JOURNAL OF VIROLOGY, Feb. 1992, p. 848-856 Vol. 66, No. 2 22-538X/92/2848-9$2./ Copyright 1992, Amerian Soiety for Mirobiology Identifiation and Charaterization of a Neutralization Site within the Seond Variable Region of Human Immunodefiieny Virus Type 1 gpl2 MICHAEL S. C. FUNG,1* CECILY R. Y. SUN,' WAYNE L. GORDON,' RUEY-SHYAN LIOU,1 TSE WEN CHANG,' WILLIAM N. C. SUN,' ERIC S. DAAR,2 AND DAVID D. HO3 Tanox Biosystems, In., 131 Stella Link, Houston, Texas 77251; Division of Infetious Diseases, Cedars-Sinai Medial Center, University of California at Los Angeles, Shool of Mediine, Los Angeles, California 9482; and Aaron Diamond AIDS Researh Center, New York University Shool of Mediine, New York, New York 1163 Reeived 13 August 1991/Aepted 31 Otober 1991 Two monolonal antibodies designated BAT85 and G3-136 were raised by immunizing BALB/ mie with gpl2 purified from human immunodefiieny virus type 1 (HIV-1) IIIB-infeted H9 ell extrats. Among three HIV-1 laboratory isolates (IIIB, MN, and RF), BAT85 neutralized only IIIB infetion of CEM-SS ells, whereas G3-136 neutralized both IIIB and RF. These antibodies also neutralized a few primary HIV-1 isolates in the infetion of ativated human peripheral blood mononulear ells. In indiret immunofluoresene assays, BAT85 bound to H9 ells infeted with IIIB or MN, while G3-136 bound to H9 ells infeted with IIIB or RF, but not MN. Using sequene-overlapping syntheti peptides of HIV-1 IIIB gpl2, the binding site of BAT85 and G3-136 was mapped to a peptidi segment in the V2 region (amino aid residues 169 to 183). The binding of these antibodies to immobilized gpl2 was not inhibited by the antibodies direted to the prinipal neutralization determinant in the V3 region or to the CD4-binding domain of gpl2. In a ompetition enzyme-linked immunosorbent assay, soluble CD4 inhibited G3-136 but not BAT85 from binding to gpl2. Deglyosylation of gpl2 by endo-o-n-aetylgluosaminidase H or redution of gpl2 by dithiothreitol diminished its reativity with G3-136 but not with BAT85. These results indiate that the V2 region of gpl2 ontains multiple neutralization determinants reognized by antibodies in both a onformationdependent and -independent manner. Intensive efforts in AIDS researh have been foused on the development of an effetive vaine for protetion of individuals at risk for human immunodefiieny virus type 1 (HIV-1) infetion (4, 45). It has beome evident that humoral responses play ruial roles in protetive immunity against infetion by HIV-1 and its progression to disease states. In this respet, Wendler et al. (49) showed that higher neutralizing antibody titers orrelated with better prognosis and low or dereasing neutralizing antibody titers orrelated with disease progression. Jakson et al. (19) and Karpas et al. (22, 23) demonstrated that AIDS-related omplex and AIDS patients treated with hyperimmune plasma from asymptomati HIV-1-seropositive individuals showed linial improvement, whih orrelated with redution in HIV-1 viremia. More reently, preliminary studies by Emini et al. (7) showed that himpanzees injeted with a mouse-human himeri HIV-1-neutralizing monolonal antibody (MAb) were proteted from infetion by the virus. The envelope proteins of HIV-1, gp12 and gp4l, are known to be the major targets for HIV-1-neutralizing ativity in the sera of HIV-1-infeted individuals (13, 18, 33, 48) and of animals immunized with the viral proteins (3, 28, 4). Several neutralization sites have been identified in gpi2, inluding an immunodominant epitope in the V3 region (amino aid residues 38 to 322), ommonly known as the prinipal neutralization determinant (PND) (9, 21, 24, 27, 32, 43, 44). Antibodies raised against this ontinuous epitope * Corresponding author. 848 usually exhibit type-speifi neutralizing ativity against the homologous viral isolates (11, 12, 33, 4, 43). However, reent studies by Javaherian et al. (2) and White-Sharf et al. (5) showed that antibodies raised in animals against the relatively onserved portion of the PND displayed broadened reativity with multiple HIV-1 isolates. Anti-PND antibodies do not blok virus binding to CD4 but prevent virus internalization subsequent to CD4 binding (31, 44). Another ontinuous neutralization site on gpl2 is loated in the C4 region (amino aid residues 413 to 447), whih is found to be involved in CD4-gpl2 interation (29, 47). Antibodies to this region exhibit broad neutralizing ativity against divergent HIV-1 isolates, but this neutralization site appears to be rather immunosilent in HIV-1-infeted individuals (47). Sera from most HIV-1-infeted individuals usually have low levels of broad neutralizing ativity against HIV-1 (2, 13, 48), whih were found to be assoiated with antibodies direted to onformational or disontinuous epitopes on gpl2 (13). Consistent with this finding, we have reported a human MAb (I5e) whih reognizes a disontinuous epitope on gpl2 involved in CD4 binding and in antibody neutralization of multiple HIV-1 isolates (16). We also generated a HIV-1-neutralizing murine MAb (G3-4), whih identified another distint onformational epitope on gpl2 involved in the interation with CD4 (15). In this report, we desribe the generation and haraterization of two murine MAbs designated BAT85 and G3-136 whih define a unique neutralization domain in the V2 region of gpl2.
VOL. 66, 1992 NEUTRALIZATION SITE IN THE V2 REGION OF HIV-1 gpl2 849 MATERIALS AND METHODS MAb reagents and polylonal antibodies. Murine HIVneutralizing MAb BAT123 immunoglobulin Gl(K) [IgGl(K)] binds to the PND in the V3 region of IIIB gpl2 (amino aid residues 38 to 322) (32). Murine MAb G3-519 [IgGl(K)] reognizes a omponent of the CD4-binding domain in the C4 region of gpl2 (amino aid residues 423 to 437) (47). Murine MAb G3-4 [IgG2a(K)] reognizes a onformational neutralizing epitope on HIV-1 gp12 (15). I5e [IgGl(K)] is a HIV-1-neutralizing human MAb (a gift from J. Robinson, Louisiana State University Shool of Mediine, New Orleans) whih also reognizes a distint onformational epitope on gpl2 (16). The murine MAbs were produed from asiti fluid of BALB/ mie (Harlan Sprague-Dawley, In., Houston, Tex.) injeted with the hybridomas and affinity purified by using reombinant protein A-Sepharose (Repligen, Boston, Mass.). I5e was produed from ulture supernatant of the hybridoma and purified by protein A affinity hromatography. Polylonal anti-hiv-1 gpl2 antibodies (D625) (International Enzymes, In., Fallbrook, Calif.) were generated by immunizing sheep with a syntheti peptide enompassing a highly onserved peptidi segment (APTKAKRRVVQREKR) in the arboxy terminus of HIV-1 (BH-1 strain) gpl2 (amino aid residues 497 to 511) (35-37). The antibodies were isolated from the sheep hyperimmune sera by affinity hromatography using the speified syntheti peptide oupled to Sepharose. Cells and HIV-1 isolates. CEM-SS ells, a synytiumsensitive lone of CEM ells (a gift from P. Nara, National Caner Institute), were maintained in RPMI 164 medium (GIBCO Laboratories, Grand Island, N.Y.) supplemented with 1% heat-inativated fetal bovine serum (GIBCO), 1 U of peniillin per ml, and 1 p.g of streptomyin per ml. H9 ells hronially infeted with HIV-1 IIIB, MN, or RF, were maintained in RPMI 164 medium with 15% heat-inativated fetal bovine serum. These infeted ells were derived from single-ell lones whih were tested to be almost 1% expressing gpl2 on the ell surfae by immunofluoroytometri methods. The ulture supernatants from these HIV- 1-infeted ells ontaining high reverse transriptase ativities were kept at - 16 C and used for HIV neutralization assays. Peripheral blood mononulear ells (PBMCs) were isolated from HIV-1-seronegative donors by Fioll-Paque density gradient entrifugation, and were stimulated for 3 days with 2 jig of phytohemagglutinin (PHA) (Sigma Chemial Co., St. Louis, Mo.) per ml in RPMI 164 medium supplemented with 15% heat-inativated fetal bovine serum, 1% human interleukin-2 (Cellular Produts, In., Buffalo, N.Y.), 1 U of peniillin per ml, and 1,ug of streptomyin per ml. After stimulation, PBMCs were washed and maintained in the omplete growth medium for infetivity assays. Primary HIV-1 isolates (LS, RP, CC, LL, TB, and AC) were obtained from plasma speimens from HIV-1-seropositive individuals, as previously desribed (5, 16, 17). The JR-CSF isolate was kindly provided by I. Chen (University of California at Los Angeles, Shool of Mediine, Los Angeles). Generation and sreening of MAbs against HIV-1 gpl2. Affinity-purified HIV-1 IIIB gpl2 was used to immunize male BALB/ mie. Our standard proedures for purifiation of gpl2 from infeted H9 ell lysates, immunization of mie, and generation of hybridomas were previously desribed in detail (1, 47). Briefly, 3 days after the final immunization, the mie were sarified, and spleen ells were isolated and fused with Sp2/ myeloma ells. After seletion using hypoxanthine-aminopterin-thymidine (HAT) medium, ulture supernatants from the wells of mirodilution plates with hybridomas were tested for reativity with gpl2 as oating antigen by enzyme-linked immunosorbent assays (ELISAs). Culture supernatants showing strong gpl2 reativity were tested for staining live HIV-1 IIIBinfeted ells by an indiret immunofluoresene method (47). Hybrids from seleted fusion wells were expanded and loned. MAbs from hybridomas seleted for further haraterization were produed in mouse asiti fluid and purified by protein A affinity hromatography (47). gp12 ELISA for sreening of hybridomas. The proedures were similar to those desribed earlier (1, 47). Wells of Immulon 2 mirodilution plates (Dynateh, Chantilly, Va.) were oated overnight at room temperature with 1,ul of purified gpl2 (.1 p.g/ml) in phosphate-buffered saline (PBS). The wells were then treated with 5% BLOTTO in PBS for 1 h at room temperature and washed with PBS ontaining.5% Tween 2 (PBST). Next, 1,ul of test ulture supernatant was added to eah well for 1 h at room temperature. The wells were then washed and inubated for 1 h at room temperature with horseradish peroxidase-onjugated goat anti-mouse IgG (Jakson ImmunoResearh Laboratories, In., West Grove, Pa.) diluted in PBST ontaining 1% bovine serum albumin (PBSTB). After another washing step, bound antibodies were deteted by reation with.1% tetramethyl benzidine and.3% hydrogen peroxide as substrates. The optial density at 45 nm (OD45) of the reation solution was read with an ELISA reader. Neutralization assays of HIV-1 laboratory isolates using CEM-SS ells. The synytium-forming miroassays using CEM-SS ells were performed as previously desribed (1, 39). Briefly, 5,ul of diluted MAb was mixed with 5,ul of viral ulture supernatant ontaining 2 synytium-forming units (SFU) of HIV-1 IIIB, MN, or RF and inubated for 1 h at room temperature. The mixtures were added into miroulture wells ontaining 5 x 14 DEAE-dextrantreated CEM-SS ells, and the ell ultures were maintained in 5% CO2 at 37 C for 3 to 4 days. The synytia were then enumerated under an inverted mirosope. The neutralizing ativity was expressed as ID5, defined as the onentration required to ahieve 5% inhibition of the infetion (i.e., Vn/Vo = 5%), where Vn is the number of SFU in the test wells and Vo is the number of SFU in the ontrol without test antibodies. In the study of the epitope speifiity using peptide V1SP, the neutralizing ativity of BAT85 at 12.5,ug/ml (ID5) was tested for inhibition by inubating the antibody with various onentrations of the peptide for 2 min at room temperature, prior to the addition of a viral inoulum of 1 SFUs of HIV-1 IIIB. The irrelevant peptide a-endorphin (Peninsula Laboratories, Belmont, Calif.) was used for a ontrol. Neutralization assays of primary HIV-1 isolates using PHAativated PBMCs. The titration of primary HIV-1 isolates and their neutralization assays were performed by our standard proedures (5, 17). An inoulum of 5 5% tissue ulture infetive doses was inubated with various onentrations of MAbs (.1 to 25,ug) in 1 ml of final ulture for 3 min at 37 C. The ontrol ontained no MAb. On day 4 to 7, depending on the repliation kinetis of the primary HIV-1 isolates, the HIV-1 p24 antigen levels of the antibody-treated and untreated ell ultures were measured with Abbott HIV-1 Antigen Kits (Abbott Laboratories, North Chiago, Ill.). The neutralizing ativity of the antibody is expressed as the perent redution of p24 antigen levels in the antibodytreated wells ompared with that of the ontrol.
85 FUNG ET AL. Flow ytometry. The binding of BAT85 and G3-136 to H9 ells infeted with HIV-1 IIIB, MN, or RF was analyzed by flow ytometri methods using a Coulter EPIC ell analyzer (Coulter Eletronis, Hialeah, Fla.) as previously desribed (25). The bound MAbs were deteted by fluoresein-onjugated goat anti-mouse IgG (Jakson ImmunoResearh Laboratories, In.). Generation of HIV-1 gpl2 peptides. All peptides used in epitope mapping were synthesized by using the RaMPs peptide synthesis system (Du Pont Co., Wilmington, Del.) by the 9-fluorenylmethoxyarbonyl synthesis protool desribed in the manual (6a). The purity of the RaMPSsynthesized peptides was haraterized by high-performane liquid hromatography, and the predited struture was assessed by fast-atom bombardment mass spetrometry analysis. Amino aid sequenes of the syntheti peptides for IIIB (HXB2 lone), MN, and RF isolates of HIV-1 were obtained from Myers et al. (38). Epitope mapping using syntheti HIV-1 gpl2 peptides. Preliminary epitope mapping was performed by using nitroellulose strips impregnated with sequene-overlapping syntheti oligopeptides (9 to 15 amino aid residues) enompassing the entire length of HIV-1 IIIB gpl2 polypeptide hain (24) (from S. Petteway formerly of Du Pont Co.). The amino aid sequenes were derived from HIV-1 lone HXB2 (38). The reativities of anti-hiv-1 gpl2 MAbs with these peptides were determined by the methods previously desribed (32). Briefly, the nitroellulose strips were inubated overnight at room temperature with test MAbs at 1,ug/ml in 5% BLOTTO. The strips were then washed with PBST. The bound antibody was then deteted by inubating the strips with diluted horseradish peroxidase-onjugated goat antimouse IgG (Jakson ImmunoResearh Laboratories, In.) for 1 h at room temperature. The strips were washed, and the olor reation was developed with a substrate solution ontaining.3% 4-hloro-1-naphthol and.5% hydrogen peroxide. Further fine epitope mapping was performed by peptide ELISA. Wells of Immulon 2 mirotest plates were oated overnight at room temperature with 1,ul of the syntheti peptides (1,ug/ml) in PBS. The wells were then treated with 5% BLOTTO for 1 h at room temperature. After the wells were washed with PBST, 1 pal of BAT85 or G3-136 (1 jig/ml) was added to the wells for reation for 1 h at room temperature. The wells were then washed, and the bound MAb was then deteted by horseradish peroxidase-onjugated goat anti-mouse IgG. The olor reation was developed by inubation with the peroxidase substrate solution as desribed above. Reativity was regarded as positive if the OD45 was twofold higher than that of the wells oated with the irrelevant peptide a-endorphin. Competition ELISA. The effets of anti-hiv-1 gpl2 MAbs, peptide V15P and reombinant soluble CD4 (scd4) on the binding of BAT85 and G3-136 to HIV-1 IIIB gpl2 were determined by the proedures previously desribed (35-37). scd4 was obtained from Biogen, In., Cambridge, Mass. Wells of Immulon 2 mirotest plates were oated overnight at room temperature with 5,u of a solution ontaining 5,ug of affinity-purified sheep anti-hiv-1 gpl2 (D625, International Enzymes, In.) per ml. The wells were then treated with 5% BLOTTO for 1 h at room temperature. Fifty miroliters of purified gpl2(.5 pug/ml) was then added to eah well for inubation for 2 h at room temperature. The plates were then washed. One hundred miroliters of diluted biotinylated MAb was added to the wells with or without various amounts of ompeting MAbs (inluding BAT85, J. VIROL. G3-136, G3-4, BAT123, G3-519, and ISe), peptide V15P or scd4. The biotinylated MAbs were used at dilutions whih gave half-maximal binding to the aptured gpl2. The inubation time was 1 h at room temperature. The plates were washed and reated with 1 Ll of horseradish peroxidaseonjugated streptavidin (diluted 1:2, in PBSTB) (Jakson ImmunoResearh Laboratories, In.) for 1 h at room temperature. The plates were washed again and allowed to reat with the peroxidase substrate solution as desribed above. The degree of inhibition was expressed as the perent derease in OD45 in the test wells ompared with the OD of ontrol wells without the inhibitors. RIPA. The radioimmunopreipitation (RIPA) proedure was arried out as desribed earlier (34, 47). Briefly, HIV-1 IIIB-infeted H9 ells were metabolially labeled for 4 h with [35S]ysteine and [35S]methionine (1,uCi/ml; ICN Pharmaeutials, In., Irvine, Calif.) and suspended in RIPA lysing buffer (5 mm Tris hydrohloride [ph 8.], 15 mm NaCl, 1% Triton X-1, 1% sodium deoxyholate,.1% sodium dodeyl sulfate, 1 mm phenylmethylsulfonyl fluoride). To examine the effet of antibody binding to gpl6 and gpl2 after redution of disulfide linkages, the metabolially labeled ell lysates was treated with.1 M dithiothreitol in PBS at 37 C for 1 h. Dithiothreitol was then removed by dialysis. The effet of deglyosylation of gpl6 and gpl2 on antibody binding was also examined by treating the ell lysates with 1 mu of endo-o-n-aetylgluosaminidase H for 2.5 h at 37 C. The treated or untreated lysates were preleared with protein A-Sepharose bound to rabbit antiserum to mouse kappa light hain (K-PAS) for 3 h at room temperature. RIPA was performed by adding 3,ug of purified MAb (BAT85, G3-136, or G3-4) and.2 ml of a 1% suspension of K-PAS to 2,ul of labeled and larified lysate. The samples were inubated for 18 h at 4 C, and the beads were washed with RIPA lysing buffer. The pellets were suspended in eletrophoresis sample buffer and boiled for 3 min. Proteins were analyzed by sodium dodeyl sulfatepolyarylamide gel eletrophoresis, followed by autoradiography. RESULTS Generation and haraterization of MAbs BAT85 and G3-136. Affinity-purified gpl2 prepared from lysates of HIV-1 IIIB-infeted H9 ells was used to immunize mie. The ulture supernatants of hybrids were tested for reativity with gpl2by ELISA. Antibodies exhibiting strong gpl2 reativity were tested for staining live HIV-1 IIIB-infeted H9 ells by an indiret immunofluoresene method in order to determine whether the antibodies reated with gpl2 on infeted ells. The MAbs from loned hybrids were tested for neutralizing ativity against HIV-1 IIIB using a synytium-forming assay. Two neutralizing MAbs designated BAT85 and G3-136 [both IgGl(K)] were identified for further haraterization. In the flow ytometri assay shown in Fig. 1A, BAT85 at 1 p.g/ml stained 87 and 35% of H9 ells infeted with HIV-1 IIIB and MN, respetively. It did not stain H9 ells infeted with HIV-1 RF. G3-136 stained 73 and 91% of H9 ells infeted with HIV-1 IIIB and RF, respetively (Fig. 1B) but did not stain H9 ells infeted with HIV-1 MN. In the in vitro neutralization assays against HIV-1 laboratory isolates using CEM-SS ells, BAT85 neutralized IIIB with ID5 of 12.5 p,g/ml (Fig. 2A). It did not neutralize MN and RF. G3-136 neutralized IIIB and RF with ID5 of 16 and 2 pg/ml, respetively (Fig. 3A). It did not neutralize MN.
VOL. 66, 1992 NEUTRALIZATION SITE IN THE V2 REGION OF HIV-1 gp12 851 1. A --..P- - -- - - -.1 IA: *.1 1 1 BAT85, sg/mi 1 1. 8 ir 1W 1W 1W ILO SICZI7'Nn () 8 6 4 4. 2 FIG. 1. Binding ativities of BAT85 (A) and of G3-136 (B) to H9 ells infeted with HIV-1 IIIB ( *), MN (...), or RF (.Uninfeted H9 ells ( )were used for a ontrol. The H9 ells (17ells per ml) were inubated with or without the antibodies for 3 min on ie. After the ells were washed, they were inubated with goat anti-mouse IgG oupled with fluoresein isothioyanate for 3 min, washed, and fixed with paraformaldehyde. Fluoresene intensity was analyzed by flow ytometry. Neutralizing ativities against the infetivity of several primary HIV-1 isolates were also tested by using PHA-ativated PBMCs. BAT85 neutralized two (LS and RP) of six isolates tested with an ID5 of less than 1,ug/ml (Fig. 2B); G3-136 neutralized three (AC, LS, and TB) of seven isolates with ID5 ranging from.3 to.4,ug/ml (Fig. 3B). Epitope mapping of BAT85 and G3-136. To delineate the epitopes on gpl2 that these two MAbs reognize, we first tested the effet of other previously haraterized neutralizing anti-hiv-1 gpl2 MAbs on the binding of biotinylated BAT85 and G3-136 to gp12. These other MAbs inlude BAT123, G3-519, G3-4, and ISe; the last two MAbs reognize distint onformational neutralizing epitopes on gp12. In the ompetition ELISA, purified HIV-1 IIIB gpl2 was aptured by affinity-purified sheep anti-hiv-1 gpl2 antibodies whih are speifi for a highly onserved peptidi segment in the arboxy terminus of gp12. The immobilization of gpl2 did not hinder the binding of BAT123, G3-519, G3-4, I5e, BAT85, and G3-136. Figure 4A shows that the binding of biotinylated BAT85 to gpl2 was partially inhibited by G3-4 (8%) and G3-136 (65%) at 4,g/ml. In ontrast, BAT123, G3-519, and ISe had no effet. Unonjugated BAT85 at 4,ug/ml ompletely abolished binding. When the binding of biotinylated G3-136 to gpl2 was tested, G3-4 inhibited the binding as effetively as G3-136 did (Fig. 4B). However, BAT85 inhibited the binding by only 33% even at 4,ug/ml, and BAT123, G3-519, and ISe had no effet. These results were onfirmed by examining the ef-.1.1 1. 1. 1. BAT85, sg/ml FIG. 2. Neutralization of HIV-1 by BAT85. (A) Inhibition of infetion of CEM-SS ells by laboratory HIV-1 isolate IIIB (), MN (), and RF (A). A viral inoulum of 1 SFU was used to infet the ells in the presene or absene of various onentrations of BAT85. Vn is the mean number of synytia in dupliate test wells, and Vo is the mean number of synytia in the dupliate ontrol wells. (B) Inhibition of infetion of PHA-ativated PBMCs from HIV-1-seronegative donors by primary HIV-1 isolates LS (), RP (V), CC(O), JR-CSF (O), LL(A), and TB (A). A viral inoulum of 5 5% tissue ulture infetive doses was used to infet the ells in the presene or absene of various onentrations of BAT85. The neutralizing ativity of the antibody is expressed as the perent redution of p24 antigen levels in the antibody-treated wells ompared with that of the ontrol. fets of BAT85 and G3-136 on the binding of biotinylated G3-4, BAT123, G3-519, and ISe to gpl2. As expeted, BAT85 and G3-136 ompeted with G3-4 but not with BAT123, G3-519, and ISe (16; also unpublished results). To define the binding epitope(s) of BAT85 and G3-136 on gp12, the MAbs were examined by using nitroellulose strips impregnated with sequene-overlapping syntheti peptides enompassing the entire length of HIV-1 IIIB gpl2 polypeptide hain. It was found that BAT85 and G3-136 reated speifially with a peptidi segment, VQKEYAFFY KLDIIP (referred to as V15P) in the V2 region of gp12, amino aid residues 169 to 183. To onfirm this observation, overlapping oligopeptides enompassing the peptidi segment (amino aid residues 159 to 198) in HIV-1 IIIB gp12 (Table 1) were synthesized for ELISAs and liquid-phase ompetition assays. When these peptides were used as the solid-phase antigens at 1,ug/ml, BAT85 reated strongly with V15P (OD45 = 1.4), and weakly with L2T (OD45 =.13). It did not reat with F1SY, V1OK, and AlOP, giving OD45s less than twofold greater than the negative ontrol a-endorphin (OD45 =.1). G3-136 also reated with V15P
852 FUNG ET AL. J. VIROL. 1.() 1 8.14. -C -. L- 6 4 2. 1... ~. 1 1 1 G3-136, yg/ml.1 1. 1. Competing antibodies, ug/ml C) a- 1 8 6 4 2 B * a.1.1 1. 1. 1. G3-136. /g/ml FIG. 3. Neutralization of HIV-1 by G3-136. (A) Inhibition of infetion of CEM-SS ells by laboratory HIV-1 isolates IIIB (), MN (), and RF (A). (B) Inhibition of infetion of PHA-ativated PBMCs from HIV-1-seronegative donors by primary HIV-1 isolates AC (), LS (), LL (A), TB (A), CC (O), JR-CSF (O), and RF (V). The neutralizing ativities of the antibody were determined as desribed in the legend to Fig. 2. (OD45 =.44) but did not reat with F15Y, L2T, V1OK, and A1OP. In the determination of reativity with different HIV-1 isolates, BAT85 did not reat with peptides M15S and T15P, whih orrespond to the same region of peptide V15P. *t_. a- 1 8 6 4 2.1 1. 1. Competing antibodies, p,g/ml FIG. 4. Inhibition of the binding of biotinylated BAT85 (A) and biotinylated G3-136 (B) to aptured HIV-1 IIIB gpl2 in ELISA by BAT85 (A), G3-136 (), G3-4 (), BAT123 (A), G3-519 (l), and 15e (). The ompetition ELISA proedure is desribed in Materials and Methods. in the gpl2 of HIV-1 MN and RF, respetively. G3-136 reated with T15P (OD45 =.23), but not with M15S. In the liquid-phase ompetition binding assay examining the effets of peptides on the binding of biotinylated BAT85 to aptured gpl2, V15P inhibited the binding of BAT85 to TABLE 1. Epitope mapping of BAT85 and G3-136 to the V2 region of HIV-1 gpl2 by peptide ELISA Viral peptide Sequene (amino aids 159-198") Reativity (OD45) and antigen BAT85 G3-136 HIV HXB2 16 17 18 19 F15Y FNISTSIRGKVQKEY.1.3 V15P VQKEYAFFYKLDIIP 1.4.44 L2T LDIIPIDNDTTSYSLTSCNT.13.2 V1OK VQKEYAFFYK.2.1 AlOP AFFYKLDIIP.3.1 HIV MN M15S M----LL----VS.1.2 HIV RF T15P T--K--L- -- -VV-.1.23 a Dedued amino aid sequenes are obtained from Myers et al. (38). Homologous amino aids are represented by dashes. The OD45 of the blank was between.1 and.3.
VOL. 66, 1992 NEUTRALIZATION SITE IN THE V2 REGION OF HIV-1 gp12 853..r_._- - ) - ) 1 8 6 4 2 7 6 5- z 4-. 3-2 r~ 1.1.1 1. 1. Competing peptides, tg/ml FIG. 5. Competition of the binding between biotinylated BAT85 and aptured HIV-1 IIIB gpl2 by the epitope peptide V15P () and the irrelevant peptide a-endorphin (). The amino aid sequene of V15P is shown in Table 1. The ELISA proedure is desribed in Materials and Methods. gp12 in a dose-dependent manner (Fig. 5). The a-endorphin peptide had no effet, even at a onentration of 4 p.g/ml. The binding speifiity of V15P to BAT85 was also examined in the neutralization assay of HIV-1 IIIB using CEM-SS ells. Figure 6 shows that V15P exhibited a dose-dependent inhibition of the neutralizing ativity of BAT85, while a-endorphin had no effet. Either peptide alone did not affet the infetivity of HIV-1 IIIB. Effets of scd4 on the binding of BAT85 and G3-136 to gpl2o. To examine whether the neutralizing epitope mapped to the V2 region was involved in the interation between CD4 and gpl2, the effet of scd4 on the binding of the biotinylated MAbs to aptured gp12 was tested. scd4 inhibited the binding of G3-136, G3-4, and G3-519 to gpl2 (Fig. 7), although not ompletely even at a onentration as high as 4,ug/ml. The binding of BAT85 and BAT123 to gpl2 is not affeted by scd4. These results suggest that the E -._- ( o ' 5 4 3 2 1-1'~~~~~~~~~~~,~.1.1 1. 1. Competing peptides, ug/ml FIG. 6. Effets of oligopeptides on the neutralizing ativity of BAT85 against HIV-1 IIIB. As desribed in Materials and Methods, CEM-SS ells (5 x 14) were infeted with 1 SFU of HIV-1 IIIB in the presene of BAT85 at 12.5,ug/ml equivalent to ID5 ( >)- Various onentrations of the epitope peptide V15P () or the irrelevant peptide oa-endorphin () were mixed with the antibody as desribed in Materials and Methods. The perent inhibition of synytium formation is defined as the perent derease in the number of synytia in the test wells ompared with that of the ontrol wells without the antibody..1.1 1. 1. scd4, /g/mi FIG. 7. Inhibitory ativity of scd4 on the binding of biotinylated MAbs to a,ptured HIV-1 IIIB gpl2 in ELISA. These biotinylated antibodies, inluding BAT85 (D), G3-136 (A), G3-4 (A), G3-519 (O), and BAT123 () were used at dilutions whih gave halfmaximal binding to the aptured gpl2. Various onentrations of scd4 were mixed with the labeled antibodies for binding to the gpl2 for 1 h at room temperature as desribed in Materials and Methods. epitopes reognized by BAT85 and G3-136 may be different, and that of G3-136 is near a CD4-binding domain of gpl2. Effet of redution and deglyosylation of gpl2 on its binding by BAT85 and G3-136. To study whether the binding of BAT85 and G3-136 to gp12 was affeted by onformational hanges of gpl2, the effets of redution of disulfide linkages and deglyosylation of gpl2 were examined. As summarized in Table 2, the reativity of G3-136 with gpl2 in a RIPA was redued when the viral lysate was first denatured by dithiothreitol. When a radiolabeled HIV-1 lysate was subjeted to endo-p-n-aetylgluosaminidase H digestion, the partially deglyosylated produt of 9 to 1 kda was weakly reognized by G3-136 (Table 2). In parallel experiments, the reativity of BAT85 with redued or deglyosylated gp12 was unaltered, whereas the reativity of G3-4 with gp12 was ompletely abolished under these onditions. DISCUSSION We report here that a unique neutralization domain in the V2 region of HIV-1 gp12 has been identified by two murine MAbs, BAT85 and G3-136, raised against HIV-1 IIIB gpl2. These two MAbs an neutralize in vitro the infetivity of ertain laboratory and primary HIV-1 isolates. The immunologi and biologi properties of these MAbs are summarized in Table 2. By using sequene-overlapping syntheti peptides enompassing the entire length of the HIV-1 gpl2 polypeptide hain, the binding site of BAT85 and G3-136 was mapped to the peptidi segment V15P (amino aid residues 169 to 183) in HIV-1 IIIB gp12 (Table 1). Confirming our findings, Stephens et al. (46) reently reported that antisera to the syntheti peptide defining the orresponding segment in the V2 region of HIV-1 SF2 gpl2 (amino aid residues 171 to 185) neutralized in vitro the infetivity of the virus. The speifiity of peptide V15P for BAT85 has been onfirmed by the liquid-phase ompetition assays as illustrated in Fig. 5 and 6. The absene of reativity of these MAbs with the shorter peptides (VlOK and AlOP) of this region indiates
854 FUNG ET AL. J. VIROL. TABLE 2. Properties of neutralizing MAbs to the V2 region of HIV-1 gpl2a Neutralization Staining ells Effet of degly- Effet of disulfide Binding to epitope MAb of HIV infeted with HIV scd4 inhibition of osylation on bond redution on peptide (amino IIIB MN RF IIIB MN RF gpl2 binding gpl2 binding gpl2 binding aids 169-183) BAT85 [IgGl(K)] + - - ++ + - - None None ++ G3-136 [IgGl(K)] + - ++ + - ++ + Redued Redued + G3-4 [IgG2a(K)] + - ++ + - + + + Abolished Abolished - a Symbols: +, reative; -, unreative; ++, strongly reative. that the atual binding site may be omposed of several amino aid residues interspersed within peptide V15P. The identities of the ruial amino aid residues remain to be eluidated. Despite the fat that both BAT85 and G3-136 an bind to peptide V15P, only G3-136 reats with peptide T15P of HIV-1 RF and strongly stains H9 ells infeted with this viral isolate. It is still unlear whih amino aid residues in these peptides atually determine the binding speifiities of different HIV-1 isolates. Inasmuh as the binding of G3-136 to gpl2 is partially sensitive to deglyosylation and redution (Table 2), its binding epitope is likely to be onformation dependent. In ontrast, the binding of BAT85 to gpl2 is not sensitive to onformational hanges of gpl2. This subtle differene in the binding harateristis between G3-136 and BAT85 is onsistent with the inability of BAT85 to ompete effetively with the binding of G3-136 to gpl2 (Fig. 4B). We reently reported that a murine HIV-1-neutralizing MAb, G3-4, reognized a onformational epitope in HIV-1 IIIB gpl2 (Table 2) (15). Although G3-4 did not reat with peptide V15P, it effetively inhibited the binding of G3-136 but not BAT85 to gpl2, suggesting that the binding epitopes of G3-4 and G3-136 are overlapping (Table 2). The FIG. 8. Shemati representation of the epitopes for BAT85, G3-136, G3-4, G3-519, BAT123, and ISe on gpl2 and their spatial relationship with one another and with the CD4-binding site as determined by ompetition ELISA. binding of G3-4 to gpl2 is stritly onformation dependent, beause its binding is ompletely abolished when gpl2 is deglyosylated or redued (15). On the basis of these findings, we onlude that the V2 region of gpl2 may ontain multiple neutralization determinants onstituted by either ontinuous or disontinuous strutures. Results from this study also indiate that the neutralization determinants in the V2 region reognized by BAT85 and G3-136 are likely to be topographially removed from the PND in the V3 region and the CD4-binding domain in the C4 region, sine the binding of BAT85 or G3-136 to gpl2 annot be inhibited by BAT123 and G3-519 in ELISA. In addition, BAT85 and G3-136 do not inhibit the gpl2 binding of ISe, whih reognizes a distint onformational epitope yet to be defined (16). This observation provides redene to the possibility of ombining HIV-1 MAbs and various neutralization epitopes in passive immunotherapy to onfer improved effiay and broadened viral reativity. In vitro experiments are in progress in our laboratories to address the funtional interations between these neutralization sites. It is interesting to note that G3-136 neutralizes a heterologous HIV-1 isolate (RF) more effiiently than the homologous isolate (IIIB) (Fig. 3A), although its immunoreaitivity is stronger with peptide V15P of IIIB than with the orresponding peptide T15P of RF. A possible explanation for this observation is that the binding site on native HIV-1 IIIB gpl2 is not as aessible as that of HIV-1 RF. The aessibility of the neutralization sites on gpl2 an be affeted by the intriate folding of gpl2, as well as the variable extent of glyosylation of gpl2 (8, 14, 3). The juxtaposition of disontinuous regions of gpl2 has been impliated in the assoiation between gpl2 and gp4l (26) and in the esape of HIV-1 variants from neutralization by antibodies (41, 51). In line with this, reent studies by Davis et al. (6) indiated that glyosylation of gpl2 ould affet the binding of antipeptide antibodies to gpl2. Another possible explanation for the higher neutralizing ativity of G3-136 against HIV-1 RF is that sine the binding of G3-136 to gpl2 is partially onformation dependent (Table 2), G3-136 may fit the threedimensional onfiguration of the binding site on the gpl2 of the RF isolate better. In order to understand the possible mehanisms of neutralization mediated by BAT85 and G3-136, we examined whether their binding to gpl2 an affet CD4-gpl2 interation. Indeed, like G3-519, the binding of G3-136 and G3-4 to gpl2 an be inhibited by scd4. In ontrast, the binding of BAT85 to gpl2 is not affeted in the presene of scd4. These data suggests that the neutralizing ativities of G3-136 and G3-4 may be mediated by interfering with the interation between CD4 and gpl2, whereas BAT85 may have a distint and as yet undefined mehanism. The inability of scd4 to ompletely abolish the binding of G3-136, G3-4, or G3-519 to gpl2 an be explained by the fat that the
VOL. 66, 1992 NEUTRALIZATION SITE IN THE V2 REGION OF HIV-1 gp12 855 strutural determinants reognized by these antibodies are either only sterially proximal to the atual CD4 binding sites or are forming parts of the multiple disontinuous regions responsible for CD4 binding (1, 42). The spatial relationship of the newly identified V2 neutralization epitopes relative to the other previously defined neutralization sites (V3, C4, and the I5e onformational determinant) and to the CD4-binding site is shematially shown in Fig. 8. On the basis of the findings in this study, we onlude that the V2 region of gpl2 bears multiple and overlapping neutralization determinants. To better design ative and passive immunization strategies, it is important to study whether neutralizing antibodies to this region play an ative role in the protetive immunity against HIV-1 infetion in humans. ACKNOWLEDGMENTS We thank N. Chang for advie, J. Robinson for antibody 15e, S. Petteway for gp12 peptide-oated nitroellulose strips, N. Jones for tehnial assistane in peptide synthesis, and J. Wright for preparation of the manusript. This study was supported in part by grants from the National Institutes of Health (AI25541, A128747, A127742, and A1243), the Ernst Jung Foundation, and the Aaron Diamond Foundation. REFERENCES 1. Ardman, B., M. Kowalski, J. Bristol, W. Haseltine, and J. Sodroski. 199. Effets on CD4 binding of anti-peptide sera to the fourth and fifth onserved domains of HIV-1 gpl2. J. Aquired Immune Defi. Syndr. 3:26-214. 2. Berkower, I., G. E. Smith, G. Chad, and D. Murphy. 1989. Human immunodefiieny virus 1: predominane of a groupspeifi neutralizing epitope that persists despite geneti variation. J. Exp. Med. 17:1681-1695. 3. Berman, P. W., J. E. Groopman, T. Gregory, P. R. Clapham, R. A. Weiss, R. Ferriani, L. Riddle, C. Shimasaki, C. Luas, L. A. Lasky, and J. W. Eihberg. 1988. Human immunodefiieny virus type 1 hallenge of himpanzees immunized with reombinant envelope glyoprotein gpl2. Pro. Natl. Aad. Si. USA 85:52-524. 4. Bolognesi, D. P. 1989. Prospets for prevention of and early intervention against HIV. JAMA 261:37-313. 5. Daar, E. S., X. L. Li, T. Moudgil, and D. D. Ho. 199. High onentrations of reombinant soluble CD4 are required to neutralize primary human immunodefiieny virus type 1 isolates. Pro. Natl. Aad. Si. USA 87:6574-6578. 6. Davis, D., D. M. Stephens, C. Willers, and P. J. Lahmann. 199. Glyosylation governs the binding of antipeptide antibodies to regions of hypervariable amino aid sequene within reombinant gpl2 of human immunodefiieny virus type 1. J. Gen. Virol. 71:2889-2898. 6a.Du Pont Company. 1987. Du Pont atalog. Du Pont Company, Wilmington, Del. 7. Emini, E. A., W. A. Shleif, K. Murthy, Y. Eda, S. Tokiyoshi, S. D. Putney, S. Matsushita, J. H. Nunberg, and J. W. Eihberg. 1991. Passive immunization with a monolonal antibody direted to the HIV-1 gpl2 prinipal neutralization determinant onfers protetion against HIV-1 hallenge in himpanzees, p. 72. Th.A.64. Abstr. VII Int. Conf. AIDS, vol. 2. Istituto Superiore Di Sanita, Rome. 8. Fenouillet, E., J. C. Glukman, and E. Bahraoui. 199. Role of N-linked glyans of envelope glyoproteins in infetivity of human immunodefiieny virus type 1. J. Virol. 64:2841-2848. 9. Fung, M. S. C., C. Sun, N. C. Sun, N. T. Chang, and T. W. Chang. 1987. Monolonal antibodies that neutralize HIV-1 virions and inhibit synytium formation by infeted ells. Bio/ Tehnology 54:94-946. 1. Fung, M. S. C., C. R. Y. Sun, R. S. Liou, W. Gordon, N. T. Chang, T. W. Chang, and N. C. Sun. 199. Monolonal antiidiotypi antibody mimiking the prinipal neutralization site in HIV-1 gp12 indues HIV-1 neutralizing antibodies in rabbits. J. Immunol. 145:2199-226. 11. Goudsmit, J., C. Debouk, R. H. Meloen, L. Smit, M. Bakker, D. M. Asher, A. V. Wolff, C. J. Gibbs, Jr., and D. C. Gajdusek. 1988. Human immunodefiieny virus type 1 neutralization epitope with onserved arhiteture eliits early type-speifi antibodies in experimentally infeted himpanzees. Pro. Natl. Aad. Si. USA 85:4478-4482. 12. Goudsmit, J., C. Thiriant, L. Smit, C. Bruk, and C. J. Gibbs. 1988. Temporal development of ross-neutralization between HTLV-IIIB and HTLV-IIIRF in experimentally infeted himpanzees. Vaine 6:229-232. 13. Haigwood, N. L., C. B. Barker, K. W. Higgins, P. V. Skiles, G. K. Moore, K. A. Mann, D. R. Lee, J. W. Eihberg, and K. S. Steimer. 199. Evidene for neutralizing antibodies direted against onformational epitopes of HIV-1 gp12, p. 313-32. In R. A. Lerner (ed.), Vaine 9. Cold Spring Harbor Laboratory, Cold Spring Spring Harbor, N.Y. 14. Hansen, J.-E. S., H. Clausen, C. Nielsen, L. S. Teglbjaerg, L. L. Hansen, C. M. Nielsen, E. Dabelsteen, L. Mathiesen, S.-I. Hakomori, and J.. Nielsen. 199. Inhibition of human immunodefiieny virus (HIV) infetion in vitro by antiarbohydrate monolonal antibodies: peripheral glyosylation of HIV envelope glyoprotein gp12 may be a target for virus neutralization. J. Virol. 64:2833-284. 15. Ho, D. D., M. S. C. Fung, Y. Cao, X. L. Li, C. Sun, T. W. Chang, and N. C. Sun. 1991. Another disontinuous epitope on gpl2that is important in human immunodefiieny virus type 1 neutralization is identified by a monolonal antibody. Pro. Natl. Aad. Si. USA 88:8949-8952. 16. Ho, D. D., J. A. MKeating, X. L. Li, T. Moudgil, E. S. Daar, N. C. Sun, and J. E. Robinson. 1991. Conformational epitope on gpl2 important in CD4 binding and human immunodefiieny virus type 1 neutralization identified by a human monolonal antibody. J. Virol. 65:489-493. 17. Ho, D. D., T. Moudgil, and M. Alam. 1989. Quantitation of human immunodefiieny virus type 1 in the blood of infeted persons. N. Engl. J. Med. 321:1621-1625. 18. Ho, D. D., M. G. Sarngadharan, M. S. Hirsh, R. T. Shooley, T. R. Rota, R. C. Kennedy, T. C. Chanh, and V. L. Sato. 1987. Human immunodefiieny virus neutralizing antibodies reognize several onserved domains on the envelope glyoproteins. J. Virol. 61:224-228. 19. Jakson, G. G., J. T. Perkins, M. Rubenis, D. A. Paul, M. Knigge, J. C. Despotes, and P. Spener. 1988. Passive immunoneutralization of human immunodefiieny virus in patients with advaned AIDS. Lanet ii:647-651. 2. Javaherian, K., A. J. Langlois, G. J. LaRosa, A. T. Profy, D. P. Bolognesi, W. C. Herlihy, S. D. Putney, and T. J. Matthews. 199. Broadly neutralizing antibodies eliited by the hypervariable neutralizing determinant of HIV-1. Siene 25:159-1593. 21. Javaherian, K., A. J. Langlois, C. MDanal, K. L. Ross, L. I. Ekler, C. L. Jellis, A. T. Profy, J. R. Rushe, D. P. Bolognesi, S. D. Putney, and T. J. Matthews. 1989. Prinipal neutralizing domain of the human immunodefiieny virus type 1 envelope protein. Pro. Natl. Aad. Si. USA 86:6768-6772. 22. Karpas, A., I. K. Hewlett, F. Hill, J. Gray, N. Byron, D. Gilgen, V. Bally, J. K. Oates, B. Gazzard, and J. E. Epstein. 199. Polymerase hain reation evidene for human immunodefiieny virus 1 neutralization by passive immunization in patients with AIDS and AIDS-related omplex. Pro. Natl. Aad. Si. USA 87:7613-7617. 23. Karpas, A., F. Hill, M. Youle, V. Cullen, J. Gray, N. Byron, F. Hayhoe, M. Tenant-Flowers, L. Howard, D. Gilgen, J. K. Oates, D. Hawkins, and B. Gazzard. 1988. Effets of passive immunization in patients with the aquired immunodefiieny syndrome-related omplex and aquired immunodefiieny syndrome. Pro. Natl. Aad. Si. USA 85:9234-9237. 24. Kenealy, W. R., T. J. Matthews, M. C. Ganfield, A. J. Langlois, D. M. Waselefsky, and S. R. Petteway, Jr. 1989. Antibodies from human immunodefiieny virus-infeted individuals bind to a short amino aid sequene that eliits neutralizing antibodies in animals. AIDS Res. Human Retroviruses 5:173-182.
856 FUNG ET AL. 25. Kim, Y. W., M. S. C. Fung, N. C. Sun, C. R. Y. Sun, N. T. Chang, and T. W. Chang. 199. Immunoonjugates that neutralize HIV virions kill T ells infeted with diverse strains of HIV-1. J. Immunol. 144:1257-1262. 26. Kowalski, M., J. Potz, L. Basiripour, T. Dorfman, W. C. Goh, E. Terwilliger, A. Dayton, C. Rosen, W. Haseltine, and J. Sodroski. 1987. Funtional regions of the envelope glyoprotein of human immunodefiieny virus type 1. Siene 237:1351-1355. 27. LaRosa, G. J., J. P. Davide, K. Weinhold, J. A. Waterbury, A. T. Profy, J. A. Lewis, A. J. Langlois, G. R. Dreesman, R. N. Boswell, P. Shadduk, L. H. Holley, M. Karplus, D. P. Bolognesi, T. J. Matthews, E. A. Emini, and S. D. Putney. 199. Conserved sequene and strutural elements in HIV-1 prinipal neutralizing determinant. Siene 249:932-935. 28. Lasky, L. A., J. E. Groopman, C. W. Fennie, P. M. Benz, D. J. Capon, D. J. Dowbenko, G. R. Nakamura, W. M. Nunes, M. E. Renz, and P. W. Berman. 1986. Neutralization of the AIDS retrovirus by antibodies to a reombinant envelope protein. Siene 233:29-212. 29. Lasky, L. A., G. M. Nakamura, D. H. Smith, C. Fennie, C. Shimasaki, E. Patzer, P. Berman, T. Gregory, and D. J. Capon. 1987. Delineation of a region of the human immunodefiieny virus type I gpl2 glyoprotein ritial for interation with the CD4 reeptor. Cell 5:975-985. 3. Leonard, C. K., M. W. Spellman, L. Riddle, R. J. Harris, J. N. Thomas, and T. J. Gregory. 199. Assignment of intrahain disulfide bonds and haraterization of potential glyosylation sites of the type 1 reombinant human immunodefiieny virus envelope glyoprotein (gp12) expressed in Chinese hamster ovary ells. J. Biol. Chem. 265:1373-1382. 31. Linsley, P. S., J. A. Ledbetter, E. Kinney-Thomas, and S.-L. Hu. 1988. Effets of anti-gp12 monolonal antibodies on CD4 reeptor binding by the env protein of human immunodefiieny virus type 1. J. Virol. 62:3695-372. 32. Liou, R. S., E. M. Rosen, M. S. C. Fung, W. N. C. Sun, C. Sun, W. Gordon, N. T. Chang, and T. W. Chang. 1989. A himeri mouse-human antibody that retains speifiity for HIV gpl2 and mediates the lysis of HIV-infeted ells. J. Immunol. 143:3967-3975. 33. Matthews, T. J., A. J. Langlois, W. G. Robey, N. T. Chang, R. C. Galo, P. J. Fishinger, and D. P. Bolognesi. 1986. Restrited neutralization of divergent human T-lymphotropi virus III isolates by antibodies to the major envelope glyoprotein. Pro. Natl. Aad. Si. USA 83:979-9713. 34. MDougal, J. S., J. K. A. Niholson, G. D. Croso, S. P. Cort, M. S. Kennedy, and A. C. Mawle. 1986. Binding of the human retrovirus HTLV-III/LAV/ARV/HIV to the CD4 (T4) moleule: onformation dependene, epitope mapping, antibody inhibition, and potential for idiotypi mimiry. J. Immunol. 137:2937-2944. 35. Moore, J. P. 199. Simple methods for monitoring HIV-1 and HIV-2 gpl2 binding to scd4 by ELISA: HIV-2 has a 25-fold lower affinity than HIV-1 for scd4. AIDS 3:297-35. 36. Moore, J. P., J. A. MKeating, I. M. Jones, P. E. Stephens, G. Clements, S. Thomson, and R. A. Weiss. 199. Charaterization of reombinant gpl2 and gpl6 from HIV-1: binding to monolonal antibodies and scd4. AIDS 4:37-315. 37. Moore, J. P., L. A. Wallae, E. A. C. Follett, and J. A. MKeating. 1989. An enzyme-linked immunosorbent assay for antibodies to the envelope glyoproteins of divergent strains of HIV-1. AIDS 3:155-163. J. VIROL. 38. Myers, G., S. F. Josephs, A. B. Rabson, T. F. Smith, and F. Wong-Staal (ed.). 1989. Human retroviruses and AIDS. Los Alamos National Laboratories, Los Alamos, N.Mex. 39. Nara, P. L., W. C. Hath, N. M. Dunlop, W. G. Robey, L.. Arthur, M. A. Gonda, and P. J. Fishinger. 1987. Simple, rapid, quantitative, synytium-forming miroassay for the detetion of human immunodefiieny virus neutralizing antibody. AIDS Res. Human Retroviruses 3:283-32. 4. Nara, P. L., W. G. Robey, S. W. Pyle, W. C. Hath, N. W. Dunlop, J. W. Bess, J. C. Kelliher, L.. Arthur, and P. J. Fishinger. 1988. Purified envelope glyoproteins from human immunodefiieny virus type 1 variants indue individual, typespeifi neutralizing antibodies. J. Virol. 62:2622-2628. 41. Nara, P. L., L. Smit, N. Dunlop, W. Hath, M. Merges, D. Waters, J. Kelliher, R. C. Gallo, P. J. Fishinger, and J. Goudsmit. 199. Emergene of viruses resistant to neutralization by V3-speifi antibodies in experimental human immunodefiieny virus type 1 IIIB infetion of himpanzees. J. Virol. 64:3779-3791. 42. Olshevsky, U., E. Helseth, C. Furman, J. Li, W. Haseltine, and J. Sodroski. 199. Identifiation of individual human immunodefiieny virus type 1 gpl2 amino aids important for CD4 reeptor binding. J. Virol. 64:571-577. 43. Palker, T. J., M. E. Clark, A. J. Langlois, T. J. Matthews, K. J. Weinhold, R. R. Randall, D. P. Bolognesi, and B. F. Haynes. 1988. Type-speifi neutralization of the human immunodefiieny virus with antibodies to env-enoded syntheti peptides. Pro. Natl. Aad. Si. USA 85:1932-1936. 44. Skinner, M. A., A. J. Langlois, C. B. MDanal, J. S. MDougal, D. P. Bolognesi, and T. J. Matthews. 1988. Neutralizing antibodies to an immunodominant envelope sequene do not prevent gpl2 binding to CD4. J. Virol. 62:4195-42. 45. Sonigo, P., M. Girard, and D. Dormond. 199. Design and trials of AIDS vaines. Immunol. Today 11:465-471. 46. Stephens, D. M., D. Davis, and P. J. Lahmann. 1991. The seond variable region of HIV-1 external envelope glyoprotein ontains a neutralizing epitope, p. 72, TH.A.66. Abstr. VII Int. Conf. AIDS, vol. 2. Istituto Superiore Di Sanita, Rome. 47. Sun, N.-C., D. D. Ho, C. R. Y. Sun, R.-S. Liou, W. Gordon, M. S. C. Fung, X.-L. Li, R. C. Ting, T.-H. Lee, N. T. Chang, and T.-W. Chang. 1989. Generation and haraterization of monolonal antibodies to the putative CD4-binding domain of human immunodefiieny virus type 1 gpl2. J. Virol. 63:3579-3585. 48. Weiss, R. A., P. R. Clapham, J. N. Weber, A. G. Dalgleish, L. A. Lasky, and P. W. Berman. 1986. Variable and onserved neutralization antigens of human immunodefiieny virus. Nature (London) 324:572-575. 49. Wendler, I., V. Bienzle, and G. Hunsman. 1987. Neutralizing antibodies and the ourse of HIV-indued disease. AIDS Res. Human Retroviruses 3:157-163. 5. White-Sharf, M. E., A. Conley, B. J. Potts, E. A. Emini, J. R. Rushe, and S. Silver. 1991. Monolonal antibodies direted to the HIV-1 gpl2 V3 loop exhibit patterns of broad HIV-1 neutralization in vitro, p. 176, W.A.1341. Abstr. VII Int. Conf. AIDS. vol. 2. Istituto Superiore Di Sanita, Rome. 51. Willey, R. L., E. K. Ross, A. J. Bukler-White, T. S. Theodore, and M. A. Martin. 1989. Funtional interation of onstant and variable domains of human immunodefiieny virus type 1 gpl2. J. Virol. 63:3595-36.