Peptides which elicit a high neutralizing antibody titer, cytotoxic T lymphocyte response and T helper cell response in a broad range of MHC type recipients Number:7,094,405 from the United States Patent and Trademark Office (PTO) owispatent

Title: Peptides which elicit a high neutralizing antibody titer, cytotoxic T lymphocyte response and T helper cell response in a broad range of MHC type recipients

Abstract: Peptide constructs comprised of multideterminant T helper peptides from the envelope glycoprotein of HIV previously identified to induce proliferative responses in four different haplotypes of mice and IL-2 responses in 52-73% of HIV positive, flu positive patients (cluster peptides), were co-linearly synthesized with the peptide 18 of the V3 loop of HIV-1 gp 160, corresponding to the principal neutralizing determinant of HIV-IIIB and also shown to contain a dominant CTL epitope. Cognate help for peptide 18 antibody was elicited following a single immunization in all strains of mice which had previously responded to a T cell epitope encompassed by the peptides. In two strains of mice, the level of neutralizing antibody achieved was comparable to levels adequate for protection from homologous viral challenge in chimpanzees. After a single boost, much higher antibody titers for 90% neutralization in the range of 1:1000 to 1:16,000 were achieved. Spleen cells from mice of three distinct MHC haplotypes sharing the D.sup.d class I MHC molecule but with different class II molecules, immunized with the compound peptides, exhibited enhanced gp160-specific CTL activity.

Patent Number: 7,094,405 Issued on 08/22/2006 to Berzofsky,   et al.

---

Inventors:	Berzofsky; Jay A. (Bethesda, MD), Ahlers; Jeffrey D. (Kensington, MD), Pendleton; C. David (Bethesda, MD), Nara; Peter (Frederick, MD), Shirai; Mutsunori (Kagawa, JP)
Assignee:	The United States of America as represented by the Secretary of the Department of Health and Human Services (Washington, DC) N/A (
Appl. No.:	09/455,076
Filed:	December 6, 1999

---

Related U.S. Patent Documents

---

	Application Number	Filing Date	Patent Number	Issue Date
	08060988	May., 1993
	07751998	Aug., 1991
	07847311	Mar., 1992	5976541
	07492318	Feb., 1990	5081226
	07148692	Jan., 1988
	07014430	Feb., 1987
	06947935	Dec., 1986

---

Current U.S. Class:	424/188.1 ; 424/192.1; 424/208.1; 530/324; 530/325; 530/326
Current International Class:	A61K 39/21 (20060101)
Field of Search:	422/188.1,208.1

---

Foreign Patent Documents

	0362909		Apr., 1990		EP
	0448095		Sep., 1991		EP
	WO9304697		Mar., 1993		WO

Other References

Ahlers et al., J. Immunol. 150(12):5647-5665, 1993. cited by other . Berzofsky et al., J. Clin. Invest, 88:876-884, 1991. cited by other . Bhardwaj et al., Eur. J. Immunol. 22:2009-2016, 1992. cited by other . Defoort et al., Proc. Natl. Acad. Sci. 89:3879-3883, 1992. cited by other . Del Val et al., Cell 66:1145-1153, 1991. cited by other . Eisenlohr et al., J. Exp. Med. 175:481-487, 1992. cited by other . Engelhard, Ann. Rev. Immunol. 12:181-207, 1994. cited by other . Greenstein et al., J. Immunol. 148:3970-3977, 1992. cited by other . Hansen et al., in fundamental Immunology, Paul, ed., Raven Press; New York, pp. 609 and 612, 1993. cited by other . Hart et al., Proc. Natl. Acad. Sci. 88:9448-9452, 1991. cited by other . Hart et al., J. Immunol. 145:2677-2685, 1990. cited by other . Javaherian et al., Proceedings of the Natl. Acad. of Sciences of the USA 86(17):6768-6772, 1989. cited by other . Lasarte et al., Cellular Immunol. 141:211-218, 1992. cited by other . Palker et al., J. Immunol. 142:3612-3619, 1989. cited by other . Partidos et al., Immunol. 77:262-266, 1992. cited by other . Rovinski et al., J. Virol. 66(7):4003-4012, 1992. cited by other . Sarobe et al., Eur. J. Immunol. 21:1553-1558, 1991. cited by other . Schrier et al., J. Virol. 62(8):2531-2536, 1988. cited by other . Strognin, in Laboratory Diagnosis of Viral Infections, Lennette, ed., Marcel Dekker, Inc., New York, pp. 211-219, 1992. cited by other.

Primary Examiner: Parkin; Jeffrey S.
Attorney, Agent or Firm: Towsend and Townsend and Crew LLP

---

Parent Case Text

---

This application is a divisional of and claims the benefit of U.S. patent application Ser. No.08/060,988, filed May 14, 1993; which is a continuation-in-part of U.S. patent application Ser. No. 07/751,998, filed Aug. 29, 1991, now abandoned and a continuation-in-part of U.S. patent application Ser. No. 07/847,311, filed Mar. 6, 1992 now U.S. Pat. No. 5,976,541; which is a continuation-in-part of U.S. patent application Ser. No. 07/148,692, filed Jan. 26, 1988, now abandoned. Priority application Ser. No. 07/751,998 is also a continuation in part of U.S. patent application Ser. No. 07/148,692, filed on Jan. 26, 1988, now abandoned, and a continuation in part of U.S. patent application Ser. No. 07/492,318, filed on Feb. 28, 1990 (now U.S. Pat. No. 5,081,226), which is a continuation of U.S. patent application Ser. No. 07/014,430, filed on Feb. 12, 1987 now abandoned, which is a continuation in part of U.S. patent application Ser. No. 06/947,935, filed on Dec. 30, 1986 now abandoned. All of the above noted priority applications and patents are incorporated herein by reference in their entirety for all purposes.

---

Claims

---

What is claimed is:

1. An isolated multideterminant co-linear polypeptide consisting of a cluster peptide comprising overlapping T helper epitopes from within HIV envelope protein gp160 and a peptide 18(P18) peptide comprising a cytotoxic T cell epitope and an epitope which elicits a neutralizing antibody response to HIV, wherein said epitopes are located in the third hypervariable region (V3 loop) of the HIV envelope protein gp 160, wherein said cluster peptide is amino terminal to the P18 peptide, and wherein said multideterminant polypeptide can induce a T helper response, a cytotoxic T lymphocyte response and a neutralizing antibody response.

2. The polypeptide of claim 1, wherein said cluster peptide is PCLUS1 (amino acid residues 1-20 of SEQ ID NO:27), PCLUS3 (amino acid residues 1-24 of SEQ ID NO:28), PCLUS4 (amino acid residues 1-24 of SEQ ID NO:29), PCLUS6 (amino acid residues 1-33 of SEQ ID NO:30) or PCLUS6.1 (amino acid residues 1-27 of SEQ ID NO:31).

3. A polypeptide according to claim 1, wherein said P18 peptide is derived from HIV-1 gp160.

4. The polypeptide of claim 1, wherein said P18 peptide has the amino acid sequence SITKGPGRVIYATGQ (SEQ ID NO:37); SIHIGPGRAFYATGD (SEQ ID NO:38); SLSIGPGRAFRTREI (SEQ ID NO:39); SISIGPGRAFFATTID (SEQ ID NO:40); SIYIGPGRAFHTTGR (SEQ ID NO:41); GIAIGPGRTLYAREK (SEQ ID NO:42); RVTLGPGRVWYTTGE (SEQ ID NO:43); SIRIGPGKVFTAKGG (SEQ ID NO:44); GIHFGPGQALYTTGI (SEQ ID NO:45); STPIGLGQALYTTRG (SEQ ID NO:46); STPIGLGQALYTTRI (SEQ ID NO:47); or RTPTGLGQSLYTTRS (SEQ ID NO:48).

5. An immunogenic composition comprising a polypeptide of claim 1 in a pharmaceutically acceptable carrier.

6. A method for eliciting, in a mouse or a human, an antigen specific cytotoxic T lymphocyte response and an antigen specific high titer neutralizing antibody response comprising administering to said mammal mouse or human as the antigen a polypeptide of claim 1 in a pharmaceutically acceptable carrier.

7. The polypeptide of claim 2, wherein said P18 peptide has the amino acid sequence SITKGPGRVIYATGQ (SEQ ID NO:37); SIHIGPGRAFYATGD (SEQ ID NO:38); SLSIGPGRAFRTREI (SEQ ID NO:39); SISIGPGRAFFATTID (SEQ ID NO:40); SIYIGPGRAFHTTGR (SEQ ID NO:41); GIAIGPGRTLYAREK (SEQ ID NO:42); RVTLGPGRVWYTTGE (SEQ ID NO:43); SIRIGPGKVFTAKGG (SEQ ID NO:44); GIHFGPGQALYTTGI (SEQ ID NO:45); STPIGLGQALYTTRG (SEQ ID NO:46); STPIGLGQALYTTRI (SEQ ID NO:47); or RTPTGLGQSLYTTRS (SEQ ID NO:48).

---

Description

---

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention relates to peptides which comprise covalently linked T helper (Th) epitopes, cytoxic T lymphocyte (CTL) epitopes and epitopes which elicit a neutralizing antibody response (AbN) to an infectious agent, especially a parasitic or viral pathogen. Specific examples focus on application of the invention against Human Immunodeficiency Virus (HIV). The peptides have the further characteristic of evoking all three of these responses in hosts having a broad range of Major Histocompatibility Complex (MHC) types.

The invention is also directed to diagnostic methods for immune function in individuals infected with HIV which utilize the above-described peptides and is further directed to prophylactic or therapeutic vaccines which use the above-described peptides as a component of, or perhaps, as the sole active ingredient in the vaccine composition.

2. Description of the Related Art

Immune responses to HIV antigens elicited during natural infection may be a balance between those regulating viral infection and those antagonistic to the integrity of immune function (1-3). The determinants which weigh favorably or unfavorably upon this balance are not certain. The initial immune responsiveness of the host seems to influence the course of persistent HIV infection leading to progressive debilitating disease associated with increasing immunologic dysfunction (4-7). The virus may contain structures enabling it to evade the immune system, such as suppressive epitopes or masking carbohydrates, structures inducing clonal restriction (8-10), or structures that elicit deleterious effects such as antibodies which enhance viral infectivity (11-16) or autoreactive antibodies or T cells that contribute to the immunodeficiency (17-20).

The principal neutralizing determinant (PND) of the HIV envelope is located in the third hypervariable region or V3 loop between cysteine residues 301 and 331 (21-23). Antibodies to this region were initially demonstrated to be type specific in their neutralizing properties and more cross-reactive when examined by peptide binding ELISA (23-26), although more broadly neutralizing antibodies to the V3 loop have also been observed (27). Fortunately for synthetic vaccine development, such antibodies can be raised by immunization with short peptides (21,28,29). The protective efficacy of V3-specific antibodies to homologous cell-free virus challenge has been shown in chimpanzee challenge studies (15,30,31) and most recently protection against viral challenge was achieved by passive transfer into chimpanzees of a mouse-human IgG1 chimeric monoclonal antibody specific for the V3 loop (32). Sequence variation in the viral envelope protein in and outside (but affecting) this region results in both neutralization escape mutants, potential CTL escape mutants, and altered cellular tropism (33-37).

A high degree of genetic variability in HIV isolates can be found in infected individuals (38-40). HIV isolates from a given individual appear to change ver the course of disease. Under immune pressure the virus appears to exhibit differences in phenotypic characteristics such as cytopathicity, replication rates, and cellular tropism during the course of infection. Evidence that the virus may be replicating continuously at low levels during infection and never achieve a state of "true latency" supports the view that HIV-1 produces a chronic active infection and selective mechanisms play an important role in viral persistence (33). Multiple distinct V3 regions encoding the PND of the envelope protein have been detected in isolates of HIV derived from peripheral blood mononuclear cells (PBMC), suggesting that positive selection leads to much diversity of HIV env genes in vivo (40). Nevertheless there is evidence that the PND contains conserved epitopes that are the targets of neutralizing antibodies generated by sequence divergent isolates and that a limited number of peptides from the PND can elicit neutralizing antibodies recognizing multiple isolates, albeit at lower titer, and probably lower affinity (24,25,41).

The criteria for an effective vaccine must be not only that it is safe, i.e., does not contain epitopes that elicit autoimmune or virus enhancing responses, but also that it is capable of eliciting both a cellular immune response and a neutralizing antibody response to all the potential HIV variants prevalent in the infected population. In addition, since the MHC molecule of a given individual will bind and recognize only a subset of potential antigenic determinants recognized by the species as a whole, a synthetic peptide vaccine must also incorporate enough antigenic determinants to elicit recognition by T cells of most HLA types.

Accordingly, in a previous study we constructed six synthetic peptides of 20-33 residues each that correspond to six multideterminant T helper regions of the HIV envelope (42). Called cluster peptides, these span clusters of distinct but overlapping T helper epitopes recognized by proliferating T cells of three or four haplotypes of mice. These cluster peptides were tested for their ability to stimulate T cell responses in mice immunized with recombinant gp 160 (rgp160) and in peripheral blood lymphocytes of humans infected with HIV. Mice were also immunized with the cluster peptides to test for the induction of T cells responding to intact gp160 in vitro. Cluster peptides 3, 4 , and 6 (see sequences in Table I) stimulated T cells from mice of all four MHC haplotypes immunized with rgp160; and when mice were immunized with the cluster peptide, elicited T cell responses capable of recognizing the whole envelope protein in vitro. Cluster peptide 1, also used in this current study, stimulated proliferation strongly in only one strain of mice, despite the fact that the three other strains recognized components of the multideterminant region from which this peptide was made. Thus, the whole had less activity than the sum of its parts (42). Cluster peptides 1, 3, 4, and 6 stimulated significant IL-2 responses in peripheral blood lymphocytes of HIV-positive, influenza positive humans in 64, 73, 52, and 58% of tested cases respectively. It is of interest to note that these high responses were observed despite the fact that the subjects tested were presumably infected with a large number of different substrains of HIV. Cluster peptides 1, 3, and 4 have sequences relatively conserved among North American and European isolates of HIV, and cluster peptide 6 spans the boundary between conserved and variable sequences (43).

A successful peptide vaccine should be capable of eliciting T helper (Th) and cytotoxic T lymphocytes (CTL) responses as well as a neutralizing antibody response in vaccinees of multiple HLA types. Major histocompatibility complex (MHC) class I-restricted CTL appear to play a central role in the recovery from viral infection (81). Although exogenous lymphokines can substitute for T-cell help in the maturation of CTL precursors in vitro, the role of Th in priming CTL in vivo still remains poorly understood, compared to Th-B-cell collaboration. Although much evidence for a helper requirement in CTL induction exists (82-90), there is also evidence for CTL responses independent of help (85,91-95). Further, no study to date has shown a necessity for help requiring covalent linkage of a helper antigenic determiant to a CTL determinant, analogous to the linkage of carrier to hapten in cognate help for B cells. This lack of evidence may be due to the fact that the targets of CTL are whole cells, and immunization until recently required whole cells (or tissue grafts) or live viruses. The closest one could come to suggesting determinant linkage was to show that the helper determinant and CTL determinant had to be on the same skin graft to induce rejection (89), but this could not be explored further at the molecular level. Now that the possibility of peptide immunization for CTL induction has been demonstrated (96-100), it becomes feasible to address this question using peptides comprising both helper and CTL determinants. Although recent evidence indicated that a helper site is beneficial (90,101,102), it was not clear if the helper and CTL sites needed to be linked. Indeed, uncoupled helper and CTL epitope peptides were effective in two studies (90, 102) and not tested in the other (101), but in the former studies, the mixture of helper and CTL determinant peptides was administered in incomplete Freund's adjuvant emulsion, which sequesters the two peptides in the same microenvironment, or was given at high dose for multiple immunizations.

SUMMARY OF THE INVENTION

The present invention is directed to peptides which provide a broad immune response to an antigen expressed by a pathogen. The antigen is typically one derived from a viral or parasitic pathogen. Our strategy for peptide design was to link each cluster peptide to a short synthetic peptide (peptide 18), previously identified to be an immunodominant site recognized by CD8 cytotoxic T cells in association with class I molecules, and found within the V3 loop or principal neutralizing determinant region of the HIV-IIIB envelope protein.

Immunization of hosts having a broad range of MHC types, (the H-2 loci in mice, equivalent to the HLA loci of humans) with a peptide of our invention results in an immune response having both humoral and cellular components. On the humoral side, a high titer neutralizing antibody response is observed. With respect to the cellular immune response, both cytotoxic T lymphocytes and T helper cells are elicited.

Proper choice of the epitopes employed evokes such a response that is broadly specific for a number of strains of a pathogen. This is particularly important if there is a great divergence in antigen structure among strains of the target pathogen, for example, as is observed for HIV. One method for designing peptides so as to produce a broadly specific response to a number of strains of HIV is described in co-pending U.S. patent application Ser. No. 07/760,530, hereby incorporated in its entirety by reference.

Accordingly, it is one object of the present invention to provide peptides which evoke all of a T helper response, a cytotoxic T lymphocyte response and a high titer of neutralizing antibody in a plurality of hosts expressing a broad range of MHC types.

The peptides of the present invention are also useful in a diagnostic context. For instance the particular peptides disclosed in the examples can be used in a variety of assay formats to assess the immune function of the T helper cells, cytotoxic T lymphocytes and B cells (both B-cell precursors and mature plasma cells) in individuals infected with HIV. Thus, it is also an object of the present invention to provide diagnostic methods and immune function assays which employ the peptides described herein as reagents.

By virtue of the broad immune response which is elicited in a host immunized with the peptides of the present invention, it is a further object of the present invention to provide vaccines, of either or both of a prophylactic and therapeutic nature, against a parasitic or viral infection, and against HIV-1 infection especially.

It is yet a further object of the present invention to provide a method of immunization of a mammalian host which elicits a broad immune response against a parasitic or viral pathogen, especially the HIV-1 virus.

The peptides of the present invention are comprised of a covalent linkage of a peptide having a multideterainant T helper epitope, such as described in U.S. patent application Ser. No. 751,998, a peptide having a cytotoxic T lymphocyte (CTL) epitope, preferably one which elicits CTL that are cross-reactive with a variety of strains of the target virus, such as described in U.S. patent application Ser. No. 07/847,311 or U.S. patent application Ser. No. 07/148,692 and a peptide having a determinant which elicits a neutralizing antibody (a principal neutralizing determinant (PND)). The epitopes in each case are those which can be shown to be recognized by hosts having a broad range of major histocompatibility complex antigens (MHC). The MHC are also called HLA in humans and are the cell surface proteins which determine, in part, whether or not tissue transplants are accepted or rejected by the host. The MHC proteins are involved in presentation of antigens to the immune system in early stages of an immune response. By virtue of the fact that the peptides of the present invention are recognized by a number of different MHC or HLA types, they are expected to be efficacious in a large portion of the host population.

The peptides of the present invention further demonstrate the property of eliciting a high titer of neutralizing antibody against the antigen from which their sequences are derived.

The methods for assessing the immune function of an individual having a viral infection that employ the peptides of the present invention are in vitro tests which measure the response of is lated cells from an individual to incubation of the cells with the peptide. For instance, Th activity can be assessed by measurement of cytokines released specifically in response to incubation of peripheral blood cells from a patient with a peptide of the present invention. A preferred cytokine to be measured is interleukin-2 (IL-2). The method of measurement can be any of the techniques known to the art, for instance, measuring proliferation of an interleukin-2 dependent cell line in supernatants of cultures of the incubated peripheral blood cells. Alternatively, ELISA assay of the IL-2 (or other cytokine) can be performed (see U.S. Ser. No. 07/751,998, hereby incorporated by reference). The methods for immunization with the peptide of the present invention can be quite simple, such as intravenous injection of a sterile composition comprised of one or more of the peptides of the present invention and a pharmaceutically acceptable carrier solution or adjuvant. Alternatively, the peptides can be administered bound to the surface of irradiated antigen-presenting cells, as is described in co-pending U.S. patent application Ser. No. 08/031,494 (hereby incorporated in its entirety by reference).

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1A, 1B P18 specific antibody response of mice of four different MHC haplotypes following immunization with cluster peptide 6-18. (a) the primary antibody response 31 days after immunization with 20 nanomoles of peptide (levels at day 21 were lower; see Results). (b) an anamestic response to a boost of 10 nanomoles of peptide 37 and 49 weeks after primary immunization. Symbols correspond to individual mice except for + which indicates a prebleed pool.

FIGS. 2A-H HIV-1 IIIB neutralization profiles of four strains of mice 31 days following a single immunization with peptide PCLUS 6-18.

FIGS. 3A-D HIV-1 IIIB neutralization profiles of four strains of mice 10 days following a single boost with 10 nanomole of PCLUS 6-18 39-42 weeks post primary immunization. Vn/Vo is plotted versus reciprocal dilution for each numbered serum as in FIG. 2. Animal numbers and symbols represented correspond to those in FIG. 2 for the primary response. Note that the abscissa for mouse strain BALB/c is different from the other strains, such that the endpoint dilution for BALB/c is 1:32,768 whereas for the other strains it is 1:4096.

FIGS. 4A, 4B Competitive binding curves using p18 and rgp 120 as competitor, to assess affinity of antibodies for peptide and whole protein. Solid symbols represent sera with >90% neutralizing activity at one of the dilutions tested and the open symbols represent sera with <90% neutralizing activity at the lowest dilution tested. The data are representative of 3-4 experiments.

FIGSS. 5A, 5B. Fine specificity of neutralizing vs. non-neutralizing sera in PCLUS 3-18 and PCLUS 6-18 immunized mice. Neutralizing sera (solid bars) and non-neutralizing sera (open bars) were tested in an ELISA assay on wells coated with P18 substituted peptides. Fifteen peptides with single amino acid substitutions from the HIV-1 IIIB sequence (RIQRGPGRAFVTIGK; SEQ ID NO:7) toward the HIV-1 RF sequence (**TKGPGRVIYATGQ; SEQ ID NO:8) where used to coat wells (See Table V; SEQ ID NOs:7-23). Where the two sequences were identical, an Ala was substituted. Peptides were called 18-1 through 18-15, where the second number indicates the position in the sequence that was substituted. The letter under the number in each group indicates the amino acid from the RF sequence (or Ala) that was substituted at that position in the corresponding P18 IIIB sequence. An asterisk denotes a deletion. Sera are compared at a dilution of 1:1000.

FIGS. 6A, 6B Binding to P18 variants substituted within the central V3 loop region. Peptide 18 variants substituted at positions 3-10 (as shown in Table V) were used to coat microtiter wells, and sera were tested for binding in an ELISA assay. The letter under the number in each graph indicates the amino acid from the RF sequence (or Ala) that was substituted at that position in the corresponding P18 IIIB sequence. Solid bars represent neutralizing sera and open bars represent nonneutralizing sera. Columns represent the mean absorbance ratio of binding to substituted peptide versus P18 at 405 nm of duplicate readings for individual sera, identified by number, from animals immunized with PCLUS 3-18 (A, upper panel) or PCLUS 6-18 (B, lower panel).

FIGS. 7A-C. Induction of HIV-1 envelope gp160-specific CTL activity by immunization with compound peptides in QS21 adjuvant. Because the standard error of the mean (SEM) of triplicate wells was consistently less than 8% of the mean, error bars are omitted for clarity.

FIGS. 8A-8D. The requirement for linkage between helper and CTL determinants for priming of CTL. B10.D2 (FIG. 8A) and (FIG. 8C) or B10.A(5R) (FIG. 8B) and (FIG. 8D).

FIGS. 9A, 9B Phenotype of the CTL effectors (A), and the helper T cells (B) induced by immunization with the compound peptide constructs.

FIGS. 10A-H shows HIV neutralizing activity of PCLUS3-18MN boosted sera and PCLUS6-18MN boosted sera.

FIGS. 11A-H shows HIV neutralizing activity of PCLUS6-18MN boosted sera and PCLUS6.1-18MN boosted sera.

FIGS. 12A, 12B show CTL response elicited by immunization with P18-MN or PCLUS3-18MN in a variety of adjuvants.

FIGS. 13A, 13B show CTL response following two immunizations of peptide in a variety of adjuvants.

DETAILED DESCRIPTION OF THE INVENTION

In an effort to provide help for an enhanced neutralizing antibody response, we have directly linked cluster peptides to peptide 18 (P18), which is contained within the PND. P18 consists of amino acid residues 308-322 of HIV-1 IIIB gp160 (sequence numbering according to the Los Alamos database (43), which is 7 less than the numbering of Ratner et al. (44) that we used previously (42)). P18 also contains an immunodominant cytotoxic T cell site (45,46). It is to be understood that the region of gp160 envelope protein homologous to the P18 region, from other strains of HIV than IIIB, can be employed in a similar manner. For instance, Example III shows results obtained when the P18 region of strain MN is used as the CTL epitope in the immunogen peptide. The peptides which are representative of the P18 region from various strains of HIV are disclosed in co-pending U.S. patent application Ser No. 07/847,311.

For example, P18 from isolate RF has the amino acid sequence SITKGPGRVIYATGQ(SEQ ID NO:37); P18 from isolate SC has the amino acid sequence SIHIGPGRAFYATGD (SEQ ID NO:38); P18 from isolate WMJ-2 has the amino acid sequence SLSIGPGRAFRTREI (SEQ ID NO:39); P18 from isolate Z321 has the amino acid sequence SISIGPGRAFFATTD (SEQ ID NO:40); P18 from isolate SF2 has the amino acid sequence SIYIGPGRAFHTTGR (SEQ ID NO:41); P18 from isolate NYS has the amino acid sequence GIAIGPGRTLYAREK (SEQ ID NO:42); P18 from isolate CDC4 has the amino acid sequence RVTLGPGRVWYTTGE (SEQ ID NO:43); P18 from isolate Z3 has the amino acid sequence SIRIGPGKVFTAKGG (SEQ ID NO:44); P18 from isolate MAL has the amino acid sequence GIHFGPGQALYTTGI (SEQ ID NO:45); P18 from isolate ZE has the amino acid sequence STPIGLGQALYTTRG (SEQ ID NO:46); P18 from the isolate JY1 has the amino acid sequence STPIGLGQALYTTRI (SEQ ID NO:47); and P18 from the isolate ELI has the amino acid sequence RTPTGLGQSLYTTRS (SEQ ID NO: 48).

The immunogenicity of haptenic peptides has been shown to be increased by linear polymerization or coupling to T helper determinants (47-49). The cluster peptides should provide help in multiple MHC haplotypes. Remarkably high neutralizing titers were obtained in mice of several MHC types after just a single boost with some of these peptides. We further attempted to examine the fine specificity and affinity of neutralizing antibody directed against peptide 18. This approach can be used in designing peptides for a synthetic peptide vaccine for the immunoprophalaxis and immunotherapy of HIV infection (50).

Some of the materials and methods employed in the Examples described below are used in more than one of the Examples. These materials and methods are described as general materials and methods.

General Materials and Methods

Synthesis of peptides. The cluster peptide-peptide-18 and T helper-peptide-18 constructs were synthesized on an automated peptide synthesized (No. 430A; Applied Biosystems, Foster City, Calif.) utilizing t-boc chemistry (51) according to the sequences shown in Table I (SEQ ID NOs:1-6). The peptides were cleared from the resin with HF and initially purified by size inclusion chromatography. (P4 Biogel; BioRad Laboratories, Mountain View, Calif.). Purification to single peaks was achieved by reverse-phase HPLC on .mu.bondapack reverse-phase C18 analytical and preparative columns (Waters Associates, Milford, Mass.). Peptide 55-18 was synthesized with an extra Ala at the N-terminus to avoid an N-terminal Gln, which would cyclize to form pyroglutamic acid.

TABLE-US-00001 TABLE I Sequences of T Helper Sites Linked to Peptide 18 HIV-1 IIIB Peptides Sequences* Peptide 18 MW (daltons) ##STR00001## ##STR00002## ##STR00003## *HIV-1 IIIB numbering is according to the Los Alamos Protein sequence data base (43). Previous reference to these peptides (42) used the Ratner numbering system (44). .sup..dagger.Peptide 55-18 is shown with an added alanine at the N-terminus to avoid formation of pyroglutamic acid.

The peptides containing the individual or multideterminant epitopes can be joined together by synthesizing them as a colinear peptide as described above. Alternatively, side-chain carboxylic acid and amino groups can be used to form peptide bonds; connection of the peptides through the side-chains provides an immunogen having a branched structure. In a third embodiment, the peptides can be joined by non-peptide conjugations. Several methods for conjugating peptides are well-known in the art. One such method is set forth in U.S. Pat. No. 4,886,782.

Mice. H-2 congenic mice on the B10 background and BALB/c mice were purchased from Jackson Laboratories (Bar Harbor, Me.) or were bred in our colony at BioCon Inc., Rockville, Md. Mice used in this study were 8-20 weeks old.

ELISA. Wells of round bottom flexible PVC microtiter plates (#3912 Falcon Labware, Oxnard, Calif.) were coated overnight at 4.degree. C. with 100 .mu.l of 10 .mu.M Peptide 18, substituted peptide 18, cluster peptide 3, cluster peptide 6, 0.2 .mu.g/ml recombinant gp 120 (ABT, Advanced Biotechnologies, Mass.) or 2 .mu.M sperm whale myoglobin in 0.1 M carbonate buffer pH 9.6. Recombinant gp120 plates were coated with 100 .mu.l of a 0.2 .mu.g/ml afftinity-purified rgp120 (ABT) in carbonate buffer. The plates were blocked with 1% BSA in phosphate buffered saline (PBS) for 1-1.5 h. at 4.degree. C. and washed with PBS containing 0.05% Tween 20 and 1% BSA (PBSTB). Next, 100 .mu.l of test serum was added to duplicate wells and incubated for 1-1.5 h. at 4.degree. C. Test sera were assayed at 10 fold dilutions in PBSTB ranging from 1:100-1:10,000. The wells were then washed with 200 .mu.l PBSTB 10 times using an automatic plate washer (BioRad Model 1550) and incubated for 1 h at 4.degree. C. with 100 .mu.l alkaline phosphatase-conjugated goat anti-mouse IgG (Promega, Madison, Wis.) diluted 1:7500 in PBSTB. After 10 washes, bound antibodies were detected by the addition of 100 .mu.l of 1 mg/ml paranitrophenyl phosphate as substrate. The optical density at 405 nm was read with an ELISA reader. Specific absorbance was determined as the mean optical density at 405 nm on the relevant antigen coated wells minus the optical density 405 nm on nonrelevant sperm whale myoglobin coated wells. IgM specific for peptide 18 was detected using goat anti-mouse .mu. chain specific antibody (Sigma Chemical Co., St Louis, Mo.) followed by alkaline phosphatase conjugated anti goat. Total IgG and isotypes IgG1 and IgG2a were detected using biotin conjugated rat anti mouse monoclonals LO-MG1-13 for IgG1 and LO-MG2a-3 for IgG2a (BioSource International, Westlake Village, Calif.). Following the addition of strepavidin alkaline phosphatase (Sigma Immunochemicals, St. Louis, Mo.) substrate was added and plates read on the ELISA reader.

The invention is illustrated in detail by the Examples presented below. The Examples are presented as illustrations of preferred embodiments of the invention and are not meant to be limiting of the invention in any manner.

EXAMPLE I

Immunization of Mice to Produce High Titer Neutralizing Antibody Against HIV-1

The peptides of the present invention can be used to produce a high titer of AbN against a target antigen by immunization of a mammalian host with the peptide. In this Example, peptides derived from the sequence of the HIV-1 envelope glycoprotein gp 120 are used to elicit high titers of AbN in mice.

Immunizations. Mice were immunized intraperitoneally with 20 nanomoles of each peptide emulsified 1:1 in Freund's Complete Adjuvant (CFA). At days 21 and 31 post immunization blood was drawn by retroorbital bleed, allowed to clot and the serum removed and frozen at -20.degree. C. Because antibody levels were found to still be rising between day 21 and day 31 after primary immunzation, the results from 31-day sera are reported. Selected groups of animals were boosted 36-52 weeks post primary immunization with 10 nanomoles of peptide in CFA intraperitoneally and bled 10-11 days later, when the secondary response was generally found to be optimal.

Neutralization assays of HIV-1. The quantitative infectivity microassay using CEM-SS cells was performed as described previously (52). Briefly, serial two-fold dilutions of 50 .mu.l heat inactivated (56.degree. C., 30 min.) test serum mixed with 50 .mu.l of culture supernatant containing 200 syncytium forming units (SFU) of HIV-1-IIIB, or HIV-1 MN strain grown optimally from logarithmic cultures of H9 cells and previously cryopreserved and titered were incubated for 30 min at room temperature. The mixtures were added to duplicate wells containing 5.times.10.sup.4 DEAE-dextran treated CEM-SS cells for 1 h at 37.degree. C., after which the virus-antibody mixtures were removed and replaced with medium, and the cells cultured only in complete medium for 5 days (for HIV-1 IIIB) or 4 days (for HIV-1 MN, determined to be optimal for each virus strain) at 37.degree. C. in 5% CO.sub.2. Units of infectious virus were quantitated by subsequent syncytia formation of infected cells under an inverted microscope. The reciprocal geometric mean neutralization titer was expressed as the serum dilution capable of inhibiting HIV-1 foci by greater than 90% of control CEM-SS/HIV-1 infected cells infected with the indicated strain of HIV-1 (i.e. Vn/Vo<0.1). The assay measures neutralization of cell-free virus in the first incubation, not inhibition of syncytium formation, which is only the readout for enumerating infectious virus. No cytostatic or toxic properties of the serum alone on the CEM-SS cells were observed at the hightest concentration tested. Also, the heat inactivation at 56.degree. C. for 30 min has been shown to eliminate the nonspecific neutralizing activity of mouse sera (53).

Determination of direct binding by immunofluorescence. To assess binding of peptide-induced antibodies to native gp 120, serial ten-fold dilutions of selected neutralizing and nonneutralizing sera were tested for binding to viral gp 120 expressed on the surface of HIV-1 IIIB productively infected cells in a live cell immunofluorescence assay (IFA) (52).

Competition ELISA binding curves. The binding tests were performed by mixing in sterile polypropylene tubes 125 .mu.l of different dilutions of p18 (0-20 .mu.M) or rgp120 (0-160 nM) in phosphate buffered saline pH 7.2, 1% ovalbumin, 0.05% Tween 20 with 125 .mu.l of a constant dilution of antisera determined to be in the linear range of absorbance versus antibody dilution in the same buffer. After overnight incubation at 4.degree. C. with gentle shaking, a volume of 100 .mu.l was added to duplicate wells of a microtiter plate coated with the respective competing antigen p18 or gp120, incubated at 4.degree. C. for 20 min. and a standard ELISA assay performed. Binding curves were generated using a four parameter logistic function of log serum dilution versus absorbance using a commercial software program (Biometalics Inc., Princeton, N.J.) and an estimated dissociation constant (K.sub.d value) for individual sera was determined.

Binding to P18 substituted peptides. The specificity of neutralizing and nonneutralizing sera were tested in a standard ELISA assay for binding to fifteen substituted p18 peptides (37) coated onto plastic microtiter wells.

The synthetic multideterminant peptides (Table I), called "cluster peptides" (abbreviated PCLUS in the names of specific constructs), each constitute clusters of overlapping, but distinct, shorter T helper determinants identified in previous studies (42,54,55). Three cluster peptides, PCLUS 3, 4, and 6 used in this study were chosen because they fulfilled the criteria of eliciting proliferative responses in four independent MHC haplotypes of mice that differ in both an I-A and I-E molecule, and also in humans of multiple HLA types (42,56,57). PCLUS 1 was strongly recognized by only one strain of mice, B10.BR, yet stimulated IL-2 production in 23 of 36 HIV seropositive, flu-positive donors. Peptide HP53 (residues 827-841 in the Los Alamos database (43) numbering, which is 7 less than that of Ratner (44) used previously (42,54,55)) and peptide HP55 (residues 834-848) have previously been identified as T helper epitopes of the HIV IIIB envelope sequence in mice of the A.sup.kE.sup.k and A.sup.bE.sup.b haplotypes and the A.sup.kE.sup.k, A.sup.bE.sup.b, A.sup.dE.sup.d, and A.sup.sE.sup.s haplotypes respectively, and are contained within the longer PCLUS peptides. (E.sup.b and E.sup.s are used here to indicate the expressed E.sub..beta..sup.bE.sub..alpha. and E.sub..beta..sup.sE.sub..alpha. molecules, respectively. Although the nonpolymorphic E.sub..alpha. is not expressed in pure H-2.sup.b and H-2s haplotypes, we used recombinant strains that express E.sub..alpha.). Peptide HP53 (also referred to as env TH 4.1) was also previously shown to elicit IL-2 production in peripheral blood lymphocytes of asymptomatic HIV-seropositive human patients (57). Peptide 18 (residues 308-322) is a B cell epitope located within the hypervariable V3 loop region of the HIV-1 IIIB envelope known as the principal neutralizing determinant (PND), and is the major immunodominant cytotoxic T cell epitope in mice (45), as well as being recognized by human CTL (46).

Synthetic peptide vaccine constructs were prepared by synthesizing peptide 18 at the carboxy terminus of the cluster peptide. Immunization of mice with 20 nanomoles of these constructs produced enhanced peptide 18 specific antibodies, whereas no peptide 18 specific antibody was detected in mice immunized with peptide 18 alone (Table II). The orientation of T helper cell and B cell epitopes proved crucial for the immunogenicity of the construct since although cluster peptide 3-18 elicited an antibody response in all four strains tested, the reverse polarity construct, P18-cluster peptide 3 in which the helper site was C-terminal to P18, elicited an antibody response to peptide 18 in only one strain, B10.HTT, and at a significantly lower level.

TABLE-US-00002 TABLE II HIV IIIB Peptide 18 Specific Antibody Response and Neutralizing Activity in Four Strains of Mice 31 days Following a Single Immunization with 20 nanomole T Helper-P18 Peptide BALB/c or B10.D2 B10.BR or B10.A B10.HTT B10.A(5R) Mouse Strain (A.sup.dE.sup.d) (A.sup.kE.sup.k) (A.sup.aE.sup.a) (A.sup.bE.- sup.b) Peptide ELISA* Neutralization.sup..dagger. ELISA Neutralization ELISA Neut- ralization ELISA Neutralization P18 0.00 -- 0.00 -- 0.00 NT.sup..dagger-dbl. 0.00 NT (0/5) (0/5) 53-18.sup. 0.05 .+-. 0.3 NT 0.13 .+-. .05 NT 0.78 .+-. .60 -- 2.30 .+-. .03 32.0 (0/5) (1/5) 55-18.sup. 0.48 .+-. .15 -- 0.00 NT 2.47 .+-. .25 22.6 0.32 .+-. .23 NT (0/5) (2/5) PCLUS 1-18 0.12 .+-. .03 NT 0.68 .+-. .04 16.0 0.13 .+-. .05 NT 0.07 .+-. .04 NT (2/5) PCLUS 3-18.sup..sctn. 0.39 .+-. .14.sup. 10.6 0.50 .+-. .19.sup. -- 2.70 .+-. .30 32.0 0.86 .+-. .16 -- (5/5) (0/5) (1/6) (0/5) 0.45 .+-. .17 9.5 0.28 .+-. .08 -- 0.51 .+-. .22 -- 0.03 .+-. .01 NT (4/5) (0/5) (0/5) PCLUS 4-18 0.64 .+-. .19 8.0 0.38 .+-. .07 NT 0.61 .+-. .16 -- 0.16 .+-. .04 NT (1/4) (0/3) PCLUS 6-18.sup..sctn. 0.84 .+-. .12 9.5 0.94 .+-. .24 19.0 0.88 .+-. .23 11.3 0.63 .+-. .23 8 (4/5) (3/5) (4/5) (1/5) 0.77 .+-. .13 42.2 0.80 .+-. .06 32.0 0.16 .+-. .07 -- 0.87 .+-. .08 20.2 (5/5) (4/5) (0/5) (3/5) *Mean p18 Specific Absorbance .+-. S.E.M. of 5 mouse sera tested at a 1:1000 serum dilution .sup..dagger.Reciprocal Geometric Mean IIIV IIIB Neutralizing Antibody Titer expressed as the serum dilution capable of inhibiting HIV IIIB specific focl by greater than 90% of control CEM-SS/HIV IIIB infected cells. (No. of mice with positive neutralizing titers/total tested.) .sup..dagger-dbl.NT Not tested .sup. B10.D2 and/or B10.A mouse strains used. .sup..sctn.For PCLUS 3-18 and PCLUS 6-18, two different experiments are shown.

In most cases the T helper peptide component of our constructs provided help for peptide 18-specific antibody in the strains in which these T cell epitopes elicited proliferative responses in previous studies (54,58). In the case of peptide 53-18 and 55-18, the A.sup.bE.sup.b and A.sup.kE.sup.k haplotypes responded to peptide 53-18 and the A.sup.sE.sup.s, A.sup.dE.sup.d, and A.sup.bE.sup.b haplotypes responded to peptide 55-18 (Table II). In addition, A.sup.sE.sup.s mice responded to peptide 53-18, but A.sup.kE.sup.k mice failed to make anti-P18 antibodies to peptide 55-18 even though they were previously shown to proliferate in response to P55 .(54). PCLUS 1-18, which encompasses T helper cell epitopes recognized by all four strains of mice used in this study, was able to elicit a strong antibody response in only one strain, B10.BR, and a marginal one in two other strains, B10.D2 and B10.HTT. This was not surprising since in our previous study (42) we showed that cluster peptide 1 was not simply the sum of its parts, but failed to stimulate proliferation in some strains in which a smaller component of the cluster peptide did. The larger peptide may fold back upon itself and hinder interaction with the MHC or T cell receptor or may undergo different processing which would destroy a component epitope. PCLUS 3-18, 4-18, and 6-18 elicited strong peptide 18 antibody responses in all the strains of mice tested in at least one experiment, as measured by ELISA. These results show that by linking a B cell epitope to the carboxy terminus of a cluster of immunodominant T cell epitopes, in most cases the processing of individual epitopes is patent and the T cells elicited are capable of providing help to the B cell for production of specific antibody. Since a single immunization with these cluster peptide-peptide-18 constructs could induce high levels of P18-specific antibodies, we tested to see if a boost, which might represent a viral challenge, would produce a characteristic anamnestic response resulting in enhanced peptide 18 specific antibody (FIG. 1). After the primary immunization, the response to which is slower than a secondary immunization, sera were obtained on days 21 and 31, but the antibody levels were still rising on day 31. Therefore, we show in panel A individual mean peptide 18 specific absorbance readings for animals immunized with 20 nanomoles of PCLUS 6-18 and bled 31 days post immunization and compare in panel B individual absorbance readings for animals boosted with 10 nanomole of PCLUS 6-18, 37 and 49 weeks post primary immunization and bled 11 days later, when the secondary response usually peaks. The increase in antibody response observed after boosting was between 2.5 to 12 fold.

To assess usefulness of these synthetic peptide constructs in a vaccine, it was necessary to determine if the antibody elicited to peptide 18 following a single immunization with these constructs was capable of neutralizing the virus in vitro. Neutralizing activity, expressed as the reciprocal geometric mean titer capable of inhibiting cell-free infectious units of HIV IIIB by greater than 90% compared to control CEM-SS/HIVIIIB infected cells was elicited in peptide 53-18, 55-18, and PCLUS 1-18 in the strains which responded by a specific antibody response. However antibody was neutralizing in only one of five animals immunized with 53-18 and only two of five immunized with 55-18 and PCLUS 1-18 (Table II). This occurred despite equal or higher levels of peptide 18 specific antibodies by ELISA in other animals within the group. The finding that total peptide 18-specific antibody levels did not correlate with neutralization was extended by an interstrain comparison in animals immunized with PCLUS 4-18. Although all strains responded with significant levels of antibody to peptide 18 by ELISA compared to prebleed controls, only one of four animals of the BALB/c strain generated neutralizing antibody at the lowest dilution tested (Table II). The lack of correlation between specific peptide 18 antibody response and neutralizing activity was especially apparent in animals immunzed with PCLUS 3-18. Significant levels of peptide 18 specific antibody were elicited as determined by ELISA in all strains of mice immunized with PCLUS 3-18, and the animals of the A.sup.sE.sup.s haplotype showed the strongest antibody response to this construct. Nevertheless, only animals of the H-2.sup.d haplotype made antibodies capable of neutralizing virus in vitro (9 of 10 animals), despite lower levels of antibody by ELISA. This finding suggests that the in vivo induction of neutralizing antibody by the B cell epitope (peptide 18) immunogen depends on other factors in addition to the level of help, such as the specificity of helper T cells, or other MHC-linked regulatory factors.

PCLUS 6-18 reproducibly elicited neutralizing antibody in all strains tested in which it elicited a significant antibody response by ELISA (Tables II and III). The geometric mean neutralizing antibody titers achieved in BALB/c (42.2) and B10.BR (32.0) correspond to levels of neutralizing activity directed to the V3 loop that have been found sufficient to protect chimpanzees from a live homologous viral challenge (30). Neutralization titration profiles from two separate experiments for each group of animals following a single immunization with PCLUS 6-18 are shown in FIG. 2.

Each animal received 20 nanomoles of synthetic peptide emulsified in CFA (1:1) intraperitoneally in a volume of 0.1 ml. Neutralizing activity is determined in a microculture syncytium-forming assay using the HIV-1 IIIB HX3 strain and is expressed as Vn/Vo where Vn is the mean number of syncytia forming units (SFU) in duplicate test wells and Vo the number of SFU in control wells incubated without test sera. Each curve represents serial two fold dilutions of individual mouse serum (designated by animal number) except the prebleed pool which includes all animals within a group. The two columns represent two separate experiments.

It is interesting to note that three of five BALB/c mice shown in the first panel had 90% neutralizing titers greater than 64, which was the highest dilution tested. In addition, four of ten B10.BR mice shown in panels 3 and 4 exhibited neutralizing titers greater than 64. Over half ( 22/40) of all animals immunized with cluster peptide 6-18 demonstrated 50% neutralization of live virus at a dilution of 1:64 and all but one neutralized 50% of the virus at one of the dilutions tested. This single animal number 6269 B10.HTT had a negligible antibody response in a group that appeared to be poorly immunized. The mice that responded to the primary immunization were given a single boost with the same construct in CFA 37 or 49 weeks after the first immunization. This single boost produced remarkably high titers of neutralizing antibody with 90% neutralization occurring out to 1:2048-1:4096 in many animals and 1:16,384 in some (FIG. 3 and Table IV). These neutralizing titers against the homologous viral strain after just two immunizations are at least four to eight fold higher than the highest titers of other polyclonal sera induced by any immunization that we have ever observed. (8,59). Moreover, the timing indicates that memory from the primary immunization lasted at least 11 months. Furthermore, in three of four strains boosted with PCLUS6-18 IIIB, the sera were also capable of neutralizing the HIV-1 MN strain, albeit at much lower titer (Table IV). Although mice of the H-2.sup.d haplotype produced some of the highest neutralization titers against HIV-1 IIIB (e.g. 1:16,384) following a single boost, none of these sera showed cross-neutralizing activity against the MN strain.

In order to attempt to explain why some sera had high neutralizing activity and other sera with similar ELISA titers for the same short sequence did not, we compared affinity, isotype, fine specificity and other properties of neutralizing and nonneutralizing antibodies generated by each of the cluster peptide-peptide-18 immunizations (summarized in Table III). It is important to note that all the sera tested, with the exception of that of one animal (B10.BR # 9770), were capable of neutralizing the virus at some dilution as assessed by 50% inhibition compared to control sera, even if they were nonneutralizing by the criterion of 90% inhibition. Therefore it may be more difficult to differentiate specificity differences, since sera that are nonneutralizing at 90% inhibition may possess low levels of neutralizing antibodies which would blur the distinction.

TABLE-US-00003 TABLE III Properties of Neutralizing Antibodies Generated By Cluster Peptide P18 Immunization Cluster Neutralization IIIB/119 ELISA Peptide/ IIIB MN Binding PCLUS PCLUS IgM IgM IC 50* Mouse Strain Serum # 90% 50% 90% (IFA) p18 gp120 3 6 (P18) (pg120) IgG1 p18 gp120 PCLUS 1-18 B10.BR 6256 16 >64 0.81 1.66 0.29 0.16 0.59 0.63 4597 16 >64 0.67 1.69 0.14 0.09 4599 -- 8 0.75 1.66 0.15 0.09 Prebleed -- -- 0.00 Pool PCLUS 3-18 BALB/c 6242 8 >64 Negative 0.49 1.17 0.09 0.35 0.38 6243 8 32 0.44 1.17 0.06 0.42 0.33 0.91 9.30E-07 7.19E-10 6744 -- 32 0.08 0.41 0.33 6245 16 >64 Negative 0.98 1.68 0.05 0.02 0.41 0.33 6246 8 16 0.79 1.57 0.05 0.09 0.52 0.39 Prebleed -- -- Negative 0.13 0.20 0.03 0.02 Pool B10.BR 6252 -- 16 0.66 1.19 0.05 0.01 7803 -- 8 0.31 1.19 0.04 0.07 0.43 0.59 0.63 7804 -- 8 0.46 7806 -- 16 0.10 6253 -- 32 0.10 Prebleed -- -- 0.12 0.20 0.05 0.02 Pool B10.HTT 7128 -- 32 0.75 7129 -- 8 0.00 7130 -- 16 Negative 0.84 1.58 0.05 0.09 0.56 0.56 9.25E-07 5.14E-10 7131 -- 16 0.33 0.85 0.08 0.13 0.41 0.54 7132 -- 16 1.26 1.67 0.08 0.12 0.56 0.57 Prebleed -- -- 0.12 0.16 0.03 0.02 0.47 Pool B10.A(5R) 4832 -- 8 0.86 0.32 0.27 H3043 -- 8 0.51 0.37 0.40 Prebleed -- -- 0.00 Pool PCLUS 4-18 BALB/c 9748 -- 8 0.63 9749 -- 8 0.69 9750 8 32 1.19 1.34 0.06 0.01 9752 -- -- 0.06 B10.A(5R) 9776 -- -- 0.18 0.21 0.02 0.02 Prebleed -- -- 0.09 Pool PCLUS 6-18 BALB/c 6247 32 >64 Negative 0.44 1.56 0.06 1.20 0.47 0.58 6248 8 32 0.75 6249 8 Negative 0.78 6250 16 >64 1.29 6251 -- >64 0.95 0.41 0.58 0.82 6.77E-07 3.88E-10 9753 >64 >64 0.36 9754 32 >64 0.85 9755 >64 >64 0.47 9756 >64 >64 0.99 1.50 0.05 1.15 9757 16 32 1.17 Prebleed -- -- 0.08 Pool B10.BR 6359 16 >64 1.10 2.07 0.05 0.87 0.44 0.43 6254 -- 64 0.37 1.22 0.00 0.46 0.38 0.36 6255 8 32 Negative 1.14 1.33 0.00 1.23 7367 16 >64 1.21 9768 >64 >64 0.79 9769 -- 16 0.63 1.43 0.06 0.81 0.50 0.71 9770 -- -- 1:100 1.13 1.54 0.19 8.17 9771 16 >64 0.97 0.44 0.74 9772 32 >64 0.94 1.69 0.09 1.53 0.55 0.81 7.44E-07 2.07E-09 7309 >64 >64 1:1000 0.33 Prebleed -- -- 0.11 0.22 0.07 0.04 Pool B10.HTT 7827 8 >64 1.03 1.58 0.08 0.23 0.41 0.63 7828 16 64 1.10 1.54 0.12 0.96 0.49 0.51 7829 16 >64 1:10 0.90 1.58 0.03 1.02 0.69 0.50 7830 8 32 1:100 1.32 1.49 0.03 0.40 0.90 0.59 7822 -- 32 1.00 6267 -- >64 0.19 0.19 0.08 0.11 0.31 0.36 Prebleed -- -- 0.08 0.18 0.02 0.02 Pool B10.A(5R) 4578 -- 16 0.99 1.47 0.02 0.59 0.37 0.49 4579 8 >64 0.97 1.51 0.06 1.29 0.59 0.63 9777 >64 >64 1.10 1.64 0.08 0.25 0.44 0.74 2.73E-10 9778 -- 16 0.59 9779 -- 16 0.57 1.39 0.15 0.86 9780 8 32 0.86 9781 32 32 0.76 Prebleed Neg- 0.05 0.20 0.03 0.01 Pool ative IC 50 = concentration of peptide producing 50% inhibition in a competitive ELISA, as an estimate of Kd (inverse avidity)

TABLE-US-00004 TABLE IV Properties of Sera Boosted with 10 nanomole Cluster Peptide-P18 Neutralization Cluster Peptide/ ELISA IIIB MN Mouse Strain Serum # 1:1,000 1:10,000 90% 50% 90%* 50% PCLUS 3-18 BALB/c 6243 0.44 0.08 8 32 -- -- 6244 0.64 0.12 256 >512 6246 1.48 0.54 128 >512 PCLUS 6-18 BALB/C 6247 2.27 0.42 4068 >32,768 6251 1.49 0.66 64 512 -- -- 9753 1.45 1.41 16,384 >32,768 9754 1.47 0.72 16,384 >32,768 -- -- 9755 0.12 0.10 32 128 9757 1.75 1.32 16,384 >32,768 -- -- B10.BR 7309 1.12 0.17 4096 16,384 7367 1.25 0.34 4096 >32,768 9768 1.64 0.41 1024 2048 -- >512 9769 1.54 0.17 256 2048 -- -- 9770 1.50 0.36 256 2048 9771 1.65 0.86 128 512 9772 1.23 0.26 1024 >4096 -- >512 B10.HTT 7822 1.77 1.15 1024 >4096 128 512 7828 1.42 1.10 2048 >4096 128 >512 7829 1.39 0.68 512 >4096 128 >512 7830 1.19 0.43 16,384 >32,768 B10.A(5R) 9777 1.49 0.76 2048 >4096 -- 128 9778 1.23 1.02 128 512 -- >512 9779 1.55 0.41 512 2048 -- 64 9780 1.28 0.59 256 1024 -- 256

ELISA readings measured at 405 nm and reciprocal dilution neutralization titers of individual mouse sera 39-52 weeks following a single boost with 10 nanomole of cluster peptide 3 or 6 in four different strains of mice. Neutralization of homologous virus at 90% and 50% endpoints is indicated in the column labeled IIIB indicating HIV-1 IIIB and cross neutralization in the column MN representing HIV-1 MN. - indicates negative at the lowest dilution tested, 1:64.

Results in column 4 show direct binding to HIV IIIB-infected cells by immunofluorescence. Three of five sera from animals immunized with cluster peptide 6-18 and capable of neutralizing virus by 90% in the syncytium-forming assay exhibited binding to virus infected cells, although there was no correlation between neutralizing titer and IFA titer. Two neutralizing sera from mice immunized with PCLUS 3-18, #6245, and PCLUS 6-18, #6249, failed to bind infected cells by IFA. In addition, one of two nonneutralizing sera, from mouse #9770 immunized with PCLUS 6-18, exhibited binding to virus-infected cells despite the absence of any neutralizing activity at the level of 50% inhibition in the synctium forming assay. In this small sample we were not able to detect a difference between neutralizing and nonneutralizing sera by IFA.

Sera from animals immunized with the cluster peptides were tested for binding to rgp 120 in an ELISA assay. Note that it is only relevant to compare absorbance reading from the ELISA assays in Table III within columns and not between columns, since different reagents and assay conditions are employed for each type of ELISA. All immune sera tested bound rgp120 whereas prebleed control sera did not, and there was no significant difference in level found between neutralizing and nonneutralizing sera. In addition no difference between sera could be found in IgM levels which bound P18 or rgp 120. In a small sample, the isotype of antibodies elicited to P18 was determined to be IgG.sub.1. None of the sera tested showed any antibodies of the isotype IgG.sub.2a. No correlation was found between isotype and neutralizing activity.

Since antibodies elicited to the T helper sites may play a role in a subsequent viral infection, it was important to determine if sera from animals immunized with the cluster peptide-peptide-18 constructs contained any antibodies to the cluster peptides themselves. Also, if antibodies to the helper sites were neutralizing, these could account for the lack of correlation between binding to peptide 18 and neutralization. We tested this possibility in an ELISA using plates coated with cluster peptide 3 alone or cluster peptide 6. None of the animal sera tested showed binding to PCLUS 3. In contrast, all animals immunized with PCLUS 6-18 produced antibodies that were reactive to the T helper site and the level of antibody produced was proportional to their peptide 18 response, whereas none of the control animals produced anti-cluster peptide 6 antibodies. However, it is unlikely that neutralizing activity of these sera was due to antibodies to cluster peptide 6, because of its location in the intracytoplasmic tail of gp 41, and the lack of cross-neutralization of the MN strain (Table III), as discussed further below. Therefore, the neutralizing activity musts be primarily directed to the P18 portion of the construct.

It remained possible that differences in neutralizing ability were due to differences in affinity of antibodies. To test this possibility, sera selected on the basis of their ability to neutralize virus in the quantitative infective syncytium-plaque forming assay were mixed with various concentrations of peptide 18 or rgp 120 and allowed to achieve solution phase equilibrium. Dilutions of competitor were allowed to reach solution phase equilibrium in an overnight incubation at 4.degree. C. with a 10.sup.-3 dilution of each antiserum. Free antibody was then determined by short term incubation on competing peptide 18 or rgp 120 coated plates respectively, in an ELISA assay. Concentrations required for 50% maximal competition (IC.sub.50), as estimates of K.sub.d, were determined for each serum tested from the binding curves. Each experiment was repeated three to four times and representative results are shown in FIG. 4. Binding avidities (reciprocal of K.sub.d) of all sera tested were over two logs higher (i.e. IC.sub.50 two logs lower) when tested with rgp 120 than with peptide 18 (Table III). Binding avidities of neutralizing and nonneutralizing sera to peptide 18 were comparable, and any differences in avidity for rgp 120 were equivocable. In one case neutralizing serum from an animal immunized with cluster peptide 6-18, B10.BR mouse number 9772, showed a 5-fold lower binding avidity for rgp 120 than a corresponding nonneutralizing serum, BALB/c number 6251, whereas another neutralizing serum in this same group, B10.A(5R) mouse number 9777, showed only a 1.4 fold lower binding avidity than corresponding nonneutralizing serum 6251. The fact that the binding curves were not as steep as expected reveals that these sera were not homogeneous or monoclonal in the population of antibodies binding to either peptide 18 or rgp 120, and heterogeneity may have influenced these results. However, the fact that neutralizing sera often had lower avidities than nonneutralizing sera suggests that neutralization does not correlate with higher average avidity to peptide or to recombinant gp 120.

To explore the possibility that fine specificity differences might explain differences in neutralizing activity of sera with comparable peptide binding activity, we used peptides with single amino acid substitutions, in which single residues from the HIV-1-RF sequence replaced residues in the HIV-1-IIIB sequence, (see Table V, SEQ ID NOs:7-23) to test the effect of each residue on the binding of neutralizing and non-neutralizing sera from animals immunized with PCLUS 3-18 and PCLUS 6-18 (FIG. 5).

TABLE-US-00005 TABLE V Substituted Peptides Used for Determination of Binding Specificity (37) 315 329 18IIIB(P18) R I Q R G P G R A F V T I G K 18RF * * T K G P G R V I Y A T G Q 18-1 * 18-2 * 18-3 T 18-4 K 18-5 A 18-6 A 18-7 A 18-8 A 18-9 V 18-10 I 18-11 Y 18-12 A 18-13 T 18-14 A 18-15 O *Indicates a deletion .dagger-dbl. The underlined amino acids are substitutions in the 18 IIIB sequence.

Substitutions at neither the amino nor the carboxy-terminus of peptide 18 seemed to affect binding by either neutralizing sera, represented by the solid columns, or nonneutralizing sera, represented by the open columns, from animals immunized with PCLUS 3-18 or PCLUS 6-18. In fact, binding was enhanced over peptide 18 control when a tyrosine was substituted for a valine at position number 11 and substitutions at positions number 12, 13, 14, and 15 revealed a similar enhancement. Binding of both groups of sera was reduced when substitutions were made in the central loop region of the peptide 18 sequence PGRAF. This was not surprising since the sequence GPGR has been shown to be the binding site for neutralizing antibodies and maintains a well defined .bet