Patent Grant
7071322
1. Field of the Invention
This invention relates to HIV envelope polypeptides and vaccines containing the polypeptides.
2. Description of the Related Art
Acquired immunodeficiency syndrome (AIDS) is caused by a retrovirus identified as the human immunodeficiency virus (HIV). There have been intense efforts to develop a vaccine that induces a protective immune response based on induction of antibodies or cellular responses. Recent efforts have used subunit vaccines where an HIV protein, rather than attenuated or killed virus, is used as the immunogen in the vaccine for safety reasons. Subunit vaccines generally include gp120, the portion of the HIV envelope protein which is on the surface of the virus.
The HIV envelope protein has been extensively described, and the amino acid and nucleic acid sequences encoding HIV envelope from a number of HIV strains are known (Myers, G. et al., 1992. Human Retroviruses and AIDS. A compilation and analysis of nucleic acid and amino acid sequences. Los Alamos National Laboratory, Los Alamos, N. Mex.). The HIV envelope protein is a glycoprotein of about 160 kd (gp160) which is anchored in the membrane bilayer at its carboxyl terminal region. The N-terminal segment, gp120, protrudes into the aqueous environment surrounding the virion and the C-terminal segment, gp41, spans the membrane. Via a host-cell mediated process, gp160 is cleaved to form gp120 and the integral membrane protein gp41. As there is no covalent attachment between gp120 and gp41, free gp120 is sometimes released from the surface of virions and infected cells.
The gp120 molecule consists of a polypeptide core of 60,000 daltons which is extensively modified by N-linked glycosylation to increase the apparent molecular weight of the molecule to 120,000 daltons. The amino acid sequence of gp120 contains five relatively conserved domains interspersed with five hypervariable domains. The positions of the 18 cysteine residues in the gp120 primary sequence, and the positions of 13 of the approximately 24 N-linked glycosylation sites in the gp120 sequence are common to all gp120 sequences. The hypervariable domains contain extensive amino acid substitutions, insertions and deletions. Sequence variations in these domains result in up to 30% overall sequence variability between gp120 molecules from the various viral isolates. Despite this variation, all gp120 sequences preserve the ability of the virus to bind to the viral receptor CD4 and to interact with gp41 to induce fusion of the viral and host cell membranes.
gp120 has been the object of intensive investigation as a vaccine candidate for subunit vaccines, as the viral protein which is most likely to be accessible to immune attack. At present, clinical trials using gp120 MN strain are underway. However, to date no human vaccine trial has been of sufficient size to confirm or refute vaccine efficacy.
The development of candidate HIV-1 vaccines is burdened by the lack of in vivo or in vitro models of HIV-1 infection that accurately approximate the conditions of natural infection in humans. Several candidate HIV-1 vaccines [Berman et al.; J. Virol. 7:4464–9 (1992); Haigwood et al.; J. Virol. 66:172–82 (1992); Salmon-Ceron et al.; AIDS Res. and Human Retroviruses 11:1479–86 (1995)] have been described that elicit broadly cross-reactive antibodies able to neutralize a variety of diverse HIV-1 isolates in vitro. However, the relevance of in vitro assays to protective immunity in vivo is uncertain. Although several vaccines have provided chimpanzees with protection from challenge by homologous and heterologous strains of HIV-1, protection has not always correlated with in vitro neutralization assays carried out in T cell lines, or in lectin and cytokine activated peripheral blood mononuclear cells (PBMCs) [Berman et al.; Nature 345:622–5 (1990); Bruck et al.; Vaccine 12(12):1141–8 (1994); El-Amad et al.; AIDS 9:1313–22 (1995); Girard et al.; J. Virol. 69:6239–48 (1995); and Fulz et al; Science 256:1687–1690 (1992)]. While successful protection of chimpanzees is encouraging and has historically proved to be a reliable indicator of vaccine efficacy, the conditions of infection in all experimental models of HIV-1 infection differ significantly from natural infection in humans.
Experimental HIV-1 infection in vivo and in vitro both suffer from the limitation that the in vitro amplification of HIV-1, which is required to prepare virus stocks for in vitro or in vivo infectivity experiments, imposes a genetic selection that results in a spectrum of virus quasi-species that differ from the spectrum of variants present in the clinical specimens used to establish the culture [Kusumi et al.; J. Virol. 66:875 (1992); Meyerhans et al.; Cell 58:901–10 (1989)]. Because of these uncertainties, and even greater uncertainties related to the amount of virus transmitted, the site and cell type involved in initial replication, and the kinetics of virus dissemination, the ability of currently available in vitro or in vivo assays to reliably predict vaccine efficacy is questionable.
One of the candidate HIV-1 vaccines that have entered human clinical trials is recombinant gp120 prepared in Chinese hamster ovary (CHO) cells from the MN strain of HIV-1 (MN-rgp120) (Berman et al.; J. Virol. 7:4464–9 (1992)). To date, approximately 499 adults have participated in Phase 1 and 2 immunogenicity and safety trials of this vaccine. The data collected thus far suggest that MN-rgp120 is safe, immunogenic, and elicits high titers of neutralizing antibodies in greater than 95% of individuals immunized according to a 0, 1, and 6 month immunization schedule [Belshe et al.; JAMA 272(6):475–80 (1994); McElrath; Seminars in Cancer Biol. 6:1–11 (1995)]. However, during the course of these trials, nine vaccinees who received MN-rgp120 have become infected with HIV-1 through high risk behavior. Small trials, such as these, in populations with low rates of infection and minimally sized placebo control groups do not have sufficient statistical power to confirm or refute vaccine efficacy.
However, effective vaccines based on gp120 or another HIV protein for protection against additional strains of HIV are still being sought to prevent the spread of this disease.
Recombinant subunit vaccines are described in Berman et al., PCT/US91/02250 (published as number WO91/15238 on 17 Oct. 1991). See also, e.g. Hu et al., Nature 328:721–724 (1987) (vaccinia virus-HIV envelope recombinant vaccine); Arthur et al., J. Virol. 63(12): 5046–5053 (1989) (purified gp120); and Berman et al., Proc. Natl. Acad. Sci. USA 85:5200–5204 (1988) (recombinant envelope glycoprotein gp120).
Numerous sequences for gp120 are known. The sequence of gp120 from the IIIB substrain of HIV-1LAI referred to herein is that determined by Muesing et al., “Nucleic acid structure and expression of the human AIDS/lymphadenopathy retrovirus,” Nature 313:450–458 (1985). The sequences of gp120 from the NY-5, Jrcsf, Z6, Z321, and HXB2 strains of HIV-1 are listed by Myers et al., “Human Retroviruses and AIDS; A compilation and analysis of nucleic acid and amino acid sequences,” Los Alamos National Laboratory, Los Alamos, N. Mex. (1992). The Thai isolate CM244 is described by McCutchen et al., “Genetic Variants of HIV-1 in Thailand,” AIDS Res. And Human Retroviruses 8:1887–1895 (1992). The MN1984 clone is described by Gurgo et al., “Envelope sequences of two new United States HIV-1 isolates,” Virol. 164:531–536 (1988). As used herein, MN, MN-rgp120, the MN clone or isolate refers to MHGNE. The MNGNE amino acid sequence is Sequence ID NO:41.
Each of the above-described references is incorporated herein by reference in its entirety.
Oligonucleotide sequences encoding gp120 polypeptides from breakthrough isolates of vaccine trials using MN-rgp120 and the encoded gp120 polypeptides are provided. Use of the gp120 polypeptides from one or more of the isolates in a subunit vaccine, usually together with MN-rgp120, can provide protection against HIV strains that are sufficiently different from the vaccine strain (e.g.; MN-rgp120) that the vaccine does not confer protection against those strains. Antibodies induced by the polypeptides are also provided.
The present invention provides gp120 polypeptides from breakthrough isolates of HIV vaccine trials. Novel oligonucleotide sequences encoding gp120 from breakthrough isolates which can be used to express gp120 are also provided. Use of gp120 polypeptides from one or more of the isolates in a subunit vaccine, usually together with MN-rgp120, can provide protection against HIV strains that are sufficiently different from the vaccine strain (e.g.; MN-rgp120) that the vaccine does not confer protection against those strains.
In one embodiment, the vaccine is based on the use of the MN-rgp120 polypeptide (Sequence ID NO: 41) and gp120 polypeptides from MN-like viruses that include neutralizing epitopes that are not present in the initial vaccine strain, and are sufficiently different from those of the vaccine strain, to have been able to cause HIV-1 infections in MN-rgp120 vaccinated individuals (i.e.; to result in breakthrough infections). Use of the initial vaccine strain empirically determines the viruses present in the population that contain additional neutralizing epitopes sufficiently different from those of the vaccine strain to escape protection induced by the vaccine strain. Use of an initial representative gp120 polypeptide in a vaccine acts as a sieve so that viruses that are not effectively protected against by the vaccine strain breakthrough the vaccine, empirically resulting in determination of additional strains in a given geographic region that are not is protected against by the initial vaccine strain. Use of gp120 from those breakthrough isolates complements the vaccine isolate by providing additional neutralizing epitopes not present in the initial vaccine strain, therefore creating a more complete vaccine that confers protection against multiple different virus strains in the region.
Prior HIV-1 vaccine strategies were based on selection of appropriate candidate vaccine polypeptides based on homology alignment studies. However, since some of the neutralizing epitopes are conformation-dependent and the location of all of these epitopes is not known, this approach necessarily cannot determine all of the neutralizing epitopes that should be included in a vaccine for a particular region. In contrast, the present approach uses a selected representative strain and empirically determines strains that are sufficiently different and therefore breakthrough the barrier of protection provided by the initial vaccination program. Those strains can be included in the vaccine to confer more complete protection from HIV strains in the region. In addition, those strains can be used alone to confer protection against the breakthrough virus.
In another embodiment, the invention comprises a vaccine containing a first HIV gp120 polypeptide sequence and a breakthrough isolate HIV gp120 polypeptide sequence from a vaccinee vaccinated with a vaccine including the first HIV gp120 polypeptide sequence, the HIV gp120 polypeptide sequences being in a suitable carrier. Fragments of one or both HIV gp120 polypeptide sequences can be substituted for one or both of the corresponding HIV gp120 polypeptide sequences.
Preferably, the first gp120 polypeptide sequence contains neutralizing epitopes found in one or more gp120 polypeptides present in isolates from the geographical region where the initial vaccine (i.e., the vaccine that gives rise to the breakthrough isolate) is administered. More preferably, the first gp120 polypeptide sequence contains at least one of the more common neutralizing epitopes for the region, and most preferably the first gp120 polypeptide sequence contains at least one of the three most common neutralizing epitopes.
gp120 polypeptide sequences suitable for use as the first gp120 polypeptide sequence include gp120 MN, the Thai isolate CM244 sequence (hereinafter “gp120 CM244”), gp120 MN-GNE6 (Sequence ID NOs: 43 and 44; also known in the art as “gp120 GNE6”), and gp120 MIN-GNE8 (Sequence ID NO: 46; also known in the art as “gp120 GNE8”), and the like. gp120 MN, gp120 MN-GNE6, and gp120 MN-GNE8 are especially preferred for use as the first gp120 polypeptide sequence in initial vaccines for North America. gp120 CM244 is especially preferred for use as the first gp120 polypeptide sequence in initial vaccines for Thailand.
In a variation of this embodiment, the vaccine includes two different (i.e., first and second) gp120 polypeptide sequences, or fragments thereof, in combination with a breakthrough isolate HIV gp120 polypeptide sequence. The latter can be from a vaccinee vaccinated with either or both of the first and second HIV gp120 polypeptide sequences.
Exemplary vaccines include those containing combinations of gp120 MN, gp120 CM244, gp120 MN-GNE6 (Sequence ID NOs: 43 and 44), and gp120 MN-GNE 8 (Sequence ID NO: 46). Combinations of gp120 MN and gp120 CM244 or gp120 MN-GNE8 (Sequence ID NO: 46) with a breakthrough isolate HIV gp120polypeptide sequence are especially preferred.
In vaccines containing gp120 MN, the breakthrough isolate HIV gp120 polypeptide sequence can be an HIV gp120 polypeptide sequence selected from the group consisting of Sequence ID NOs: 2, 5, 8, 10, 12, 16, 19, 23, 25, 28, 31, 33, 36, and 39, and fragments thereof.
The term “subunit vaccine” is used herein, as in the art, to refer to a viral vaccine that does not contain virus, but rather contains one or more viral proteins or fragments of viral proteins. As used herein, the term “multivalent”, means that the vaccine contains gp120 from at least two HIV isolates having different amino acid sequences.
The term “breakthrough isolate” or “breakthrough virus” is used herein, as in the art, to refer to a virus isolated from a vaccinee.
The terms “amino acid sequence”, “polypeptide sequence”, and “polypeptide” are used interchangeably herein as in the art, as are the terms “nucleic acid sequence”, “nucleotide sequence”, and “oligonucleotide”.
Polypeptides from Breakthrough Isolates
The gp120 polypeptides of this invention correspond to the amino acid sequences of seven breakthrough isolates which are illustrated below in Table 1. A polypeptide of this invention includes an HIV gp120 amino acid sequence illustrated in Table 1 (Sequence ID NOs: 1, 4, 7, 9, 11, 15, 18, 22, 24, 27, 30, 32, 35, and 38) and fragments thereof. The polypeptides of this invention can include fused sequences from two or more HIV gp120 or gp160 amino acid sequences.
The polypeptide can also be joined to another viral protein, such as a flag epitope amino acid sequence. The term “flag epitope” is used herein, as in the art, to denote an amino acid sequence that includes an epitope recognized by a monoclonal antibody. Flag epitopes facilitate using single monoclonal antibody affinity purification of a plurality of different recombinant proteins, each having the flag epitope recognized by the monoclonal antibody. Numerous amino acid sequences can function as flag epitopes. The N-terminal sequences of Herpes Simplex Virus Type 1 (HSV-1) glycoprotein D (gD-1) is conveniently used as the flag epitope and its use is described in detail in the examples. The flag epitope is conveniently fused to the N terminus of the HIV gp120 polypeptide sequence. Alternatively, however, monoclonal antibodies that recognize neutralizing epitopes in the rgp120 sequences can be used to affinity purify the amino acid sequences, and a flag epitope can be omitted.
In addition, various signal sequences can be joined to a polypeptide of this invention. Although rgp120 is secreted to some extent in HIV cultures, the amount of the envelope glycoprotein released from (secreted by) the host cells varies widely from strain to strain. Various signal sequences can be introduced into the polypeptide by joining a nucleotide sequence encoding the signal sequence to the nucleotide sequence encoding the rgp120 to facilitate secretion of rgp120 from the cells. For example, Chiron HIV gp120 polypeptides include a signal sequence from tissue plasminogen activator (TPA) that provides good secretion of rgp120. Additional signal sequences are well known and include the N-terminal domain of murine leukemia virus surface protein gp70 described by Kayman et al., J. Virol. 68:400–410 (1984).
Table 1 illustrates the nucleotide and deduced amino acid sequences for two clones of each the seven breakthrough isolates of this invention. The clones are: C6.1; C6.5; C8.3; C8.6; C15.2; C15.3; C7.2; C7.10; C11.5; C11.7; C10.5; C10.7; C17.1; and C17.3. These sequence are SEQ. ID. NOs: 1–40. The amino acid sequence for MN and the nucleotide and deduced amino acid sequences for MN-GNE6 and MN-GNE8 are illustrated in the sequence listing hereinafter. In the listing for MIN-GNE6, a stop codon appears at amino acid residue position 51. This stop codon can be replaced with a codon encoding the corresponding amino acid from MN or MN-GNE8 or another isolate.
In addition to the listing in Table 1,
In one embodiment, a gp120 polypeptide of this invention has the same amino acid sequence as the sequence of one of the breakthrough isolates. In another embodiment, the amino acid sequence is truncated, as described in detail hereinafter. In another embodiment, a gp120 polypeptide sequence of this invention contains a substitution, insertion, or deletion (alteration) of one or more amino acids in the sequence of a breakthrough isolate. Usually, with the exception of amino acids that are not present in a truncated amino acid sequence and eliminate an epitope, a gp120 polypeptide of this invention will include alterations in the amino acid sequence of a breakthrough isolate that do not alter the polypeptide's ability to induce the same neutralizing antibodies as the amino acid sequence of the isolate.
In general, substitutions in the amino acid sequence of a gp120 polypeptide of this invention are conservative substitutions, particularly for amino acid residues in the V2, V3, and C4 domains of gp120, which domains contain neutralizing epitopes. However, non-conservative substitutions, particularly in domains that do not contain neutralizing epitopes are contemplated.
Conservative substitutions replace an amino acid with an amino acid of similar size and character. For example, a hydrophobic residue or hydrophilic residue is replaced with another hydrophobic residue or hydrophilic residue, respectively. Amino acids can be divided into the following groups: positively charged residues (K, R and H); negatively charged residues (D and E); amides (N and Q); aromatics (F, Y, and W); hydrophobics (P, G, A, V, L, I, and M); and uncharged residues (S and T). Usually, residues within a group are replaced with another member of the group.
In one embodiment, critical amino acid residues in the V2, V3, and C4 domains of gp120 are identical to the corresponding residues in a breakthrough-isolate sequence. Critical amino acid residues in the V2, V3, and C4 domains of gp120 are described in the experimental section. In another embodiment, all amino acid residues in the V2, V3, and C4 domains of gp120 are identical to corresponding residues in a breakthrough isolate sequence.
Oligonucleotide Encoding gp120 from Breakthrough Isolates
The present invention also provides novel oligonucleotides encoding gp120 from the breakthrough isolates which can be used to express gp120. An oligonucleotide of this invention encodes a polypeptide of this invention. The oligonucleotide can be DNA or RNA, usually DNA. Although numerous nucleotide sequences can encode the same amino acid sequence due to the degeneracy of the genetic code, conveniently, the oligonucleotides of this invention include a nucleotide sequence of a breakthrough isolate as illustrated in Table 1 (Sequence ID NOs: 2, 5, 8, 10, 12, 16, 19, 23, 25, 28, 31, 33, 36). Usually, an oligonucleotide of this invention is less than about 5 kilobases (kb), preferably less than about 3 kb.
To express the encoded amino acid sequence, the oligonucleotide can be inserted into a transcription unit. The transcription unit can be inserted into a plasmid for production of cell lines, inserted into a virus (e.g.; vaccinia) or can be used directly as a DNA vaccine. Suitable transcription units for production of vaccine proteins are well known. A preferred expression vector, designated psvI6B5, is illustrated in Sequence ID NO: 45. The vector includes an HSV-1 gDl signal sequence joined to a linker sequence. The gp120 nucleotide sequence to be expressed starts with the Kpn I site of the gene. Since all gp120 or gp160 sequences contain this site, any gp120 nucleotide sequence can be analogously inserted into the vector and expressed. The vector ends with a poly A tail from SV40.
In addition to being useful to express a polypeptide sequence of this invention, the oligonucleotides of this invention can also be used in diagnostics to detect HIV isolates. For example, the oligonucleotide or a portion thereof encoding a neutralizing epitope can be used in branched chain DNA diagnostics or as a probe in in situ hybridization studies.
Vaccine Preparation
A gp120 polypeptide of this invention from a selected breakthrough isolate(s) in a suitable carrier is used to make a subunit vaccine. The polypeptide can be used alone, but is generally administered in a multivalent subunit vaccine that includes gp120 MN. In addition to one or more gp120 polypeptides of this invention, the vaccine generally includes the MN polypeptide (hereinafter, MN-rgp120). The vaccine usually includes about 3 to about 5 different gp120 polypeptides, but 30 or more different gp120 polypeptides can be used.
Preparation of gp120 polypeptides for use in a vaccine is well known and is described hereinafter. With the exception of the use of the selected HIV isolate, the gp120 subunit vaccine prepared in the method does not differ from gp120 subunit vaccines of the prior art.
As with prior art gp120 subunit vaccines, gp120 at the desired degree of purity and at a sufficient concentration to induce antibody formation is mixed with a physiologically acceptable carrier. A physiologically acceptable carrier is nontoxic to a recipient at the dosage and concentration employed in the vaccine. Generally, the vaccine is formulated for injection, usually intramuscular or subcutaneous injection. Suitable carriers for injection include sterile water, but preferably are physiologic salt solutions, such as normal saline or buffered salt solutions such as phosphate-buffered saline or ringer's lactate. The vaccine generally contains an adjuvant. Useful adjuvants include QS21 (Quillaja saponaria, commercially available from Cambridge Biotech, Worcester, Mass.), which stimulates cytotoxic T-cells, and alum (aluminum hydroxide adjuvant). Formulations with different adjuvants which enhance cellular or local immunity can also be used. In particular, immunopotentiators such as cytokines can be included in the vaccine. Examples of suitable immunopotentiating cytokines include interleukins, such as interleukin-2 (IL-2) and interleukin-12 (IL-12), and tumor necrosis factor-alpha (TNF-α).
Additional excipients that can be present in the vaccine include low molecular weight polypeptides (less than about 10 residues), proteins, amino acids, carbohydrates including glucose or dextrans, chelating agents such as EDTA, and other excipients that stabilize the protein or inhibit growth of microorganisms.
The vaccine can also contain other HIV proteins. In particular, gp41 or the extracellular portion of gp41 or HIV-1 core proteins such as P24, P17, and P55 can be present in the vaccine. Although the amino acid sequence of gp41 is more conserved than that of gp120, gp41 contains neutralizing epitopes. Preferably, any gp41 present in the vaccine is from an HIV isolate present in the vaccine. gp160 from an isolate used in the vaccine can replace gp120 in the vaccine or be used together with gp120 from the isolate. Alternatively, gp160 from a different isolate than those in the vaccine can additionally be present in the vaccine.
Vaccines according to the invention can also contain one or more soluble gp120 polypeptide sequences, or fragments thereof, in combination with an engineered virus specifically designed to express proteins that induce a cytotoxic T-cell response. Suitable engineered viruses are derived from, for example, Canary Pox virus, vaccinia viruses, attenuated human herpes viruses (such as, e.g., herpes simplex viruses), and Varicella Zoster. Exemplary engineered viruses are modified to express any HIV protein capable of inducing a cytotoxic T-cell response, such as those described above. Typically, immunization with the gp120/engineered virus vaccine is followed by administration of one or more doses of the gp120 polypeptide sequence(s) to boost the immune response. If desired, viruses can be engineered to express one or more gp120 polypeptide sequences of the invention, or fragments thereof, and used in vaccines with or without soluble gp120 polypeptide sequences.
Vaccine formulations generally include a total of about 300 to 600 .mu.g of gp120, conveniently in about 1.0 ml of carrier. Preferred formulations include use of twice the weight of a gp120 polypeptide in twice as 600 .mu.g alum. However, formulations having smaller amounts (e.g.; 50 .mu.g per dose) are also used, generally with alum or other adjuvants. The amount of gp120 for any isolate present in the vaccine will vary depending on the immunogenicity of the gp120. For example, gp120 from some strains of HIV may be less immunogenic than gp120 from the MN strain (Sequence ID NO: 41). If two strains having different immunogenicity are used in combination, empirical titration of the amount of each virus would be performed to determine the percent of the gp120 of each strain in the vaccine. For isolates having similar immunogenicity, approximately equal amounts of each isolate's gp120 would be present in the vaccine. For example, in a preferred embodiment, the vaccine includes gp120 from the MN and a strain of this invention at concentrations of about 300 .mu.g per strain in about 1.0 ml of carrier. When the vaccine includes gp120 from about 30 isolates, about 10 to about 50 .mu.g can be used. Methods of determining the relative amount of an immunogenic protein in multivalent vaccines are well known and have been used, for example, to determine relative proportions of various isolates in multivalent polio vaccines.
The vaccines of this invention are administered in the same manner as prior art HIV gp120 subunit vaccines. In particular, the vaccines are generally administered at 0, 1, and at 6, 8 or 12 months, depending on the protocol. A preferred protocol includes administration at 0, 1, 6, and 12 months. Following the immunization procedure, annual or bi-annual boosts can be administered. However, during the immunization process and thereafter, neutralizing antibody levels can be assayed and the protocol adjusted accordingly.
The vaccine is administered to uninfected individuals. In addition, the vaccine can be administered to seropositive individuals to augment immune response to the virus, as with prior art HIV vaccines. It is also contemplated that DNA encoding the strains of gp120 for the vaccine can be administered in a suitable vehicle for expression in the host. In this way, gp120 can be produced in the infected host, eliminating the need for repeated immunizations. Preparation of gp120 expression vehicles is described hereinafter.
Although the gp120 isolates described herein can be used as a vaccine as described above, the amino acid sequences can also be used alone or in combinations in the same type of formulation for use as an immunogen, to induce antibodies that recognize the isolate(s) present in the immunogen. Immunogens are formulated in the same manner as vaccines and can include the same excipients, etc. Antibodies induced by the immunogens can be used in a diagnostic to detect the HIV strain in the immunogen or to affinity purify the strain.
gp120 Polypeptide Sequences and Chemokine Receptors
While CD4 is the primary cellular receptor for HIV-1, it is not sufficient for entry of HIV-1 into cells. Co-receptors required in conjunction with CD4 have been identified. These co-receptors are members of the chemokine receptor family of seven-transmembrane G-protein coupled receptors. The chemokine superfamily is subdivided into two groups based on the amino terminal cysteine spacing. The CXC chemokines are primarily involved in neutrophil-mediated inflammation, and the CC chemokines tend to be involved in chronic inflammation. At least five CC chemokine receptors, designated CC-CKR1–5 (also known in the art as CCR1–5), and at least four CXC chemokine receptors, designated CXC-CKR1–4 (also known as CXCR-1–4), have been identified.
CXC-CKR-4 (CXCR-4), which has also been called the alpha-chemokine receptor fusin, serves as an entry cofactor for T-cell-tropic HIV-1 strains. CC-CKR-5 (CC-RS), which has been called beta-chemokine receptor, together with its related family members, such as CC-CKR-2b and CC-CKR3, serve as entry cofactors for macrophage-tropic HIV-1 strains. T-cell-tropic strains can infect primary T-cells and T-cell lines, but not macrophages, whereas macrophage-tropic strains can infect macrophages and primary T-cells, but not T-cell lines. T-cell- and macrophage-tropic strains are discussed more fully in Deng et. al., Nature 381:661–666 (1996), which is hereby incorporated by reference in its entirety. Examples of T-cell-tropic strains include laboratory isolates, such as IIIB and MN. Macrophage-tropic strains include primary isolates, including but not limited to CM244, GNE6, GNE8, and breakthrough viruses from vaccinees immunized with gp120-based vaccines. Dual-tropic strains can, use both types of co-receptors, entering cells via CXC-CKR-4 or via one or more CC-CKR family members, preferably CC-CKR-5, CC-CKR-2b, or CC-CKR-3. While the present invention is not intended to be bound or limited by any one theory, the entry of T-cell tropic and macrophage-tropic HIV-1 strains is believed to provide a unifying explanation of the differences in cell tropism between viral strains, the resistance to HIV-1 infection by many CD4-transfected nonprimate cells, and the HIV-1-infection resistance of a portion of the human population.
Accordingly, in one embodiment is a vaccine containing (1) a first gp120 polypeptide sequence, or fragment thereof, from a macrophage-tropic HIV-1 strain and/or a second gp120 polypeptide sequence, or fragment thereof, from a T-cell tropic strain, in combination with (2) a breakthrough isolate HIV gp120 polypeptide sequence, or fragment thereof, from a vaccinee vaccinated with the first and/or second HIV gp120 polypeptide sequence. Preferably, the vaccine includes at least two gp120 polypeptide sequences that bind to different chemokine receptors. In one embodiment, the vaccine includes first and second gp120 polypeptide sequences that bind to different chemokine receptors. In addition, the breakthrough isolate gp120 polypeptide sequence can bind to a different chemokine receptor than the chemokine receptor(s) bound by either or both of the first and second gp120 polypeptide sequence(s).
A preferred T-cell tropic strain is a laboratory isolate, most preferably MN. Preferred macrophage-tropic viruses for use in the invention are GNE6 and GNE8, which are representative of the breakthrough viruses disclosed herein and differ from MN in that their gp120s induce the formation of antibodies that recognize the gp120 sequences (e.g., the V3 domain) involved in binding to CC chemokine receptors, such as CXC-CKR-5.
In one embodiment, HIV infection is prevented by administering one or more chemokine receptor-binding gp120 polypeptide sequences, or fragment(s) thereof containing appropriate chemokine receptor-binding domains, in a vaccine, such as those described above. Preferably, the vaccine also includes one or more CD4-binding gp120 polypeptide sequences or appropriate fragments thereof. Such vaccines induce anti-HIV antibodies that inhibit viral gp120-chemokine receptor or -CD4 binding. In addition, such gp120 polypeptides can directly inhibit HIV infection by binding to one or more co-receptors for HIV infection, such as CD4 or a chemokine receptor, thus providing a prophylactic or therapeutic effect in treating HIV infection. Preferably, gp120 polypeptide sequences useful in this regard contain the T-cell binding (TCB) domain.
Various uses of chemokine receptor-binding gp120 polypeptides are discussed below with regard to the CC chemokine receptor family. However, those skilled in the art recognize that this discussion applies equally to CXC chemokine receptors that act as cofactors in HIV infection.
The gp120 polypeptides can be used as a composition containing one or more gp120 polypeptides, as described for use as a vaccine or immunogen. The composition can be administered, prophylactically or therapeutically, to a patient at risk of infection or in need of such treatment using the dosages and routes and means of administration described herein. However, chronic administration may be preferred and dosages can be adjusted accordingly. It is noted that in vivo administration can also induce antibodies that bind viral gp120, further inhibiting virus binding to CC-CKR.
The gp120 polypeptides can also be used in screening assays to identify antagonists of CC-CKR. For example, candidate antagonists can be screened for inhibition of binding of gp120 to a CC-CKR CC-CKR receptor that is isolated and attached to a surface (e.g., plastic dish) or recombinantly or naturally expressed on the surface of a cell. Antagonists can either bind gp120 or bind receptor. Preferred candidate antagonists include gp120 compounds, small gp120 peptides (5 to 20 amino acids in length, preferably 7 to 10 amino acids in length) or peptidomimetics of gp120 that bind receptor, monoclonal antibodies that bind gp120, and small organic molecules that bind either gp120 or receptor.
The antibodies induced by the gp120 polypeptides can also be used to induce anti-idiotype antibodies that bind CC chemokines. These anti-idiotype antibodies can be screened for binding to an anti-gp120 polypeptide antibody and inhibiting gp120 from binding CC-CKR receptor. Such anti-idiotype antibodies mimic gp120 by binding to CC-CKR receptor. Such antibodies, preferably human antibodies, can be obtained in a number of ways, such as human antibodies from combinatorial libraries (e.g., Burton et al. Adv. Immunolo. (1994) 57:191–280). It is now possible to produce transgenic animals (e.g., mice) that are capable, upon immunization, of producing a full repertoire of human antibodies in the absence of endogenous immunoglobulin production. For example, homozygous deletion of the antibody heavy-chain joining region (JH) gene in chimeric and germ-line mutant mice results in complete inhibition of endogenous antibody production. Transfer of the human germ-line immunoglobulin gene array in such germ-line mutant mice results in the production of human antibodies upon antigen challenge as described in Jakobovitis et al., Proc. Natl. Acad. Sci. USA 90: 2551 (1993); Jakobovits et al., Nature 362:255–258 (1993); Bruggermann et al., Year in Immuno. 7: 33 (1993).
Alternatively, phage display technology as described by McCafferty et al., Nature 348:552–553 (1990) can be used to produce human antibodies and antibody fragments in vitro from immunoglobulin variable (V) domain gene repertoires from unimmunized donors. According to this technique, antibody V domain genes are closed in-frame either into either a major or minor coat protein gene of a filamentous bacteriophage, such as M13 or fd, and displayed as functional antibody fragments on the surface of the phage particle. Because the filamentous particle contains a single-stranded DNA copy of the phage genome, selections based on the functional properties of the antibody also result in selection of the gene encoding the antibody exhibiting those properties. Phage display can be performed in a variety of formats as reviewed by, for example, Johnson, et al., Current Opinion in Structural Biology 3:564–571 (1993).
Several sources of V-gene segments can be used for phage display. Clackson et al., Nature, 352: 624–628 (1991) isolated a diverse array of anti-oxazolone antibodies from a small random combinatorial library of V genes derived from the spleens of immunized mice. A repertoire of V genes from unimmunized human donors (or embryonic cells) can be constructed. It has been demonstrated that antibodies to a diverse array of antigens (including self-antigens) can be isolated essentially following the techniques described by Marks et al., J. Mol. Biol., 222: 581–597 (1991), or Griffith et al., EMBO J., 12: 725–734 (1993).
In a natural immune response, antibody genes accumulate mutations at a high rate (somatic hypermutation). Some of the changes introduced confer higher affinity, and B cells displaying high-affinity surface immunoglobulin are preferentially replicated and differentiated during subsequent antigen challenge. This natural process can be mimicked by employing the technique known as “chain shuffling” (Marks et al., Bio/Technol. 10:779–783 [1992]). In this method, the affinity of “primary” human antibodies obtained by phage display can be improved by sequentially replacing the heavy and light chain V region genes with repertoires of naturally occurring variants (repertoires) of V domain genes obtained from unimmunized donors. This technique allows the production of antibodies and antibody fragments with affinities in the nM range. A strategy for making very large phage antibody repertoires has been described by Waterhouse et al., Nucl. Acids Res., 21: 2265–2266 (1993).
Accordingly, antibodies that bind CC-CKR can be obtained by screening antibodies or fragments thereof expressed on the surface of bacteriophage in combinatorial libraries or in other systems as described above with a gp120 monoclonal antibody that inhibits gp120 binding to receptor.
In addition to screening antibodies with a gp-120 antibody, random or combinatorial peptide libraries can be screened with either a gp120 antibody or the gp120 compounds of the invention. Approaches are available for identifying peptide ligands from libraries that comprise large collections of peptides, ranging from 1 million to 1 billion difference sequences, which can be screened using monoclonal antibodies or target molecules. The power of this technology stems from the chemical diversity of the amino acids coupled with the large number of sequences in a library. See for example, Scott et al., Cur. Open Biotechnol. 5(1):40–8 (1994); Kenan et al. Trends Biochem. Sci. (1994) 19(2):57–64. Accordingly, the monoclonal antibodies, preferably human monoclonals or fragments thereof, generated as discussed herein, find use in treatment by inhibiting or treating HIV infection or disease progression, as well as in screening assays to identify additional pharmaceuticals.
Production of gp120
gp120 for a vaccine can be produced by any suitable means, as with prior art HIV gp120 subunit vaccines. Recombinantly-produced or chemically synthesized gp120 is preferable to gp120 isolated directly from HIV for safety reasons. Methods for recombinant production of gp120 are described below.
Oligonucleotides encoding gp120 from breakthrough isolates and capable of expressing gp120 can be prepared by conventional means. For example, the nucleotide sequence can be synthesized. Alternatively, another HIV nucleotide sequence encoding gp120 can be used as a backbone and altered at any differing residues as by site-directed mutagenesis. Site-directed mutagenesis is described in Kunkel et al, Proc. Natl. Acad. Sci. (USA) 82:488–492 (1985) and Zoller et al, Nuc. Acids Res. 10:6487–6500 (1982) and is well known.
In a preferred embodiment, the nucleotide sequence is present in an expression construct containing DNA encoding gp120 under the transcriptional and translational control of a promoter for expression of the encoded protein. The promoter can be a eukaryotic promoter for expression in a mammalian cell. In cases where one wishes to expand the promoter or produce gp120 in a prokaryotic host, the promoter can be a prokaryotic promoter. Usually a strong promoter is employed to provide high-level transcription and expression.
The expression construct can be part of a vector capable of stable extrachromosomal maintenance in an appropriate cellular host or may be integrated into host genomes. Normally, markers are provided with the expression construct which allow for selection of a host containing the construct. The marker can be on the same or a different DNA molecule, desirably, the same DNA molecule.
The expression construct can be joined to a replication system recognized by the intended host cell. Various replication systems include viral replication systems such as those from retroviruses, simian virus, bovine papilloma virus, or the like. In addition, the construct may be joined to an amplifiable gene, e.g. the DHFR gene, so that multiple copies of the gp120 DNA can be made. Introduction of the construct into the host will vary depending on the construct and can be achieved by any convenient means. A wide variety of prokaryotic and eukaryotic hosts can be employed for expression of the proteins.
Preferably, the gp120 is expressed in mammalian cells that provide the same glycosylation and disulfide bonds as in native gp120. Expression of gp120 and fragments of gp120 in mammalian cells as fusion proteins incorporating N-terminal sequences of Herpes Simplex Virus Type 1 (HSV-1) glycoprotein D (gD-1) is described in Lasky, L. A. et al., 1986 (Neutralization of the AIDS retrovirus by antibodies to a recombinant envelope glycoprotein) Science 233: 209–212 and Haffar, O. K. et al., 1991 (The cytoplasmic tail of HIV-1 gp160 contains regions that associate with cellular membranes.) Virol. 180:439–441, respectively. A preferred method for expressing gp120 is described in the examples. In the examples, a heterologous signal sequence was used for convenient expression of the protein. However, the protein can also be expressed using the native signal sequence.
An isolated, purified gp120 polypeptide having one of the amino acid sequences illustrated in Table 1 can be produced by conventional methods. For example, the proteins can be chemically synthesized. In a preferred embodiment, the proteins are expressed in mammalian cells using an expression construct of this invention. The expressed proteins can be purified by conventional means. A preferred purification procedure is described in the examples.
gp120 Fragments
The present invention also provides gp120 fragments that are suitable for use in inducing antibodies for use in a vaccine formulation. A truncated gp120 sequence, as used herein, is a fragment of gp120 that is free from a portion of the intact gp120 sequence beginning at either the amino or carboxy terminus of gp120. A truncated gp120 sequence of this invention is free from the Cs domain. The CS domain of gp120 is a major immunogenic site of the molecule. However, antibodies to the region do not neutralize virus. Therefore, elimination of this portion of gp120 from immunogens used to induce antibodies for serotyping is advantageous.
In another embodiment, the truncated gp120 sequence is additionally free from the carboxy terminal region through about amino acid residue 453 of the gp120 V5 domain. The portion of the V5 domain remaining in the sequence provides a convenient restriction site for preparation of expression constructs. However, a truncated gp120 sequence that is free from the entire gp120 V5 domain is also suitable for use in inducing antibodies.
In addition, portions of the amino terminus of gp120 can also be eliminated from the truncated gp120 sequence. In particular, the truncated gp120 sequence can be free from the gp120 signal sequence. The truncated gp120 sequence can be free from the carboxy terminus through amino acid residue 111 of the gp120 C1 domain, eliminating most of the C1 domain but preserving a convenient restriction site. However, the portion of the C1 domain through the V2 cysteine residue that forms a disulfide bond can additionally be removed, so that the truncated gp120 sequence is free from the carboxy terminus through amino acid residue 117 of the gp120 C1 domain. In a preferred embodiment, the truncated gp120 sequence is free from the amino terminus of gp120 through residue 111 of the C1 domain and residue 453 through the carboxy terminus of gp120.
The truncated gp120 sequences can be produced by recombinant engineering, as described previously Conveniently, DNA encoding the truncated gp120 sequence is joined to a heterologous DNA sequence encoding a signal sequence.
It is understood that the application of the teachings of the present invention to a specific problem or situation is within the capabilities of one having ordinary skill in the art in light of the teachings contained herein. Examples of the products of the present invention and representative processes for their isolation, use, and manufacture appear-below, but should not be construed to limit the invention. All literature citations herein are expressly incorporated by reference.
Specimen collection from human volunteers. Blood was collected from MN-rgp120-immunized individuals who were infected with HIV-1 while participating in Phase I (NIH Protocol AVEG 016) and Phase II (NIH Protocol AVEG 201) HIV-1 vaccine trials sponsored by the National Institutes of Health (NIH). The demographics of the subjects in the study, and the study design have been described in McElrath; Seminars in Cancer Biol. 6:1–11 (1995); McElrath et al.; Abstracts from Eighth Annual Meeting of the National Cooperative Vaccine Development Groups for AIDS. Bethseda, Md. 216 (1996). Specimens were obtained according to an informed consent protocol approved by the institutional review boards of the participating institutions. In the experimental section, the time of HIV-1 infection is specified with regard to data provided by the NIH AIDS Vaccine Evaluation Network where PCR (RNA) and/or serologic assays were used to detect HIV-1 infection.
Sample preparation for cloning HIV-1 envelope glycoproteins. Peripheral blood mononuclear cells (PBMCS) from HIV-1 infected vaccinees were prepared from heparinized venous blood by FICOLL-HYPAQUE gradient centrifugation. Cell number and viability were determined. After separation, PBMCs were washed twice in phosphate-buffered saline and suspended at a cell density of 6×106 cells/ml in PCR lysis buffer (50 mM KCl, 10 mM Tris (pH 8.4), 2.5 mM MgCl2, 0.1 mg/ml gelatin (Sigma), 0.45% NONIDET P40 detergent, 0.45% TWEEN 20 detergent (both detergents are commercially available from United States Biochemical Corp.) and 0.06 mg/ml Proteinase K (Gibco BRL) to lyse the cells. The lysate was incubated at 50–60° C. for 1 hour, followed by inactivation of the Proteinase K at 95° C. for 10 minutes. Lysates were shipped frozen and stored at −70° C. until use.
Polymerase chain reaction (PCR) amplification. Samples were subjected to two rounds of PCR amplification using the nested primers described below. In the first round, 25 μl aliquots of PBMC lysates (containing about 1 μg genomic DNA) were mixed with an equal volume of a PCR reaction mix containing 400 μM each dNTP, 200 μg/ml BSA (Sigma Chemical Corporation, RIA grade) and about 100 pmoles of each primer in 50 mM KCl, 20 mM Tris (pH 8.4) and 3 mM MgCl2. After an initial 10 minute denaturation step at 95° C., 5 units of Taq polymerase (AMPLITAQ, Perkin Elmer Cetus) were added during an 55° C. soak step, and samples were overlayed with mineral oil.
The PCR profile was as follows: 2 cycles having 1 minute at 55° C., 2.5 minutes at 72° C. and 1 minute at 94° C., followed by 28 cycles with 30 seconds at 55° C., 2.5 minutes at 72° C. and 45 seconds at 94° C., and an extension step at 72° C. for 5 minutes.
Aliquots of 10 .mu.l from the first-round reactions were re-amplified with appropriate nested primers in a final reaction volume of 100 .mu.l, using either the reagents and profile described above or the reagents and profile described in the PCR Optimizer Kit (Invitrogen.) PCR reaction products were purified using QIAQUICK-spin columns (Qiagen Inc.) The primer pair used in the first round was either 120.os.F (5′-gggaattcggatccAGAGCAGAAGACAGTGGCAATGA with homologous sequence at position 6248–6270 of HIVPV22) (SEQ. ID. NO: 47) or JM11A (5′-ctcgag-CTCCTGAAGACAGTCAGACTCATCAAG at position 6048–6074) (SEQ. ID. NO: 48) in the forward direction [Kusumi et al.; J. Virol. 66:875 (1992)] combined with 120.os.R (5′-ggtctagaagctttaGCCCATAGTGCTCCTGCTGCT-CC at position 7836–7859) (SEQ. ID. NO: 49) in the reverse direction. The internal nested primers were 120.BX.F (5′-gggcggatcctcgaGGTACCTGTRTGGAAAG-AAGCA at position 6389–6410; R: A or G) (SEQ. ID. NO: 50) and 120.is.R (5′-ggtctagaagctttaTGCTCCYAAGAACCCAAGGAACA at position 7819–7841; Y: T or C) (SEQ. ID. NO: 51). Heterologous primer sequences are shown in lower case letters.
Subcloning of PCR products and the expression of recombinant envelope glycoproteins as fusion proteins. The HIV-1 envelope glycoprotein gp120 sequences were cloned and expressed as chimeric genes and fusion proteins, where the signal sequence and 27 amino acids from the mature N terminus of herpes simplex virus type 1 (HSV-1) were fused to the N-terminal sequences of the gp120 genes, corresponding to amino acid 13 of the mature gp120 sequence. PCR products containing gp120 sequences from the breakthrough specimens were cloned into pRK5 expression plasmid as chimeric genes using combinations of restrictions sites engineered into the heterologous PCR primer tails and the Xho I site engineered into the N-terminal sequence of HSV-1 gD.
The resulting double-stranded DNA was sequenced with Sequenase and the dGTP Reagent Kit (United States Biochemical Corp.). Sequences from glycoprotein D were provided to enhance expression and to provide a flag epitope to facilitate protein analysis, as described in Berman et al.; J. Virol. 7:4464–9 (1992); Nakamura et al.; AIDS and Human Retroviruses 8:1875–85 (1992); and Nakamura et al.; J. Virol. 67:6179–91 (1993).
Briefly, isolated DNA fragments generated by the PCR reaction were ligated into a plasmid (pRK.gD-5, pRKgDstop) designed to fuse the gp120 fragments, in frame, to the 5′ sequences of the glycoprotein D (gD) gene of Type 1 Herpes Simplex Virus (gD-1) and the 3′ end to translational stop codons. The fragment of the gD-1 gene encoded the signal sequence and 25 amino acids of the mature form of HSV-1 protein. To allow for expression in mammalian cells, chimeric genes fragments were cloned into the pRK5 expression plasmid (Eaton et al., Biochemistry 291:8343–8347 (1986)) that contained a polylinker with cloning sites and translational stop codons located between a cytomegalovirus promotor and a simian virus 40 virus polyadenylation site.
The resulting plasmids were transfected into the 293s embryonic human kidney cell line (Graham et al., J. Gen. Virol. 36:59–77 (1977)) using a calcium phosphate technique (Graham et al., Virology 52:456–467 (1973)). Growth conditioned cell culture media was collected 48 hr after transfection, and the soluble proteins were detected by ELISA or by specific radioimmunoprecipitation where metabolically labeled proteins from cell culture supernatants were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (PAGE) and visualized by autoradiography as described in Berman et al., J. Virol. 63:3489–3498 (1989) and Laemmli Nature 227:680–685 (1970).
Serologic assays. Sera were assayed for antibodies to rgp120, antibodies to synthetic gp120 V3 domain peptides corresponding to sequences from the gp120 V3 domain, and antibodies able to inhibit the binding of MN-rgp120 to cell surface CD4 using serologic assays described in Berman et al.; J. Virol. 7:4464–9 (1992); Nakamura et al.; AIDS and Human Retroviruses 8:1875–85 (1992); and Nakamura et al.; J. Virol. 67:6179–91 (1993). Endpoint titers of antibody binding to gp120 and V3 peptides were determined using three fold-serial dilutions of sera. The endpoint dilution titer was defined as the last dilution that produced an optical density value that was two times higher than the mean of the optical densities of 1:50 diluted, pooled, normal human sera. Antibody titers were calculated by a computer program that interpolated values between antibody dilutions. The inter-assay coefficient of variation of positive control standard sera was 35%.
Binding of monoclonal antibodies to rgp120 from breakthrough viruses. An ELISA similar to that described by Moore et al.; AIDS 3:155–63 (1989) was used to measure the binding of various monoclonal antibodies (MAbs) to rgp120s from breakthrough viruses. Briefly, Nunc-Immuno plates (Maxisorp, certified) were coated (100 μl at 5 μg/ml in PBS at 4° C. overnight) with an affinity-purified sheep polyclonal antiserum to a peptide at the C terminus of gp120 (D7324, International Enzymes, Fallbrook, Calif.). After washing once with PBS-0.05% TWEEN-20 detergent, the plates were blocked with PBS-1.0% BSA for 30–60 minutes at room temperature. Cell culture supernatants from 293s cells, diluted to contain equivalent amounts of the gD-rgp120 fusion protein, were added and incubated for 2 hours at room temperature followed by three washes with PBS-0.05% TWEEN-20 detergent. Various MAbs were diluted in PBS-1.0% BSA and 100 μL of the diluted MAbs were added to each well and incubated for 1 hour at room temperature.
The plates were washed 3 times and incubated with 100 μl of a horseradish peroxidase-conjugated second antibody (goat anti-mouse or anti-human IgG, Cappel) for 1 hour at room temperature. After 3 washes the plates were developed and the OD492 (optical density at 492 nm) read in a plate reader. Growth conditioned cell culture supernatants were normalized by dilution based on binding by MAb 5B6 which is specific for HSV-1 glycoprotein D fusion protein.
Virus neutralization assays. The ability of vaccinee sera to inhibit infection of MT4 cells by HIV-1MN was measured in a cytopathicity assay where cell viability was quantitated using a calorimetric indicator dye, as described in Robertson et al.; J. Virol. Methods 20:195–202 (1988). Briefly, a virus stock of HIV-1MN (obtained from Dr. Michael Norcross, U.S. Food and Drug Administration) was prepared as the clarified supernatant from chronically infected H9/HIV-1MN cell culture. H9 cells chronically infected with HIV-MN were pelleted and resuspended in one-tenth the original volume of medium. Cell-associated virus was released by the mechanical shearing effects of rapid vortexing of the cells as described in Wrin et al.; J. Virol. 69:39–48 (1995).
An amount of virus sufficient to ensure complete cell lysis killing in 7 days was incubated with three-fold serial dilutions of test antisera, and then used to challenge MT4 T-lymphoid cells in 10% FCS/RPMI-1640 cell culture media. The cultures were incubated for 7 days at 37° C. in 5% CO2, and then cell viability was tested by the dye MTT, as described by Robertson et al.; J. Virol. Methods 20:195–202 (1988). Virus neutralization endpoints were quantitated by measurement of OD at 570–650 nm, and then the endpoint titers were calculated as the reciprocal of the antiserum dilution giving a signal that was two-fold above the control signal with unprotected (killed) cells. These titers were typically twice those calculated at 50% protection.
Immunization history of infected subjects. Since 1992, 499 adults have been immunized with MN-rgp120 in Phase I trials in low or moderate risk individuals and in a Phase II clinical trial involving moderate to high risk individuals. The studies described herein entail the genetic and immunologic characterization of the first seven of nine individuals who became infected with HIV-1 through high risk behavior during the course of these trials. A listing of the trials and summary of the status of the vaccinees is presented in Table 2A. A listing of the analysis of the vaccinees is presented in Table 2B.
Interval:
indicates interval between last immunization and detection of HIV-1 infection.
Three of the infections occurred in a Phase I trial (NIH Protocol AVEG 201) that compared the safety and immunogenicity of MN-rgp120 formulated in two different adjuvants (alum and QS21), and four of the infections occurred in a Phase II trial aimed at establishing the safety and immunogenicity of MN-rgp120 in various high risk groups (e.g., intravenous drug users, homosexual and bisexual males, and partners of HIV-1 infected individuals).
Of the seven infections studied (Table 3), two (C6 and C8) occurred after two injections, three (C7, C10 and C15) occurred after three injections, and two (C11 and C17) occurred after receiving the four scheduled injections. The interval between receiving the last immunization and becoming infected was 2 to 13.5 months.
413
4460
9707
Antibody response to gp120 in vaccinated individuals. The magnitude and specificity of the antibody response to MN-rgp120 was measured by ELISA in all infected individuals throughout the course of the immunization regime (
The antibody response observed in C7 (
An a typical antibody response was also seen in subject C15 (
Retrospective analysis of serum and plasma from subjects C6 (
Antibody titers to the V3 domain. To further characterize the antibody response to gp120, antibody titers were measured to a synthetic V3 domain peptide of MN-rgp120 containing the principal neutralizing determinant (PND). Five of the seven subjects developed good V3 titers (1:400 to 1:4000) after the second immunization, however two subjects (C7 and C15) required three immunizations before developing significant tiers (
The results obtained indicate that the ability to form antibodies reactive with the V3 domain at various time-points prior to HIV-1 infection is not a valid correlate of protective immunity against all strains of HIV-1.
CD4 Inhibition titers. Antibodies that block the binding of gp120 to CD4 represent a heterogeneous class of virus neutralizing antibodies. Some are known to bind to the C4 domain of gp120 [Nakamura et al.; J. Virol. 67:6179–91 (1993); Anderson et al.; J. Infect. Dis. 160:960–9 ((1989)], and some are known to recognize conformation dependent discontinuous epitopes [Berman et al.; J. Virol. 7:4464–9 (1992); Nakamura et al.; J. Virol. 67:6179–91 (1993); McKeating et al.; AIDS Research and Human Retroviruses 8:451–9 (1992); Ho et al.; J. Virol. 65:489–93 (1991); Barbas et al.; Proc. Natl. Acad. Sci. USA 91:3809–13 (1994)].
One way to detect antibodies to both types of epitopes is to measure the ability of vaccinee sera to prevent the binding of [125I]-labeled gp120 to cell surface CD4 [[Nakamura et al.; AIDS and Human Retroviruses 81875–85 (1992); Nakamura et al.; J. Virol. 67:6179–91 (1993)]. CD4 blocking titers were detected in all seven of the vaccinees prior to infection (
Virus neutralizing activity. The virus neutralizing activity of antisera from MN-rgp120-immunized subjects was measured using a calorimetric assay that measured the viability of MT-4 cells after incubation with antibody treated virus (HIV-1MN). Since the actual date of infection was not known for any of the breakthrough infections, and serum samples were collected infrequently, the magnitude of the neutralizing antibody response at the time of infection is not known for any of the vaccinees.
Of the seven infections examined, the serum sample closest to the time of infection was that obtained from C7, where a neutralizing titer of 1:15 to HIV-1MN was present three weeks prior to detection of HIV-1 infection (Table 4). In all other cases, however, the interval between the last injection and the time of infection was 10 to 25 weeks.
When sera from the two early infections were examined (Table 4), one individual (C6) had a peak neutralizing titer of 1:10 ten weeks prior to detection of HIV-1 infection, whereas the other individual (C8) had a neutralizing titer of 1:80 ten weeks prior to detection of HIV-1 infection. Subject C15, who was immunized according to an accelerated immunization schedule, developed a neutralizing titer of 1:35 after the third injection, 14 weeks prior to HIV-1 infection. Subject C10, who had a peak neutralizing titer of 1:200 following the third immunization (week 24), had no detectable titer at week 52, six months prior to the first indication of HIV-1 infection (week 78).
Subject C11 possessed a neutralizing titer of 1:90 at fourteen weeks prior to detection of HIV-1 and a peak titer of 1:500 following the third immunization. Similarly vaccinee C17 had a neutralizing titer of 1:150 fourteen weeks prior to infection and a peak titer of 1:400 at two weeks after the third immunization.
Based on the rate of decay of the gp120 response of approximately two months [Belshe et al.; JAMA 272(6):475–80 (1994)], as well as the observation that neutralizing titers of 1:150 decayed to 1:10 in 10 weeks in vaccinees C10 and C17, it appears that neutralizing titers in C8, C15, C11, and C17 could have declined to 1:10 or less in the intervals between the last pre-infection serum sample and the time of HIV-1 detection.
The results of these studies demonstrated that all vaccinees developed some level of virus-neutralizing antibodies at some time prior to HIV-1 infection, and that the magnitude of the neutralizing response was probably low at the time of infection. In general, the magnitude of the virus-neutralizing response observed in the individuals that became infected with HIV-1 was comparable to that seen in non-infected vaccinees as described in Belshe et al.; JAMA 272(6):475–80 (1994).
Sequences of Viruses. To evaluate the similarity of the breakthrough viruses with the vaccine antigen, nucleotide sequences for gp120 from all seven breakthrough viruses were determined. Envelope glycoprotein genes were amplified from proviral DNA using the polymerase chain reaction. Sequences were obtained by direct amplification of DNA from lysates of gradient-purified lymphocytes obtained directly from patient blood without any intermediate tissue culture or amplification step.
A listing of the complete gp120 sequences (two clones per specimen) is provided in
Interestingly, a high percentage (four of seven) of the breakthrough viruses differed from MN-rgp120 by 25–30% [Myers et al.; Retroviruses and AIDS Database, Los Alamos National Laboratory (1992 and 1995)]. Historically this degree of sequence variation is typical of inter-subtype (intra-clade) variation rather than intra-subtype variation which is expected to be in the 10–20% range [Myers et al.; Retroviruses and AIDS Database, Los Alamos National Laboratory (1992 and 19950]. Of the viruses with the greatest homology to MN-rgp120, two (C6 and C8) occurred as early infections, prior to complete immunization, and one (C17) occurred as a late infection.
Polymorphism in the V3 Domain. Of particular interest were polymorphisms in regions known to contain epitopes recognized by virus neutralizing antibodies. The best characterized neutralizing epitope, the principal neutralizing determinant (PND), occurs at the tip of the V3 loop. In subtype B viruses, approximately 60% possess the MN serotype-defining signature sequence, IGPGRAF (SEQ. ID. NO: 52), based on identity with the prototypic MN strain of HIV-1 [Berman et al.; Virol. 7:4464-9 (1992); Myers et al.; Retroviruses and AIDS Database, Los Alamos National Laboratory (1992 and 1995); La Rosa et al.; Science 249:932–5 (1990)].
Three of the viruses (C6, C8, and C17) possessed the MN serotype signature sequence (
The prevalence of viruses with PND sequences matching the breakthrough viruses ranged from a high of 1.3% (C7) to a low of 0.2% (C11) in a listing of 519 subtype B sequences compiled by the Los Alamos National Laboratory [Myers et al.; Retroviruses and AIDS Database, Los Alamos National Laboratory (1992 and 1995)]. Similarly low frequencies were observed in three other independently derived data sets (Table 6). The occurrence of these sequences did not differ significantly between data sets collected prior to 1985 [La Rosa et al.; Science 249:932–5 (1990)] and data collected 1992, or from a set of 160 epidemiologically unlinked individuals (B. Korber, personal communication). All four sets of data agreed that the prevalence of viruses with MN-like PND sequences was in the range of 60%. Based on this data, four of the seven breakthrough infections were determined to be caused by viruses that fell outside of the spectrum of viruses that the vaccine was expected to prevent.
Other features of breakthrough virus V3 domains. Like MN-rgp120, the V3 domains of all of the breakthrough viruses were 36 amino acids in length. However, all seven viruses differed from MN-rgp120 with respect to the number of glycosylation sites and with respect to the syncytium-inducing (SI) signature sequence.
The sequence of MN-rgp120 is somewhat unusual [Myers et al.; Retroviruses and AIDS Database, Los Alamos National Laboratory (1992 and 1995)] in that it lacks an N-linked glycosylation site at position 306 in the V3 domain. The lack of this glycosylation site does not appear to be antigenically significant since antisera to MN-rgp120 are known to neutralize a variety of viruses (e.g. SF-2, DU6587-5, DU4489-5, CC) that possess a glycosylation site at this position [Berman et al.; J. Virol. 7:4464–9 (1992)]
In addition, the V3 domain of MN-rgp120 possessed sequence polymorphisms (R at position 311, K at position 324, K at position 328) typical of syncytium inducing viruses [Fouchier et al.; J. Virol. 66:3183–87 (1992)], whereas all seven breakthrough viruses possessed sequences associated with non-syncytium-inducing viruses. Syncytium-inducing viruses have been associated with rapid disease progression [Tersmette et al.; J. Virol. 62:2026–32 (1988)] and T cell tropism [O'Brien et al.; Nature (London) 348:69–73 (1990); Shioda et al.; Nature (London) 349:167–9 (1991)]. To date viruses with these properties have not been recovered from any of the MN-rgp120 immunized volunteers.
Polymorphism in the V1, V2 and C4 domains. Previous investigations have identified additional neutralizing epitopes in the V1, V2 and C4 domains of gp120 [Nakamura et al.; J. Virol. 67:6179–91 (1993); McKeating et al.; AIDS Research and Human Retroviruses 8:451–9 (1992); Ho et al.; J. Virol. 65:489–93 (1991); Barbas et al.; Proc. Natl. Acad. Sci. USA 91:3809–13 (1994); McKeating et al.; J. Virol. 67:4932–44 (1993); Moore et al.; J. Virol. 67:6136–6151 (1993); Davis et al.; J. Gen. Virol. 74:2609–17 (1993)].
The best characterized of these neutralizing epitopes is in the C4 domain which has attracted special attention because antibodies binding to this area are known to block the binding of gp120 to CD4 [Moore et al.; AIDS 3:155–63 (1989); McKeating et al.; AIDS Research and Human Retroviruses 8:451–9 (1992)]. Because the epitope is located in a conserved (C) domain, naturally-occurring polymorphism in this region is far more limited than in other neutralizing epitopes. Nakamura et al.; J. Virol. 67:6179–91 (1993) reported that the binding of a number of neutralizing MAbs was dependent on K at position 429.
Comparison of the sequence of MN-rgp120 with other strains of HIV-1 showed that a common polymorphism, involving the substitution of E for K, occurs at this position. Indeed, substrains of the same virus isolate often show polymorphism at this position. The HXB2 substrain of HIV-1LAI contains K at position 429, whereas the BH10, IIIB, and LAV substrains of the HIV-1LAI contain E at this position [Nakamura et al.; J. Virol. 67:6179–91 (1993)]. Similarly, the 1984 isolate of HIV-1MN exhibited E at this position, while the 1990 isolate of HIV-1MN, used to produce MN-rgp120, possessed K at this position.
When the sequences of the infected vaccine recipients were examined (
Studies with monoclonal antibodies have defined neutralizing epitopes in the V1 and V2 domains of gp120 [McKeating et al.; J. Virol. 67:4932–44 (1993); Moore et al.; J. Virol. 67:6136–6151 (1993); Davis et al.; J. Gen. Virol. 74:2609–17 (1993)]. Like the polymorphisms that occur in the C4 domain, the V2 domains exhibit several common polymorphisms that affect the binding of virus neutralizing antibodies. One such polymorphism occurs at position 171 which is critically important for the binding of murine MAb 1025, whereas residue 187 is important for the binding of MAb several MAbs represented by 1088.
When the V2 domain sequences were examined (
Other neutralizing epitopes have been reported in the V1 domain of gp120 [O'Brien et al.; Nature (London) 348:69–73 (1990); McKeating et al.; J. Virol. 67:4932–44 (1993)]. Although the neutralizing epitopes in the V1 domain of MN-rgp120 have not been characterized, the polymorphism seen among the breakthrough viruses in this region was interesting. Particularly striking (
Although examination of sequence databases suggest that variation in the V2 region is comparable to the V1 region, the V2 region of the breakthrough viruses showed less variation than expected. Specifically, the length of the V2 region ranged from 36 amino acids (C7) to 39 amino acids in length, with six of seven viruses containing three N-linked glycosylation sites in this domain. A high degree of polymorphism was found in the V4 region where sequences ranged from 26 (C10) to 33 (C15, C7) amino acids in length and contained either 4 or 5 N-linked glycosylation sites.
Antigenicity of envelope glycoproteins from breakthrough viruses. To determine the significance of sequence variation on glycoprotein antigenicity, recombinant gp120 was prepared from the viruses of all seven infected vaccinees (
In control experiments, the binding of MAb 5B6 (which is specific for the HSV gD-1 flag epitope fused to the N terminus of all of the rgp120 protein) was used to standardize the amount of gp120 from each isolate (
The antigenic structure of the V3 domain was examined using the 1034 MAb (isolated from mice immunized with MN-rgp120 as described in Nakamura et al.; J. Virol. 67:6179–91 (1993) and the 50.1 MAb (prepared from mice immunized with a synthetic V3 domain peptide as described in Rini et al.; Proc. Nati. Acad. Sd. USA 90:6325–9 (1993). Both MAbs are known to exhibit potent virus neutralizing activity. When binding to the recombinant proteins was examined, the MAb binding to MN-rgp120 was at least 10-fold greater than to any of the breakthrough virus envelope proteins (
A distinction in the epitopes recognized by these MAbs was evident since C6-rgp120 and C8-rgp120 gave comparable binding with 50.1, whereas 1034 bound better to the C6-derived protein than the C8-derived protein. The poorest MAb reactivity was with rgpl20s from C11 and C15. This result was consistent with sequence analysis demonstrating that these two viruses both possessed the radical substitution of S for R at position 18 in the V3 domain. Surprisingly, both of these MAbs exhibited better than expected binding to rgp120 from the C7 and C10 viruses. Like MN-rgp120, both proteins contained the penta-peptide, IGPGR sequence (SEQ. ID. NO: 57) in the V3 loop, but differed from MN-rgp120 in that V and L replaced A and F at positions 319 and 320 in gp120 from C10, and W replaced F at position 320 in gp120 from C7. These results indicate that R at position 318 is essential for the binding of these two MAbs, and that the epitopes recognized by 1034 and 50.1 are not completely destroyed by the hydrophobic substitutions at positions 319 and 320.
As predicted from the sequence data, there was little if any binding to the breakthrough virus rgp120s using MAbs (1088 and 1025) directed to the V2 region of MN-rgp120 Also consistent with sequence data was the observation that MAb 1024 directed to the C4 domain of MN-rgp120 gave some reactivity with C17-rgp120 which, like MN-rgp120 contained K at position 429, but gave no reactivity with the other isolates that contained E at residue 429.
Together, these studies demonstrated that the antigenic structure of all seven breakthrough viruses differed from the vaccine immunogen at three well characterized neutralizing epitopes.
A totally different pattern of reactivity was observed with the human hybridoma, MAb 15e, prepared from an HIV-1 infected individual as described in Ho et al.; J. Virol. 65:489–93 (1991). With this MAb, the greatest binding was achieved with MN-rgp120 and rgp120 from C7, and the poorest reactivity was seen with the two clones of rgp120 from the C11. Moderate, but comparable reactivity was seen with rgp120s from the C10 and C17.
These results demonstrate that the 15e epitope is polymorphic, and that the epitope is conserved on MN-rgp120 and rgp120 from C7, but has been lost on rgp120s from C11. Interestingly, the two different clones of gp120 derived from C6 gave strikingly different patterns of antibody binding. Thus, rgp120 from clone C6.5 exhibited strong reactivity with this antibody, whereas rgp120 from clones C6.1 exhibited significantly weaker activity with this MAb. Comparison of sequence data (
In these studies, the viruses and immune responses in seven of nine vaccinees who became infected with HIV-1 vaccine through high risk activity while participating in Phase I or Phase 2 trials of MN-rgp120, a candidate HIV-1 vaccine were analyzed. Such infections would be expected to occur for one of two reasons: 1) lack of sufficient immune response at the time of infection; or 2) infection with viruses that fall outiside of the antigenic spectrum expected to be covered by the vaccine immunogen. The data indicate that both explanations may be involved with the infections observed (Table 8).
Two of the infections occurred in individuals who failed to receive the minimum three doses of vaccine typically required for the induction of protective immunity with protein subunit vaccines (e.g. hepatitis B virus formulated in alum adjuvant as described in Francis et al.; Ann. Int. Med. 97:362–6 (1982). Two additional breakthrough infections occurred in vaccinees who had weak or undetectable primary (C7) and booster (C15) responses. Of the three individuals who became infected with HIV-1 after receiving three or more productive immunizations (C10, C11, and C17), at least two, and possibly all three, appear to have become infected more than six months after receiving their last immunization. Because antibody titers to MN-rgp120 typically decay with a half-time of 2 to 2.5 months [Belshe et al.; JAMA 272(6):475–80 (1994); Berman et al.; AIDS 8:591–601 (1994)], antibody titers would be expected to have decayed at least eight-fold and possibly as much as sixty four-fold at the time of infection. Thus, the lack of a sufficient immune response at the time of infection represents a potential explanation for at least six of the seven breakthrough infections.
Data from vaccine efficacy studies in gp160 immunized chimpanzees [McElrath et al.; Longitudinal Vaccine-Induced Immunity and Risk Behavior of Study Participants in AVEG Phase II Protocol 201. In: Abstracts from Eighth Annual Meeting of the National Cooperative Vaccine Development Groups for AIDS. Bethseda, Md. 1996:216] challenged with HIV-1, and gp120-immunized rhesus macaques challenged with a chimeric SIV/HIV-1 virus (SHIV) suggest that the magnitude of the neutralizing antibody response at the time of infection is a critical correlate of protective immunity. If maintaining neutralizing antibody titers proves to be a valid correlate of protective immunity in humans, then formulations (e.g. novel adjuvants) or immunization regimes (frequent boosting) designed to maximize the antibody responses may be required to achieve long lasting protection. Use of a booster every six months may be advantageous.
The other likely explanation for the late infections is the antigenic difference between the vaccine and the breakthrough virus envelope glycoproteins. This explanation is supported by the observation that four of the seven breakthrough viruses possessed envelope glycoproteins that differed from the MN-rgp120 by 25–30% at the amino acid level. Differences of this magnitude have historically [Myers et al.; Retroviruses and AIDS Database, Los Alamos National Laboratory (1992 and 1995)] been associated with inter-subtype variation and far exceeds the average 10–20% variation expected for viruses within the same subtype.
Although the biologic significance of sequence variation in many regions of the envelope glycoprotein is unclear, polymorphism at neutralizing epitopes is an important factor that affects vaccine efficacy. Previous studies [Salmon-Ceron et al.; AIDS Res. and Human Retroviruses 11:1479–86 (1996); Javaherian et al.; Science 250:1590—3 (1990)] have demonstrated that the breadth of neutralizing activity that could be elicited by HIV-1 envelope derived vaccines was critically dependent on the sequence of epitopes in the V3 domain (e.g.; the PND). Thus, candidate vaccines based on the LAI strain of HIV-1 (the prototypic “non-MN-like” subtype B virus), exhibited little or no cross neutralizing activity with subtype B viruses, whereas vaccines that contained the “MN-like-” PND sequence (IGPGRAF) (SEQ. ID. NO: 52) exhibited broad cross neutralizing activity. That four of the seven breakthrough viruses possessed envelope glycoproteins with radical amino acid substitutions in the PND is consistent with the explanation that differences in antigenic structure explain some of these infections.
Over the last few years, it has become clear that polymorphism among “MN-like” viruses occurs at neutralizing epitopes outside of the PND. The best example occurs in the C4 domain where two antigenically distinct variants are distinguished by the presence of either K or E at position 429 [Moore et al.; AIDS 3:155–63 (1989)]. Because six of the seven breakthrough viruses differed from the vaccine strain in that they contained E rather than K at position 429, antibodies raised to the C4 domain of MN-rgp120 were unlikely to neutralize the viruses infecting in six of the seven vaccinees.
Other neutralizing epitopes are known to be present in the V1 and V2 domains of gp120. Although these regions are highly variable, due to insertions and deletions, neutralizing epitopes have been described by McKeating et al.; J. Virol. 67:4932–44 (1993); Moore et al.; J. Virol. 67:6136–6151 (1993); and Davis et al.; J. Gen. Virol. 74:2609–17 (1993). Several of these epitopes overlap an amino terminal sequence of the V2 domain containing the tri-peptide sequence RDK at positions corresponding to 142 to 144 of MN-rgp120 [McKeating et al.; J. Virol. 67:4932–44 (1993); Moore et al.; J. Virol. 67:6136–6151 (1993)]. Like the C4 epitope, variation in this sequence is known to occur between different substrains derived from the same parental isolate. Since all seven breakthrough viruses differed from MN-rgp120 in that they possessed the RDK sequence, rather than the GDK sequence present in the vaccine antigen, neutralizing antibodies to the V2 domain of MN-rgp120 would not have been expected neutralize any of the viruses recovered from the vaccinees immunized with MN-rgp120.
Although polymorphisms at neutralizing epitopes might account for the lack of protection in most of the infections, this does not appear to explain the infection of vaccinee C17, who was infected by a virus that matched MN-rgp120 in the V3 and C4 domains. If a difference in sequence was responsible for the lack of protection in this case, the critical difference might relate to the unusual sequence in the V1 domain of gp120 from this breakthrough virus. Several studies have shown that the V1 domain possesses epitopes recognized by virus neutralizing monoclonal antibodies [McKeating et al.; J. Virol. 67:4932–44 (1993); Davis et al.; J. Gen. Virol. 74:2609–17 (1993); Kayman et al.; J. Virol. 68:400–410 (1994)].
Although far less is known about the V1 epitopes relative to other neutralizing sites, the V1 epitopes appear to be conformation-dependent, and antisera from HIV-1 infected individuals recognize epitopes in the V1 and V2 domains [McKeating et al.; J. Virol. 67:4932–44 (1993); Kayman et al.; J. Virol. 68:400–410 (1994)]. The V1 sequence of the virus from C17 is noteworthy because it is smaller and contains fewer N-linked glycosylation sites than that of MN-rgp120 or any of the other breakthrough viruses. By the same token, the envelope glycoproteins from C11 and C6 are noteworthy because they are significantly larger and contain more glycosylation sites than MN-rgp120 or the other breakthrough viruses.
While differences in amino acid sequence can provide clues to differences in antigenic structure, the consequences of such polymorphism can only be proven through antibody binding studies. To correlate differences in sequence with differences in antigenic structure, gp120 from two clones each of all seven breakthrough viruses was expressed and the antigenicity of the clones with a panel of monoclonal antibodies was examined. As predicted from the sequence data, none of the breakthrough virus envelope glycoproteins reacted with neutralizing MAbs to the V2 domain of MN-rgp120. When MAbs to the C4 domain were examined, only the C17 envelope glycoprotein (that matched MN-rgp120 with respect to K429) showed significant, albeit lower, binding. Surprisingly, the three breakthrough envelope glycoproteins that contained the subtype B PND consensus sequence, IGPGRAF (SEQ. ID. NO: 52), gave poor reactivity with all three PND directed MAbs, even though they possessed PND sequences closely related to the vaccine immunogen. Thus, all three of the vaccinee isolates appeared to possess changes outside of the recognition site that interfered with MAb binding.
It has been known for many years that resistance to neutralization in vitro can sometimes be attributed to mutations in remote sequences that alter the conformation of neutralizing epitopes and interfere with recognition by virus neutralizing antibodies [Nara et al.; J. Virol. 64:3779–91 (1990); Cordonnier et al.; Nature 340:571–4 (1989)]. Together, these results indicate that the antigenic structure of the envelope glycoproteins recovered from the breakthrough viruses differed significantly from that of the vaccine antigen.
A novel result was the localization of residues in the C3 domain that appeared to affect the binding of the virus neutralizing human MAb, 15e. This MAb is known to recognize a discontinuous epitope, block CD4 binding, and neutralize a variety of laboratory and primary isolates of HIV-1 [Ho et al.; J. Virol. 65:489–93 (1991); Thali et al.; J. Virol. 66:5635–5641 (1992); Moore et al.; AIDS Res. Hum. Retroviruses 9:1179–1187 (1993)].
Comparative binding to envelope glycoproteins from the breakthrough viruses indicated that recognition by this antibody is critically dependent on residues in the C3 or C4 domains of gp120. The unique occurrence of a positively charged K at position 351 in the C3 domain provides a common explanation for the inability of the C11.5, C11.7 and C6.1 strains of HIV-1 to bind to 15e. Alternatively, it is possible that different amino acid substitutions in different locations account for the failure of 15e to bind to rgp120s from the C6 and C11 clones. The only obvious positions where substitutions of this type occur are in the C4 domain where T replaces M at 434 (C11) and T replaces I at 439.
The present studies demonstrate that the current formulation of MN-rgp120 is less than 100% effective against HIV-1 infection. Based on previous in vitro and in vivo studies with MN-rgp120, protection from natural HIV-1 infection in humans is expected to depend on a threshold concentration of virus-neutralizing antibodies, and antigenic similarity between the vaccine immunogen and the challenge virus.
In this regard, only one of the seven breakthrough infections (C17) was unexpected. This individual received a full course of immunizations yet became infected with a virus similar to MN-rgp120 at least two important neutralizing epitopes (V3 and C4 domains). This infection might be related to the magnitude of the antibody response at the time of infection, or antigenic differences between the breakthrough virus and the vaccine strain, or circumstances of infection (e.g., ulcerative lesions, infection by donor with acute infection or high viremia), not monitored in this protocol. Alternatively this individual may represent a true vaccine failure, without clear explanation.
On balance, the analysis of breakthrough infections described herein did not uncover any data that would discourage the continued development of MN-rgp120 as a vaccine to prevent HIV-1 infection. The results support speculation that enhancing vaccine immunogenicity (as by additional booster immunizations) may be required to maintain long term protective immunity, and that the addition of rgp120 from other antigenically different strains of virus in addition to MN-rgp120 are useful to expand the breadth of protection.
The availability of viruses and viral glycoproteins derived from breakthrough infections may provide an important means to streamline the process of identifying new antigens for inclusion into a multivalent vaccine. Recombinant viral glycoproteins prepared from breakthrough viruses, by definition, possess antigenic structures that are significantly different from MN-rgp120, and are be representative of viruses currently being transmitted. Thus, combining rgp120 from breakthrough viruses with MN-rgp120 is an effective way complement and significantly expand antigenic complexity and increase breadth of cross neutralizing activity.
This application is a divisional of application Ser. No. 09/419,362, filed Oct. 15, 1999, now U.S. Pat. No. 6,585,979 which is a divisional of application Ser. No. 08/889,841 (now U.S. Pat. No. 6,090,392), filed Jul. 8, 1997, which claims the benefit of U.S. Provisional Application No. 60/069,891 filed Jul. 8, 1996 abandoned.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4725669 | Essex et al. | Feb 1988 | A |
| 5166050 | Shriver et al. | Nov 1992 | A |
| 5420030 | Reitz, Jr. et al. | May 1995 | A |
| 5576000 | Reitz, Jr. et al. | Nov 1996 | A |
| 5792459 | Haigwood | Aug 1998 | A |
| Number | Date | Country |
|---|---|---|
| A-3332089 | Nov 1989 | AU |
| 0 187 041 | Jul 1986 | EP |
| 0 327 180 | Aug 1989 | EP |
| 0 335 635 | Oct 1989 | EP |
| 0 339 504 | Nov 1989 | EP |
| 0 527 760 | Feb 1993 | EP |
| 0 394 386 | Sep 1995 | EP |
| 0 187 041 | May 1996 | EP |
| 0 279 688 | Apr 1997 | EP |
| WO 8912095 | Dec 1989 | WO |
| WO 9002196 | Mar 1990 | WO |
| WO 9003984 | Apr 1990 | WO |
| WO 9104273 | Apr 1991 | WO |
| WO 9113906 | Sep 1991 | WO |
| WO 9115238 | Oct 1991 | WO |
| WO 9115512 | Oct 1991 | WO |
| WO 9320104 | Oct 1993 | WO |
| WO 9428929 | Dec 1994 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20040052821 A1 | Mar 2004 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60069891 | Jul 1996 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 09419362 | Oct 1999 | US |
| Child | 10371472 | US | |
| Parent | 08889841 | Jul 1997 | US |
| Child | 09419362 | US |
