Share this article on:

Cross-clade HIV-1 neutralization by an antibody fragment from a lupus phage display library

Karle, Sangeetaa; Planque, Stephaniea; Nishiyama, Yasuhiroa; Taguchi, Hiroakia; Zhou, Yong-Xina; Salas, Mariab; Lake, Douglasc; Thiagarajan, Perumala; Arnett, Franka; Hanson, Carl Vb; Paul, Sudhira

Research Letters

aChemical Immunology Research Center, Departments of Pathology and Medicine, University of Texas, Houston Medical School, Houston, TX, USA; bViral and Rickettsial Disease Laboratory, California Department of Health Services, Richmond, CA, USA; and cDepartment of Microbiology and Immunology, University of Arizona, Tucson, AZ, USA.

Received: 16 May 2003; revised: 4 June 2003; accepted: 14 July 2003.

A single-chain fragment containing antibody V domains (scFv) isolated from a lupus antibody library displayed the ability to bind gp120 and the conserved gp120 determinant composed of residues 421–436. The scFv neutralized R5 and X4-dependent HIV-1 strains from clades B, C and D. The lupus repertoire may be useful as a source of neutralizing antibodies to HIV.

The mutability of the immunodominant epitopes of HIV envelope gp120 and its distinctive conformational characteristics allow the virus to escape neutralization by most anti-env antibodies. The host cell receptor binding site of gp120, the CD4 cell binding site, is comparatively conserved. Antibodies to the CD4 cell binding site are infrequent, and only one antibody to this site with comparatively broad neutralizing activity has been reported (clone b12) [1]. Several other CD4 cell binding site antibodies fail to neutralize HIV broadly. Such antibodies may represent an immune response to monomer gp120 shed from the viral surface, the CD4 cell binding site of which could assume a conformation distinct from that found in trimeric gp120–gp41 complexes on the virus [1]. Sera from uninfected systemic lupus erythematosus patients and mouse strains predisposed to autoimmune disease contain antibodies that bind the conserved gp120 determinant composed of residues 421–436 [2,3], which contributes important contact sites for binding to CD4 receptors [4,5]. This is of interest because of the rare coexistence of lupus and HIV/AIDS [6]. Whether antibodies to HIV-1 found in lupus patients neutralize HIV and protect against the virus is not known.

These considerations prompted us to study gp120 binding single-chain fragment containing antibody V domain (scFv) constructs (VL domain attached to the VH domain by a 16-residue peptide linker) from a phage display library from the peripheral blood lymphocytes of lupus patients. scFv-displaying phages were packaged from the library and subjected to selection by affinity chromatography on recombinant gp120 (1013 phages; gp120 from strain SF2, Austral; immobilized on Biorad Affigel-10; 1 ml gel; 47 μg gp120/ml gel; see Paul et al. [7] for library preparation and phage handling procedures). Phagemid DNA from the pH 2.7 eluate [8] was grown in HB2151 cells and soluble scFv constructs in the periplasmic extracts were screened for binding to immobilized gp120 (MN strain; Immunodiagnostics, Inc., Woburn, MA, USA) and the bovine serum albumin conjugate of synthetic Cys-gp120 (421–436) (KQIINMWQEVGKAMYA, consensus sequence in clade B) by enzyme-linked immunosorbent assay procedures [9]. Fifty-four per cent of selected scFv clones displayed binding to gp120 (A490 0.16–2.39; N = 13 clones) and 31% of the clones displayed binding to gp120 (421–436) (A490 0.37–2.68).

One of the scFv clones (clone JL413) observed to bind both full-length gp120 and synthetic gp120 (421–436) was analysed further. The scFv purified to electrophoretic homogeneity by metal affinity chromatography [7] displayed dose-dependent binding of gp120 and gp120 (421–436) (A490 0.02–0.26 and 0.05–0.59, respectively, at scFv concentrations 7.4 nM–1 μM). Comparison of the complementary DNA sequence of this clone (GenBank no. AF329461) indicated the closest germline VL and VH genes to be L5 and VH4-59, respectively. Extensive amino acid replacements in the VL and VH regions of scFv were evident (replacements in VL residues 7–88 and VH residues 7–91: 12 and 7, respectively). The replacement/silent mutation ratio in the complementarity determining regions was greater than in the framework regions (9/1 and 4/5, respectively), suggesting that the V genes have been matured adaptively by mutational processes [10].

HIV-1 infection was studied using phytohemagglutinin-stimulated peripheral blood mononuclear cells (PBMC) with measurement of the p24 antigen as the index of infection [11]. Four independent culture replicates were analysed individually for p24 concentration (means ± s.e.m.). HIV-1 was treated with an equal volume of increasing concentrations of metal affinity-purified scFv in PBS (1 h; TCID50 for virus = 100). Phytohemagglutinin-stimulated PBMC from healthy human donors (0.25 million) were added and cultures incubated for 3 days (37 °C), the cells were washed twice with PBS and once with RPMI1640, incubated in fresh RPMI for 24 h, lysed with Triton X-100, and p24 in the supernatants was measured. Virus stocks were titered with each batch of donor PBMC to identify working dilutions giving optimum TCID50 and p24 signal in the linear range of the enzyme immunoassay (5–3200 pg/ml) after 4 days (enzyme immunoassay kit; Beckman Coulter, Fullerton, CA, USA).

After incubation with increasing concentrations of purified scFv JL413, dose-dependent neutralization of the following primary HIV-1 isolates drawn from clades B, C and D was observed (Fig. 1 and Table 1; co-receptor and clade information in parentheses): ZA009 (R5, C); BR004 (R5, C); Ug046 (X4, D); SF-162 (R5, B); and, 23135 (co-receptor not known, B) (from the National Institutes of Health AIDS Reagent Program, except strain 23135, which was from S. Levine, University of Southern California). Assays using independent scFv preparations from several bacterial cultures yielded reproducible HIV-1 neutralization (strain ZA009, N = 3, IC50 0.2 ± 0.1 μg/ml s.d.; strain Ug046, N = 2, IC50 2.1 and 5.5 μg/ml). The inhibition of infection was not caused by a cytotoxic effect. PBMC viability after treatment with diluent and scFv JL413 (27 μg/ml, 72 h) in the absence of HIV-1 was 81.2 ± 2.8%, and 76.8 ± 3.9%, respectively (150–200 cells counted; staining with acridine orange and ethidium bromide as in Brawn et al. [12]). In additional control studies, no loss of HIV infectivity (strain ZA009) was evident in the presence of identically purified extracts of bacteria-harboring vector without an antibody insert [7] or identically purified irrelevant scFv and L chains (control scFv clone JL610 in Fig. 1; not shown, scFv clone JL611, 2.5 μg/ml). In side-by-side comparisons, the neutralization of strain BR004 was observed at lower concentrations of scFv JL413 compared with IgG b12, a broadly neutralizing anti-CD4 cell binding site antibody [1] (Fig. 1; kindly provided by Dr D. Burton).

Fig 1. Concentration-dependent HIV-1 neutralization by purified single-chain antibody fragment (seFv clone JL413).

Fig 1. Concentration-dependent HIV-1 neutralization by purified single-chain antibody fragment (seFv clone JL413).

Table 1

Table 1

The cross-clade neutralizing activity of the scFv is consistent with the conserved character of the recognition site, determinant 421–436, and the contribution by this determinant of contact sites involved in host cell CD4 binding [4,5]. The ability to recognize a linear CD4 cell binding site component distinguishes scFv from anti-CD4 cell binding site antibodies that bind conformational determinants, most of which do not display cross-clade neutralizing activity [1]. Understanding the distinctive epitope specificity of scFv requires identification of the inciting immunogen in lupus patients, which remains to be accomplished. Gp120 determinant 421–436 expresses no known homologies with conventional human autoantigens. The presence and expression of endogenous retroviral gene sequences has been linked to autoimmune disease [13], but the possibility that polypeptide products of such sequences may induce protective antibodies to HIV has not been explored. As rare antibodies are readily identified by phage display methods, no inferences can be drawn from the present study about the statistical association between lupus and HIV-neutralizing antibodies. Howevwever, antibodies capable of binding gp120 determinant 421–436 have previously been reported in lupus sera but not in non-autoimmune individuals [2,3], suggesting that production of the antibodies is not a generalized occurrence. Evaluating the anti-viral protection afforded by the antibodies requires a study of larger groups of lupus patients. Regardless of caveats concerning their functional role in lupus, the identification of a broadly neutralizing recombinant antibody fragment is significant with regard to developing immunotherapeutic reagents against HIV.

Back to Top | Article Outline


1. Kwong PD, Doyle ML, Casper DJ, Cicala C, Leavitt SA, Majeed S, et al. HIV-1 evades antibody-mediated neutralization through conformational masking of receptor-binding sites. Nature 2002; 420:678–682.
2. 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; 10:1071–1077.
3. Zhou YX, Karle S, Taguchi H, Planque S, Nishiyama Y, Paul S. Prospects for immunotherapeutic proteolytic antibodies. J Immunol Methods 2002; 269:257–268.
4. 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; 64:5701–5707.
5. Kwong PD, Wyatt R, Robinson J, Sweet RW, Sodroski J, Hendrickson WA. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 1998; 393:648–659.
6. Palacios R, Santos J, Valdivielso P, Marquez M. Human immunodeficiency virus infection and systemic lupus erythematosus. An unusual case and a review of the literature. Lupus 2002; 11: 60–63.
7. Paul S, Tramontano A, Gololobov G, Zhou YX, Taguchi H, Karle S, et al. Phosphonate ester probes for proteolytic antibodies. J Biol Chem 2001; 276:28314–28320.
8. Tyutyulkova S, Gao QS, Thompson A, Rennard A, Paul S. Efficient vasoactive intestinal polypeptide hydrolyzing autoantibody light chains selected by phage display. Biochim Biophys Acta 1996; 1316:217–223.
9. Karle S, Nishiyama Y, Zhou YX, Luo J, Planque S, Hanson C, et al. Carrier-dependent specificity of antibodies to a conserved peptide determinant of gp120. Vaccine 2003; 21:1213–1218.
10. Nossal GJ. B lymphocyte physiology: the beginning and the end . Ciba Found Symp 1997; 204:220–230.
11. D'Souza MP, Milman G, Bradac JA, McPhee D, Hanson CV, Hendry RM. Neutralization of primary HIV-1 isolates by anti-envelope monoclonal antibodies. AIDS 1995; 9:867–874.
12. Brawn RJ, Barker CR, Oesterle AD, Kelly RJ, Dandliker WB. An improved fluorescence probe cytotoxicity assay. J Immunol Methods 1975; 9:7–26.
13. Talal N, Flescher E, Dang H. Are endogenous retroviruses involved in human autoimmune disease? J Autoimmun 1992; 5 (Suppl. A):61–66.
© 2004 Lippincott Williams & Wilkins, Inc.