PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Autoimmun Rev. Author manuscript; available in PMC 2009 June 1.
Published in final edited form as:
PMCID: PMC2527403
NIHMSID: NIHMS57180

Catalytic antibodies to HIV: Physiological role and potential clinical utility

Abstract

Immunoglobulins (Igs) in uninfected humans recognize residues 421–433 located in the B cell superantigenic site (SAg) of the HIV envelope protein gp120 and catalyze its hydrolysis by a serine protease-like mechanism. The catalytic activity is encoded by germline Ig variable (V) region genes, and is expressed at robust levels by IgMs and IgAs but poorly by IgGs. Mucosal IgAs are highly catalytic and neutralize HIV, suggesting that they constitute a first line of defense against HIV. Lupus patients produce the Igs at enhanced levels. Homology of the 421–433 region with an endogenous retroviral sequence and a bacterial protein may provide clues about the antigen driving anti-SAg synthesis in lupus patients and uninfected subjects. The potency and breadth of HIV neutralization revives hopes of clinical application of catalytic anti-421–433 Igs as immunotherapeutic and topical microbicide reagents. Adaptive improvement of anti-SAg catalytic Igs in HIV infected subjects is not customary. Further study of the properties of the naturally occurring anti-SAg catalytic Igs should provide valuable guidance in designing a prophylactic vaccine that amplifies protective catalytic immunity to HIV.

Keywords: Catalytic antibodies, B cell superantigen, HIV, lupus, passive immunotherapy, vaccination

Take-home messages

1. Adaptive immune responses do not generally control HIV infection because of the mutability of immunodominant gp120 epitopes. In contrast, pre-existing catalytic Igs in uninfected humans directed to the gp120 B cell SAg site neutralize HIV and probably provide partial protection against the virus.

2. The catalytic Igs recognize residues 421–433, which are essential for host CD4 receptor binding. Mucosal IgAs express highest catalytic activity. Serum IgG is poorly catalytic, suggesting distinctive B cell differentiation pathways favoring catalytic Ig synthesis.

3. Amplification of catalytic anti-421–433 Ig synthesis in HIV infected subjects is generally proscribed. However, catalytic Igs are produced at enhanced levels by uninfected lupus patients and after prolonged infection in subjects who do not develop AIDS. Sequence homology of the 421–433 epitope with a HERV sequence and microbial proteins may help shed light on host-virus evolutionary relationships.

4. Guidance for design of candidate peptide vaccines can be obtained by: (a) Determining the neutralization-relevant 421–433 conformation complexed with an anti-SAg Ig; and (b) identifying the antigen driving catalytic Ig synthesis in uninfected humans.

5. Catalytic Igs to the 421–433 epitope neutralize HIV more potently and broadly than conventional Igs. They are candidate passive immunotherapeutic reagents and vaginal microbicides.

1. Introduction

Infection with human immunodefiency virus-1 (HIV) causes acquired immunodeficiency syndrome (AIDS), characterized by depletion of CD4+ T cell, hyperglobulinemia and B cell hyperplasia. The major host cells are T cells and macrophages. Their infection occurs via binding of the coat protein gp120 trimer to CD4 receptors and chemokine co-receptors (mostly CCR5 and CXCR4).1 In addition, monomeric gp120 induces neuronal and CD4 T cell death, and the monomer shed from the virus may play a role in disease pathogenesis. Progression of infected humans to AIDS varies from a year to more than two decades.2 Some repeatedly exposed humans remain free of infection.3 The inability of the adaptive immune system to prevent and control infection derives from the structural variability of the HIV envelope. gp120 is composed of five comparatively constant (C) regions and five highly variable (V) regions. Most adaptive responses are directed against V domain immunodominant epitopes, which mutate rapidly. This allows emergence of escape viral mutants.4 Adaptive cytotoxic T cell and neutralizing Ig responses only provide transient protection.5, 6 gp120 structural variability also underlies the failure to develop an effective HIV vaccine. gp120 V domain sequences expressed by diverse HIV strains found around the world are highly variable. Protein and DNA vaccination strategies aimed at inducing protective T cell and Ig responses have been unsuccessful.7, 8

We review here catalytic Igs to a conserved epitope in the B cell superantigenic site (SAg) of gp120. The Igs are found at variable levels in humans without infection and are likely to provide partial protection against HIV. The suitability of the SAg epitope as a target for HIV immunotherapy and prophylactic vaccination are discussed.

2. Superantigenic character of gp120

B cell SAgs are antigens recognized by Ig V domains without the requirement of adaptive sequence diversification. The paired V domains of the heavy (VH) and light (VL) chain subunits of physiological IgG, IgA and IgM class Igs bind9 and catalyze the hydrolysis10, 11 of gp120 by recognizing its SAg site (Fig. 1A). Conserved framework regions (FRs) are involved in gp120 SAg recognition, assessed by V domain homology analysis and FR/complementarity determining region (CDR) swapping studies.12, 13 B cell receptor (BCR; surface Ig complexed to signal transducing proteins) engagement by SAgs is thought to downregulate B cells without dependence on T cells.14 The effect may be mediated by modulation of CD79b expression, a BCR associated signal transducing protein.15 Most reported B cell SAg binding Igs contain VH3 family V domains found in the majority of expressed Igs of healthy adults. The binding is characterized by moderate affinity.16 VH3 usage, however, is not the sole determinant of gp120 SAg recognition. A monoclonal catalytic VH2 family IgM from a human without infection recognizes the gp120 SAg site.10 Rare L chain Ig subunits isolated from phage display libraries can bind and hydrolyze the gp120 SAg site independent of the H chain.17 Igs with the most potent anti-viral activity are likely to be formed by pairing of VL and VH domains with the greatest gp120 binding capability. Ig catalytic sites are usually located in VL domains.18 VH pairing with catalytic VL domains will also enhance anti-viral potency. A single catalyst molecule hydrolyzes thousands of antigen molecules and hydrolysis of polypeptides usually results in loss of their biological activity.19 This is exemplified by observations of more rapid gp120 hydrolysis and superior HIV neutralization by salivary IgA from uninfected subjects compared to other Ig classes (Fig. 1B).

Fig. 1
Cleavage of biotinylated gp120 by salivary IgA, serum IgM, IgA and IgG from humans without HIV infection

Synthetic gp120 peptide studies have suggested that residues 421–433 are components of the SAg site.13 This peptide region also contains amino acids that are essential for binding to host CD4 receptors determined by previous structural studies.20 As CD4 recognition is obligatory for virus entry into host cells, the 421–433 epitope is relatively conserved in diverse HIV strains. IgMs,10 IgAs11 and isolated L chain subunits17 from uninfected humans hydrolyzed the 432–433 peptide bond located within this epitope. Catalytic selectivity was indicated by lack of hydrolysis of irrelevant proteins. Proteolytic Igs utilize nucleophilic mechanisms akin to serine proteases.18 We prepared a 421–433 peptide analog containing the strongly electrophilic phosphonate group (E-421–433) that binds covalently to the catalytic sites of serine proteases and proteolytic Igs.21 E-421–433 specifically and irreversibly inhibited the activity of the proteolytic Igs. IgA neutralizing activity was inhibited by E-421–433.11 These observations indicate noncovalent recognition of the peptide epitope coordinated with nucleophilic attack on the polypeptide backbone as the mechanism of catalysis and HIV neutralization.

3. What drives catalytic Ig formation?

Ig catalysis22 and SAg recognition23 are germline V gene encoded functions. The expression of these functions is subject to deterioration or improvement due to biased germline gene usage; V-D-J/V-J sequence diversification; the VH/VL combination; and somatic hypermutation occurring at the FRs/CDRs. IgMs and IgAs preferentially express gp120 SAg hydrolyzing activity, with only low level activity evident in IgGs.11 The activity levels of polyclonal Igs from different humans of individual monoclonal Igs are highly variable. No obvious antigenic stimulus driving the synthesis of the catalytic Igs exists in humans who are not infected with HIV. Occam’s razor suggests consideration of a random distribution of the SAg-selective catalytic activity that is expressed fortuitously as the Igs mature adaptively in response to irrelevant antigenic stimuli. However, studies summarized below suggest that the gp120 SAg hydrolyzing (and neutralizing) activity is open to selection and improvement under certain circumstances.

Polyclonal IgGs from patients with lupus but no HIV infection and mouse lupus models24 bind the 421–433 gp120 region at levels superior to controls without autoimmune disease. We isolated Ig light chain subunits (IgLs) and HIV neutralizing single chain Fv fragments that recognize the 421–433 epitope from lupus Ig libraries.17, 25 Mutations are frequent in the FRs and CDRs of these clones. Some of the Ig fragments catalyze the cleavage of gp120.17 We and others reported the tendency towards enhanced catalytic Ig synthesis in autoimmune disease.26, 27 Several groups have noted that HIV infection is infrequent in lupus patients.28, 29 Increased 421–433 binding and hydrolyzing Igs may be the reason. A generalized increase of polyreactive Igs to microbial proteins is an unlikely explanation, as the lupus Ig binding, hydrolytic and neutralizing activities are specific for the 421–433 region.

Concerning the antigenic stimulus driving synthesis of the 421–433 recognizing Igs without HIV infection, we found no human proteins in the databanks with sequence similarity to the gp120 421–433 region. However, 27 of 39 nucleotides that code for the consensus clade B residues 421–433 (CCGTATGTAACGAAAAGGATGAAAGACGGTGTACAAATA) are identical to a human endogenous retroviral sequence (HERV; TTAGATCTGATGAAAAGGATGAAAGAAATTTTTCAAAAA; identities underlined; nucleotides 303–341, rv 012650, family HERVL47).17 Expression of human endogenous retroviral sequences (HERVs), which constitute up to 8% of the human genome, is increased in lupus.30 Interestingly, certain anti-gp120 Igs are reported to recognize a HERV expressed by placental tissue.31 S. gordonii is a commensal bacterium that colonizes mucosal sites at varying densities. A probable transcription regulator-like (PRL) protein from this bacterium contains a peptide region homologous to gp120 residues 421–432 as shown below:

gp120421KQ-IINMWQEVGK432
S.gordonii PRL protein77KQFIINMSQNVGK89

Regrettably, amplification of protective catalytic Igs to the SAg is not routine in HIV infected subjects. However, we observed that prolonged infection for 5 years in a subpopulation of HIV infected individuals who remained AIDS-free is associated with a modest increase of catalytic IgAs to gp120.11 No increase was observed at an earlier stage in infection (6 months). In contrast, robustly increased conventional Igs with binding activity directed to the non-SAg gp120 epitopes were evident at this time. Evidently, therefore, immunological pathways that generate antigen binding Igs do not favor the synthesis of anti-SAg catalytic Igs.

Binding of conventional antigens to the BCR drives B cell division. BCR-catalyzed gp120 cleavage will result in release of the gp120 fragments, depriving the cells of the proliferative signal. In the conventional immune response, therefore, the catalytic activity can improve adaptively only to the extent that product release is slower than the proliferative transmembrane signal. IgAs/IgMs display superior proteolytic activity compared to IgGs. This is understandable if µ/α-BCRs generate more rapid transmembrane signals compared to γ-BCRs, a hypothesis that remains to be tested experimentally. A second scenario is that BCR catalysis itself induces cell proliferation. Peptide bond hydrolysis is highly exothermic and liberates large amount of energy. In theory, the catalytic energy could be utilized to induce favorable BCR signaling, e.g., by inducing a productive change in the BCR conformation or by increasing BCR diffusion in the lipid bilayer, thus increasing the probability of BCR cross-linking. Theoretically, BCR engagement by conventional antigens and SAgs could result in opposite cellular signaling, in that BCR catalyzed SAg hydrolysis and product release may lessen the down-regulatory and apoptotic effects of prolonged BCR occupancy by the SAg. Accordingly, adaptive maturation of B cells producing proteolytic Igs may be less restricted than of cells expressing BCRs that bind the SAg site stably.

4. Implications for host-virus evolutionary relationship (Fig. 2)

Fig. 2
A plausible phylogenetic explanation for the B cell superantigenic (SAg) character of modern HIV gp120 and (b) sequence similarity of the SAg site to an extant human endogenous retroviral sequence (HERV)

HIV is thought to have originated via zoonotic transmission from monkeys to humans a few decades ago.32 The sequence of HIV gp120 residues 421–433 is also conserved in simian immunodeficiency virus (SIV) envelope sequences in Los Alamos database. No virus homologs of HIV in lower organisms have been identified thus far. The ability of Igs to recognize SAgs is encoded by germline V genes, and HIV is neutralized as a consequence of Ig recognition. Modern B cell SAgs have probably helped shape the structure of germline Ig genes, in that the SAg recognition capability was presumably developed by Darwinian evolution over millions of years as a defense against microbial infection. These arguments support a hypothetical ancient homolog of HIV that drove the evolution of human germline-encoded anti-gp120 SAg Igs. Additional supporting evidence for the primordial HIV homolog is the existence of a HERV sequence homologous to the modern 421–433 region. HERVs are thought to be relics of viral sequences integrated into the host genome. They may fulfill a useful function in the evolution of novel host proteins with improved functions.33 However, HERVs also hold the potential of serving as building blocks for novel pathogenic viruses if expressed as functional genomes with appropriately packaged capsid and infective envelope structures.

5. Clinical prospects

We turn now to the utilitarian features of the catalytic anti-HIV Igs. gp120 SAg site binding by IgGs from uninfected humans (which are poorly catalytic) correlates inversely with the frequency of subsequent HIV infection.9 The protective effects of infusion of IgG from uninfected monkeys against SIV34 and of pooled IgG from uninfected humans (IVIG) against HIV has been investigated.35 Proteolytic Igs inactivate antigens irreversibly and with potency superior to conventional Igs.36 Catalytic salivary IgAs from uninfected humans neutralized primary, R5-dependent clade B and C strains. Saliva from healthy humans is available in abundant amounts, and pooled salivary IgA is a candidate for development as a passive immunotherapy formulation. HIV is transmitted mainly across mucosal surfaces. There is agreement that a safe and effective vaginally applied microbicide will help slow heterosexual HIV transmission. The salivary catalytic IgAs could be formulated as a vaginal microbicide. Similar considerations apply to the homogeneous single chain Fv constructs from the lupus library. These constructs display cross-clade HIV neutralizing activity, and their potency is superior to monoclonal Igs that have been advanced as candidate anti-HIV reagents.25, 37 This revives hopes for effective passive immunotherapy of HIV infection.

Developing a prophylactic vaccine is the highest priority in HIV research. Synthetic peptides are notorious for assuming conformations that differ from the corresponding epitope in native proteins. Structural elucidation of synthetic 421–433 analogs bound to the anti-SAg neutralizing Igs will be useful to reveal the neutralizing relevant conformation of the peptide. Constraining candidate peptide vaccines into the correct conformation will enhance the probability of inducing protective Igs. B cell differentiation pathways that favor adaptive improvement of Ig catalytic activity are virtually uncharted. Their elucidation will help design immunogens that induce protective catalytic immunity. Monoclonal Igs raised by immunization with an electrophilic analog of full-length gp120 (E-gp120) catalyze the hydrolysis of gp12038 and display cross-clade HIV neutralizing activity.39 Some of these Igs recognize the 421–433 region. However, E-gp120 contains numerous antigenic epitopes, and the overall polyclonal Ig response is insufficiently focused at the 421–433 epitope. A KLH-conjugate of residues 421–436 induces Igs that recognize native gp120,40 indicating that at least some peptide molecules assume the neutralization-relevant conformation of this epitope. The mucosal milieu supports the synthesis of IgAs with superior catalytic and neutralizing activities compared to serum Igs.11 Mucosal immunization, therefore, may help improve the protective quality of the Ig response.

Acknowledgments

Our work was funded by National Institutes of Health Grants, AI058865, AI067020, AI062455, AI071951 and RR024148 (CTSA), and by the Texas Higher Education Coordinating Board.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. Haase AT. Perils at mucosal front lines for HIV and SIV and their hosts. Nat Rev Immunol. 2005;5(10):783–792. [PubMed]
2. Cao Y, Qin L, Zhang L, Safrit J, Ho DD. Virologic and immunologic characterization of long-term survivors of human immunodeficiency virus type 1 infection. N Engl J Med. 1995;332(4):201–208. [PubMed]
3. Devito C, Hinkula J, Kaul R, et al. Cross-clade HIV-1-specific neutralizing IgA in mucosal and systemic compartments of HIV-1-exposed, persistently seronegative subjects. J Acquir Immune Defic Syndr. 2002;30(4):413–420. [PubMed]
4. Richman DD, Little SJ, Smith DM, Wrin T, Petropoulos C, Wong JK. HIV Evolution and Escape. Trans Am Clin Climatol Assoc. 2004;115:289–303. [PubMed]
5. Moore JP, Cao Y, Ho DD, Koup RA. Development of the anti-gp120 antibody response during seroconversion to human immunodeficiency virus type 1. J Virol. 1994;68(8):5142–5155. [PMC free article] [PubMed]
6. Soudeyns H, Pantaleo G. The moving target: mechanisms of HIV persistence during primary infection. Immunol Today. 1999;20(10):446–450. [PubMed]
7. Pitisuttithum P, Berman PW, Phonrat B, et al. Phase I/II study of a candidate vaccine designed against the B and E subtypes of HIV-1. J Acquir Immune Defic Syndr. 2004;37(1):1160–1165. [PubMed]
8. Sekaly RP. The failed HIV Merck vaccine study: a step back or a launching point for future vaccine development? J Exp Med. 2008;205(1):7–12. [PMC free article] [PubMed]
9. Townsley-Fuchs J, Kam L, Fairhurst R, et al. Human immunodeficiency virus-1 (HIV-1) gp120 superantigen-binding serum antibodies. A host factor in homosexual HIV-1 transmission. J Clin Invest. 1996;98(8):1794–1801. [PMC free article] [PubMed]
10. Paul S, Karle S, Planque S, et al. Naturally occurring proteolytic antibodies: selective immunoglobulin M-catalyzed hydrolysis of HIV gp120. J Biol Chem. 2004;279(38):39611–39619. [PubMed]
11. Planque S, Mitsuda Y, Taguchi H, et al. Characterization of gp120 Hydrolysis by IgA antibodies from humans without HIV infection. AIDS Res Hum Retroviruses. 2007;23(12):1541–1554. [PubMed]
12. Karray S, Juompan L, Maroun RC, Isenberg D, Silverman GJ, Zouali M. Structural basis of the gp120 superantigen-binding site on human immunoglobulins. J Immunol. 1998;161(12):6681–6688. [PubMed]
13. Goodglick L, Zevit N, Neshat MS, Braun J. Mapping the Ig superantigen-binding site of HIV-1 gp120. J Immunol. 1995;155(11):5151–5159. [PubMed]
14. Juompan L, Lambin P, Zouali M. Selective deficit in antibodies specific for the superantigen binding site of gp120 in HIV infection. Faseb J. 1998;12(14):1473–1480. [PubMed]
15. Patke CL, Shearer WT. gp120- and TNF-alpha-induced modulation of human B cell function: proliferation, cyclic AMP generation, Ig production, and B-cell receptor expression. J Allergy Clin Immunol. 2000;105(5):975–982. [PubMed]
16. Sasano M, Burton DR, Silverman GJ. Molecular selection of human antibodies with an unconventional bacterial B cell antigen. J Immunol. 1993;151(10):5822–5839. [PubMed]
17. Nishiyama Y, Karle S, Planque S, Taguchi H, Paul S. Antibodies to the superantigenic site of HIV-1 gp120: hydrolytic and binding activities of the light chain subunit. Mol Immunol. 2007;44(10):2707–2718. [PubMed]
18. Gao QS, Sun M, Rees AR, Paul S. Site-directed mutagenesis of proteolytic antibody light chain. J Mol Biol. 1995;253(5):658–664. [PubMed]
19. Paul S, Nishiyama Y, Planque S, et al. Antibodies as defensive enzymes. Springer Semin Immunopathol. 2005;26(4):485–503. [PubMed]
20. Olshevsky U, Helseth E, Furman C, Li J, Haseltine W, Sodroski J. Identification of individual human immunodeficiency virus type 1 gp120 amino acids important for CD4 receptor binding. J Virol. 1990;64(12):5701–5707. [PMC free article] [PubMed]
21. Taguchi H, Burr G, Karle S, et al. A mechanism-based probe for gp120-Hydrolyzing antibodies. Bioorg Med Chem Lett. 2002;12(21):3167–3170. [PubMed]
22. Gololobov G, Sun M, Paul S. Innate antibody catalysis. Mol Immunol. 1999;36(18):1215–1222. [PubMed]
23. Silverman GJ. B cell superantigens: possible roles in immunodeficiency and autoimmunity. Semin Immunol. 1998;10(1):43–55. [PubMed]
24. Bermas BL, Petri M, Berzofsky JA, Waisman A, Shearer GM, Mozes E. Binding of glycoprotein 120 and peptides from the HIV-1 envelope by autoantibodies in mice with experimentally induced systemic lupus erythematosus and in patients with the disease. AIDS Res Hum Retroviruses. 1994;10(9):1071–1077. [PubMed]
25. Karle S, Planque S, Nishiyama Y, et al. Cross-clade HIV-1 neutralization by an antibody fragment from a lupus phage display library. Aids. 2004;18(2):329–331. [PubMed]
26. Paul S, Volle DJ, Beach CM, Johnson DR, Powell MJ, Massey RJ. Catalytic hydrolysis of vasoactive intestinal peptide by human autoantibody. Science. 1989;244(4909):1158–1162. [PubMed]
27. Shuster AM, Gololobov GV, Kvashuk OA, Gabibov AG. Dna-specific catalytic antibodies in human blood sera. Dokl Akad Nauk SSSR. 1991;318(5):1262–1264. [PubMed]
28. Sekigawa I, Lee S, Kaneko H, et al. The possible role of interleukin-16 in the low incidence of HIV infection in patients with systemic lupus erythematosus. Lupus. 2000;9(2):155–156. [PubMed]
29. Barthel HR, Wallace DJ. False-positive human immunodeficiency virus testing in patients with lupus erythematosus. Semin Arthritis Rheum. 1993;23(1):1–7. [PubMed]
30. Urnovitz HB, Murphy WH. Human endogenous retroviruses: nature, occurrence, and clinical implications in human disease. Clin Microbiol Rev. 1996;9(1):72–99. [PMC free article] [PubMed]
31. Lyden TW, Johnson PM, Mwenda JM, Rote NS. Expression of endogenous HIV-1 crossreactive antigens within normal human extravillous trophoblast cells. J Reprod Immunol. 1995;28(3):233–245. [PubMed]
32. Holmes EC. On the origin and evolution of the human immunodeficiency virus (HIV) Biol Rev Camb Philos Soc. 2001;76(2):239–254. [PubMed]
33. Mayer J, Meese E. Human endogenous retroviruses in the primate lineage and their influence on host genomes. Cytogenet Genome Res. 2005;110(1–4):448–456. [PubMed]
34. Haigwood NL, Watson A, Sutton WF, et al. Passive immune globulin therapy in the SIV/macaque model: early intervention can alter disease profile. Immunol Lett. 1996;51(1–2):107–114. [PubMed]
35. Mouthon L, Lortholary O. Intravenous immunoglobulins in infectious diseases: where do we stand? Clin Microbiol Infect. 2003;9(5):333–338. [PubMed]
36. Berisha HI, Bratut M, Bangale Y, Colasurdo G, Paul S, Said SI. New evidence for transmitter role of VIP in the airways: impaired relaxation by a catalytic antibody. Pulm Pharmacol Ther. 2002;15(2):121–127. [PubMed]
37. Binley JM, Wrin T, Korber B, et al. Comprehensive cross-clade neutralization analysis of a panel of anti-human immunodeficiency virus type 1 monoclonal antibodies. J Virol. 2004;78(23):13232–13252. [PMC free article] [PubMed]
38. Paul S, Planque S, Zhou YX, et al. Specific HIV gp120-cleaving antibodies induced by covalently reactive analog of gp120. J Biol Chem. 2003;278(22):20429–20435. [PubMed]
39. Nishiyama Y, Karle S, Mitsuda Y, et al. Towards irreversible HIV inactivation: stable gp120 binding by nucleophilic antibodies. J Mol Recognit. 2006;19(5):423–431. [PubMed]
40. Karle S, Nishiyama Y, Taguchi H, et al. Carrier-dependent specificity of antibodies to a conserved peptide determinant of gp120. Vaccine. 2003;21(11–12):1213–1218. [PubMed]