produced and maintained. Along with peptides derived from other catabolized somatic proteins, naturally processed IgE peptides play a key role in the assembly and expression on the cell surface of MHC-I molecules by IgE-committed B-lymphocytes and plasma cells. Cells displaying such naturally processed IgE peptides, presented at the cell surface by MHC-I molecules, can become the targets for cytotoxic T-lymphocytes as an intrinsic physiological process of immunosurveillance. Since the IgE heavy chain does not vary isotypically in a given mammalian species, peptides naturally processed and presented from it can serve as a universal vaccine against IgE-mediated diseases in that mammalian species. Highly differentiated cell types, such as IgE-producing B-lymphocytes, display such IgE heavy chain peptides, and are lysed by these IgE-specific cytotoxic T-lymphocytes. Existing anti-IgE therapy using anti-IgE antibodies aims at neutralizing circulating IgE only after the IgE response has been induced, but is not capable of inhibiting IgE production. Small molecule therapeutics, such as anti-histamines or leukotriene antagonists, are aimed at events downstream of the IgE response, and are believed to block the effect of mediators of allergen-IgE activated mast cells and basophils. The present invention has distinct advantages over these two earlier therapeutic approaches by interrupting or diminishing IgE-mediated allergic inflammation at its inception, that is, at the level of IgE production. The present invention's immunoglobulin E-specific cytotoxic T-lymphocyte response decreases undesirably high levels of circulating IgE by directly targetting IgE production by lysing IgE-producing cells, which are the presumptive source of circulating IgE in serum and mucosal fluid. FIG. 2 illustrates three stage-specific IgE-producing targets of the IgE-specific CTL response: (1) pre-switched B-lymphocytes (Stage I B-lymphocytes), (2) IgE-switched B- lymphocytes (Stage π B-lymphocytes), and (3) IgE-producing plasma cells (Stage 111 B- lymphocytes). Pre-switched B-lymphocytes are mature cell surface IgM/IgD-positive B- lymphocytes expressing IgE heavy chain germ line transcripts under the influence of allergenic stimulations such as allergens and cytokines (Gauchat, 1990, J. Exp. Med., 172:463). They may express low levels of IgE heavy chain prior to gene rearrangement (Yaoita et al, 1982, Nature, 297:697). The transition from Stage I to Stage II, is preceded by switching from germ-line IgE transcript to differentially spliced IgE messages (Jung et al, 1993, Science, 259:984). Naturally processed IgE peptides are already present in this transition process. The present invention's IgE-specific cytotoxic T-lymphocyte response impedes the transition from Stage I to Stage II. Stage II B-lymphocytes exhibit a high density of surface IgE anchored to the cell membrane via IgE membrane exons 1 and 2 (Achatz et al, 1997, Science, 276:409), and are lysed by the present invention's IgE-specific cytotoxic T-lymphocytes. Following allergen and IL-4 stimulation, Stage II B-lymphocytes are transformed into Stage in, IgE-secreting plasma cells. The present invention's IgE-specific cytotoxic T-lymphocytes lyse allergen-activated Stage II B- lymphocytes in the process of transforming to Stage HI cells. Stage HI cells are IgE-secreting plasma cells. These cells expressed the highest levels of secretory IgE and thus the highest levels of naturally processed and presented IgE on MHC-I molecules. These are therefore the most conspicuous cellular targets for the present invention's IgE-specific cytotoxic T-lymphocytes. The present invention's immunoglobulin E-specific cytotoxic T-lymphocyte response also targets other cells that display naturally processed and presented IgE peptides, such as mast cells or basophils that have IgE bound to their surface via the FceRI receptor, and thus decreases the level of mast cells or basophils in blood and in allergy-affected tissues. FIG. 1 and FIG. 2 depict the exocytic pathway of IgE peptides processing and presentation on MHC-I molecules. IgE peptides recognized as epitopes by cytotoxic T-lymphocytes can also be generated by the endocytic pathway (Watts, 1997, Annu. Rev. Immunol, 15:821; Rock and Gamble, 1990, Science, 249:918). Mast cells are known to process exogenous antigens on MHC-I molecules (Malaviya et al, 1996, J. Immunol, 156:1490). Mast cells and basophils that bind IgE-allergen complexes via the FceRI receptor can also display naturally processed and presented IgE peptides on MHC-I molecules, and are thus susceptible to lysis by the present invention's IgE-specific cytotoxic T-lymphocytes.

The present invention concerns, among other things, novel approaches to the therapy or prophylaxis of immunoglobulin E-mediated disease conditions. The present invention provides a method for identifying peptides that induce a cytotoxic T-lymphocyte response against immunoglobulin E.

The present invention also provides compositions that include at least one isolated immunogenic peptide identified by a method of the present invention, such as those able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. The present invention further provides compositions that include at least one isolated polynucleotide encoding the sequence for an immunogenic peptide identified by a method of the present invention, such as those able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. The present invention further provides compositions that include antigen-presenting cells that recognize at least one isolated immunogenic peptide identified by a method of the present invention, such as those able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides.

The present invention also provides methods for modulating an immunoglobulin E- mediated condition in a mammal, including providing to the mammal a composition that elicits in the mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, in an amount sufficient to elicit a cytotoxic T-lymphocyte response in the mammal. These compositions can include one or more peptides, nucleic acid molecules, and cells, and combinations thereof, optionally with one or more co-stimulatory factors.

DETAILED DESCRIPTION OF THE INVENTION

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Generally, the nomenclature used herein and the manufacture or laboratory procedures described below are well known and commonly employed in the art. Conventional methods are used for these procedures, such as those provided in the art and various general references. Where a term is provided in the singular, the inventors also contemplate the plural of that term. The nomenclature used herein and the laboratory procedures described below are those well known and commonly employed in the art. Where there are discrepancies in terms and definitions used in references that are incorporated by reference, the terms used in this application shall have the definitions given herein. Other technical terms used herein have their ordinary meaning in the art that they are used, as exemplified by a variety of technical dictionaries (Delves and Roitt, "Encyclopedia of Immunology'', 2nd ed., Academic Press, New York, NY, 1998; Rosen et al, "Dictionary of Immunology", Stockton Press, New York, NY, 1989). The inventors do not intend to be limited to a mechanism or mode of action. Reference thereto is provided for illustrative purposes only.

I. METHODS FOR IDENTIFYING PEPTIDES

The present invention provides methods for identifying peptides that induce a cytotoxic T-lymphocyte response against immunoglobulin E. The peptides identified by such methods are useful in the production of active vaccines for therapy or prophylaxis in a mammal of immunoglobulin E-mediated conditions, including, for example, atopic hypersensitivity conditions and non-atopic hypersensitivity conditions. The method can include the steps of: a) providing at least one test peptide suspected of being able to bind to at least one major histocompatibility complex class I molecule, and b) evaluating the at least one test peptide for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. Preferably, a peptide that induces such a response is identified.

TEST PEPTIDE

The test peptide can be provided by any suitable means or method known in the art. The test peptide can be naturally occurring, synthesized, combinatorially synthesized, or biologically or recombinantly produced, or an active fragment or active fragments thereof. An active fragment is a fragment of a peptide that retains at least one activity of a parent peptide such as but not limited to the ability to bind to a major histocompatibility class I molecule or the ability to elicit in a mammal a cytotoxic T-lymphocyte response, preferably the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed immunoglobulin E peptides that are presented on a major histocompatibility class I molecule. Such active fragments are part of the present invention. A naturally occurring peptide may be isolated from any natural source, such as for example from whole blood, plasma, serum, or bone marrow. The test peptide can be made from digests of such naturally occurring peptides or other fragments such as active fragments thereof. The test peptide can be synthesized, for example chemically synthesized by conventional peptide synthesis. The test peptide can be synthesized by combinatorial synthesis, which may be based on conventional organic chemical synthesis, or on biochemical synthesis (using for example enzymes to catalyze reactions), or on both. The test peptide can also be biologically produced, for example by recombinant expression of the test peptide using a vector in a host cell by means known in the art (Sambrook et al, Molecular Biology: A Laboratory Manual (Third Edition) Cold Spring Harbor Press, Cold Spring Harbor, New York, (2001)), or by use of directed evolution methods such as yeast two-hybrid systems, protein fragment complementation assay, phage display, ribosome display, yeast surface display, and bacterial surface display techniques as described for example by Mδssner and Plϋckthun (Mδssner and Plϋckthun, 2001, Chimia, 55:324). Test peptides can contain amino acids that commonly occur in nature, that is to say those such as alanine, arginine, aspartic acid, asparagine, cysteine, glutamic acid, glutamine, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, and valine. Test peptides can contain amino acids that rarely occur or do not occur in nature, such as but not limited to 2- aminoadipic acid, 3-aminoadipic acid, beta-alanine (beta-aminopropionic acid), 2-aminobutyric acid, 4-aminobutyric acid (piperidinic acid), 6-aminocaproic acid, 2-aminoheptanoic acid, 2- aminoisobutyric acid, 3-arninoisobutyric acid, 2-aminopimelic acid, 2,4-diaminobutyric acid, desmosine, 2,2'-diaminopimelic acid, 2,3-diaminopropionic acid, N-ethylglycine, N- ethylasparagine, hydroxylysine, //o-hydroxylysine, 3-hydroxyproline, 4-hydroxyproline, isodesmosine, a;Z/o-isoleucine, N-methylglycine (sarcosine), N-methylisoleucine, N-methylvaline, norvaline, norleucine and ornithine. Test peptides can contain all L-amino acids, all D-amino acids, or a mixture of L- and D-amino acids. Test peptides can be additionally modified, for example by acylating the amino terminus or amidating the carboxy terminus to increase stability or half-life (Lipton, U. S. Patent 5,028,592). Test peptides can be optionally labelled with a detectable moiety. Any suitable detectable moiety that renders the test peptide detectable can be used. Examples of detectable moieties include but are not limited to radioactive isotopes, non- radioactive isotopes, biotin, antigens detectable by specific antibodies, enzymes, particles (such as metal, polymer, latex, or magnetic particles), chromophores, fluorophores, or dyes. Such detectable moieties can be directly attached to the test peptide, or attached through a linker, or attached by non-covalent methods as are known in the art.

The test peptide can be a peptide of between about five to about two thousand amino acids in length, or a peptide of between about five to about one thousand four hundred amino acids in length, or a peptide of between about five to about one thousand amino acids in length, or a peptide of between about five to about five hundred amino acids in length, or an oligopeptide of between about five to about forty amino acids in length, or an oligopeptide of between about five to about seventeen amino acids in length. Shorter peptides (oligopeptides), such as those of between about five to about forty amino acids in length, or between about five and about twenty amino acids in length, in particular those of six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, or seventeen amino acids in length, or between about five and about ten amino acids in length, can be more suitable to binding assays since the MHC-I binding site usually binds a peptide of between eight to ten amino acids in length (Pamer and Cresswell, 1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). However, the addition of amino acids to the N-terminus or to the C-terminus of such an oligopeptide can be advantageous, for example, by increasing the peptide' s stability and half-life in vitro or in circulation in vivo. Longer peptides can be more useful as vaccines. Test peptides can be linear or non-linear (for example, circular or branched).

The test peptide can contain an amino acid sequence derived from immunoglobulin E, for example, a sequence derived from the heavy chain of immunoglobulin E. For developing a vaccine against IgE for a particular species such as humans, the test peptide preferably contains an amino acid sequence derived from any of the constant region domains (CHI, CH2, CH3, or CH4 domains), or from the membrane exons of that species' IgE heavy chain, such as human IgE heavy chain. Peptides identified as useful for one species can be used as guidance for identifying peptides useful for another species. For example, peptides identified in a mouse can be used to identify peptides as presumptively useful in humans. Identification of such presumptive peptides can be guided by, for example, predictive algorithms, such as those described herein or known in the art. The usefulness of such peptides can be confirmed by screening in vitro, in vivo, or ex vivo, or combinations thereof, using methods described herein. Such confirmed peptides can be further modified in sequence or structure and further screened for desirable activity. Such modification of sequence or structure can be guided by, for example, predictive algorithms, such as those described herein or known in the art.

MAMMAL

Mammals that can be used in the identification of the test peptide, or in which the test peptide may be used for prophylaxis or therapy of an IgE-mediated condition, can be any mammal with an immunoglobulin E and a major histocompatibility complex class I molecule of interest. Particularly useful in this regard are mammals of economic importance such as those used in food production (including cattle, swine, sheep, and goats), or for mammals of domestic importance such as those that are common companion animals or pets (including dogs, cats, horses, ferrets, mice, rats, rabbits, guinea pigs, gerbils, and hamsters). Also particularly useful are mammals that are non-human primates, such as lemurs, lorises, tarsiers, monkeys, and apes. Humans are included as mammals. Humans are also particularly useful for identification of the test peptide where the test peptide is to be used for prophylaxis or therapy of an IgE-mediated condition specific to humans. MAJOR HISTOCOMPATIBILITY COMPLEX CLASS I MOLECULE The major histocompatibility complex class I (MHC-I) molecule to which the test peptide is suspected to bind can be any MHC-I molecule, including a mammalian MHC-I molecule, or an MHC-I molecule specific to humans such as a human leukocyte antigen (HLA). Preferably, the human leukocyte antigen is one that corresponds to a specific HLA haplotype, such as a particular HLA- A, HLA-B, HLA-C, and HLA-Cw serotype, or more specifically, a particular HLA genotype where more than one HLA genotype exists for a HLA serotype. Lists of known HLA serotypes and genotypes are publicly available, for example on the Internet website hiv.basic.nwu.edu HLA/Reports/DoSeraGenoList.cfm. The binding of the test peptide to the major histocompatibility complex class I molecule can optionally be predicted by an algorithm when the binding motif for that particular MHC-I molecule is known (Schueler-Furman et al, 2000, Protein Sci., 9:1838; Andersen et al, 2000, Tissue Antigens, 55:519; Gulukota et α/., 1997, J. Mol. Biol, 267:1258; Brusic et α/., 2002, Immunol. Cell Biol, 80:280; Dzuris et al, 2000, J Immunol, 164:283; Rammensee et α/., 1999, Immunogenetics, 50:213; Chelvanayagam, 1996, Immunogenetics, 45:15; Stryhn et al, 1996, Eur. J. Immunol, 26:1911; Hammer, 1995, Curr .Opin. Immunol, 7:263; Falk et al, 1993, Semin. Immunol, 5:81). For example, about 180 binding motifs are known to date for the human MHC-I molecules HLA- A, HLA-B, HLA-C, and HLA-Cw. These HLA binding motifs are publicly available, for example on the Internet website hiv.basic.nwu.edu/HLA/Reports/DoMotifList.cfm. The algorithms are also publicly available, for example on the Internet websites bimas.dcrt.nih.gov/molbio/hla_bind/index.html, hiv.basic.nwu.edu/HLA/MotifScanner.cfin, and www.uni-tuebingen.de/uni/kxi/. A similar analysis can be performed for any MHC-I molecule for which the binding motif and algorithm is known. Other algorithms known in the art can also be used for this purpose.

EVALUATION OF TEST PEPTIDE— A

The methods of the present invention identify those test peptides that can be useful in the production of active vaccines for therapy or prophylaxis in a mammal of immunoglobulin E- mediated conditions. Test peptides are evaluated for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. A variety of approaches can be used to evaluate a test peptide for this IgE-specific, CTL-inducing ability, examples of which follow. One approach to evaluate a test peptide for the ability to elicit in a mammal a cytotoxic T- lymphocyte response to naturally processed and presented immunoglobulin E peptides includes the steps of: a) contacting the test peptides with mammalian cells defective in the expression of a chosen cell surface MHC-I molecule, and b) selecting from the test peptides those that cause expression of that MHC-I molecule on the cell surface of the defective cells. The defect in the mammalian cells can be for example a mutation of the transporter associated with antigen processing (TAP) proteins such as but not limited to TAPl or TAP2.

Another approach to evaluate a test peptide for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides includes the steps of: a) contacting the test peptides with a chosen MHC-I molecule, and b) selecting from the test peptides those that bind to the MHC-I molecule. In practice the MHC-I molecule to be used in such a direct binding assay can be an isolated "MHC-I Moiety". This MHC-I Moiety can be for example a native protein (for example, an isolated intact MHC-I molecule), a complexed protein (for example, an MHC-I molecule complexed with beta-2 microglobulin, or an MHC-I molecule that is non-covalently bound to avidin), or an engineered protein (for example, a chimeric protein consisting in part of an MHC-I molecule), or active fragment thereof, so long as the MHC-I binding site is present and available for binding by the test peptide. Alternatively, the MHC-I molecule can be expressed on a cell surface. The test peptides can be tested for the ability to bind to such MHC-I expressing cells in vivo, ex vivo, or in vitro, or combinations thereof, using methods described herein or as known in the art. The MHC-I molecule expressed by these cells can be endogenous or exogenous (for example, as the result of transfecting the cell with the gene for a foreign MHC-I molecule). The ability of a test peptide to bind to a MHC-I molecule can be tested by any suitable technique (see for example Phizicky and Fields, 1995, Microbiol Rev., 59:94). For example, the equilibrium dissociation constant K_& can be calculated from measuring labelled test peptide that directly binds to the MHC-I molecule. As one alternative, unlabelled test peptide can be competed with the corresponding labelled test peptide for binding to the MHC-I molecule, and the binding affinity of each test peptide to the MHC-I molecule can be calculated from the inhibition concentrations (IC₅0). As another alternative, the on rate and off rate of unlabelled test peptide binding to the MHC-I molecule can be measured, for example by surface plasmon resonance (SPR) techniques.

A further approach to evaluate a test peptide for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides includes the steps of: a) contacting the test peptides with mammalian cells defective in the expression of a chosen cell surface MHC-I molecule, b) selecting from the test peptides those that cause expression of that MHC-I molecule on the cell surface of the defective cells, c) preparing cytotoxic T-lymphocytes that recognize the selected peptide(s), and d) testing these cytotoxic T-lymphocytes for the ability to lyse target cells. The defect in the mammalian cells may be for example a mutation of the transporter associated with antigen processing (TAP) proteins such as but not limited to TAP1 or TAP2.

Preparation of Cytotoxic T-Lymphocytes Cytotoxic T-lymphocytes (CTL) can be prepared by any of a number of ways. Illustrative methods are described herein, although alternative methods can be used. For example, cytotoxic T-lymphocytes can be prepared by immunizing a mammal that expresses the chosen cell surface MHC-I molecule with a selected peptide, and harvesting cytotoxic T-lymphocytes from the immunized mammal. The mammal can express the MHC-I molecule from an endogenous gene or from an exogenous gene. The immunization can include, in addition to the selected peptide, at least one co-stimulatory factor (Frauwirth and Thompson, 2002, J. Clin. Invest, 109:295). Suitable co-stimulatory factors include, for example, cytokines, mitogens, antibodies, antigen- presenting cells (Mayordomo et al, 1997, Stem Cells, 15:94), and peptides derived from a helper T-lymphocyte epitope foreign to the immunized mammal. Co-stimulatory factors may be delivered together with the peptide used for immunization, for example as a fusion with the peptide, or separately, for example, as a peptide encoded by a nucleic acid molecule. Cytotoxic T-lymphocytes can also be prepared by immunizing peripheral blood lymphocytes with a selected peptide, and harvesting cytotoxic T-lymphocytes from the immunized peripheral blood lymphocytes. As is the case for immunizing a mammal, the immunization of peripheral blood lymphocytes may include, in addition to the selected peptide, at least one co-stimulatory factor as described above, such as peptides derived from a helper T-lymphocyte epitope foreign to the immunized peripheral blood lymphocytes. Co-stimulatory factors may be delivered together with the peptide used for immunization, for example as a fusion with the peptide, or separately, for example, as a peptide encoded by a nucleic acid molecule.

Target Cells The cytotoxic T-lymphocytes prepared by the above methods can be tested for their ability to lyse target cells. Target cells are any cell type that is recognized by the cytotoxic T- Iymphocytes as having an immunoglobulin E molecule or fragment of an immunoglobulin E molecule, such as an IgE peptide that is naturally processed and presented by an MHC-I molecule on the cell's surface. The target cells and the cytotoxic T-lymphocytes can each be derived from the same mammalian species, or can each be derived from different mammalian species. One example of suitable target cells are mammalian cells defective in the expression of the chosen cell surface MHC-I molecule, where the cells have been contacted with the selected peptide or peptides, and where the selected peptide(s) cause(s) the MHC-I molecule to be expressed on the surface of the cells. The defect in the mammalian cells can be, for example, a mutation of the transporter associated with antigen processing (TAP) proteins TAPl or TAP2. Another example of suitable target cells are mammalian cells that express the chosen cell surface MHC-I molecule, where the cells have been contacted with the selected peptide or peptides, and where the selected peptide(s) bind to the cell surface MHC-I molecule. The MHC-I molecule expressed by these cells can be expressed from an endogenous gene or from an exogenous gene. Another example of suitable target cells are immunoglobulin E-bearing cells, such as transitional (Stage I to Stage H) B-lymphocytes and Stage JJ B-lymphocytes. Another example of suitable target cells are immunoglobulin E-producing cells, such as IgE-secreting plasma cells or IgE-secreting tumor cells (for example, IgE myeloma cells). Particularly useful in developing a vaccine against IgE for humans are target cells that produce human immunoglobulin E. Another example of suitable target cells are cells that have immunoglobulin E bound to their surface, such as IgE-activated mast cell or basophils.

EVALUATION OF TEST PEPTIDE— B

Yet another approach to evaluate a test peptide for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides utilizes steps that resemble procedures outlined above, but includes the additional element of establishing a binding motif for the chosen MHC-I molecule. This approach includes the steps of: a) identifying and selecting test peptides that bind to the chosen MHC-I molecule, b) determining the amino acid sequence of the selected peptides, c) establishing an amino acid binding motif for the chosen MHC-I molecule from these determined amino acid sequences, d) providing second generation test peptides that have amino acid sequences derived from immunoglobulin E and that are predicted by an algorithm to bind to the established amino acid binding motif, and e) testing the second generation test peptides for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, wherein a peptide that induces such a response is identified.

Identifying and Selecting Step a) of the above approach can use any suitable process to identify and select test peptides that bind to the chosen MCH-I molecule. Such processes are described herein, although other approaches can be used. One example of a suitable process to identify and select test peptides that bind to the chosen MCH-I molecule includes the steps of: a) contacting the test peptides with mammalian cells defective in the cell surface expression of the chosen MHC-I molecule, and b) selecting from the test peptides those that causes expression of the MHC-I molecule at the surface of the cells. The defect in the mammalian cells can be for example a mutation of the transporter associated with antigen processing (TAP) proteins such as but not limited to TAPl or TAP2.

Another example of a suitable process to identify and select test peptides that bind to the chosen MCH-I molecule includes the steps of: a) contacting the test peptides with a chosen MHC-I molecule, and b) selecting from the test peptides those that bind to the MHC-I molecule. The MHC-I molecule to be used in such a direct binding assay can be any isolated MHC-I Moiety as described above, or active fragment thereof, so long as the MHC-I binding site is present and available for binding by the test peptide. Alternatively, the MHC-I molecule can be expressed on a cell surface from an exogenous or an endogenous gene as described above. The test peptides can be tested for the ability to bind to such MHC-I expressing cells in vivo, ex vivo, or in vitro. The ability of a test peptide to bind to a MHC-I molecule can be tested as described herein or by any suitable technique such as those known in the art.

Another example of a suitable process to identify and select test peptides that bind to the chosen MCH-I molecule includes the steps of: a) contacting a display scaffold that displays the test peptide with mammalian cells defective in the cell surface expression of the chosen MHC-I molecule, and b) selecting from these test peptides those that cause expression of the MHC-I molecule at the surface of the cells. The display scaffold that displays the test peptide can be any suitable supporting macromolecule or macromolecular assemblage, such as a polypeptide (Kamb, et al, U. S. Patent 6,025,485; Christmann et al, 1999, Protein Eng, 12:797; Abedi et al, 1998, Nucleic Acids Res., 26:623; Peelle et al, 2001, J. Protein Chem., 20:507), a phage (He, 1999, J. Immunol. Methods, 231:105; Smith, 1985, Science, 228:1315), a ribosome (Schaffϊtzel et al, 1999, J. Immunol. Methods, 231:119; Roberts, 1999, Curr. Opin. Chem. Biol, 3:268), an mRNA (Wilson et al, 2001, Proc. Natl. Acad. Sci., 98:3750), or a yeast cell surface (Yeung and Wittrup, 2002, Biotechnol. Prog., 18:212; Shusta et al, 1999, J. Mol. Biol, 292:949), a bacterial cell surface (Leenhouts et al, 1999, Antonie Van Leeuwenhoek, 76:367; Christmann et al, 2001, J. Immunol. Methods, 257:163), or a bacterial spore surface (Wittrup, 2001, Curr. Opin. Biotechnol, 12:395; Boder and Wittrup, 1998, Biotechnol Prog., 14:55). For example, test peptides can be displayed as inserts within exposed loops of a peptide such as a conformationally constrained polypeptide such as green fluorescent protein (Abedi et al, 1998, Nucleic Acids Res., 26:623), displayed as inserts within the phage coat proteins PHI or PNJJI of a filamentous bacteriophage, such as M13, fd, or fl phage (Smith, 1985, Science, 228:1315), displayed as an mRΝA-peptide fusion (Wilson et al, 2001, Proc. Natl. Acad. Sci, 98:3750), or displayed as a heterologous or chimeric peptide expressed on the surface of bacterial cells (Gunneriusson et al, 1999, Appl. Env. Microbiol, 65:4134). The defect in the mammalian cells can be for example a mutation of the transporter associated with antigen processing (TAP) proteins TAPl or TAP2.

Another example of a suitable process to identify and select test peptides that bind to the chosen MCH-I molecule includes the steps of: a) contacting a display scaffold that displays the test peptide with the chosen MHC-I molecule, and b) selecting from these test peptides those that bind to the MCH-I molecule. Any suitable display scaffold as described above can be used to display the test peptides. The MHC-I molecule to be used in such a direct binding assay can be any isolated MHC-I Moiety as described above, or active fragment thereof, so long as the MHC-I binding site is present and available for binding by the test peptide. Alternatively, the MHC-I molecule can be expressed on a cell surface from an exogenous or an endogenous gene as described above. The test peptides can be tested for the ability to bind to such MHC-I expressing cells in vivo, ex vivo, or in vitro. The ability of a test peptide to bind to a MHC-I molecule can be tested as described herein or by any suitable technique such as those known in the art.

Testing

Step-e) of the above approach can use any suitable method to test the second generation test peptides for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, wherein a peptide that induces such a response is identified. Such methods are described herein, although other approaches can be used.

One suitable method of testing the second generation test peptides includes the steps of: a) preparing cytotoxic T-lymphocytes that recognize a second generation test peptide, and b) testing these cytotoxic T-lymphocytes for their ability to lyse target cells. Cytotoxic T- lymphocytes that recognize a second generation test peptide can be prepared by any of a number of ways. Cytotoxic T-lymphocytes can be prepared by immunizing a mammal that expresses the chosen cell surface MHC-I molecule with a selected second generation test peptide, and harvesting cytotoxic T-lymphocytes from the immunized mammal. The mammal can express the MHC-I molecule from an endogenous gene or from an exogenous gene. The immunization can include, in addition to the selected second generation test peptide, at least one co-stimulatory factor (Frauwirth and Thompson, 2002, J. Clin. Invest, 109:295). Suitable co-stimulatory factors include, for example, cytokines, mitogens, antibodies, antigen-presenting cells (Mayordomo et al, 1997, Stem Cells, 15:94), and peptides derived from a helper T-lymphocyte epitope foreign to the immunized mammal. Co-stimulatory factors may be delivered together with the second generation peptide used for immunization, for example as a fusion with the second generation peptide, or separately, for example, as a peptide encoded by a nucleic acid molecule. Cytotoxic T-lymphocytes that recognize a second generation test peptide can also be prepared by immunizing peripheral blood lymphocytes with a selected second generation test peptide and harvesting cytotoxic T-lymphocytes from the immunized peripheral blood lymphocytes. As is the case for immunizing a mammal, the immunization of peripheral blood lymphocytes can include, in addition to the selected second generation test peptide, at least one co-stimulatory factor as described above, such as peptides derived from a helper T-lymphocyte epitope foreign to the immunized peripheral blood lymphocytes. Co-stimulatory factors may be delivered together with the second generation peptide used for immunization, for example as a fusion with the second generation peptide, or separately, for example, as a peptide encoded by a nucleic acid molecule.

The cytotoxic T-lymphocytes that recognize a second generation test peptide are tested for their ability to lyse target cells. Target cells are any cell type that is recognized by the cytotoxic T-lymphocytes as having an immunoglobulin E molecule or fragment of an immunoglobulin E molecule, as described above. One example of suitable target cells are mammalian cells defective in the expression of the chosen cell surface MHC-I molecule, where the cells have been contacted with the selected second generation test peptide or peptides, and where the selected second generation test peptide(s) cause(s) the MHC-I molecule to be expressed on the surface of the cells. The defect in the mammalian cells can be for example a mutation of the transporter associated with antigen processing (TAP) proteins such as but not limited to TAPl or TAP2. Another example of suitable target cells are mammalian cells that express the chosen cell surface MHC-I molecule, where the cells have been contacted with the selected second generation test peptide or peptides, and where the selected peptide(s) bind to the cell surface MHC-I molecule. The MHC-I molecule expressed by these cells can be expressed from an endogenous gene or from an exogenous gene. Other examples of suitable target cells are immunoglobulin E-bearing cells, immunoglobulin E-producing cells (such as IgE-secreting plasma cells or IgE-secreting tumor cells), and cells that have immunoglobulin E bound to their surface (such as IgE-activated mast cell or basophils). Particularly useful in developing a vaccine against IgE for humans are target cells that produce human immunoglobulin E.

Another suitable method of testing the second generation test peptides includes the steps of: a) contacting the second generation test peptide the chosen MHC-I molecule, and b) selecting from these second generation test peptides those that bind to the chosen MHC-I molecule. The MHC-I molecule to be used in such a direct binding assay can be any isolated MHC-I Moiety as described above, or active fragment thereof, so long as the MHC-I binding site is present and available for binding by the second generation test peptide. Alternatively, the MHC-I molecule can be expressed on a cell surface from an exogenous or an endogenous gene as described above. The second generation test peptides can be tested for the ability to bind to such MHC-I expressing cells in vivo, ex vivo, or in vitro. The ability of a second generation test peptides to bind to a MHC-I molecule can be tested as described herein or by any suitable technique such as those known in the art. II. COMPOSITIONS ELICITING IgE-SPECIFIC CYTOTOXIC T-LYMPHOCYTES

IMMUNOGENIC PEPTIDES

The present invention also provides compositions that include at least one isolated immunogenic peptide identified by a method of the present invention as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T- lymphocyte response to naturally processed and presented immunoglobulin E peptides. Such compositions can be provided in pharmaceutically acceptable carriers, and can be provided to be administered using an appropriate dose, route of administration, and regime, to provide an appropriate end-point, such as prophylaxis, palliation, or amelioration of symptoms of immunoglobulin E-mediated disease conditions. Such dose, route of administration, regime, and end-point can be determined using methods known in the art. Appropriate toxicity and efficacy evaluation can also be evaluated using methods known in the art. Appropriate in vitro, in vivo, or ex vivo methods, or combinations thereof, can be utilized.

Pharmaceutically Acceptable Carriers Compositions of the present invention can include, in addition to the immunogenic peptide or peptides, a pharmaceutically acceptable carrier. Suitable pharmaceutically acceptable carriers include diluents, excipients, or carrier materials, selected according to the intended form of administration and consistent with conventional pharmaceutical practice. Examples of suitable carriers include water, physiological saline, phosphate-buffered saline, saline buffered with a physiologically compatible salt, a water-in-oil emulsion, and an oil-in-water emulsion. The pharmaceutically acceptable carrier can also include appropriate stabilizers, disintegrating agents, binders, preservatives, flavoring agents, or coloring agents, as is consistent with conventional pharmaceutical practice ("Remington: The Science and Practice of Pharmacy", 20^th edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000; Miller et al, U. S. Patent 6,355,619; Adams et al, U. S. Patent 6,342,220). The immunogenic peptide or peptides of the present invention can be formulated and used as tablets, capsules or elixirs for oral administration, suppositories for rectal administration, ointments, creams, or lotions for topical administration, sprays, aerosols, or inhalants, sterile solutions, suspensions or injectables, and the like. Injectables can be prepared in conventional forms either as liquid solutions or suspensions, as solid forms suitable for solution or suspension in liquid prior to injection, or as emulsions. Suitable excipients include, for example, water, saline, dextrose, mannitol, lactose, lecithin, albumin, sodium glutamate, cysteine hydrochloride, and the like. In addition, if desired, the injectable pharmaceutical compositions can contain minor amounts of nontoxic auxiliary substances, such as wetting agents, pH buffering agents and the like. If desired, an absorption enhancing preparation, such as liposomes, can be used.

Co-stimulatory Factors The compositions can include, in addition to the immunogenic peptide or peptides, at least one co-stimulatory factor (Frauwirth and Thompson, 2002, J. Clin. Invest, 109:295). Suitable co-stimulatory factors include, for example, such molecules as B7 and CD40, cytokines, mitogens, antibodies, antigen-presenting cells (Carreno and Collins, 2002, Annu. Rev. Immunol, 20:29; Mayordomo et al, 1997, Stem Cells, 15:94), and peptides derived from a helper T- lymphocyte epitope foreign to the immunized mammal. Co-stimulatory factors can be delivered together with the immunogenic peptide used for immunization, for example as a fusion with the peptide, or separately, for example, as a peptide encoded by a nucleic acid molecule. In the case where the co-stimulatory factor is a peptide, the co-stimulatory peptide and the immunogenic peptide can be provided together on one or more "polypeptide framework" (Skerra, 2000, J. Mol. Recognit, 13:167). Suitable polypeptide frameworks preferably contain at least one exposed loop region that is exposed to the hydrophilic environment, preferably is thermodynamically stable, and preferably can accept insertion of the immunogenic peptide or of the co-stimulatory peptide. Examples of such polypeptide frameworks include green fluorescent protein (GFP), staphylococcal nuclease, fibronectin, immunoglobulin (lg), and heat shock protein (HSP). Immunogenic peptides The immunogenic peptide or peptides used in the compositions of the present invention can be naturally occurring, synthesized, combinatorially synthesized, or biologically or recombinantly produced, or an active fragment or active fragments thereof, as described herein. An active fragment is a fragment of a peptide that retains at least one activity of a parent peptide such as but not limited to the ability to bind to a major histocompatibility class I molecule or the ability to elicit in a mammal a cytotoxic T-lymphocyte response, preferably the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed immunoglobulin E peptides that are presented on a major histocompatibility class I molecule. A naturally occurring immunogenic peptide may be isolated from any natural source, such as for example from whole blood, plasma, serum, or bone marrow. The immunogenic peptide can be made from digests of such naturally occurring peptides or other fragments such as active fragments thereof. A naturally occurring immunogenic peptide can be isolated from any natural source, such as for example from whole blood, plasma, serum, or bone marrow. The immunogenic peptide can be synthesized, for example chemically synthesized by conventional peptide synthesis. The immunogenic peptide can be synthesized by combinatorial synthesis, which may be based on conventional organic chemical synthesis, or on biochemical synthesis (using for example enzymes to catalyze reactions), or on both. The immunogenic peptide can also be biologically produced, for example by recombinant expression of the immunogenic peptide using a vector in a host cell by means known in the art. Immunogenic peptides can contain amino acids that commonly occur in nature, amino acids that rarely occur or do not occur in nature, or a mixture. Immunogenic peptides can contain all L-amino acids, all D-amino acids, or a mixture of L- and D- amino acids. Immunogenic peptides can be additionally modified as described herein.

The immunogenic peptide can be a peptide of between about five to about two thousand amino acids in length, or a peptide of between about five to about one thousand four hundred amino acids in length, or a peptide of between about five to about one thousand amino acids in length, or a peptide of between about five to about five hundred amino acids in length, or an oligopeptide of between about five to about forty amino acids in length, or an oligopeptide of between about five to about seventeen amino acids in length. Shorter peptides (oligopeptides), such as those of between about five to about forty amino acids in length, or between about five and about twenty amino acids in length, in particular those of six, seven, eight, nine, ten, eleven, • twelve, thirteen, fourteen, fifteen, sixteen, or seventeen amino acids in length, or between about five and about ten amino acids in length, can be more suitable to binding assays since the MHC-I binding site usually binds a peptide of between eight to ten amino acids in length (Pamer and Cresswell, 1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). However, the addition of amino acids to the N-terminus or to the C-terminus of such an oligopeptide can be advantageous, for example, by increasing the peptide's stability and half-life in vitro or in circulation in vivo. The length of the immunogenic peptide takes into consideration the overall in vivo immugenicity of the peptide, which can be influenced by the peptide's stability and half-life in vivo. Thus, while short oligopeptides can bind more readily to a major histocompatibility complex class I molecule in in vitro binding assays, longer immunogenic peptides can be more useful as vaccines. Longer immunogenic peptides can be cleaved in vivo to yield the actively immunogenic sequence. Immunogenic peptides can be linear or non-linear (for example, circular or branched), or can be inserted into a suitable polypeptide framework as described herein. Immunogenic peptides inserted into a polypeptide framework, such as an immunoglobulin, can be processed through the endogenous degradative pathway and are presented to T-lymphocytes in the context of MHC class I molecules (Billetta et al, 1995, Eur. J. Immunol, 25:776).

The immunogenic peptide can contain an amino acid sequence derived from immunoglobulin E, for example, a sequence derived from the heavy chain of immunoglobulin E. For developing a vaccine against IgE for a particular species such as humans, the immunogenic peptide preferably contains an amino acid sequence derived from any of the constant region domains (CHI, CH2, CH3, or CH4 domains), or from the membrane exons of that species' IgE heavy chain, such as human IgE heavy chain.

The immunogenic peptide or peptides of the composition are preferably able to bind to a major histocompatibility complex class I molecule, which can be any MHC-I molecule, including a mammalian MHC-I molecule. For use as a vaccine against IgE in humans, the immunogenic peptide preferably binds to an MHC-I molecule specific to humans such as a human leukocyte antigen (HLA). Preferably, the human leukocyte antigen is one that corresponds to a specific HLA haplotype, such as a particular HLA- A, HLA-B, HLA-C, and HLA-Cw serotype, or more specifically, a particular HLA genotype where more than one HLA genotype exists for a HLA serotype. For use as a vaccine, the immunogenic peptide can be delivered as a complexed peptide, for example, as an immunogenic peptide complexed with an MHC-I Moiety such as is describe herein, whereby the immunogenic peptide is protected from proteases and thus exhibits a desirable gain in half-life. For cell- or tissue-specific delivery, the immunogenic peptide or complexed immunogenic peptide can optionally be covalently or non-covalently bound to a targetting or "homing" molecule, that is to say, a molecule, such as an integrin, that targets or "homes to" a particular type of tissue or cell, such as an antigen-presenting cell.

Formulations The compositions containing the immunogenic peptide or peptides can be provided in any formulation suitable to the intended form of administration and consistent with conventional pharmaceutical practice ("Remington: The Science and Practice of Pharmacy", 20^th edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000). Examples of suitable formulations include a solid, a powder, a gel, a solution, a suspension, an emulsion, liposomes, microspheres, injectable particles, inhalable particles, or a dissolvable matrix (Miller et al, U. S. Patent 6,355,619).

POLYNUCLEOTIDES

The present invention further provides compositions that include at least one isolated polynucleotide encoding the sequence for an immunogenic peptide identified by a method of the present invention as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. Such compositions can be provided in pharmaceutically acceptable carriers, and can be provided to be administered using an appropriate dose, route of administration, and regime, to provide an appropriate end-point, such as prophylaxis, palliation, or amelioration of symptoms of immunoglobulin E-mediated disease conditions. Such dose, route of administration, regime, and end-point can be determined using methods known in the art. Appropriate toxicity and efficacy evaluation can also be evaluated using methods known in the art. Appropriate in vitro, in vivo, or ex vivo methods, or combinations thereof, can be utilized.

Pharmaceutically Acceptable Carriers Compositions of the present invention can include, in addition to the isolated polynucleotide encoding the sequence for an immunogenic peptide, a pharmaceutically acceptable carrier. Suitable pharmaceutically acceptable carriers include diluents, excipients, or carrier materials, selected according to the intended form of administration and consistent with conventional pharmaceutical practice. Examples of suitable carriers include water, physiological saline, phosphate-buffered saline, saline buffered with a physiologically compatible salt, a water- in-oil emulsion, and an oil-in- water emulsion. The pharmaceutically acceptable carrier can also include appropriate stabilizers, disintegrating agents, binders, preservatives, flavoring agents, or coloring agents, as is consistent with conventional pharmaceutical practice ("Remington: The Science and Practice of Pharmacy", 20^l edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000). The isolated polynucleotide encoding the sequence for an immunogenic peptide of the present invention can be formulated and used as tablets, capsules or elixirs for oral administration, suppositories for rectal administration, ointments, creams, or lotions for topical administration, sprays, aerosols, or inhalants, sterile solutions, suspensions or injectables, and the like. Injectables can be prepared in conventional forms either as liquid solutions or suspensions, as solid forms suitable for solution or suspension in liquid prior to injection, or as emulsions. Suitable excipients include, for example, water, saline, dextrose, mannitol, lactose, lecithin, albumin, sodium glutamate, cysteine hydrochloride, and the like. In addition, if desired, the injectable pharmaceutical compositions can contain minor amounts of nontoxic auxiliary substances, such as wetting agents, pH buffering agents and the like. If desired, an absorption enhancing preparation, such as liposomes, can be used.

Polynucleotide "Polynucleotide" refers to a polymeric form of nucleotides of at least ten bases in length, either ribonucleotides or deoxyribonucleotides or a modified form of either type of nucleotide. The polynucleotides of the present invention can be a DNA molecule or an RNA molecule. Compositions of the invention where the polynucleotide is a DNA molecule are termed DNA vaccines (Feigner et al, U. S. Patent 6,413,942; Content et al, U. S. Patent 6,384,018; Feigner et al, U. S. Patent 6,214,804; Huebner et al, U. S. Patent 5,846,946; Feigner et al, U. S. Patent 5,703,055; Feigner et al, U. S. Patent 5,589,466; Hasan et al, 1999, J Immunol Methods, 229:1; Lewis and Babiuk, 1999, Adv. Virus Res., 54:129; Gurunathan et al, 2000, Annu. Rev. Immunol, 18:927). Compositions of the invention where the polynucleotide is a RNA molecule are termed RNA vaccines (McKenzie et al, 2001, Immunol. Res., 24:225; Hoerr et al, 2000, Eur. J. Immunol, 30:1; Ying et al, 1999, Nature Med., 5:823). The polynucleotide of the invention can be single-stranded or double-stranded or any combination of both. The polynucleotide of the invention can be a naked polynucleotide or a complexed polynucleotide (Pachuk et al, 2000, Curr. Opin. Mol. Then, 2:188; Hoerr et al, 2000, Eur. J. Immunol, 30:1). The polynucleotide of the invention can include a suitable vector, such as a mammalian expression vector or a viral vector. Suitable vectors include, but are not limited to, a phage, cosmid, retrovirus, vaccinia, adenovirus, adeno-associated virus, herpes simplex virus, papillomavirus, Epstein Barr virus (EBV), and the like (Nabel et al, U. S. Patent 5,910,488). Preferably, the vector is defective in that it lacks functional virulence genes, such that it is not infective after introduction into the target cell (Zeng et al, 1998, Cell Biol Toxicol, 14:105; Sorma et al, 1999, Nature Biotechnol, 17:224; Stratford-Perricaudet et al, 1992, J. Clin. Invest, 90:626; Samulski et al, 1989, J. Virol, 63:3822; Kaplitt et al, 1991, Molec. Cell. Neurosci., 2:320). Alternatively, polynucleotides of the present invention may be introduced by lipofection in vivo using liposomes (Feigner et al, 1987, Proc. Natl Acad. Sci, 84:7413). Liposomes may be targeted to particular tissues or cell types by coupling lipids to other molecules, for example, receptor ligands or antibodies that bind to a particular cell type. Polynucleotides of the present invention, in vectors or not in vectors, may also be applied to or formulated within a matrix, such as a polymeric solid matrix, a semisolid or gel, or a membrane, which is introduced into or applied externally to the mammal or cell to be treated.

Polynucleotides of the present invention for gene therapy, in viral vectors, in liposomes, or not in a vector, may be also be introduced into cells ex vivo. Target cells may be removed from the body, the polynucleotides of the present invention, in or not in a vector, may be introduced into the cells by any appropriate method, such as by infection, as a calcium phosphate precipitate, by lipofection, elecfroporation, or other methods known or developed in the art. After introduction of the nucleic acid into the cells, the cells can be reintroduced into the body.

Co-stimulatory Factors The compositions can include, in addition to the polynucleotide encoding the sequence for an immunogenic peptide, at least one co-stimulatory factor (Frauwirth and Thompson, 2002, J. Clin. Invest, 109:295). Suitable co-stimulatory factors include, for example, such molecules as B7 and CD40, cytokines, mitogens, antibodies, antigen-presenting cells (Mayordomo et al, 1997, Stem Cells, 15:94), and peptides derived from a helper T-lymphocyte epitope foreign to the immunized mammal. Co-stimulatory factors can be delivered together with the polynucleotide used for immunization, for example as a fusion with the polynucleotide, or separately, for example as a peptide or a nucleic acid molecule encoding a peptide. These co-stimulatory factors can be delivered as genes, for example, as genes for a co-stimulatory cytokine or other co- stimulatory factor (Scheerlinck, 2001, Vaccine, 19:2647; Cohen et al, 1998, FASEBJ., 12:1611).

Encoded Immunogenic Peptide The immunogenic peptide encoded by the polynucleotide of the present invention can be a peptide of between about five to about two thousand amino acids in length, or a peptide of between about five to about one thousand four hundred amino acids in length, or a peptide of between about five to about one thousand amino acids in length, or a peptide of between about five to about five hundred amino acids in length, or an oligopeptide of between about five to about forty amino acids in length, or an oligopeptide of between about five to about seventeen amino acids in length. Shorter peptides (oligopeptides), such as those of between about five to about forty amino acids in length, or between about five and about twenty amino acids in length, in particular those of six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, or seventeen amino acids in length, or between about five and about ten amino acids in length, can be more suitable to binding assays since the MHC-I binding site usually binds a peptide of between eight to ten amino acids in length (Pamer and Cresswell, 1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). However, the addition of amino acids to the N-terminus or to the C-terminus of such an oligopeptide can be advantageous, for example, by increasing the peptide's stability and half-life in vitro or in circulation in vivo. The length of the immunogenic peptide encoded by the polynucleotide of the present invention takes into consideration the overall in vivo immugenicity of the peptide, which may be influenced by the peptide's stability and half-life in vivo. Thus, while short oligopeptides can bind more readily to a major histocompatibility complex class I molecule in in vitro binding assays, longer immunogenic peptides can be more useful as vaccines. Longer encoded immunogenic peptides can be cleaved in vivo to yield the actively immunogenic sequence, immunogenic peptides encoded by the polynucleotide of the present invention can be linear or non-linear (for example, circular or branched), or can be inserted into a suitable polypeptide framework itself encoded by the polynucleotide. Examples of such polypeptide frameworks include but are not limited to green fluorescent protein (GFP), staphylococcal nuclease, fibronectin, immunoglobulin (lg), and heat shock protein (HSP). Immunogenic peptides inserted into a polypeptide framework, such as an immunoglobulin, can be processed through the endogenous degradative pathway and are presented to T-lymphocytes in the context of MHC-I molecules (Billetta et al, 1995, Eur. J. Immunol, 25:776). The peptides encoded by the polynucleotides of the present invention can be screened for immunogenicity by methods described herein. A polynucleotide of the present invention can also encode a fusion protein that includes at least a portion of the immunogenic peptide of the present invention and a polypeptide of interest. A polypeptide of interest can be any polypeptide, for example a detectable polypeptide, such as green fluorescent protein (GFP), or a sequence that aids in the purification of a polypeptide, such as the FLAG epitope. A polynucleotide that encodes a fusion protein can be made by operably linking a nucleic acid molecule that encodes a polypeptide of interest with a nucleic acid molecule that encodes at least a portion of the immunogenic peptide of the present invention. The operable linking can be direct or indirect, such as in the case where a linker connects the immunogenic peptide of the present invention with a polypeptide of interest. The polynucleotide of the present invention and the nucleic acid that encodes a polypeptide of interest are preferably operably linked in frame such that an immunogenic polypeptide of the present invention and an functional polypeptide of interest are translated from the nucleic acid, but that need not be the case.

The immunogenic peptide encoded by the polynucleotide of the invention can contain an amino acid sequence derived from immunoglobulin E, for example, a sequence derived from the heavy chain of immunoglobulin E. For use as a vaccine against IgE for a particular species such as humans, the polynucleotide of the invention preferably encodes an immunogenic peptide that contains an amino acid sequence derived from any of the constant region domains (CHI, CH2, CH3, or CH4 domains), or from the membrane exons of that species' IgE heavy chain, such as human IgE heavy chain.

The immunogenic peptide encoded by the polynucleotide of the present invention is preferably able to bind to a major histocompatibility complex class I molecule, which may be any MHC-I molecule, including a mammalian MHC-I molecule. For use as a vaccine against IgE in humans, the polynucleotide of the invention preferably encodes an immunogenic peptide that binds to an MHC-I molecule specific to humans such as a human leukocyte antigen (HLA). Preferably, the human leukocyte antigen is one that corresponds to a specific HLA haplotype, such as a particular HLA- A, HLA-B, HLA-C, and HLA-Cw serotype, or more specifically, a particular HLA genotype where more than one HLA genotype exists for a HLA serotype. Formulations Compositions of the present invention containing the polynucleotide encoding an immunogenic peptide can be provided in any formulation suitable to the intended form of administration and consistent with conventional pharmaceutical practice ("Remington: The Science and Practice of Pharmacy", 20^th edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000). Examples of suitable formulations include a solid, a powder, a gel, a solution, a suspension, an emulsion, liposomes, microspheres, injectable particles, inhalable particles, or a dissolvable matrix (OΗagan, 1998, J. Pharm. Pharmacol, 50:1; Pachuk et al, 2000, Curr. Opin. Mol. Ther., 2:188; Gurunathan et al, 2000, Annu. Rev. Immunol, 18:927; Cohen et al, 1998, FASEB J., 12:1611). Optionally, a suitable adjuvant, such as an immunostimulatory oligonucleotide can be included (Norman et al, 1999, Meth. Mol. Med., 29:185; Klinman et al, 1999, Vaccine, 17:19; Davis et al, U. S. Patent 6,406,705).

ANTIGEN-PRESENTING CELLS

The present invention further provides compositions that include antigen-presenting cells that recognize at least one isolated immunogenic peptide identified by a method of the present invention as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented ^' immunoglobulin E peptides. Such compositions can be provided in pharmaceutically acceptable carriers, and can be provided to be administered using an appropriate dose, route of administration, and regime, to provide an appropriate end-point, such as prophylaxis, palliation, or amelioration of symptoms of immunoglobulin E-mediated disease conditions. Such dose, route of administration, regime, and end-point can be determined using methods known in the art. Appropriate toxicity and efficacy evaluation can also be evaluated using methods known in the art. Appropriate in vitro, in vivo, or ex vivo methods, or combinations thereof, can be utilized. Pharmaceutically Acceptable Carriers Such compositions can include, in addition to the antigen-presenting cells that recognize the immunogenic peptide, a pharmaceutically acceptable carrier. Suitable pharmaceutically acceptable carriers include diluents, excipients, or carrier materials, selected according to the intended form of administration and consistent with conventional pharmaceutical practice. Examples of suitable carriers include water, physiological saline, phosphate-buffered saline, saline buffered with a physiologically compatible salt, cell culture medium, serum, or plasma. The pharmaceutically acceptable carrier can also include appropriate stabilizers, disintegrating agents, binders, preservatives, flavoring agents, or coloring agents, as is consistent with conventional pharmaceutical practice ("Remington: The Science and Practice of Pharmacy", 20^th edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000). The isolated antigen-presenting cells that recognize the immunogenic peptide of the present invention can be formulated and used as suspensions for oral administration, suppositories or suspensions for rectal administration, ointments, creams, or lotions for topical administration, sprays, aerosols, or inhalants, sterile suspensions or injectables, and the like. Injectables can be prepared in conventional forms as suspensions, as solid forms suitable for suspension in liquid prior to injection, or as emulsions. Suitable excipients include, for example, water, saline, dextrose, mannitol, lactose, lecithin, albumin, sodium glutamate, cysteine hydrochloride, and the like. In addition, if desired, the injectable pharmaceutical compositions can contain minor amounts of nontoxic auxiliary substances, such as wetting agents, pH buffering agents and the like. If desired, an absorption enhancing preparation, such as liposomes, can be used.

Antigen-Presenting Cells The antigen-presenting cells of the present invention can be any cells that recognize the immunogenic peptide and can elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. Preferably, the antigen-presenting cells are dendritic cells (Steinman and Pope, 2002, J Clin. Invest, 109:1519; Bhardwaj, 2001, Trends Mol Med., 7:388; Mitchell and Nair, 2000, Curr. Opin. Mol. Ther., 2:176; Morse and Lyerly, 2000, Curr. Opin. Mol. Ther., 2:20).

Co-stimulatory Factors The compositions can include, in addition to the antigen-presenting cells that recognize the immunogenic peptide, at least one co-stimulatory factor (Frauwirth and Thompson, 2002, J. Clin. Invest, 109:295). Suitable co-stimulatory factors include, for example, such molecules as B7 and CD40, cytokines, mitogens, antibodies, other antigen-presenting cells (Mayordomo et al,

1997, Stem Cells, 15:94), and peptides derived from a helper T-lymphocyte epitope foreign to the immunized mammal. Co-stimulatory factors can be delivered together with the antigen- presenting cells used for immunization, or separately, for example as a peptide or a nucleic acid molecule encoding a peptide.

Recognized Immunogenic Peptide The immunogenic peptide recognized by the antigen-presenting cells of the present invention can be a peptide of between about five to about two thousand amino acids in length, or a peptide of between about five to about one thousand four hundred amino acids in length, or a peptide of between about five to about one thousand amino acids in length, or a peptide of between about five to about five hundred amino acids in length, or an oligopeptide of between about five to about forty amino acids in length, or an oligopeptide of between about five to about seventeen amino acids in length. Shorter peptides (oligopeptides), such as those of between about five to about forty amino acids in length, or between about five and about twenty amino acids in length, in particular those of six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, or seventeen amino acids in length, or between about five and about ten amino acids in length, can be more suitable to binding assays since the MHC-I binding site usually binds a peptide of between eight to ten amino acids in length (Pamer and Cresswell,

1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). However, the addition of amino acids to the N-terminus or to the C-terminus of such an oligopeptide can be advantageous, for example, by increasing the peptide's stability and half-life in vitro or in circulation in vivo. The length of the immunogenic peptide recognized by the antigen-presenting cells takes into consideration the overall in vivo immugenicity of the peptide, which may be influenced by the peptide's stability and half-life in vivo. Thus, while short oligopeptides can bind more readily to a major histocompatibility complex class I molecule in in vitro binding assays, longer immunogenic peptides can be more useful as vaccines. Longer immunogenic peptides recognized by the antigen-presenting cells can be cleaved in vivo to yield the actively immunogenic sequence. Immunogenic peptides recognized by the antigen- presenting cells can be linear or non-linear (for example, circular or branched), or can be inserted into a suitable polypeptide framework as described herein. Immunogenic peptides inserted into a polypeptide framework, such as an immunoglobulin, can be processed through the endogenous degradative pathway and are presented to T-lymphocytes in the context of MHC class I molecules (Billetta et al, 1995, Eur. J. Immunol, 25:776).

The immunogenic peptide recognized by the antigen-presenting cells of the invention can contain an amino acid sequence derived from immunoglobulin E, for example, a sequence derived from the heavy chain of immunoglobulin E. For use as a vaccine against IgE for a particular species such as humans, the antigen-presenting cells of the invention preferably recognize an immunogenic peptide that contains an amino acid sequence derived from any of the constant region domains (CHI, CH2, CH3, or CH4 domains), or from the membrane exons of that species' IgE heavy chain, such as human IgE heavy chain.

The immunogenic peptide recognized by the antigen-presenting cells of the present invention is preferably able to bind to a MHC-I molecule, which can be any MHC-I molecule, including a mammalian MHC-I molecule. For use as a vaccine against IgE in humans, the antigen-presenting cells preferably recognize an immunogenic peptide that binds to an MHC-I molecule specific to humans such as a human leukocyte antigen (HLA). Preferably, the human leukocyte antigen is one that corresponds to a specific HLA haplotype, such as a particular HLA- A, HLA-B, HLA-C, and HLA-Cw serotype, or more specifically, a particular HLA genotype where more than one HLA genotype exists for a HLA serotype. Formulations The compositions containing the antigen-presenting cells that recognize an immunogenic peptide can be provided in any formulation suitable to the intended form of administration and consistent with conventional pharmaceutical practice ("Remington: The Science and Practice of th Pharmacy", 20 edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000). The antigen-presenting cells can be intact cells, or cells that have been further processed, for example, into membranes preparations or subcellular fractions. Examples of suitable formulations include a cell suspension, a suspension of crude membrane extracts, a suspension of purified membranes, a suspension of subcellular membrane organelle fractions, and membrane preparations mixed with liposomes.

III. METHODS FOR MODULATING IgE-MEDIATED CONDITIONS

The present invention also provides methods for modulating an immunoglobulin E- mediated condition in a mammal, comprising providing to the mammal a composition that elicits in the mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, in an amount sufficient to elicit a cytotoxic T-lymphocyte response in the mammal.

Compositions The compositions used by these methods can be any of the compositions described above, that is to say, any of the following: a) compositions that include at least one isolated immunogenic peptide identified by a method of the present invention, such as those as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, b) compositions that include at least one isolated polynucleotide encoding the sequence for an immunogenic peptide identified by a method of the present invention, such as those as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, and c) compositions that include antigen-presenting cells that recognize at least one isolated immunogenic peptide identified by a method of the present invention, such as those as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. The compositions used by these methods may optionally include one or more co-stimulatory factor. The compositions used by these methods may also be a combination of the immunogenic peptides, nucleic acid molecules, or antigen-presenting cells described herein, optionally with one or more co-stimulatory factors.

Modulation

The methods of the present invention can modulate immunoglobulin E-mediated conditions in a mammal preferably by decreasing the level of circulating secreted immunoglobulin E in the mammal, by decreasing the level of circulating immunoglobulin E- bearing B-lymphocytes in the mammal, by decreasing the level of resident mucosal immunoglobulin E-bearing B-lymphocytes in tissues such as the bronchus-associated lymphoid tissues (BALT) and gut-associated lymphoid tissues (GALT), by decreasing the level of mast cells or basophilic granulocytes sensitized to immunoglobulin E in the mammal, or by decreasing the level of immunoglobulin E-producing cells in the mammal. Preferably, clinical end-points such as, but not limited to, reduction of immunoglobulin E-mediated disease symptoms (for example, itching or burning eyes, nose, or throat; vomiting, constipation or diarrhea; sneezing, rhinorrhea, coughing or wheezing; dermatitis, eczema, or hives; asthma, shortness of breath, edema, or anaphylaxis) are achieved.

Immunoglobulin E-Mediated Conditions The immunoglobulin E-mediated conditions modulated by the methods of the invention include but are not limited to atopic hypersensitivity conditions (such as allergic rhinitis, allergic asthma, food allergies, and contact allergies or atopic dermatitis), non-atopic hypersensitivity conditions (such as anaphylaxis, urticaria, and hives), and immunoglobulin E myeloma. Mammals Mammals in which modulation of an immunoglobulin E-mediated condition can be desirable include mammals of economic or domestic importance (such as cattle, swine, sheep, goats, dogs, cats, horses, ferrets, mice, rats, rabbits, guinea pigs, gerbils, and hamsters), non- human primates (such as lemurs, lorises, tarsiers, monkeys, and apes). It is particularly desirable to modulate immunoglobulin E-mediated conditions in humans.

Amounts, Route of Administration, and Regime

The amount of the composition sufficient to elicit a cytotoxic T-lymphocyte response in the mammal, and to thereby modulate an immunoglobulin E-mediated condition in that mammal, will necessarily vary according to the specific composition, the specific IgE-mediated condition, the species of the mammal, and possibly the individual mammal. The dose can be tailored to achieve a desired effect, but will depend on such factors as weight, diet, concurrent medication and other factors, which those skilled in the medical arts will recognize. In practicing the methods of the present invention, the compositions of the present invention can be used alone or in combination with one another, or in combination with other therapeutic agents (such as, but not limited to, antihistamines, leukotriene receptor antagonists, mast cell stabilizers, steroids, epinephrine, decongestants, antibiotics, and allergy desensitization treatments), or in combination with other diagnostic agents (such as, but not limited to, allergen skin tests, allergen blood tests, and histological examination). These products can be utilized in vivo, preferably in a mammalian patient, preferably in a human, or ex vivo or in vitro. In employing them in vivo, the compositions of the present invention can be administered to the patient in a variety of ways, including orally, parenterally, intravenously, subcutaneously, intramuscularly, colonically, rectally, nasally, ocularly, topically, or intraperitoneally, using a variety of appropriate doses and regimes to modulate the IgE-mediated condition. Such methods can also be used in testing bioactivities in vivo. As will be readily apparent to one skilled in the art, the useful in vivo dosage to be administered and the particular mode of admimstration will vary depending upon the age, weight, and type of patient being treated, the particular composition of the present invention that is employed, and the specific use for which the composition is employed. The determination of effective dosage levels, that is to say, the dose levels necessary to achieve the desired result, can be accomplished by one skilled in the art using routine methods known to those of skill in the art. Typically, human clinical applications of products are commenced at lower dosage levels, with the dosage level being increased until the desired effect is achieved. Alternatively, acceptable in vitro or ex vivo studies or combinations thereof can be used to establish useful doses, routes of administration, and regimes of the compositions of the present invention. In non-human animal studies, applications of a composition of the present invention can be commenced at higher dose levels, with the dosage being decreased until the desired effect is no longer achieved, or adverse side effects are reduced or disappear. The efficacy of a given combination of a dosage level, mode of administration, and adminstration regime for a composition of the present invention can be monitored by any suitable method, for example, by measuring levels of immunoglobulin E in the patient (using, for example, ELISAs for IgE), by measuring levels of mediators (such as histamine or leukotrienes) or by monitoring symptoms of allergic disease (such as the severity of atopic dermatitis or atopic rhinitis).

The dosage for the compositions of the present invention can range broadly depending upon the desired effects, the therapeutic indication, route of administration, regime, and purity and activity of the composition. Where the composition is one that includes at least one isolated immunogenic peptide identified by a method of the present invention as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T- lymphocyte response to naturally processed and presented immunoglobulin E peptides, this amount can be between about 0.01 milligrams per kilogram body weight to about 10 milligrams per kilogram body weight, or between about 0.05 milligrams per kilogram body weight to about 10 milligrams per kilogram body weight, or between about 0.05 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight, or between about 0.1 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight, or between about 0.5 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight, or between about 0.5 milligrams per kilogram body weight to about 1 milligrams per kilogram body weight of the isolated immunogenic peptide. Where the composition is one that includes at least one isolated polynucleotide encoding the sequence for an immunogenic peptide identified by a method of the present invention as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, this amount can be between about 0.01 milligrams per kilogram body weight to about 10 milligrams per kilogram body weight, or between about 0.05 milligrams per kilogram body weight to about 10 milligrams per kilogram body weight, or between about 0.05 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight, or between about 0.1 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight, or between about 0.5 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight, or between about 0.5 milligrams per kilogram body weight to about 1 milligrams per kilogram body weight of said at least one isolated polynucleotide. Where the composition is one that includes antigen-presenting cells that recognize at least one isolated immunogenic peptide identified by a method of the present invention as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, this amount can be between about 1 x 10³ antigen-presenting cells per kilogram body weight and 1 x 10 antigen-presenting cells per kilogram body weight, or between about 1 x 10⁴ antigen-presenting cells per kilogram body weight and 1 x 10⁸ antigen- presenting cells per kilogram body weight, or between about 1 x 10⁵ antigen-presenting cells per kilogram body weight and 1 x 10⁸ antigen-presenting cells per kilogram body weight, or between 1 x 10⁵ antigen-presenting cells per kilogram body weight and 1 x 10⁷ antigen-presenting cells per kilogram body weight.

The exact formulation, route of administration, regime, and dosage can be chosen by the individual physician in view of the patient's condition ("Goodman & Gilman's The Pharmacological Basis of Therapeutics", 10 edition, Hardman (ed.) and Limberd, McGraw-Hill Professional Publishing, New York, NY, 2001). It should be noted that the attending physician would know how to and when to terminate, interrupt, or adjust administration due to toxicity, organ dysfunction, or other adverse effects. Conversely, the attending physician would also know to adjust treatment to higher levels if the clinical response were not adequate. The magnitude of an administrated dose in the management of the disorder of interest will vary with the severity of the condition to be treated and to the route of administration. The severity of the condition may, for example, be evaluated in part by standard prognostic evaluation methods for that condition. Further, the dose, and perhaps the dose frequency, will also vary according to the age, body weight, and response of the individual patient, including veterinary patients.

Depending on the specific conditions being treated, compositions of the present invention can be formulated and administered systemically or locally. Techniques for formation and administration can be found in "Remington: The Science and Practice of Pharmacy" (20^th edition, Gennaro (ed.) and Gennaro, Lippincott, Williams & Wilkins, 2000). Suitable routes of administration can include oral, rectal, transdermal, otic, ocular, vaginal, transmucosal, or intestinal administration, or parenteral delivery, including intramuscular, subcutaneous, or intramedullary injections, as well as intrathecal, direct intraventricular, intravenous, infraperitoneal, intranasal, or intraocular injections.

For injection, the compositions of the present invention can be formulated in aqueous solutions, preferably in physiologically compatible buffers such as Hanks' solution, Ringer's solution, or physiological saline. For such transmucosal administration, penetrans appropriate to the barrier to be permeated are used in the formulation. Such penetrans are generally known in the art. Use of pharmaceutically acceptable carriers to formulate the compositions disclosed herein, for the practice of the methods of the invention, into dosages suitable for systemic administration is within the scope of the invention. With proper choice of carrier and suitable manufacturing practice, the compositions of the present invention, in particular, those formulated as solutions or suspensions, can be administered parenterally, such as by intravenous injection. The compositions of the present invention can be formulated readily using pharmaceutically acceptable carriers well known in the art into dosages suitable for oral administrations. Such carriers enable the compositions of the present invention to be formulated as tables, pills, capsules, liquids, gels, syrups, slurries, suspensions and the like, for oral ingestion by a patient to be treated. Compositions of the present invention that are intended to be administered intracellularly may be administered using techniques well known to those of ordinary skill in the art. For example, such compositions may be encapsulated into liposomes, then administered as described above. Substantially all molecules present in an aqueous solution at the time of liposome formation are incorporated into or within the liposomes thus formed. The liposomal contents are both protected from the external micro-environment and, because liposomes fuse with cell membranes, are efficiently delivered into the cell cytoplasm. Additionally, due to their hydrophobicity, small organic molecules such as some peptides can be directly administered intracellularly.

Compositions of the present invention suitable for use in the methods described herein include compositions wherein the active ingredients are contained in an effective amount to achieve the composition's intended purpose. Preferably, the composition is provided in an amount sufficient to achieve a desired result, such as, but not limited to, prophylaxis, palliation, or amelioration of symptoms of immunoglobulin E-mediated disease conditions (for example, itching or burning eyes, nose, or throat; vomiting, constipation or diarrhea; sneezing, rhinorrhea, coughing or wheezing; dermatitis, eczema, or hives; asthma, shortness of breath, edema, or anaphylaxis). Appropriate regimes and routes of administration to achieve such results can be determined using methods described herein or known in the art. Determination of the effective amount of a composition of the present invention is well within the capability of those skilled in the art, especially in light of the detailed disclosure provided herein. In addition to the active ingredients (such as the immunogenic peptides, polynucleotides encoding immunogenic peptides, and antigen-presenting cells recognizing immunogenic peptides described in detail herein, or combinations thereof), compositions of the present invention can contain suitable pharmaceutically acceptable carriers including but not limited to excipients and auxiliaries that facilitate processing of the active ingredients into preparations which can be used pharmaceutically. The preparations formulated for oral administration may be in the form of tables, dragees, capsules, or solutions. The compositions of the present invention can be manufactured in a manner that is itself known, for example by means of conventional mixing, dissolving, granulating, dragee-making, emulsifying, encapsulating, entrapping or lyophilizing processes. Pharmaceutical formulations for parenteral administration include aqueous solutions of active ingredients in water-soluble form.

Additionally, suspensions of the active ingredients may be prepared as appropriate oily injection suspensions. Suitable lipophilic solvents or vehicles include fatty oils such as sesame oil, or synthetic fatty acid esters, such as ethyl oleate or triglycerides or liposomes. Aqueous injection suspensions may contain substances what increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol, or dextran. Optionally, the suspension can also contain suitable stabilizers or agents that increase the solubility of the active ingredients to allow for the preparation of highly concentrated solutions. Compositions of the present invention for oral use can be obtained by combining the active ingredient with solid excipient, optionally grinding a resulting mixture, and processing the mixture of granules, after adding suitable auxiliaries, if desired, to obtain tablets or dragee cores. Suitable excipients are, in particular, fillers such as sugars, including lactose, sucrose, mannitol, or sorbitol, cellulose preparations such as, for example, maize starch, wheat starch, rice starch, potato starch, gelatin, gum fragacanth, methyl cellulose, hydroxypropylmethylcellulose, sodium carboxymethylcellulose, and/or polyvinylpyrrolidone. If desired, disintegrating agents can be added, such as cross-linked polyvinylpyrrolidone, agar, or alginic acid or a salt thereof such as sodium alginate. Dragee cores can be provided with suitable coatings. Dyes or pigments can be added to the tablets or dragee coatings for identification or to characterize different combinations of active doses. EXAMPLES

EXAMPLE 1: Induction of an immunoglobulin E-specific, major histocompatibility complex class I-restricted cytotoxic T-lymphocyte response by test peptides The following example describes the induction of an immunoglobulin E-specific, major histocompatibility complex class I-restricted cytotoxic T-lymphocyte response by test peptides.

The sequences of murine immunoglobulin E (IgE) test peptides suspected of being able to bind to specific murine major histocompatibility complex class I (MHC-I) molecules were predicted by algorithms for the MHC-I molecules publicly available at the Internet websites bimas.dcrt.nih.gov/molbio/hla_bind/index.html, www.syfpeithi.de, and hiv.basic.nwu.edu/HLA/Reports/DoMotifList.cfm. The test peptide pi for the murine MHC-I molecules H-2K^d and H-2K^b consisted of the nonameric amino acid sequence: Leu Tyr Cys Phe He Tyr Gly His He (SEQ ID NO. 1). The test peptide p2 for the murine MHC-I molecule H-2D^d consisted of the nonameric amino acid sequence: He Tyr Gly His He Leu Asn Asp Val (SEQ ID NO.2). Both test oligopeptides were synthesized by conventional 9-fluorenylmethoxycarbonyl (Fmoc) peptide chemistry.

The test peptides pi and p2 were evaluated for the ability to elicit in mice a cytotoxic T- lymphocyte response to naturally presented immunoglobulin E peptides. Eight-week old BALB/c mice in groups of four were each immunized with 30 micrograms of one of the test peptides pi or p2, together with 30 micrograms of the co-stimulatory factor OVAp (ovalbumin peptide) (Schimokevitz, 1984, J. Immunol, 133:2067), an oligopeptide consisting of the sequence He Ser Gin Ala Val His Ala Ala His Ala Glu He Asn Glu Ala Gly Arg,(SEQ ID NO. 3), in N-[l-(-2,3-dioleoyloxy)propyl]-N.NN-trimethylammonium methylsulfate (DOTAP, purchased from Boehringer Mannheim, Indianapolis, IN) cationic liposomes. Two weeks after the first immunization, mice in each respective group were immunized a second time with the appropriate test peptide (pi or p2) that had been used in the initial immunization. Two weeks after the second immunization, 5xl0⁶ spleen cells were removed from a pool of two to four mice, established in vitro, and re-stimulated with the respective test peptide in vitro for seven days. The cytotoxic T-lymphocytes thus obtained were tested using a standard chromium release assay (Colligan et al. 2002, Current Protocols in Immunology, 3.11) for their ability as effector cells to lyse target cells. The target cells consisted of P815 murine mastocytoma cells that had been contacted with the appropriate test peptide (pi or p2) and treated overnight with 50 units per milliliter interferon-gamma to facilitate processing of the test peptide. Effector cells (E) and target cells (T) were mixed at different E/T ratios for 4 hours. The specific lysis was determined by the following formula: [(experimental release - spontaneous release)/(maximal release - spontaneous release)] x 100. The results are illustrated in FIG. 3 A, which shows that both IgE- derived test peptides pi and p2, delivered with a co-stimulatory peptide and with liposomes, were able to elicit a CTL response in BALB/c mice that was cytotoxic to target cells displaying the peptides.

The length of a test peptide can influence its half-life and immunogenicity. Serine and cysteine endopeptidases and serine carboxypeptidases are known to be involved in antigen processing (Germain, 1999, Fundamental Immunology, Ch. 9, p. 263, ed. Paul; Li, 2002, Mol. Biochem. Parasitol, 120:177; Gil-Torregrosa, et al, 1998, J. Exp. Med., 188:1105). Therefore a third test peptide p3 was synthesized, which consisted of the overlapped nonameric amino acid sequences of the test peptides pi and p2, with an N-terminus sixteen amino acid extension, derived from the CH2 domain of murine immunoglobulin E, and a C-terminus ten amino acid extension (FIG. 4) (Wang et al, 1996, Eur. J. Immunol, 26:1043). The amino acid sequence of p3 was: Leu Leu His Ser Ser Cys Asp Pro Asn Ala Phe His Ser Thr He Gin Leu Tyr Cys Phe He Tyr Gly His He Leu Asn Asp Val Ser Val Ser Trp Leu Met Asp Asp Arg Glu (SEQ ID NO. 4). The serine endoproteinase Glu-C cleaves between the two residues Asp Val, which demarcate the majority of the overlapped pl/p2 sequence from the C-terminus ten amino acid extension. The test peptide p3 was similarly evaluated for the ability to elicit a cytotoxic T- lymphocyte response in mice to naturally processed and presented immunoglobulin E peptides. Eight- week old BALB/c mice in groups of four were each immunized with 30 micrograms of the test peptide p3, with or without 30 micrograms of the co-stimulatory factor OVAp, in the presence or absence of 30 micrograms DOTAP liposomes. Two weeks later, mice were immunized a second time using the same protocol as in the first immunization. Two weeks after the second immunization, 5xl0⁶ spleen cells were removed from a pool of two to four mice, established in vitro, and re-stimulated with the test peptide p3 in vitro for seven days. The cytotoxic T-lymphocytes thus obtained were tested using a standard chromium release assay for their ability as effector cells to lyse target cells. The target cells consisted of P815 murine mastocytoma cells that had been contacted with the test peptide p3 and treated overnight with 50 units per milliliter interferon-gamma. Effector cells (E) and target cells (T) were mixed at different E/T ratios for 4 hours. The specific lysis was determined by the following formula: [(experimental release - spontaneous release)/(maximal release - spontaneous release)] x 100. The results are illustrated in FIG. 3B, which shows that immunization with p3 in the presence of the co-stimulatory peptide OVAp was able to induce a CTL response in BALB/c mice that was cytotoxic to the p3-treated P815 target cells; this CTL response was augmented by delivery of the peptides in liposomes. Immunization with the test peptide p3 alone, or with the co-stimulatory peptide OVAp in liposomes, did not induce a substantial CTL-response in BALB/c mice against the target cells; delivery of p3 in liposomes also amplified the CTL response relative to that seen with p3 alone.

Induction of an IgE-specific cytotoxic T-lymphocyte response may depend on T cell co- stimulation. In the particular case of the test peptide p3, p3 alone did not induce a significant IgE-specific cytotoxic T-lymphocyte response in BALB/c adult mice. However, this test peptide was rendered immunogenic by providing a co-stimulatory factor (OVAp or OVAp entrapped in liposomes). An increased IgE-specific cytotoxic T-lymphocyte response was observed in mice immunized with p3 delivered with OVAp in liposomes. The cytotoxic T-lymphocyte response was shown to be specific for immunoglobulin E peptides since immunization with the co- stimulatory factor OVAp in liposomes alone did not elicit IgE-specific cytotoxic T-lymphocytes. The amplified IgE-specific cytotoxic T-lymphocyte response (shown by higher chromium release levels) was noted for the test peptide p3 relative to the test peptides pi and p2. This may be due to an increased immunogenicity of p3, which encompasses both pi and p2, and to increased stability due to the increased length of p3 relative to pi or p2. A fourth test peptide p4 was synthesized and evaluated for the ability to elicit a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides in mice of a different murine MHC-I haplotype. The test peptide p4 was a 17-mer peptide consisting of the overlapped nonameric amino acid sequences of the test peptides pi and p2, with an N- terminus two amino acid extension, and a C-terminus two amino acid extension (FIG. 4). The amino acid sequence of p4 was thus: He Gin Leu Tyr Cys Phe He Try Gly His He Leu Asn Asp Val Ser Val (SEQ ID NO. 5).

It was previously reported (Billeta et al, 1995, Eur. J. Immunol, 25:776) that antibodies engineered to express major histocompatibility complex class I- or class H-restricted epitopes in the complementarity-determining regions are naturally processed and presented to T- lymphocytes. The immunogenicity of the test peptide p4 as an isolated peptide was compared to the p4 amino acid sequence engineered into a polypeptide framework (specifically, the p4 amino acid sequence embedded in a loop of an immunoglobulin variable domain). Since an antibody molecule has a much longer half-life than short synthetic peptides and can target antigen- presenting cells via the Fc receptor, such a larger test peptide could be more immunogenic. The murine immunoglobulin E heavy (H) chain gene VH62 was engineered to encode the amino acid sequence of the test peptide p4 into the complementarity-determining region 3 (CDR3) of the immunoglobulin heavy chain framework to yield the fifth test peptide p5. This engineered p5 sequence was cloned into a human IgGl expression vector and expressed in murine J558L cells as an antigenized antibody composed of the chimeric (mouse/human) H chain associated with the murine lambda 2 light chain. Thus, the length of the test peptide p5 (as the engineered immunoglobulin E heavy chain) exceeds 440 amino acids, and approximates the entire length of an immunoglobulin E heavy chain (Zanetti, U. S. Patent 5,583,202). As the expressed, fully assembled antigenized antibody, the length of the test peptide p4 exceeds 1,400 amino acids in length (Billetta, 1992, Immunomethods, 1 :41; Billetta et al, 1995, EurJ.Immunol,. 25:776;

Zanetti, U. S. Patent 5,583,202, each of which are incorporated by reference in their entirety, in particular the sequences contained therein). Groups consisting each of five C57BL/6 mice were immunized with either 30 micrograms of the test peptide p4 with 30 micrograms of the co-stimulatory factor OVAp, or with 30 micrograms of the test peptide p5 delivered in 30 micrograms DOTAP liposomes. Two weeks after the first immunization, mice in each respective group were immunized a second time with the appropriate test peptide formulation (p4 plus OVAp, or p5 in DOTAP liposomes) that had been used in the initial immunization. Two weeks after the second immunization, 5x10 spleen cells were removed from a pool of two to four mice, established in vitro, and re-stimulated with the appropriate test peptide (p4 or p5) in vitro for seven days. The cytotoxic T-lymphocytes thus obtained were tested using a standard chromium release assay for their ability as effector cells to lyse target cells. The target cells consisted of murine EL-4 lymphoma cells, which display the major histocompatibility complex class I molecules H-2K and H-2D , and that had been contacted with the test peptide p4 and treated overnight with 50 units per milliliter interferon-gamma. Effector cells (E) and target cells (T) were mixed at different E/T ratios for 4 hours. The specific lysis was determined by the following formula: [(experimental release - spontaneous release)/(maximal release - spontaneous release)] x 100. The results are depicted in FIG. 5, which shows that mice immunized with either of the test peptides p4 or p5 developed comparable levels of immunoglobulin E-specific CTL response. This demonstrated that p5 had been efficiently internalized into antigen-presenting cells, naturally processed and presented through the exogenous major histocompatibility complex class I pathway to yield immunogenic p4-derived peptide-MHC-I molecule complexes.

EXAMPLE 2: Induction of an immunoglobulin E-specific, major histocompatibility complex class I-restricted cytotoxic T-lymphocyte response by antigen-presenting cells

The following example describes the use of antigen-presenting cells to elicit a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, resulting in a decrease in level of immunoglobulin E-producing cells.

Murine bone marrow-derived dendritic cells were prepared by culturing femur bone marrow cells in the presence of 1.5 nanograms per milliliter recombinant granulocyte macrophage colony stimulating factor (GM-CSF) by a method of Inaba (Inaba, 1990, J. Exp. Med., 172:631). The dendritic cells (DC) were cultured over a period of 10 days following a routine decanting procedure, and the purity of the cell preparation was determined using a monoclonal antibody against CD1 lc to be greater than 90%. These dendritic cells (1 x 10⁶ in 0.2 milliliter) were pulsed with 100 micrograms of the test peptide p3 together with 30 micrograms of the co-stimulatory factor OVAp (OVAp sequences are restricted to both LA and IA^b), with or without 30 micrograms DOTAP liposomes for 1 hour at 37 °C. Unbound p3, OVAp, and liposomes were removed by washing. The dendritic cells thus treated with p3 as an immunogenic peptide were then used to immunize mice. BALB/c mice, which display the major histocompatibility complex class I haplotype H2d, were used. Each mouse in a group of four adult BALB/c mice was immunized with 2 x 10⁴ dendritic cells (treated as described above with p3 and OVAp), injected equally via the subcutaneous and infraperitoneal routes. Two weeks after the second immunization, 5xl0⁶ spleen cells were removed from a pool of two to four mice, established in vitro, and re-stimulated with the test peptide p3 in vitro for seven days. The cytotoxic T-lymphocytes thus obtained were tested using a standard chromium release assay for their ability as effector cells to lyse target cells. Two types of target cells were used. The first target cell type consisted of P815 murine mastocytoma cells that had been contacted with the test peptide p3 and treated overnight with 50 units per milliliter interferon-gamma (FIG. 6A). The second target cell type consisted of 26.82 hybridomas, which endogenously synthesize and naturally process and present murine immunoglobulin E associated with the major histocompatibility complex class I molecule H-2d (FIG. 6B). Effector cells (E) and target cells (T) were mixed at different E/T ratios for 4 hours. The specific lysis was determined by the following formula: [(experimental release - spontaneous release)/(maximal release - spontaneous release)] x 100. Comparable levels of lysis were observed for p3-contacted P815 (FIG. 4A) and for the naturally processed and presented IgE peptide-decorated 26.82 hybridoma cells (Liu, 1980, J. Immunol, 124:2728) (FIG. 6B). These results demonstrate that the test peptide p3 served as an immunogenic peptide when presented by antigen-presenting cells (dendritic cells that recognize p3) and were able to elicit in the DC-immunized mice a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. Dendritic cells incubated with p3 and OVAp were sufficient for inducing this CTL response. The magnitude of this CTL response was not amplified by either: (a) in vivo treatment of the DC-immunized mice with the anti-murine CD40 antibody FGK-45 (Schoenberger et al , 1998, Nature, 393 :480) (24 hours after DC immunization), or (b) incubation of the dendritic cells with p3 and OVAp with DOTAP liposomes. This indicates that the cytotoxic T-lymphocyte response achieved by dendritic cells incubated with p3 and ONAp alone was at or near a ceiling.

EXAMPLE 3: Breaking self-tolerance to natural immunoglobulin E peptides.

The following example describes the breaking self-tolerance to natural immunoglobulin E peptides.

In this example, the effect of self-tolerance on induction of a cytotoxic T-lymphocyte response against immunoglobulin E was assessed. A "self immunoglobulin E peptide would tend to be considered less immunogenic in a mammal that expresses its own (self) immunoglobulin E molecules, due to the phenomenon known as self-tolerance. This example also assessed whether any type of stimulation is critically required to overcome unresponsiveness due to self-tolerance. The magnitude of the cytotoxic T-lymphocyte response against immunoglobulin E in wild-type mice (IgE +/+) that express immunoglobulin E and in mice whose immunoglobulin E gene is deficient (IgE-/-) was compared. Any CTL response against "self immunoglobulin E induced in IgE +/+ mice would represent a partial response, depending on the strength of T-lymphocyte co-stimulation. If co-stimulation reaches a supra-optimal ceiling, these two genetically disparate mouse strains should exhibit a similar magnitude of cytotoxic T-lymphocyte response following co-stimulation. Furthermore, since the pertinent targets for therapeutic purposes are cells that express naturally processed and presented immunoglobulin E peptides, immunoglobulin E-producing cells were employed as targets. In these cells, immunoglobulin E peptides are naturally processed and presented by a MHC-I molecule in a physiological, MHC-I-restricted manner. The 129 strain of mouse, which displays the major histocompatibility complex class I haplotype H2^b, was used. Mice carrying the wild-type immunoglobulin E gene (129/SvEv, IgE +/+) and immunoglobulin E gene-deficient mice (129/SvEv, IgE-/-) were equally immunized according to five different regimens, which tested different modes of co-stimulation: (a) Mice were immunized with 30 micrograms of the test peptide p3, 30 micrograms of the co-stimulatory factor OVAp, and 30 micrograms DOTAP liposomes, in a total volume of 150 microliters, injected equally via the subcutaneous and infraperitoneal routes; 24 hours after immunization, the mice were further injected intravenously with 100 micrograms of the anti-CD40 antibody FGK-45; (b) Mice were immunized with 30 micrograms of the test peptide p3, 30 micrograms of the co-stimulatory factor OVAp, and 30 micrograms DOTAP liposomes, in a total volume of 150 microliters, injected equally via the subcutaneous and infraperitoneal routes; 24 hours after immunization, the mice were further injected intravenously with 100 micrograms of the normal rat immunoglobulin G2a (IgG2a) as a control for the anti-CD40 antibody treatment;

(c) Mice were immunized with 2 x 10⁴ dendritic cells (treated with p3 and OVAp as described above in Example 2), injected equally via the subcutaneous and infraperitoneal routes;

(d) Mice were immunized with 2 x 10⁴ dendritic cells (treated with p3 and OVAp as described above in Example 2), injected equally via the subcutaneous and infraperitoneal routes; 24 hours after immunization, the mice were further injected intravenously with 100 micrograms of the anti-CD40 antibody FGK-45; and

(e) Mice were immunized with 2 x 10⁴ dendritic cells (treated with p3, OVAp, and DOTAP liposomes as described above in Example 2), injected equally via the subcutaneous and infraperitoneal routes; 24 hours after immunization, the mice were further injected intravenously with 100 micrograms of the anti-CD40 antibody FGK-45. Two weeks after the first immunization, each group of mice was immunized a second time using the formulation as in the first immunization. Two weeks to four weeks after the second immunization, 5xl0⁶ spleen cells were removed from a pool of two to four mice, established in vitro, and re-stimulated with the test peptide p3 in vitro for seven days. The cytotoxic T-lymphocytes thus obtained were tested using a standard chromium release assay for their ability as effector cells to lyse target cells. Three types of target cells were used. The first target cell type consisted of murine immunoglobulin E-secreting B4 hybridoma cells, which exhibit the major histocompatibility class I molecule H-2^b (FIG. 7A and FIG. 7C). The second target cell type consisted of murine EL-4 lymphoma cells, which display the major histocompatibility complex class I molecule H-2^b and had been contacted with the test peptide p3 and treated overnight with 50 units per milliliter interferon-gamma (FIG. 7B). The third target cell type consisted of 26.82 hybridomas, which endogenously synthesize and naturally process and present murine immunoglobulin E associated with the major histocompatibility complex class I molecule H-2^d (FIG. 7D). Effector cells (E) and target cells (T) were mixed at different E/T ratios for 4 hours. The specific lysis was determined by the following formula: [(experimental release - spontaneous release)/(maximal release - spontaneous release)] x 100. The results are shown in FIG. 7 — FIG. 7D.

An immunoglobulin E peptide-specific CTL response was observed in 129/SvEv, IgE-/- mice immunized with the test peptide p3 together with the co-stimulatory factor OVAp in DOTAP liposomes, or with dendritic cells treated with p3 and OVAp. Cytotoxic T-lymphocytes from these IgE-/- mice were able to lyse target cells that naturally processed and presented immunoglobulin E peptides bound to H-2^b, regardless of whether the IgE peptides were biosynthesized endogenously (as in the B4 hybridoma cells, FIG. 7A) or were introduced exogenously (as in the EL-4 lymphoma cells, FIG. 7B).

A similar immunoglobulin E peptide-specific CTL response was observed in 129/SvEv, IgE+/+ mice immunized with the test peptide p3 together with the co-stimulatory factor OVAp in DOTAP liposomes, or with dendritic cells treated with p3 and OVAp. That this CTL response was specific for immunoglobulin E peptides bound to a specific MHC-I haplotype is shown because these cytotoxic T-lymphocytes were able to lyse B4 hybridoma cells (which have the H- 2^b haplotype) (FIG. 7C) but not 26.82 hybridoma cells (which have the H-2^d haplotype) (FIG. 7D). Conversely, CTL raised in p3-immunized BALB/c mice (which have the H-2^d haplotype), were able to lyse IgE-secreting 26.82 hybridoma cells (see Example 2). These results show that even though the test peptide p3 was able to induce an IgE-specific response in both H-2^b and H- 2^d haplotypes, p3-specific cytotoxic T-lymphocytes induced in the H-2^b environment lyse only IgE producing B4 (H-2^b haplotype) (Rudolph et al. , 1981 , Eur. J. Immunol, 11 :527) but not

26.82 hybridomas (H-2^d haplotype). Thus, MHC-I-restriction of the CTL response is observed at both the inductive phase and effector phase.

Three additional observations were noted. Firstly, the degree of lysis of B4 hybridoma cells by CTL from 129/SvEv, IgE+/+ mice was comparable to that seen for CTL from equivalently immunized 129/SvEv, IgE-/- mice. This shows that immunization methods using different modes of co-stimulation were sufficient to overcome tolerance to self-IgE peptides in the IgE +/+ mice. Secondly, the levels of target cell lysis were similar for CTL responses induced either by immunization with the test peptide p3 together with the co-stimulatory factor OVAp in DOTAP liposomes, or by immunization with dendritic cells treated with p3 and OVAp. Immunoglobulin E test peptides, processed and presented by dendritic cells, are thus at least similar if not identical to endogenously produced, naturally processed and presented IgE peptides that are bound to MHC-I of IgE-secreting plasma cells. Finally, the CTL response induced by dendritic cell immunization was not appreciably amplified by treatment with anti-CD40 antibody. Collectively, these observations show that self-tolerance to the IgE constant region epitopes contained in the test peptide p3 can be fully overcome by appropriate immunization methods. This shows the feasibility of developing a vaccine that is efficacious in eliciting a cytotoxic T-lymphocyte response to endogenously produced, naturally processed and presented IgE in humans or in animals of domestic or economic importance.

EXAMPLE 4: Inhibition of immunoglobulin E production in vitro and in vivo by immunoglobulin E-specific cytotoxic T-lymphocytes

The following example describes the inhibition of immunoglobulin E production in vitro and in vivo by immunoglobulin E-specific cytotoxic T-lymphocytes. To determine whether cytotoxic T-lymphocytes elicited by immunoglobulin E peptide immunization could decrease immunoglobulin E production, in vitro stimulated cytotoxic T- lymphocytes were added to IgE-secreting hybridomas or to B-lymphocytes in vitro. In the first experiment, adult BALB/c (H-2^d haplotype) mice (5 per group) were immunized twice with the test peptide p3 together with the co-stimulatory factor OVAp in DOTAP liposomes and their spleen cells re-stimulated in vitro as described in the previous examples. The resulting cytotoxic T-lymphocytes were tested for the ability to lyse immunoglobulin E-secreting target cells, which were either 26.82 hybridoma cells (which have the H-2^d haplotype) or B4 hybridoma cells (which have the H-2^b haplotype), at an E/T ratio of 30 to 1. Immunoglobulin E levels in the cell culture supernatants were determined using a commercial double-sandwich ELISA assay (catalogue number 2655KI, PharMingen). The results are provided in FIG. 8A, which shows that the cytotoxic T-lymphocytes elicited by p3 immunization of H-2^d haplotypic mice decreased immunoglobulin E production by H-2^d haplotypic immunoglobulin E-secreting cells (26.82 hybridoma cells) but did not decrease IgE production by H-2^b haplotypic immunoglobulin E- secreting cells (B4 hybridoma cells).

In a second experiment, BALB/c splenic B-lymphocytes were enriched by magnetic beads coated with an antibody against the B-lymphocyte-specific cell surface antigen B220. To assess CTL effects on polyclonal IgE production, 3 x 10⁵ normal B-lymphocytes were added to either 3 x 10⁶ p3-activated spleen cells from mice immunized with test peptide p3 together with the co- stimulatory factor OVAp in DOTAP liposomes, or to 3 x 10⁶ p3-activated spleen cells from mice immunized with only the co-stimulatory factor OVAp in DOTAP liposomes as a control. The co-cultures were then stimulated with 20 micrograms per milliliter lipopolysaccharide (LPS) and 250 units per milliliter murine recombinant interleukin-4 (IL-4). Two days later, the culture supernatants were harvested. Immunoglobulin E levels were determined using a commercial double-sandwich ELISA assay (catalogue number 2655KI, PharMingen). As depicted in FIG. 8B, polyclonal immunoglobulin E production was observed in LPS and IL-4 stimulated B- lymphocyte cultures. Cytotoxic T-lymphocytes from the p3-immunized mice markedly inhibited (71 to 83% inhibition) immunoglobulin E production in vitro (22 versus 75 nanograms per milliliter; 12 versus 72 nanograms per milliliter), whereas cytotoxic T-lymphocytes obtained from the mice immunized with only OVAp as a control did not decrease immunoglobulin E production.

In a third experiment, the effect of immunoglobulin E peptide immunization on production of allergen-specific IgE in vivo was determined. Adult BALB/c mice (5/group) were immunized twice with either the test peptide p3 or the test peptide pi, together with the co- stimulatory factor OVAp in DOTAP liposomes, or with only the co-stimulatory factor OVAp in DOTAP liposomes. Mice immunized with saline served as control. Seven days after the second immunization, mice were challenged with 1 microgram keyhole limpet hemocyanin (KLH) in 2 milligrams alum intraperitoneally four times at a 10 day interval, and the allergen-specific (KLH- specific) IgE in pooled blood was then determined by passive cutaneous anaphylaxis (PCA) reaction. As depicted in FIG. 8C, allergen-specific (KLH-specific) IgE production was significantly (p < 0.01) inhibited both in mice immunized with p3 and OVAp in liposomes and in mice immunized with the test peptide pi and OVAp in liposomes (PCA = 20). In contrast, mice immunized either with only the co-stimulatory factor OVAp in liposomes or with saline exhibited relatively high titer anti-KLH IgE (PCA = 160 and PCA = 320, respectively). Reduction of allergen-specific PCA correlated directly with the elicitation of IgE peptide-specific cytotoxic T-lymphocytes in vitro (as depicted in FIG. 3B). Immunoglobulin G (IgG) responses in IgE peptide-vaccinated animals remained comparable to controls. In vitro and in vivo experiments to determine the efficacy of human MHC-I-specific IgE peptide immunization on production of a human immunoglobulin E-specific cytotoxic T- lymphocyte response can be carried out in a manner similar to that used above, using a human MHC-I molecule such as a specific HLA haplotype. The human immunoglobulin E peptide- specific cytotoxic T-lymphocytes may be prepared ex vivo, for example, using human peripheral blood lymphocytes. The efficacy of immunization with a given immunoglobulin E peptide in producing a human immunoglobulin E-specific cytotoxic T-lymphocyte response can be monitored by measuring levels of immunoglobulin E in the patient (using, for example, ELISAs for human immunoglobulin E), by measuring levels of mediators (such as histamine or leukotrienes) or by monitoring symptoms of allergic disease (such as the severity of atopic dermatitis or atopic rhinitis). This approach is advantageous because it permits the tailoring of the immunoglobulin E vaccination to an individual MHC-I haplotype profile.

EXAMPLE 5: Methods for determining therapeutically or prophylactically effective immunoglobulin E peptides with motifs for major histocompatibility complex class I molecules in humans and other mammals

The following example describes methods for determining therapeutically or prophylactically effective immunoglobulin E peptides with motifs for major histocompatibility complex class I molecules in humans and other mammals.

Four non-limiting examples of approaches for determining natural IgE peptides specific for a given mammalian MHC-I haplotype are illustrated and are provided below. Approach 1: Direct immunization of a mammal with a test peptide predicted by an algorithm to bind to an MHC-I molecule expressed on a cell surface in vivo by that mammal. Algorithms for predicting the binding of a peptide to a given major histocompatibility complex class I molecule, or that predict within a supplied peptide sequence epitopes that bind to a given MHC-I molecule, are publicly available, for example, on the following websites: bimas.dcrt.nih.gov/molbio/hla_bind/index.html, hiv.basic.nwu.edu/HLA/MotifScanner.cfm, and www.uni-tuebingen.de/uni/kxi/ or http://www.syfpeithi.de. Test peptide sequences, derived from immunoglobulin E from a particular mammalian species, are subjected to the algorithm for the desired MHC-I molecule. "Derived from" refers to having an amino acid sequence based at least partially on the linear sequence of the polypeptide (such as immunoglobulin E) from which the test peptide is derived; amino acid substitutions, deletions, or additions can be incorporated. Amino acids that commonly occur in nature, or that rarely or do not occur in nature, can be incorporated. All L-amino acids, all D-amino acids, or a mixture of L- and D-amino acids can be incorporated. Peptides can be additionally modified, and can be optionally labelled with a detectable moiety, as described herein. The test peptides showing the highest scores for the algorithm, and thus predicted to best bind to the MHC-I molecule, will be selected for further investigation. Optionally, test peptides that show medium or low scores for the algorithm may also be selected for further investigation. This approach has been taken and has proven productive in the mouse, as demonstrated in Example 1. This direct approach requires that the mammals to be immunized with the test peptide express on cell surfaces the desired MHC-I molecule, either endogenously or exogenously (as a transgene). The binding assays may be carried out in vivo or ex vivo, or in vitro on cells such as primary cell lines derived from the mammal. A large breeding animal facility may be required for performing one-step screening and ascertainment of the natural IgE peptides.

Approach 2: Direct binding of a test peptide to an MHC-I molecule.

The binding of a peptide to a particular major histocompatibility complex class I molecule haplotype can be achieved via a direct binding assay. This may be done by at least two different methods.

The first method includes the following steps: (i) The desired major histocompatibility complex class I molecule is isolated, for example, from the mammal, from an appropriate cell line, or by recombinant expression of the MHC-I gene. Recombinant expression of the MHC-I gene may be carried out on any somatic cell, as somatic cells constitutively express MHC-I genes. The full-length cDNA for an MHC-I gene can be isolated by polymerase chain reaction (PCR) techniques from the total messenger RNA by employing primers flanking the gene. The protein encoded by the isolated gene can then be produced in relatively large quantities and isolated by affinity chromatography. In practice, the major histocompatibility complex class I molecule to be used in a direct binding assay comprises an isolated MHC-I Moiety. This MHC-I Moiety can be for example a native protein (for example, an isolated intact major histocompatibility complex class I molecule), a complexed protein (for example, an MHC-I molecule complexed with beta-2 microglobulin, or an major histocompatibility complex class I molecule that is non-covalently bound to avidin), or an engineered protein (for example, a chimeric protein consisting in part of an major histocompatibility complex class I molecule), or active fragment thereof, so long as the MHC-I binding site is present and available for binding by the test peptide. (ii) Test peptides are produced and are optionally labelled with a detectable moiety.

Peptides suitable for a direct binding assay are oligopeptides of between about 8 and 17 amino acids in length, although longer or shorter peptides may be used. Any suitable means of producing the peptide may be used. Examples of production means include but are not limited to: isolation from a natural biological source, conventional organic chemical synthesis; combinatorial synthesis using conventional organic chemistry techniques or biocatalysis or both; recombinant expression; and directed evolution methods such as but not limited to phage, ribosome, yeast, and bacterial surface display. Any suitable detectable moiety that renders the test peptide detectable may be used as described herein. Such detectable moieties may be directly attached to the test peptide, or attached through a linker, or attached by non-covalent methods as are known in the art. (iii) The binding affinity of the test peptide for the major histocompatibility complex class I molecule is determined by methods described herein or as are known in the art. For example, the equilibrium dissociation constant K_d is calculated from measuring labelled test peptide that directly binds to the major histocompatibility complex class I molecule. As one alternative, unlabelled test peptide is competed with the corresponding labelled test peptide for binding to the major histocompatibility complex class I molecule. The binding affinity of each test peptide to the MHC-I molecule is calculated from the inhibition concentrations (IC5₀). As another alternative, the on rate and off rate of unlabelled test peptide binding to the major histocompatibility complex class I molecule can be measured by surface plasmon resonance (SPR) techniques. The binding affinity profile thus obtained may be compared to that obtained by a predictive algorithm. (iv) Induction of cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides is evaluated for test peptides of different affinities.

Approach 3: Test peptide-induced cell surface expression of MHC-I molecules.

The stable expression at the cell surface of a major histocompatibility complex class I molecule as part of a properly assembled MHC-I/peptide complex requires that the MHC-I molecule binds a suitable peptide (Pamer & Cresswell, 1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). Peptides associated with MHC-I molecules are predominantly generated by proteasome degradation of cytosolic proteins, followed by translocation of the resulting peptides into the endoplasmic reticulum by the transporter associated with antigen processing (TAP) protein.

This approach utilizes cells that are defective in expression of TAP proteins and thus do not stably express major histocompatibility complex class I molecules at their surface. Cell surface expression of the major histocompatibility complex class I molecule can be phenotypically corrected by incubating the defective cells with test peptides that bind to the MHC-I molecule. An example of a suitable defective cell line is the murine TAPl- and TAP2- deficient lymphoma, RMA-S, which is a cell line derived from mice displaying the H-2^b haplotype (De Bruijn, 1991, Eur. J. Immunol, 21 :2963). Another suitable defective cell line is the human cell line T2, which displays the MHC-I molecule human leukocyte antigen A2.1 (HLA-A2.1), and is similarly defective in TAPl and TAP2 expression (Pamer & Cresswell, 1998, Annu. Rev. Immunol, 16:323; Shatri et al, 2002, Annu. Rev. Immunol, 20:463). RMA-S and T2 cells fail to express, respectively, H-2^b or HLA-A2.1, on the cell surface. In the case of RMA-S cells, H-2^b cell surface expression can be corrected by incubating the cells with H-2K^b and H-2D specific test peptides. An exogenous MHC-I gene can also be introduced into RMA- S cells, and the resulting engineered cells used in the identification of peptides that bind to the exogenous MHC-I molecule. For example, a desired haplotype of the human MHC-I molecules HLA- A, HLA-B, or HLA-C can be engineered as an exogenous gene into the murine RMA-S cells (Theobald et al, 1997, J Exp Med., 185:833). Incubating the engineered cells with test peptides that bind specifically to the exogenous MHC-I molecule can induce cell surface expression of the exogenous MHC-I molecule.

Test peptides, each including an amino acid sequence derived from any of the CHI, CH2, CH3, or CH4 domains, or from the membrane exons, of human immunoglobulin E heavy chain, and that are predicted to bind to the human major histocompatibility complex class I molecule HLA-A2.1, were produced. The HLA-A2.1 haplotype is a subtype that encompasses about 95% of the HLA-A2 allele, which is itself expressed in about one-third of the Caucasian population of the United States (Mori et al, 2002, electronic publication publicly available on the Internet websites www.ashi-hla.org/publicationfiles/archives/prepr/motomi.htm, www.ashi- hla.org/publicationfiles/archives/prepr/mori_gf.htm, www.ashi- hla.org/publicationfiles/archives/prepr/mori_ab.htm, and www.ashi- hla.org/publicationfiles/archives/prepr/mori_abd.htm).

The test peptides are chosen from nonameric sequences predicted by the algorithm for peptide binding to HLA-A2.1, using software publicly available on the website bimas.dcrt.nih.gov/molbio/hla_bind7index.html. This software is provided by the Biolnformatics & Molecular Analysis Section (BIMAS), Computational Bioscience and Engineering Lab, Center for Information Technology, National Institutes of Health. To date, about 180 binding motifs are known for the human MHC-I molecules HLA- A, HLA-B, and HLA-C (see for example, Wv .basic.nwu.edu/HLA/Reports/DoMotifList.cfin), and a similar analysis can be performed for any of the haplotypes for which the binding motif and algorithm is available.

A 428 amino acid sequence of human immunoglobulin E from the human IgE myeloma U266 was submitted to the algorithm for prediction of nonameric peptide sequences that bind to HLA-A2.1. Table I lists resulting nonameric test peptides, ranked according to their predicted half-time of dissociation to HLA-A2.1; only peptides with scores greater than 10 are listed. Details of the algorithm's calculations are publicly available on the website bimas.dcrt.nih.gov/molbio la_bind/hla_motif_search_info.html#Sect5. Table I

*estimate of half time of dissociation of the test peptide from HLA-A2.1

Five test peptides (peptide numbers 539.11, 539.12, 539.8, 539.9, and 539.10, representing three of the four CH domains of human IgE) from the panel of 12 high-, moderate-, and low-affinity test peptides shown in Table I were tested in RMA-S-A2.1 cells (murine TAP- deficient RMA-S cells that are engineered to express the human MHC-I HLA-A2.1). In addition, the nonameric peptide He Leu Lys Glu Pro Val His Gly Val (SEQ ID NO. 18), which is derived from the human immunodeficiency virus (HTV), was tested as an example of a nonameric HLA- A2.1 -specific peptide that is not derived from human IgE. RMA-S-A2.1 cells (about 1 x 10⁶) were incubated in duplicate cultures in 24-wells at 37 °C overnight with the individual test peptides at a final concentration of 50 micromoles per liter, in the presence or absence of human beta-2 microglobulin. Cells were pooled from the duplicate cultures and about 1 x 10⁶ cells were incubated with 2 micrograms per milliliter fluorescein isothiocyanate-labelled anti-HLA-A2.1 antibody (catalogue number 32294x, PharMingen) at 4 °C for 30 min. Fluorescence-activated cell sorting (FACS) analysis was performed on a FACScan (Becton-Dickinson). About ten thousand events were collected and analyzed by CELLQuest software (version 1.2.2, BD Biosciences). As depicted in FIG. 9, surface HLA-A2.1 molecules were detectable on RMA-S- A2.1 cells. Increase of HLA- A2.1 -specific fluorescence intensity by one log was observed in cultures incubated with any of the five human IgE test peptides tested (FIG. 9A, FIG. 9C, FIG. 9D, FIG. 9E, and FIG 9F). In contrast, no appreciable increase of HLA- A2.1 -specific fluorescence intensity was detected in cultures incubated with medium alone (FIG. 91). The increase of HLA- A2.1 -specific fluorescence intensity is comparable among the five IgE test peptides, and between the five IgE test peptides and the HIN control peptide (FIG. 9G). The observation that the IgE peptides represent different CH domains of human IgE provides a plurality of therapeutic targets for IgE-specific cytotoxic T-lymphocytes. The addition of exogenous human beta-2 microglobulin did not substantially enhance surface expression of HLA-A2.1 (FIG. 9A versus FIG 9B; FIG. 9G versus FIG. 9H), providing that these IgE peptides, which were predicted to bind strongly to HLA-A2.1, induced a high or possibly maximal expression of HLA-A2.1 at the cell surface.

Test peptides tested as above in RMA-S-A2.1 cells can be similarly tested in the human TAP-defϊcient T2 cells for their ability to cause cell surface expression of HLA-A2.1. Test peptides that show concordance in one or both tests of binding to HLA-A2.1 can be subsequently used for immunization experiments, for example in rodents transgenically expressing HLA-A2.1, as a further evaluation of the test peptides' therapeutic or prophylactic activity. The immunogenicity of a particular peptide can reflect its binding and stabilizing capacity for HLA- A2.1. Surface expression of MHC-I molecules is correlated with the ability of a test peptide to bind to the MHC-I molecule in the endoplasmic reticulum and the Golgi apparatus.

The detection of cell surface major histocompatibility complex class I expression by flow cytometric methods, such as fluorescence-activated cell sorting (FACS), and dissociation (Kd) measurements, can be configured into a high throughput screen. The induction of cell surface MHC-I expression for test peptides can be titrated over a range, for example between about 0.1 microgram to about 100 micrograms of test peptide per milliliter. The ease of induction can correlate with therapeutic effectiveness in vivo.

Approach 4: Determining a binding motif for a major histocompatibility complex class I molecule.

At present, binding motifs (and algorithms for binding affinity) are known or characterized for many but not all human and mouse major histocompatibility complex class I molecules. Few or no binding motifs are known for MHC-I molecules specific to mammals of economic importance such as those used in food production (including cattle, swine, sheep, and goats), or for mammals of domestic importance such as those that are common pets (including dogs, cats, horses, ferrets, rats, rabbits, guinea pigs, gerbils, and hamsters). For example, at present, only a few MHC-I binding motifs are known or characterized for cattle, horses, swine, dogs, cats, and rats, and none have been reported or characterized for ferrets, gerbils, and hamsters. An example of a procedure to determine the binding motif of a specific major histocompatibility complex class I molecule, such as of an uncharacterized or of a characterized MHC-I molecule, uses phage display and includes the following steps:

(i) A combinatory library of test peptides cloned into the PHI or PVIH coat proteins of the bacterial filamentous phage Ml 3 is prepared (Smith, 1985, Science, 228:1315).

(ii) The gene encoding the desired major histocompatibility complex class I molecule is cloned by polymerase chain reaction (PCR) techniques from the total messenger RNA (or oligo-dT selected mRNAs) by employing primers flanking the gene. The protein encoded by the isolated gene is produced in relatively large quantities and isolated by affinity chromatography. The isolated MHC-I molecules are biotinylated and immobilized on avidin-coated plates, (iii) The combinatorial test peptide library is panned on the immobilized MHC-I molecules. (iv) Phages that bind to the immobilized MHC-I molecules are retained and their displayed test peptides sequenced. (v) The resulting amino acid sequences are aligned and the frequency of occurrence of individual amino acids is determined at each position of the aptamer. (vi) These amino acid occurrence frequencies are weighted according to their position in the aptamer, taking into account the major and minor anchor residue positions, and a binding motif is determined. A second example of a procedure to determine the binding motif of a specific major histocompatibility complex class I molecule, such as of an uncharacterized or of a characterized MHC-I molecule, uses ribosomal display and includes the following steps (He, 1999, J. Immunol. Methods, 231:105):

(i) A combinatory library of polynucleotides encoding test peptides cloned into a polypeptide framework was prepared by ribosomal display techniques. Suitable polypeptide frameworks preferably contain at least one exposed loop region that is exposed to the hydrophilic environment, is preferably thermodynamically stable, and preferably can accept insertion of the immunogenic peptide or of the co- stimulatory peptide. Examples of such polypeptide frameworks include green fluorescent protein (GFP), staphylococcal nuclease, fibronectin, immunoglobulin (lg), and heat shock protein (HSP). The gene encoding the chosen polypeptide framework is prepared by optionally enhancing the transcription start site and translational site for either or both eukayotic and prokaryotic systems. A test peptide aptamer library was prepared using randomized oligonucleotides, cloned into the internal loop region of the polypeptide framework of green fluorescent protein (GFP). (ii) The gene encoding the desired major histocompatibility complex class I molecule is cloned by polymerase chain reaction (PCR) techniques from the total messenger RNA (or oligo-dT selected mRNAs) by employing primers flanking the gene. The protein encoded by the isolated gene is produced in relatively large quantities and isolated by affinity chromatography. The isolated major histocompatibility complex class I molecules are biotinylated and immobilized on avidin-coated plates, (iii) The ribosomal display test peptide library is panned on the immobilized MHC-I molecules.

(iv) Aptamers that bind to the immobilized MHC-I molecules are retained and their displayed test peptides sequenced. (v) The resulting amino acid sequences are aligned and the frequency of occurrence of individual amino acids is determined at each position of the aptamer. (vi) These amino acid occurrence frequencies are weighted according to their position in the aptamer, taking into account the major and minor anchor residue positions, and a binding motif is determined. A third example of a procedure to determine the binding motif of a specific MHC-I molecule, such as of an uncharacterized or of a characterized MHC-I molecule, includes the following steps (Falk et al, 199 , Nature, 351 :290) :

(i) A library of test peptides is prepared by combinatorial synthesis. The synthesized peptides vary from 5 to 17 amino acids in length, although longer or shorter peptides can be used, (ii) The gene encoding the desired MHC-I molecule is cloned by polymerase chain reaction (PCR) techniques from the total messenger RNA (or oligo-dT selected mRNAs) by employing primers flanking the gene. The protein encoded by the isolated gene is produced in relatively large quantities and isolated by affinity chromatography. The isolated MHC-I molecules are biotinylated and immobilized on avidin-coated plates. (iii) The combinatorially synthesized test peptide library is panned on the immobilized

MHC-I molecules, (iv) Test peptides that bind to the immobilized MHC-I molecules are retained and separated from the unbound test peptides. (v) The bound test peptides are eluted from the immobilized MHC-I molecules and subjected to pooled peptide sequencing, (vii) Pooled peptide sequencing reveals strong signals for amino acids at the major binding motif positions, whereas amino acids at variable binding positions occur at low or undetectable frequencies. The obtained amino acid sequences are aligned and an occurrence frequency for an individual amino acid estimated for each position, (vi) These amino acid occurrence frequencies are weighted according to their position, taking into account the major and minor anchor residue positions, and a binding motif is determined

This approach is advantageous because it allows the development of an immunoglobulin E peptide vaccine specific to major histocompatibility complex class I molecules of previously unknown binding specificity.

EXAMPLE 6: Transgenic animals and cells

The following example describes the use of transgenic animals and cells as sources of cytotoxic T-lymphocytes, as sources of target cells, and as models of modulation of immunoglobulin E-mediated conditions.

Transgenic mice

The therapeutic or prophylactic activity of immunoglobulin E peptides that are specific for an MHC-I molecule and are obtained through approaches such as those described in Example 5, can be further evaluated using a mammal, such as a mouse, that transgenically expresses that particular MHC-1 molecule. In one variant of this approach, a hybrid MHC-1 is prepared that contains the alpha- 1 and alpha-2 domains of the particular MHC-1 for binding the test peptide, as well as the alpha-3 domain of the host mammal (for example, a mouse), which permits optimal interaction with CD 8 molecules on cytotoxic T-lymphocytes. This hybrid MHC-1, expressed on target cells obtained from the transgenic mammal, can thus bind peptide as well as receive lytic signal from cytotoxic T-lymphocytes via cognate interactions (Minev et al, 2000, Proc. Natl. Acad. Sci, 97:4796; Theobald et al, 1997, J. Exp Med., 185:833). Mice can also be engineered to be doubly transgenic, that is, transgenic for a human MHC-I molecule (HLA haplotype) and transgenic for human immunoglobulin E. The human immunoglobulin E myeloma, U266, is an example of a ready source of rearranged human immunoglobulin E genomic DNA (Neuberger, 1995, Nature, 314:268), useful in constructing such transgenic mammals. For example, mice were engineered to transgenically express both human HLA-A2.1 and human immunoglobulin E (heavy chain constant region and membrane domain, fused into the murine VH gene). The resulting animals express human HLA-A2.1 at the cell surface, and produce the chimeric human/murine immunoglobulin E (comprising human IgE heavy chain and the endogenous murine IgE light chain) as well as the endogenous murine IgE. This is illustrated in FIG. 10.

The HLA-A2.1 and human IgE double transgenic mice are a source of HLA-A2.1 haplotypic human IgE-bearing B-lymphocytes and human IgE-secreting plasma cells. These cells are an example of target cells that display naturally processed and presented human immunoglobulin E peptides bound to a particular human major histocompatibility complex class I molecule expressed on the cell surface. The ability of cytotoxic T-lymphocytes to lyse these target cells can be tested in vivo and in vitro. These double transgenic mice also provide a pseudo-natural immunophysiological environment for testing immunoglobulin E peptides for the ability to break self-tolerance. "Triple feature" mice are a further development of the HLA-A2.1 (Jackson Laboratory,

Bar Harbor, ME) and human immunoglobulin E double transgenic mice described above. These mice are constructed by crossing the double transgenic mice to mice deficient in the murine immunoglobulin E gene (Oettgen et al, 1994, Nature, 370:367). The resulting triple feature mice produce only human IgE heavy chain paired with endogenous murine IgE light chain, but no endogenous murine IgE in the HLA-A2.1 environment. These mice are useful for testing immunoglobulin E peptides in a manner similar to the double transgenic mice, and similarly also provide a source of target cells useful for testing cytotoxic T-lymphocytes. Immunoglobulin E production is diminished in the double transgenic or three feature mice in vivo when these mice are immunized with an effectively immunogenic human IgE peptide specific for HLA-A2.1, since the cytotoxic T-lymphocytes elicited in these mice target IgE producing cells.

The target cells (HLA-A2.1 haplotypic human IgE-bearing B-lymphocytes and human IgE-secreting plasma cells) derived from the double transgenic or triple feature mice are also valuable as primary cell lines or as a foundation for permanent cell lines. These derived cell lines are useful for experiments in which the B-lymphocytes are stimulated for immunoglobulin E production, or where the cells serve as naturally processed and presented IgE peptide-displaying target cells for testing in vitro the ability of cytotoxic T-lymphocytes to lyse such target cells.

Illustrative examples of preparing immunoglobulin E cDNA constructs, in a conventional eukaryotic vector and in an adenoviral vector, follow, although other methods are available.

Such constructs are useful, for example, for transfecting and producing short-term and long-term IgE-producing cell lines.

One approach of preparing a human IgE gene in a transgenic vector includes the human IgE heavy chain constant region including the CHI, CH2, CH3, and CH4 domains physiologically rearranged with a VH gene of known specificity (Neuberger, 1995, Nature,

314:268). Another approach of preparing a human IgE gene in a transgenic vector includes the human IgE heavy chain constant region including the CHI, CH2, CH3, and CH4 domains and membrane exon 1 and exon 2 (Sun et al, 1991, J Immunol, 146,199; Zhang, 1994, J. Biol Chem., 269:456), fused with murine IgE heavy chain variable region (VH domain) of a known specificity (Neuberger, 1995, Nature, 314:268). The latter construct can be particularly useful for producing transgenic animals that express the membrane anchored exogenous IgE as well as secreted exogenous IgE upon induction by a specific allergen.

In one specific example, illustrated in FIG. 11, the construction of an antigen-specific, mouse/human chimeric huIgE gene including the membrane anchor sequence was performed according to the following steps:

(i) The genomic DNA was prepared from the human cell line U266. U266 is an IgE- producing myeloma (plasmacytoma) line that displays the HLA-A2.1 haplotype. An 8.5 kb DNA fragment encompassing the IgE heavy chain constant region and membrane domains, plus the enhancer regions and promoter, was amplified by long-range PCR using the primers 5' cctggtggagcgtgagtggcc (SEQ ID NO. 19) (forward direction), and 5' cctccacacagagcccatccgtcttc (SEQ ID NO. 20) (reverse direction). (ii) The mouse IgE heavy chain variable region (VH), encompassing the 5' regulatory region and a 3' EcoRI restriction site, was amplified from genomic DNA isolated from the nitrophenol-specific (NP-specific) murine cell line JW8. The primers used for PCR were 5' tatagtcgacaccatgggatggagctgtatc (SEQ ID NO. 21) (forward direction), and 5' taccgctgaaggttttgttgtcgac (SEQ ID NO. 22) (reverse direction). (iii) The murine IgE VH segment was ligated via the Sail restriction site near the beginning of the human IgE CHI domain. The resulting chimeric construct was cloned into the modified pUC19 vector, pHSE3', which contains the MHC-I H-2 promoter and immunoglobulin heavy chain enhancer region. The expressed chimeric heavy chain is expected to exhibit anti-NP specificity, and the affinity would be expected to be further improved when chimeric heavy chain is assembled with the IgE lambda-2 light chain (Neuberger, 1995, Nature, 314:268; Engel et al, 1998, Eur. J. Immunol, 20:2289).

U266 is also a source of IgE cDNA for preparing IgE-producing cell lines. U266 myelomas express five major species of spliced IgE messages and translated products, including intact membrane IgE and secreted IgE (Batista et al, 1995, J. Immunol, 154:209). Two of the five messages encode IgE lacking a full complement of CH4 and which are not believed to be appreciably secreted, or encode IgE with only partial membrane exon sequences, which which are not believed to be appreciably assembled. These IgE isoforms produced in U266 can serve as a rich source of degraded natural IgE peptides, some of which can bind selectively to HLA-A2.1 for presentation on the cell surface as targets of cytotoxic T-lymphocytes. Such peptides are expected to be useful for sustaining the immunosurveillance process by cytotoxic T-lymphocytes that recognize naturally processed and presented IgE peptides generated in situ. U266 cells were used for the cloning of human immunoglobulin E cDNA isoforms into oviral vector as follows:

(i) Human immunoglobulin E cDNA was amplified by reverse transcriptase- polymerase chain reaction (RT-PCR) from total RNA isolated from U266 cells. The first strand cDNA was synthesized using random primers and oligo-dT

(Gibco). Full-length IgE cDNA was amplified using the following primers: 5'gagagctccgttcctcaccatgg (SEQ ID NO. 23) (forward direction) and 5'cgtcatttaccgggatttacagacacc (SEQ ID NO. 24) (reverse direction). All IgE isoforms were verified by DNA sequencing. FIG. 12A illustrates the different IgE isoforms cloned in this manner.

(ii) pSml3, encoding full-length cDNA of human immunoglobulin E membrane long form was cloned into the conventional mammalian expression vector pShuttle (Clontech) using EcoRI/XhoI and the resulting construct used to transfect human kidney 293 cells. Transient expression of the pSml3 insert was confirmed by Western blot (FIG. 13A) using goat anti-human immunoglobulin E IgG (GAHE) as the primary antibody. Rabbit anti-goat IgG antibody, conjugated with horseradish peroxidase (HRP) was used as the secondary antibody. The blot was developed using 3,3',5,5'-tetramethylbenzidine (TMB) (Kirkegaard & Perry Laboratories) as a substrate. The 65 kDa protein species, encoded by full length classical long from IgE in pShuttle was translated in 293 cells (human transformed primary embryonal kidney cells). This transcript substantially conformed to the predicted molecular weight with full membrane-encoded exon sequences.

(iii) pSml3, encoding full-length cDNA of human immunoglobulin E membrane long form was similarly cloned into the adenoviral vector pAd-lacZ using the Adeno X kit (Clontech). FIG. 12B illustrates the adenoviral IgE construct. The viral stock from primary amplification was subject to SDS-PAGE, and then followed by western blot (FIG. 13B) as described above. In apparent contrast to the results obtained from the conventional mammalian expression vector pShuttle, the heavy chain gene product expressed in adenoviral vector appeared as the same molecular weight but migrated as a broader band, suggesting adequate glycosylation.

Target cells employing cell lines, including PBMC in vitro

Transgenic mice expressing an exogenous MHC-I molecule (for example, the human MHC-I haplotype HLA-A2.1) can be immunized with a selected immunoglobulin E peptide, and the resulting cytotoxic T-lymphocytes subsequently tested for the ability to lyse IgE producing cells and to inhibit IgE production. FIG. 14 illustrates two approaches, each using a different type of target cell.

In the first approach, a cytotoxic T-lymphocyte response specific for immunoglobulin E peptides naturally processed and presented on the MHC-I molecule HLA-A2.1 is generated in mice that have been engineered to transgenically express HLA-A2.1. Cytotoxic T-lymphocytes are prepared from spleens harvested from the immunized mice. The target cells used in this first approach are cells in which the MHC class I molecule is contacted with test peptide exogenously. As shown in FIG. 14A, a specific example of such target cells are TAP deficient RMA-S cells, which transgenically express HLA-A2.1. These target cells are incubated overnight with 10 nanograms to 100 micrograms of individual test peptides (chosen from Table I) in a 10% CO₂ incubator. The target cells are then harvested and labelled for 90 minutes with ⁵¹Cr. The peptide- treated, chromium-labelled target cells are serially diluted and added to cytotoxic T-lymphocytes in 96-well plates. After a four-hour incubation at 37 °C, supernatants are removed and the chromium release is counted to indicate the extent of cell lysis.

In the second approach, the target cells that are used synthesize immunoglobulin E which is naturally processed and presented as peptides on the major histocompatibility complex class I molecule. FIG. 14B illustrates a specific example of this second approach, in which the target cells are the human immunoglobulin E-producing myeloma U266, which has the HLA-A2.1 haplotype. The U266 subclones SKO-007 and AF-10 may be similarly used as target cells.

Human peripheral blood mononuclear cells (PBMC), like the U266 cell line, express a range of spliced IgE messages and translated products (Batista et al, 1995, J. Immunol, 154:209). The information obtained by using U266 target cells regarding efficacies of HLA- A2.1 -restricted test peptides is therefore relevant for studying efficacies of test peptides in human peripheral blood. IgE-specific cytotoxic cells from HLA-A2.1 transgenic mice, such as those described above, are restricted to human HLA-A2.1, and they will not confer xenogenic reactivities to PBMC exhibiting HLA-A2.1 due to self-tolerance. The methods described herein can be used to inhibit IgE production or to lyse IgE-producing tumor cell lines (such as U266) as well as inhibiting or lysing IgE-producing cells (such as human PBMC cultures following allergen challenges or mitogen challenges in the presence of interleukin-4).

The differentially spliced IgE cDNA constructs such as those described above and illustrated in FIG. 12, cloned into a conventional eukaryotic (mammalian) vector or an adenoviral vector, can be used to transfect an appropriate cell line to yield novel target cells. For example, permanent tumor cell lines of diverse tissue origins including the lymphoid/myeloid lineages are transfected with an IgE cDNA construct. These cell lines can then be further transfected with an exogenous MHC-I gene. The immunoglobulin E peptides generated in these transfected cells are naturally processed and presented by the exogenous MHC-I molecules. Alternatively, a test primary cell culture or cell line is transfected with an IgE cDNA constructs such as those described above, as well as with an exogenous MHC-I gene. This test cell type is not limited to lymphoid cell types, so long as the transfected cells are capable of synthesizing the exogenous IgE to IgE peptides that are naturally processed and presented by the exogenous MHC-I molecules. Target cells derived from IgE producing cells from double transgenic mice

Mice expressing U266-derived IgE transgene (of an unknown VH specificity) are first immunized with IgE peptides, using appropriate co-stimulation (such as helper T-lymphocyte co- stimulatory epitopes). It has previously been shown that immunizing mice with goat antibodies against murine IgD ("GAMD") or with goat antibodies against murine IgE ("GAME") induces polyclonal IgE production in the mice, due to antigenic signals delivered by mature B cells whose surface IgD or IgE is perturbed, as well as to co-stimulation delivered by IL-4-producing Th2 specific for foreign goat IgG determinants (Katona et al, 1991, J Immunol,. 146:4215). The protective effect of immunization with IgE peptides can be evaluated by stimulating polyclonal IgE production in the U266 transgenic mice in vivo with an antigen surrogate such as GAMD or goat anti-human-IgE ("GAHE"). Upon stimulation with an anti-immunoglobulin as the antigen surrogate, mature B-lymphocytes are polyclonally activated and express IgE at high levels. As a result higher levels of naturally processed and presented IgE peptides, such as from the catabolism of IgE in the stimulated B-lymphocytes or plasma cells, are displayed on the surface of these cells. These activated B-lymphocytes and plasma cells generated in vivo are rendered more susceptible to lysis by cytotoxic T-lymphocytes specific for the IgE peptides used in the initial immunization. The downregulation of these IgE producing cells serves as a measurement of the IgE-specific cytotoxic T-lymphocyte response.

To test the effect of cytotoxic T-lymphocytes on normal IgE producing cells in vitro, single splenic cell cultures, prepared from IgE peptide immunized mice are re-stimulated with IgE peptide and appropriate co-stimulation (such as helper T-lymphocyte co-stimulatory epitopes). The activated cytotoxic T-lymphocytes thus obtained are then added to splenic cultures from double transgenic mice, stimulated in vitro with lipopolysacharide (LPS), pokeweed mitogen (PWM), antigen surrogates, or IL-4. Levels of IgE in vitro are then determined.

To test the lysis of target cells in vitro, single cell suspensions are prepared from the spleens of double transgenic mice immunized with IgE peptide as above. The cells are re- stimulated in vitro with the IgE peptide and the co-stimulatory peptide. Activated cytotoxic T- lymphocytes thus obtained are then added in various ratios to ⁵¹Cr-labeled and allergen-activated B-lymphocytes and to 51Cr-labelled and LPS/dextran sulfate-activated B-lymphocytes. The percentage of chromium release of the lysed B-lymphocytes is determined.

Target cells derived from NP-specific IgE transgenic mice

In contrast to the above transgenic mice which respond to GAME as an antigenic surrogate, transgenic mice that express nitrophenol (NP)-specific chimeric mouse/human IgE respond to nominal antigens or allergens such as nitrophenol conjugated to the carrier proteins keyhole limpet hemocyanin (KLH) or bovine serum albumin (BSA). These NP-specific chimeric IgE transgenic mice can be first immunized with IgE peptides, using appropriate co-stimulation (such as helper T-lymphocyte co-stimulatory epitopes). To determine the protective effect of immunization by the IgE peptides, the immunized mice are subsequently primed with between about 1 to about 100 micrograms NP-KLH, in preferably about 2 mg alum, but other amounts are appropriate, and then challenged with the same or similar antigens in alum. Levels of NP- specific IgE in allergen-challenged mice will be evaluated. High levels of naturally processed and presented IgE peptides are expressed by allergen-activated NP-specific IgE-bearing B- lymphocytes and NP-secreting plasma cells. Consequently, the allergen-activated NP-specific IgE-bearing B-lymphocytes and NP-secreting plasma cells are susceptible to lysis by cytotoxic T- lymphocytes specific for the IgE peptides used in the initial immunization. Inhibition of IgE production in transgenic B-lymphocytes can be a result of the illustrative methods of direct lysis of the B-lymphocytes, of inhibition of IgE production at the transcriptional and/or translational level, of skewing the ratios of normally spliced epsilon messages toward abnormally spliced messages, or of a failure of IgE to be assembled on the cell surface or to be secreted. However, the inventors do not intend to be limited to such illustrative methods or mechanisms.

Mast cells as target cells

Bone marrow-derived mast cells (BMMC) from the double transgenic mice are cultured with interleukin-3 (IL-3), granulocyte/monocyte colony stimulating factor (GM-CSF) (Franji et al, J. Immunol, 151 :6318; Chen, 2000, Immunol, 100, 471), and stem cell factor (SCF). BMMC that display naturally processed and presented IgE peptides on HLA-A2.1 can be used as targets for cytotoxic T-lymphocytes using a chromium release assay. Also, secretion of histamine, leukotrienes, and cytokines (such as GM-CSF, IL-4, interferon-gamma, and tumor necrosis factor alpha) is decreased in IgE-sensitized and allergen-challenged mast cells following incubation with cytotoxic-T lymphocytes.

In a variation of this example, BMMC cultured from HLA-A2.1 single transgenic mice, are contacted with IgE whole molecules and allergens. Immunoglobulin E and allergens form IgE-allergen complexes bound to FceRI receptors on the surface of the mast cells. The mast cells are then incubated with IgE peptide-specific cytotoxic T-lymphocytes. A chromium release assay and inhibition of secretion of small molecules and cytokines are measured as above as an indication of the cytotoxic T-lymphocytes' ability to lyse mast cells.

Inhibition of endogenous murine IgG production Since a majority of B-lymphocytes in the double transgenic mice exhibits the exogenous human IgE as well as the endogenous murine surface immunoglobulins of different isotypes, natural IgE peptide-specific cytotoxic T-lymphocytes are tested for their ability to decrease endogenous murine IgE/IgG production. In contrast, production of chimeric antigen-specific human IgE as well as endogenous murine IgG production will not be affected in mice immunized with helper T-lymphocyte peptide alone. Antigen-specific and total human IgE and murine IgG production is evaluated by ELISA or ELISA plaque assay (Chen, 1990, J. Immunol. Meth., 135:129).

Determining the optimal natural peptides in IgE and HLA double transgenic mice Mice that are doubly transgenic for human IgE and HLA-A2.1 are an alternative system for testing the "naturalness" of the test peptides such as those obtained in Example 1. Human IgE-specific cytotoxic T-lymphocytes are readily induced in mice transgenic for human IgE, because human IgE peptides are foreign to the murine system. However, such cytotoxic T- lymphocytes may be naturally tolerized in the human population that expresses both huIgE and HLA-A2.1. Like ssomatic antigens, human IgE peptides naturally presented on HLA-A2.1 in transgenic mice are considered "self antigens by the mice. IgE is a macromolecule and there should be a few natural IgE peptides that can be generated and presented by HLA-A2.1. The double transgenic mice therefore provide a pertinent model as to whether cytotoxic T- lymphocytes for a particular natural IgE peptide can be efficiently induced.

In Example 3, IgE +/+ (wild type) and IgE -/- mice showed approximately the same magnitude of IgE-specific cytotoxic T-lymphocyte response. In the case of the specific IgE peptide used in Example 3, self tolerance does not appear to play a major role in preventing induction of a CTL response. However, it is possible that different degrees of tolerance can exist with respect to a diverse spectrum of IgE natural peptides. For example, a high-affinity IgE peptide that is immunogenic in HLA-A2.1 transgenic mice could cause profound clonal deletion in double transgenic mice. In contrast, an IgE peptide of intermediate binding affinity could induce a response with appropriate modes and levels of co-stimulation. An efficacious IgE peptide vaccine strikes a balance between the IgE peptides' binding affinity and the capacity of the partially tolerant cytotoxic T-lymphocytes to respond to appropriate co-stimulation. An efficacious IgE peptide vaccine candidate can be achieved by striking a balance between the inherent properties of an antigenic epitope and the state of partial tolerance or ignorance of the immune system. This can require that (1) the IgE peptide exhibits moderate to high affinity toward the cytotoxic T-lymphocytes (but not to the extent of causing cloning inactivation of these T-lymphocytes), and (2) maintaining receptiveness of these IgE peptide-specific T-lymphocytes to concomitant T-lymphocyte co-stimulation, so that cytotoxic T-lymphocytes can be induced but are not anergized by encountering high affinity, self antigenic peptides in the absence of adequate co-stimulation. Double-transgenic mice could also serve as a useful pre-clinical animal model for assessment of the immunogenicity and inducibility of a IgE natural peptide as an IgE vaccine for humans. Managing the IgE-specific cytotoxic T-lymphocyte response by controlling delivery of costimulation

Maintaining inducibility by T-lymphocyte co-stimulation, or a state of partial tolerance of an immunoglobulin E peptide vaccine, also permits the safety feature that normal levels of IgE can be produced in vaccinated subjects during the seasons when environmental allergens are reduced. Permanent deletion of immunoglobulin E responses is not anticipated, since the IgE peptide-specific cytotoxic T-lymphocyte response is believed to be induced and maintained in the . subject when the immunogenic IgE peptide is delivered together with appropriate co-stimulation. The sustainability of an immunoglobulin E peptide antigenic stimulation is believed to be enhanced by cognate T-lymphocyte co-stimulation or by an appropriate antigen-presenting cell such as a dendritic cell. To improve the efficiency of delivering cognate co-stimulation, cytotoxic IgE natural peptides can be provided along with a co-stimulatory peptide (such as a helper T-lymphocyte epitope) in a longer polypeptide framework, or as a product of a mini-gene encoding both the immunogenic IgE peptide and the co-stimulatory peptide. Dendritic cells, which play a role in the in vivo adaptive immune response (Banchereau et al, 1998, Nature, 392:245), are a rich source of co-stimulation and are known to render self-peptides immunogenic, that is to say, they are capable of breaking self-tolerance without the need for a foreign or non-self T-lymphocyte helper epitope (Inaba, 1990, J. Exp. Med., 172:631). Tissue distribution of dendritic cells is believed to provide the spreading of immunity, and is believed to protect bodily zones not directly sensitized by antigen (Banchereau et al., 1998, Nature 392, 245). Therefore, dendritic cells that have been immunized with IgE peptides, optionally with co- stimulation, are expected to be efficient for breaking tolerance and maintaining the inducibility of the IgE peptide-specific cytotoxic T-lymphocyte response. Cytotoxic T-lymphocytes to natural immunoglobulin E peptides become dormant when uncoupled from the interaction with in vivo administered DC. EXAMPLE 7: Immunoglobulin E sequences as candidates for inducing a cytotoxic T- lymphocyte response

The following example describes examples of nucleic acid and amino acid sequences of immunoglobulin E from various mammalian species, including mammals of economic importance, mammals of domestic importance, non-human primates, and humans, and methods of identifying them, that are useful in the present invention.

The amino acid sequences for immunoglobulin E heavy chain, including complete CHI to CH4 constant region domains, have been determined for cat (Felix catus) (SEQ ID NO. 25) (Weber et al, 2000, Vet Immunol. Immunopathol, 76:299), chimpanzee (Pan troglodytes) (SEQ ID NO. 26) (Sakoyama et al, 1987, Proc. Natl Acad. Sci, 84:1080), cow (Bos taurus) (SEQ ID NO. 27) (Rabbani et al, 1996, Department of Bioscience at NOVUM, Karolinska Institute, Huddinge S-141 57, Sweden, direct submission to NCBI, accession number gi:1575497), dog (Canis familiaris) (SEQ ID NO. 28) (Patel et al, 1995, Immunogenetics, 41:282), duckbilled platypus (Orniihorhynchus anatinus) (SEQ ID NO. 29) (Aveskogh et al, 2001, Dept. of Cell and Molecular Biology, Uppsala University, BMC, Box 596, Uppsala SE-75124, Sweden, direct submission to NCBI, accession number # gi: 17223803), horse (Equus caballus) (SEQ ID NO. 30) (Navarro et al, 1995, Mol. Immunol, 32:1), mouse (Mus musculus) (SEQ ID NO. 31) (Nikaido et al, 1982, J. Biol. Chem., 257:7322), opossum (Monodelphis domestica) (SEQ ID NO. 32) (Aveskogh et al, 1998, Eur. J. Immunol, 28:2738), pig (Sus scrofa) (SEQ ID NO.33) (Vernersson et al, 1997, Immunogenetics, 46:461), rat (Rattus norvegicus) (SEQ ID NO. 34) (Kindsvogel et al, 1982, DNA, 1:335), sheep (Ovis aries) (SEQ ID NO. 35) (Engwerda et al, 1992, Vet. Immunol. Immunopathol, 34:115), brushtail possum (Trichosurus vulpecula) (SEQ ID NO. 36) (Belov et al, 1999, Mol. Immunol, 36:1255), and human (Homo sapiens) (SEQ ID NO. 37) (Kenton et al, 1982, Proc. Natl. Acad. Sci, 79:6661). The amino acid sequences for immunoglobulin E heavy chain variable region (VH) have also been partially determined for cat (SEQ ID NO. 25), and completely determined for Orniihorhynchus anatinus (duckbilled platypus) (SEQ ID NO. 29), horse (SEQ ID NO. 30), pig (SEQ ID NO. 33), sheep (SEQ ID NO. 35), and human (SEQ ID NO. 37). The amino acid alignment of the IgE heavy chain sequences for human, sheep, rat, pig, mouse, horse, duck, dog, cow, chimpanzee, and cat is depicted in FIG. 15 A, FIG. 15B, FIG. 15C, and FIG. 15D, which show sequence homology of 60% or more between IgE from different species.

A phylogenetic tree, depicted in FIG. 16, representing the inferred evolutionary divergence between the immunoglobulin E heavy chain polypeptides of human, sheep, rat, pig, mouse, horse, duck, dog, cow, chimpanzee, Trichosurus (brushtail possum), and cat, was calculated from the above sequences using the progressive multiple sequence alignment program ClustalW (Higgins et al, 1994, Nucleic Acids Res., 22:4673). This approach uses two-stage computation, that is to say, calculation of the distances between pairs of sequences and reconstruction of the phylogeny using the distance information. The phylogenetic free is built by joining the two most similar sequences first, and then adding the other sequences one by one, in order of decreasing similarity. The numerical value on top of each line indicates the evolutionary distance for that respective species from the most recent common ancestor of IgE. According to these results, the human IgE sequence is most closely related to the chimpanzee, and then more distantly related to (in order of increasing distance) cat, dog, horse, pig, sheep, cow, Trichosurus, opossum, and duckbilled platypus. Rat and mouse IgE are believed to be most primitive (that is to say, least diverged from the most recent common ancestor) with regard to IgE molecular speciation.

These results, depicted by FIG. 15 and FIG. 16, show an extent of sequence identity and homology between the IgE sequences from different species that suggests that MHC-I-specific epitopes that are recognized by IgE-specific cytotoxic T-lymphocytes can be overlapping or similar or identical between two or more mammalian species. The MHC-I specificity of each such epitope is also dependent on the peptide-binding motifs of the respective MHC-I. FIG. 17A depicts various building blocks or spliced blocks of the human immunoglobulin E heavy chain gene. The amino acid sequences of these blocks are supplied for the membrane exon sequences Ml' (SEQ ID NO. 38), M2 (SEQ ID NO. 39), and M2' (SEQ ID NO. 40), and for the constant domain sequences CH4s (SEQ ID NO. 41), CH4 (SEQ ID NO. 42), CH4' (SEQ ID NO. 43), CH5 (SEQ ID NO. 44) of human immunoglobulin E heavy chain. The corresponding cDNA sequences of these blocks are supplied for the membrane exon sequences Ml' (SEQ ID NO. 45), M2 (SEQ ID NO. 46), and M2' (SEQ ID NO. 47), and for the constant domain sequences CH4s (SEQ ID NO. 48), CH4 (SEQ ID NO. 49), CH4' (SEQ ID NO. 50), CH5 (SEQ ID NO. 51) of human immunoglobulin E heavy chain. These blocks may be spliced together in different orders, and those variants that are productively spliced (that is to say, translated into an IgE peptide product that is expressed at the cell surface, secreted, or displayed as a naturally processed and presented peptide on an MHC-I molecule at the cell surface) can serve as a source for deriving a sequence of a test peptide that preferably has the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides.

Various differentially spliced human IgE isoforms were constructed (FIG. 12A) and reported according to the nomenclature of Zhang et al. (Zhang et al, 1992, J. Exp. Med., 176:233; Zhang et al, 1994, J. Biol Chem., 269:456). These variant IgE polypeptides can incorporate a proximal 3' region (CH5) or distal 3' region (M2'). Various differentially spliced human IgE isoforms were constructed and reported according to the nomenclature of Zhang et al. (Zhang et al, 1992, J. Exp. Med., 176:233; Zhang et al, 1994, J. Biol. Chem., 269:456). These differentially spliced human IgE isoforms utilize different starting regions and combinations of membrane exons as depicted in FIG. 17A. Schematic diagrams of examples of these translated spliced human IgE messages are depicted in FIG. 17B. The abundant classical secreted human IgE isoform includes complete CHI to CH4 constant region domains, as described above for various species. A less abundant secreted human IgE isoform contains a truncated CH4 (CH4') domain, and incorporates some distal 3' sequence (CH5) preceding the regular membrane exon region, or extending into a long M2' region. A short secreted human IgE isoform contains a truncated CH4 (CH4') and CH5 domains (Zhang et al, 1994, J Biol Chem., 269:456). There are two classical plasma membrane- anchored IgE isoforms, a large membrane isoform containing the Ml ' and M2 domains, and a short membrane isoform containing the Ml and M2 domains. There also is a rare M2" membrane form that contains the CH4 domain and a truncated M2" domain (Peng et al, 1992, J. Immunol, 148:129; Zhang et al, 992, J. Exp. Med., 176:233). The fully assembled products with differentially spliced frames can result in differing translated sequences due to shifting of the reading frame, even if the spliced messages overlap a large part of the exon sequence. IgE peptides, generated by the different isoforms and differing in the amino acid sequences of CH5 and membrane exons, can provide additional therapeutic targets for cytotoxic T-lymphocytes. Individuals, such as allergic patients, may have diverse IgE isoforms, which can serve as additional targets against which a cytotoxic T-lymphocyte response can be induced.

EXAMPLE 8: Polynucleotides encoding immunogenic IgE peptides inserted into a polypeptide framework

The following example describes a polynucleotide that encodes an immunogenic peptide inserted into a polypeptide framework itself encoded by the polynucleotide. This polynucleotide is useful, for example, as a DNA vaccine. The encoded immunogenic peptide inserted into the encoded polypeptide framework is also useful as a vaccine. Green fluorescent protein as a polypeptide framework for an immunogenic peptide

Green fluorescent protein (GFP) is a 238-amino acid protein found in the jellyfish Aequorea victoria (Tsien, 1998, Annu. Rev. Biochem., 67:509; Ormδ et al, 1996 Science, 273:1392; Chalfie et al, 1994, Science, 263:802). GFP is used in the laboratory, for example as a visible marker for polypeptides or as a marker for gene expression. The GFP mutant GFPuv has the same fluorescence characteristics as the wild-type GFP (an excitation wavelength of about 395 nanometers and an emission wavelength of about 509 nanometers), but produces a 45- fold greater fluorescence signal (Crameri et al, 1996, Nature Biotech., 14:315). GFPuv was chosen because it typically has greater expression rates, intracellular yield, and infracellular solubility than the wild-type GFP (Crameri et al, 1996, Nature Biotech., 14:315). Monitoring the conformation of the GFPuv framework can be done by fluorescence measurements. GFPuv can be cloned into pCDNA3.1 for use as a DNA vaccine (Donnelly et al, 1977, Annu. Rev. Immunol, 15:617; Gurunathan et β/., 2002, Annu. Rev. Immunol, 18:927), which is effective in eliciting a cytotoxic T-lymphocyte response upon different methods of administrations. The structure of GFPuv consists of an eleven-strand beta-barrel wrapped around a central alpha-helix core. The barrel is formed by antiparallel beta-sheets interconnected by short alpha- helical "loops", which are exposed to the aqueous environment. The conformation of these loops is constrained by the overall beta-barrel structure of the protein, which is highly stable and resistant to distortion. Epitope peptide sequences, for example both cytotoxic T-lymphocyte epitopes and helper T-lymphocyte epitopes, can be inserted into these loops, where they maintain their immunogenicity (Billetta et al, 1991, Proc. Natl Acad. Sci, 88:4713; Gerloni, et al, 1997, Nature Biotechnol, 15:876; Gerloni et al, 1999, J. Immunol, 162:3782; Zaghouani et al, 1993, Science, 259:224; Zaghouani et al, 1993, Eur. J. Immunol, 23:2746). Such inserted peptide sequences are exposed to the aqueous environment and can be relatively accessible to proteolytic degradation. Insertion in these GFP loops can extend the half-live of the cytotoxic T-lymphocyte IgE peptides thus expressed. The GFP chromophore is a modified tripeptide held in the interior of the beta-barrel, and its fluorescence is predictive of the protein's unperturbed conformation (Tsien, 1998, Annu. Rev. Biochem., 61: 509; Ormδ et al, 1996 cience, 273:1392). FIG. 18 depicts the modified GFPuv vector used to express cytotoxic T-lymphocyte IgE peptides by a three-step PCR strategy. We modified the GFPuv vector rGFP-S65T (catalogue number #8364-1, Clontech, CA) by shortening the N-terminus lacZ fusion protein sequence, deleting all but the first 7 amino acids following the lacZ A C initiation codon. Additional restriction enzyme sites were inserted into selected sites of two surface-exposed, highly constrained loops. These were site 6 in loop 6 (encompassing Ala 155 to He 161) (FIG. 18A) and site 7 in loop 7 (encompassing Lys 162 to Asp 183) (FIG. 18B). Site 6 is at Gin 157 in loop 6; the antisense primer ttgtttgtctgccgtgatgtatac (SEQ ID NO. 52) and sense primer tgaggtaccgcgcgcaagaatggaatcaaacgtaacttc (SEQ ID NO. 53), which contains Kpn I and BssH I restriction sites, were used at site 6 (FIG. 18A). Site 7 is at Glu 172 in loop 7; the sense primer tgatctagactgcaggatggattcgttcaactagcag (SEQ ID NO. 54), which contains Xba I and Pst I restriction sites, and common antisense primer tgatgatgatgatgatagagctcatccatgccatgtg (SEQ ID NO. 55), which contains a Sac I restriction site, were used at site 7 (FIG. 18B). These specific restriction enzyme sites were chosen for efficient end cutting, and ligation was carried out with oligonucleotides specifying defined sequences. Primers with restriction enzyme sites of interest, flanked by a partial GFPuv open reading frame containing loops 6 and 7, were used for amplifying about half of the GFPuv polypeptide. Intact GFPuv, containing immunoglobulin E epitopes inserted at site 6 or at site 7, was then made by ligating two pieces of the amplified PCR products. The annealed product was amplified by another PCR amplification reaction to yield the final modified vector, which contained immunoglobulin E peptides cloned into GFPuv framework loops.

FIG. 19 depicts restriction enzyme digests of two clones (clone number 6/4 and clone number 6/5) obtained by the above three-step PCR strategy. These digests verified that the modified vector contained all the required sites as designed. Correct size of fragments, as predicted from the vector depicted in FIG. 18A and FIG. 18B, was observed for both clones following digestion by either Sphl and Bgl H or by Sphl and BssH I.

A series of six oligonucleotides encoding immunoglobulin E cytotoxic T-lymphocyte epitopes with Bgl H and Kpn I ends were cloned as inserts into site 6 of the GFPuv vector. Restriction digests of these site 6 clones (clones Ce3-BC, Ce3-DE, Ce3-FG, Ce2-3, Ce4-dest, and Ce4-crea) are depicted in FIG. 20A. The clones were expressed in E. coli JM109 cells and the cell lysates screened by Western blot, the results of which are depicted in FIG. 20B. The primary antibody was polyclonal rabbit anti-GFP (catalogue number #8367-2, Clontech, CA) and the secondary antibody was horseradish peroxidase labelled-goat anti-rabbit antibody (catalogue number #611-1322, Rockland Immunochemicals, PA). The cloned inserts were sequenced to verify the inserted epitopes. FIG. 20C depicts the nucleotide sequence of the six cloned immunoglobulin E cytotoxic T-lymphocyte epitopes. A parallel set of clones into site 7 of the GFPuv vector was also constructed. The site 6 and site 7 clones are tested for the ability to elicit a cytotoxic T-lymphocyte response against IgE naturally processed and presented on MHC-I molecules, for example on IgE-producing cells, or against IgE-bearing cells such as mast cells and basophils. In addition to GFP and its variants and mutants, other polypeptides (and the polynucleotides encoding them) or macromolecules can be used in an analogous manner to serve as frameworks for the IgE peptides. Examples of these include, but are not limited to, staphylococcal nuclease, fϊbronectin, immunoglobulin (lg), heat shock protein (HSP), a phage, a ribosome, mRNA, a yeast cell surface, a bacterial cell surface, or a bacterial spore surface. Purification of GFP and antigenized GFP

GFPuv contains ten histidine residues which can interact with with metals such as Ni(H). In particular, His77, Hisδ 1 , and His231 are oriented within 7.5 Angsfrøm of each other near the top of the beta-barrel and can serve as a natural "histidine tag" for metal chelation. We purified both GFPuv and GFPuv antigenized with human IgE peptides from bacterial lysates by immobilized metal affinity chromatography on a nickel column. Cell pellets prepared from 500 ml LB medium were treated with 1 milligram per milliliter lysozyme. Cytosolic supernatants of lysates without inclusion bodies were collected by centrifugation, and dialyzed in 0.05 molar phosphate-0.25 molar NaCl buffer. The material was allowed to pass through the nickel column. Fluorescent material was bound to column as a visible band and was eluted with 2 molar sodium chloride. Throughout the purification process, the GFPuv-enriched fraction was conveniently monitored by eye or visualized by illumination with a handheld UV lamp emitting at 365 nanometers. GFPuv was found to be homogenously distributed in the cytoplasm and the fluorescence-emitting material was obtained in the lysates and not in the inclusion bodies, which permitted a relatively high yield (50 milligrams protein per liter) of purified native protein. GFPuv represented about 70% of the harvested protein from the lysates. GFPuv and the antigenized GFPuv can be further purified to homogeneity by ion-exchange chromatography, such as by FPLC on a mono-Q column, or by other methods known in the art. The purifed proteins can be used to immunize a mammal and to elicit a cytotoxic T-lymphocyte response.

All publications, including patent documents and scientific articles, referred to in this application and the bibliography and attachments are incorporated by reference in their entirety for all purposes to the same extent as if each individual publication were individually incorporated by reference.

All headings are for the convenience of the reader and should not be used to limit the meaning of the text that follows the heading, unless so specified.

Claims

What is claimed is:

I . A method of identifying peptides that induce a cytotoxic T-lymphocyte response against immunoglobulin E, comprising: a. providing at least one test peptide suspected of being able to bind to at least one major histocompatibility complex class I molecule, and b. evaluating said at least one test peptide for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, wherein a peptide that induces such a response is identified. 2. The method of claim 1, wherein said at least one test peptide is naturally occurring, synthesized, combinatorially synthesized, or biologically produced. 3. The method of claim 1, wherein said mammal is selected from the group consisting of dog, cat, horse, cow, pig, sheep, goat, ferret, mouse, rat, rabbit, guinea pig, gerbil, and hamster. 4. The method of claim 1 , wherein said mammal is a non-human primate.

5. The method of claim 1, wherein said mammal is a human.

6. The method of claim 1, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen.

7. The method of claim 1, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen that corresponds to a specific human leukocyte antigen haplotype.

8. The method of claim 1 , wherein said at least one test peptide is a peptide of between about five to about two thousand amino acids in length.

9. The method of claim 8, wherein said at least one test peptide is linear. 10. The method of claim 1, wherein said at least one test peptide is an oligopeptide of between about five to about twenty amino acids in length.

I I. The method of claim 10, wherein said at least one test peptide is linear.

12. The method of claim 1, wherein said at least one test peptide comprises an amino acid sequence derived from immunoglobulin E.

13. The method of claim 1, wherein said at least one test peptide comprises an amino acid sequence derived from any of the CHI, CH2, CH3, or CH4 domains, or from the membrane exons, of human immunoglobulin E heavy chain.

14. The method of claim 1, wherein the binding of said at least one test peptide to said major histocompatibility complex class I molecule is predicted by an algorithm.

15. The method of claim 1, wherein said evaluating comprises the steps of: a. contacting said at least one test peptide with mammalian cells defective in the expression of a cell surface major histocompatibility complex class I molecule, and b. selecting from said test peptides at least one peptide that causes expression of said cell surface major histocompatibility complex class I molecule in said mammalian cells.

16. The method of claim 15, wherein said defect is due to a mutation of the TAPl or of the TAP2 loci. 17. The method of claim 1, wherein said evaluating comprises the steps of: a. contacting said at least one test peptide with said major histocompatibility complex class I molecule, and b. selecting from said test peptides at least one peptide that binds to said major histocompatibility complex class I molecule. 18. The method of claim 17, wherein said major histocompatibility complex class I molecule comprises an isolated major histocompatibility complex class I moiety. 19. The method of claim 17, wherein said major histocompatibility complex, class I molecule is expressed on a cell surface.

20. The method of claim 1, wherein said evaluating comprises the steps of: a. contacting said at least one test peptide with mammalian cells defective in the expression of a cell surface major histocompatibility complex class I molecule, b. selecting from said test peptides at least one peptide that causes expression of said cell surface major histocompatibility complex class I molecule in said mammalian cells, c. preparing cytotoxic T-lymphocytes that recognize said at least one selected peptide, and d. testing said cytotoxic T-lymphocytes for the ability to lyse target cells.

21. The method of claim 20, wherein said defect is due to a mutation of the TAPl or of the TAP2 loci.

22. The method of claim 20, wherein said cytotoxic T-lymphocytes are prepared by the steps of: a. immunizing a mammal that expresses said cell surface major histocompatibility complex class I molecule with said at least one selected peptide, and b. harvesting cytotoxic T-lymphocytes from said immunized mammal.

23. The method of claim 22, wherein said mammal expresses said cell surface major histocompatibility complex class I molecule from an endogenous gene.

24. The method of claim 22, wherein said mammal expresses said cell surface major histocompatibility complex class I molecule from an exogenous gene. 25. The method of claim 22, wherein said immunization of said mammal additionally comprises immunization with at least one co-stimulatory factor. 26. The method of claim 22, wherein said immunization of the said mammal additionally comprises immunization with at least one co-stimulatory peptide derived from a helper T- lymphocyte epitope that is foreign to said mammal.

27. The method of claim 20, wherein said cytotoxic T-lymphocytes are prepared by the steps of: a. immunizing peripheral blood lymphocytes with said at least one selected peptide, and b. harvesting cytotoxic T-lymphocytes from said immunized peripheral blood lymphocytes.

28. The method of claim 27, wherein said immunization of the peripheral blood lymphocytes additionally comprises immunization with at least one co-stimulatory factor.

29. The method of claim 27, wherein said immunization of said peripheral blood lymphocytes additionally comprises immunization with at least one co-stimulatory peptide derived from a helper T-lymphocyte epitope that is foreign to said peripheral blood lymphocytes.

30. The method of claim 20, wherein said target cells are mammalian cells defective in the expression of said cell surface major histocompatibility complex class I molecule and that have been contacted with said at least one selected peptide. 31. The method of claim 30, wherein said defect is due to a mutation of the TAPl or of the

TAP2 loci. 32. The method of claim 20, wherein said target cells are mammalian cells that express said cell surface major histocompatibility complex class I molecule and that have been contacted with said at least one selected peptide. 33. The method of claim 20, wherein said target cells are immunoglobulin E-bearing cells.

34. The method of claim 20, wherein said target cells are immunoglobulin E-producing cells.

35. The method of claim 34, wherein said immunoglobulin E-producing target cells are plasma cells.

36. The method of claim 34, wherein said immunoglobulin E-producing target cells are tumor cells.

37. The method of claim 20, wherein said target cells produce human immunoglobulin E.

38. The method of claim 20, wherein said target cells have immunoglobulin E bound to their surface.

39. The method of claim 1, wherein said evaluating comprises the steps of: a. identifying and selecting at least one test peptide that binds to a major histocompatibility complex class I molecule, b. determining the amino acid sequence of said at least one selected peptide, c. establishing from said at least one determined amino acid sequence an amino acid binding motif for said major histocompatibility complex class I molecule, d. providing at least one second generation test peptide that has an amino acid sequence derived from immunoglobulin E and that is predicted by an algorithm to bind to said established amino acid binding motif, and e. testing said at least one second generation test peptide for the ability to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, wherein a peptide that induces such a response is identified.

40. The method of claim 39, wherein the process of identifying and selecting at least one test peptide that binds to a major histocompatibility complex class I molecule comprises the steps of: a. contacting at least one test peptide with mammalian cells defective in the cell surface expression of a major histocompatibility complex class I molecule, and b. selecting from said test peptides at least one peptide that causes expression of said major histocompatibility complex class I molecule at the cell surface of said mammalian cells.

41. The method of claim 40, wherein the defect is due to a mutation of the TAPl or of the TAP2 loci.

42. The method of claim 39, wherein the process of identifying and selecting at least one test peptide that binds to a major histocompatibility complex class I molecule comprises the steps of: a. contacting at least one test peptide with a major histocompatibility complex class I molecule, and b. selecting from said test peptides at least one peptide that binds to said major histocompatibility complex class I molecule.

43. The method of claim 42, wherein said major histocompatibility complex class I molecule comprises an isolated major histocompatibility complex class I moiety. 44. The method of claim 42, wherein said major histocompatibility complex class I molecule is expressed on a cell surface.

45. The method of claim 39, wherein the process of identifying and selecting at least one test peptide that binds to a major histocompatibility complex class I molecule comprises the steps of: a. contacting at least one display scaffold that displays at least one test peptide with mammalian cells defective in the cell surface expression of a major histocompatibility complex class I molecule, and b. selecting from said test peptides at least one peptide that causes expression of said major histocompatibility complex class I molecule at the cell surface of said mammalian cells.

46. The method of claim 45, wherein the defect is due to a mutation of the TAPl or of the TAP2 loci.

47. The method of claim 39, wherein the process of identifying and selecting at least one test peptide that binds to a major histocompatibility complex class I molecule comprises the steps of: a. contacting at least one display scaffold that displays at least one test peptide with a major histocompatibility complex class I molecule, and b. selecting from said test peptides at least one peptide that binds to said major histocompatibility complex class I molecule.

48. The method of claim 47, wherein said major histocompatibility complex class I molecule comprises an isolated major histocompatibility complex class I moiety. 49. The method of claim 47, wherein said major histocompatibility complex class I molecule is expressed on a cell surface. 50. The method of claim 39, wherein said testing comprises the steps of: a. preparing cytotoxic T-lymphocytes that recognize said at least one second generation test peptide, and b. testing said cytotoxic T-lymphocytes for the ability to lyse target cells.

^■ 51. The method of claim 50 wherein said cytotoxic T-lymphocytes are prepared by the steps of: a. immunizing a mammal that expresses the said cell surface major histocompatibility complex class I molecule with at least one said selected peptide, and b. harvesting cytotoxic T-lymphocytes from said immunized mammal.

52. The method of claim 51, wherein said mammal expresses said cell surface major histocompatibility complex class I molecule from an endogenous gene.

53. The method of claim 51, wherein said mammal expresses said cell surface major histocompatibility complex class I molecule from an exogenous gene. 54. The method of claim 51, wherein said immunization of said mammal additionally comprises immunization with at least one co-stimulatory factor.

55. The method of claim 51, wherein said immunization of said mammal additionally comprises immunization with at least one co-stimulatory peptide derived from a helper T- lymphocyte epitope that is foreign to said mammal.

56. The method of claim 50, wherein said cytotoxic T-lymphocytes are prepared by the steps of: , a. immunizing peripheral blood lymphocytes with at least one said selected peptide, and b. harvesting cytotoxic T-lymphocytes from said immunized peripheral blood lymphocytes.

57. The method of claim 56, wherein said immunization of said peripheral blood lymphocytes additionally comprises immunization with at least one co-stimulatory factor.

58. The method of claim 56, wherein said immunization of said peripheral blood lymphocytes additionally comprises immunization with at least one co-stimulatory peptide derived from a helper T-lymphocyte epitope that is foreign to said peripheral blood lymphocytes. 59. The method of claim 50, wherein said target cells are mammalian cells defective in the expression of said cell surface major histocompatibility complex class I molecule and that have been contacted with said selected peptide or peptides. 60. The method of claim 59, wherein said defect is due to a mutation of the TAPl or of the TAP2 loci.

61. The method of claim 50, wherein said target cells are mammalian cells that express said cell surface major histocompatibility complex class I molecule and that have been contacted with said selected peptide or peptides.

62. The method of claim 50, wherein said target cells are immunoglobulin E-bearing cells.

63. The method of claim 50, wherein said target cells are immunoglobulin E-producing cells. 64. The method of claim 63, wherein said immunoglobulin E-producing target cells are plasma cells.

65. The method of claim 63, wherein said immunoglobulin E-producing target cells are tumor cells.

66. The method of claim 50, wherein said target cells produce-human immunoglobulin E.

67. The method of claim 50, wherein said target cells have immunoglobulin E bound to their surface.

68. The method of claim 39, wherein said testing comprises the steps of: a. contacting said at least one second generation test peptide with said major histocompatibility complex class I molecule, and b. selecting from said second generation test peptides at least one peptide that binds to said major histocompatibility complex class I molecule.

69. The method of claim 68, wherein said major histocompatibility complex class I molecule comprises an isolated major histocompatibility complex class I moiety.

70. The method of claim 68, wherein said major histocompatibility complex class I molecule is expressed on a cell surface. 71. A composition comprising at least one isolated immunogenic peptide identified by the method of claim 1 as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. 72. The composition of claim 71, further comprising a pharmaceutically acceptable carrier. 73. The composition of claim 72, wherein said pharmaceutically acceptable carrier is selected from the group consisting of water, physiological saline, phosphate-buffered saline, saline buffered with a physiologically compatible salt, a water-in-oil emulsion, or an oil-in- water emulsion. 74. The composition of claim 71, further comprising at least one co-stimulatory factor. 75. The composition of claim 71, further comprising at least one co-stimulatory peptide derived from a helper T-lymphocyte epitope that is foreign to the said mammal. 76. The composition of claim 75, wherein said co-stimulatory peptide and said immunogenic peptide are provided on at least one polypeptide framework.

77. The composition of claim 71, wherein said immunogenic peptide is naturally occurring, synthesized, combinatorially synthesized, or biologically produced.

78. The composition of claim 71, wherein said immunogenic peptide is a peptide of between about five to about two thousand amino acids in length. 79. The composition of claim 78, wherein said peptide is linear.

80. The composition of claim 71, wherein said immunogenic peptide is a oligopeptide of between about five to about twenty amino acids in length.

81. The composition of claim 80, wherein said peptide is linear.

82. The composition of claim 71, wherein said immunogenic peptide comprises an amino acid sequence derived from immunoglobulin E.

83. The composition of claim 71, wherein said immunogenic peptide comprises an amino acid sequence derived from any of the CHI, CH2, CH3, or CH4 domains, or from the membrane exons, of human immunoglobulin E heavy chain.

84. The composition of claim 71, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen.

85. The composition of claim 71, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen that corresponds to a specific human leukocyte antigen haplotype.

86. The composition of claim 71, provided in the form of a solid, a powder, a gel, a solution, a suspension, an emulsion, liposomes, microspheres, injectable particles, inhalable particles, or a dissolvable matrix.

87. A composition comprising at least one isolated polynucleotide encoding the sequence for an immunogenic peptide that has been identified by the method of claim 1 as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides.

88. The composition of claim 87, further comprising a pharmaceutically acceptable carrier.

89. The composition of claim 88, wherein said pharmaceutically acceptable carrier is selected from the group consisting of water, physiological saline, phosphate-buffered saline, saline buffered with a physiologically compatible salt, a water-in-oil emulsion, or an oil-in- water emulsion. 90. The composition of claim 87, wherein said polynucleotide is a DNA molecule.

91. The composition of claim 87, wherein said polynucleotide is an RNA molecule.

92. The composition of claim 87, further comprising at least one co-stimulatory factor.

93. The composition of claim 87, wherein said immunogenic peptide is a peptide of between about five to about two thousand amino acids in length. 94. The composition of claim 93, wherein said peptide is linear.

95. The composition of claim 87, wherein said immunogenic peptide is an oligopeptide of between about five to about twenty amino acids in length.

96. The composition of claim 95, wherein said peptide is linear.

97. The composition of claim 87, wherein said immunogenic peptide comprises an amino acid sequence derived from immunoglobulin E.

98. The composition of claim 87, wherein said immunogenic peptide comprises an amino acid sequence derived from any of the CHI, CH2, CH3, or CH4 domains, or from the membrane exons, of human immunoglobulin E heavy chain.

99. The composition of claim 87, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen.

100. The composition of claim 87, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen that corresponds to a specific human leukocyte antigen haplotype.

101. The composition of claim 87, provided in the form of a solid, a powder, a gel, a solution, a suspension, an emulsion, liposomes, microspheres, injectable particles, inhalable particles, or a dissolvable matrix.

102. A composition comprising antigen-presenting cells that recognize at least one immunogenic peptide identified by the method of claim 1 as able to bind to at least one major histocompatibility complex class I molecule and to elicit in a mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides. 103. The composition of claim 102, further comprising a pharmaceutically acceptable carrier.

104. The composition of claim 103, wherein said pharmaceutically acceptable carrier is selected from the group consisting of water, physiological saline, phosphate-buffered saline, saline buffered with a physiologically compatible salt, cell culture medium, serum, or plasma.

105. The composition of claim 102, where said antigen-presenting cells are dendritic cells.

106. The composition of claim 102, further comprising at least one co-stimulatory factor. 107. The composition of claim 102, wherein said immunogenic peptide is apeptide of between about five to about two thousand amino acids in length.

108. The composition of claim 107, wherein said peptide is linear.

109. The composition of claim 102, wherein said immunogenic peptide is a oligopeptide of between about five to about twenty amino acids in length. 110. The composition of claim 109, wherein said peptide is linear.

111. The composition of claim 102, wherein said immunogenic peptide comprises an amino acid sequence derived from immunoglobulin E.

112. The composition of claim 102, wherein said immunogenic peptide comprises an amino acid sequence derived from any of the CHI, CH2, CH3, or CH4 domains, or from the membrane exons, of human immunoglobulin E heavy chain.

113. The composition of claim 102, wherein said aj or histocompatibility complex class I molecule is a human leukocyte antigen.

114. The composition of claim 102, wherein said major histocompatibility complex class I molecule is a human leukocyte antigen that corresponds to a specific human leukocyte antigen haplotype.

115. The composition of claim 102, provided in the form of a cell suspension. 116. A method of modulating an immunoglobulin E-mediated condition in a mammal, comprising providing to said mammal a composition that elicits in said mammal a cytotoxic T-lymphocyte response to naturally processed and presented immunoglobulin E peptides, in an amount sufficient to elicit said cytotoxic T-lymphocyte response in said mammal. 117. The method of claim 116, wherein said composition is that of claim 71.

118. The method of claim 116, wherein said composition is that of claim 87.

119. The method of claim 116, wherein said composition is that of claim 102.

120. The method of claim 116, wherein the level of circulating secreted immunoglobulin E in said mammal is decreased. 121. The method of claim 11 , wherein the level of circulating immunoglobulin E- bearing B-lymphocytes in said mammal is decreased.

122. The method of claim 116, wherein the level of mast cells or basophilic granulocytes sensitized to immunoglobulin E in said mammal is decreased.

123. The method of claim 116, wherein the level of immunoglobulin E-producing cells in said mammal is decreased.

124. The method of claim 116, wherein said immunoglobulin E-mediated condition is an immunoglobulin E-mediated atopic hypersensitivity condition.

125. The method of claim 116, wherein said immunoglobulin E-mediated condition is an immunoglobulin E-mediated non-atopic hypersensitivity condition. 126. The method of claim 116, wherein said immunoglobulin E-mediated condition is immunoglobulin E myeloma.

127. The method of claim 116, wherein said mammal is selected from the group consisting of dog, cat, horse, cow, pig, sheep, goat, ferret, mouse, rat, rabbit, guinea pig, gerbil, and hamster.

128. The method of claim 116, wherein said mammal is a non-human primate. 129. The method of claim 116, wherein said mammal is a human.

130. The method of claim 117, wherein said amount is between about 0.01 milligrams per kilogram body weight to about 10 milligrams per kilogram body weight of said at least one isolated immunogenic peptide.

131. The method of claim 117, wherein said amount is between about 0.05 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight of said at least one isolated immunogenic peptide.

132. The method of claim 117, wherein said amount is between about 0.5 milligrams per kilogram body weight to about 1 milligrams per kilogram body weight of said at least one isolated immunogenic peptide. 133. The method of claim 118, wherein said amount is between about 0.01 milligrams per kilogram body weight to about 10 milligrams per kilogram body weight of said at least one isolated polynucleotide.

134. The method of claim 118, wherein said amount is between about 0.05 milligrams per kilogram body weight to about 5 milligrams per kilogram body weight of said at least one isolated polynucleotide.

135. The method of claim 118, wherein said amount is between about 0.5 milligrams per kilogram body weight to about 1 milligrams per kilogram body weight of said at least one isolated polynucleotide.

136. The method of claim 119, wherein said amount is between about 1 x 10³ said antigen-presenting cells per kilogram body weight and 1 x 10⁸ said antigen-presenting cells per kilogram body weight.

137. The method of claim 119, wherein said amount is between about 1 x 10⁵ said antigen-presenting cells per kilogram body weight and 1 x 10⁸ said antigen-presenting cells per kilogram body weight.

138. The method of claim 119, wherein said amount is between about 1 x 10⁴ said antigen-presenting cells per kilogram body weight and 1 x 10⁷ said antigen-presenting cells per kilogram body weight.