Neutralization of HIV-1 primary isolates by polyclonal and monoclonal human antibodies

Transcription

1 International Immunology, Vol. 9, No. 9, pp Neutralization of HIV-1 primary isolates by polyclonal and monoclonal human antibodies Catarina E. Hioe 1, Serena Xu 1, Padmasree Chigurupati 1,3, Sherri Burda 1, Constance Williams 2, Miroslaw K. Gorny 1 and Susan Zolla-Pazner 1,2 1 New York University and 2 Veterans Affairs Medical Centers, 423 East 23rd Street, Room N, New York, NY 10010, USA 3 Present address: Paracelsion, Inc., 222 Langhmir Labs, 95 Brown Road, Ithaca, NY 14850, USA Keywords: AIDS, immunity, neutralizing epitopes Abstract To examine antibody-mediated neutralization of HIV-1 primary isolates in vitro, we tested sera and plasma from infected individuals against four clade B primary isolates. These isolates were analyzed further for neutralization by a panel of several human anti-hiv-1 mab in order to identify the neutralizing epitopes of these viruses. Each of the HIV-1 serum and plasma specimens tested had neutralizing activities against one or more of the four primary isolates. Of the three individual sera, one (FDA-2) neutralized all of the four isolates, while the other two sera were effective against only one virus. The pooled plasma and serum samples reacted broadly with these isolates. Based on the neutralizing activities of the mab panel, each virus isolate exhibited a distinct pattern of reactivity, suggesting antigenic diversity among clade B viruses. Neutralizing epitopes were found in the V3 loop and CD4-binding domain of gp120, as well as near the transmembrane region (cluster II epitope) of gp41. A mab directed to the cluster I epitope of gp41 near the immunodominant disulfide loop weakly neutralized one primary isolate. None of the mab in the panel affected one primary isolate, US4, although this virus was sensitive to neutralization by some of the polyclonal antibody specimens. This isolate was also resistant to neutralization by a cocktail of 10 mab, most of which individually inhibited at least one of the other three viruses tested. These results suggest that neutralizing activity for this latter virus is present in certain HIV-1 sera/plasma, but is not exhibited by the mab in the panel. Thus, effective neutralizing antibodies against primary isolates can be generated by humans upon exposure to HIV-1, but not all of these antigenic specificities are represented in a large panel of human anti-hiv-1 mab. Introduction HIV-1 infection elicits a vigorous antibody response directed to polyclonal (10) and mab (11), investigators have succeeded viral envelope and internal proteins (1). The presence of the in protecting chimpanzees from HIV-1 challenge. Similar results antibodies is usually noted within 1 month post-infection, were reported from studies using the hu-pbl-scid mouse although in a few individuals seroconversion can occur up to model (12,13). Furthermore, the presence of neutralizing anti- 6 months after infection (2,3). Within a short time following bodies in HIV-1 pregnant women has been correlated with a seroconversion, neutralizing antibody activity can be detected lower risk of virus transmission to their infants (14,15). On the (4,5) and the specificities of these neutralizing antibodies have other hand, the role of neutralizing antibodies in controlling so far been mapped exclusively to the virus envelope proteins, established HIV-1 infection and disease progression is still con- gp120 and gp41 (reviewed in 6). troversial. While neutralizing antibody levels may be higher in In agreement with observations of other viruses (7 9), neutralizing HIV-1 long-term non-progressors than in patients with AIDS antibodies to HIV-1 can provide protection against the (16), no association has been demonstrated between neutraliz- initial virus infection. By passive administration of neutralizing ing antibody activity and clinical progression to AIDS (17,18). Correspondence to: S. Zolla-Pazner Transmitting editor: H. Wigzell Received 14 February 1997, accepted 27 May 1997

2 1282 Antibody neutralization of HIV-1 primary isolates Consensus concerning the ability of antibodies to neutralize that had been passaged in vitro only once or twice from the and protect against infection with primary isolates after immunization original virus stocks received from the providers. with HIV-1 vaccines has also been fraught with contro- Virus titration was performed according to the protocol of versy. Immunization of chimpanzees with some vaccine Johnson and Byington (38) using unstimulated PBMC of the preparations based on HIV-1 laboratory-adapted strains same donors that were to be used in the neutralization assays. successfully conferred protection (19 21). However, sera from For the experiments presented here, the virus stock was used these chimpanzees, as well as from human vaccinees, were neat or diluted with the assay medium such that % not able to neutralize primary isolates of HIV-1, in contrast to tissue culture infectious doses (TCID 50 ) of virus were added the high neutralizing titers detectable in the sera of these to each well, yielding p24 levels at day 6 of ng/ml. vaccine recipients for laboratory strains (22,23). These reports corroborated similar observations on the resistance of primary Polyclonal and mab isolates to neutralization by soluble CD4 (24), HIV-1 sera Five HIV-1 serum and plasma preparations were used in the (25) and most mab to HIV-1 env proteins (26), suggesting study (Table 1). Sera F and L are from single bleeds of two that HIV-1 primary isolates were more resistant to neutralization asymptomatic HIV-1 volunteers infected for at least 8 years. than were laboratory-adapted strains of HIV-1. Yet other FDA-2 obtained from the NIH AIDS Research and Reference studies demonstrated that primary isolates were neutralized Reagent Program (HIV-1 neutralizing serum 2, catalog no. by mab of different specificities (27 32), by sera from longterm 1983) is serum of an HIV-1 patient pooled from four separate non-progressors (16) and by sera from the majority of bleeds. Pool 1, pooled human plasma from HIV-1 donors, HIV-1 individuals (17,33,34). Also, using a neutralization was provided by Dr Carl Hansen (California Department of assay with enhanced sensitivity (35), primary isolate-neutraliz- Health Services). Pool 2 is a serum pool from 33 randomly ing activity could be detected in the sera of both chimpanzees picked HIV-1 subjects at the Veterans Affairs Medical Center, and humans immunized with one of several HIV-1 vaccines New York. In the experiments, an HIV-1 serum was also (36,37). included as a control. All sera were heat-inactivated at 56 C In view of the apparent problems in eliciting and detecting for 30 min prior to use. neutralizing antibody activity against HIV-1 primary isolates, Seventeen mab (six anti-v3 loop, four anti-cd4bd, two anti- we undertook a study to determine the prevalence of poly- C5 domain and five anti-gp41) tested in this study are listed clonal and mab from HIV-1-infected subjects which can in Table 2. The generation and specificities of these mab neutralize primary isolates and to compare their potency for were reported in other publications (39 45 and M. K. Gorny neutralizing four HIV-1 clade B primary isolates. These four et al., manuscript in preparation), except for mab 830-D isolates were examined for neutralization by five serum/ and 1281-D (M. K. Gorny et al., unpublished data). In the plasma specimens and 19 anti-hiv-1 mab to reveal the neutralization assays the mab were used after Protein A presence and the accessibility of neutralizing epitopes purification or as culture supernatants at the final concentrations expressed on each of the isolates. All of the five polyclonal designated in the figures and tables. A cocktail of 10 specimens tested neutralized one or more of the primary mab, consisting of 419-D, D, 782-D, 838-D, 559/64-D, isolates. Each primary isolate showed a unique pattern of 654-D, 450-D, 670-D, 1281-D and 98-6, was prepared by neutralization and sensitivity to the polyclonal and mab. mixing an equal quantity of each mab. The mab mixture was Neutralizing epitopes were found on the V3 loop and CD4- then titrated in the neutralization assay from a final total binding domain (CD4bd) of gp120, as well as on gp41. concentration of µg/ml. The human anti-gp41 mab, However, the data also suggest that other neutralizing antibodies 2F5, was purchased from Viral Testing Systems (Houston, specific for yet undefined epitopes are elicited in response TX) and the recombinant anti-cd4bd mab, IgG1b12, was to HIV-1. provided by Dr D. Burton (Scripps Institute). Preparation of PBMC and neutralization assay Methods For neutralization assays, as well as for titration and production Viruses of virus stocks, PBMC from individual random HIV-1 donors were used. Leucopaks from whole blood of individual donors, Four clade B HIV-1 primary isolates (BZ167, 91US056, each containing cells, were purchased from the 92HT593 and US4) were used in the study. BZ167 and New York Blood Center. The PBMC were isolated by Ficoll- 92HT593 are syncytium-inducing (SI) isolates, while 91US056 Hypaque centrifugation as described (35). The cells were and US4 are non-syncytium inducing (NSI) isolates. 91US056 aliquoted at cells/ml, frozen in 93% FBS, 2% penicillin/streptomycin and 92HT593 isolates were provided by Drs J. Bradac and and 5% dimethylsulfoxide, and stored at O. Sharma (NIH) through the AIDS Research and Reference 70 C until use. Reagent Program. BZ167 and US4 isolates were obtained The neutralization assays were performed with minor modifrom Dr J. Mascola (Walter Reed Army Institute of Research fications according to the previously published protocol (35). and the Henry M. Jackson Foundation for Military Medicine). Briefly, thawed PBMC were suspended in assay medium The virus stock of each isolate was expanded in phytohemag- [RPMI 1640, 10% heat-inactivated human AB serum (Sigma, glutinin (PHA)-stimulated peripheral blood mononuclear cells St Louis, MO), 2% penicillin/streptomycin and 1% L-glutamine] (PBMC), and filtered culture supernatants containing high and added to flat-bottom 96-well plates at cells/well. levels of p24 were aliquoted and stored at 70 C as previously The plates were then placed in the CO 2 incubator for at least described (35). For all of the experiments, we used viruses 1 h to allow the cells to settle. In the meantime, the antibody

3 Table 1. ID 50 of HIV-1 sera or plasma against HIV-1 primary isolates a Antibody neutralization of HIV-1 primary isolates 1283 Antibody sample Virus isolate c BZ167 91US056 92HT593 US4 Serum F 138 ( ) b 12 d ( 10 40) d ( 10 30) Serum L 100 (70 140) FDA-2 serum 127 ( ) 185 ( ) 82 (65 100) 155 (80 300) Plasma pool ( ) 84 (70 100) 35 (30 40) 10 Serum pool ( ) 158 ( ) 23 (22 25) 90 (80 100) a Reciprocal of dilution that yielded 50% reduction of p24 production relative to control. In each experiment, no inhibition was observed with HIV-negative serum. b Geometric mean titer (range) from two or three experiments. c Virus input was between 10 and 100 TCID 50 /well. d ID 50 from one representative experiment. Table 2. ID 50 of anti-hiv mab against primary isolates a mab specificity mab Virus isolate b BZ167 91US056 92HT593 US4 V3 loop 257-D 25 c V3 loop 268-D V3 loop 419-D 0.3 ( ) d 6.5 ( ) 33.2 (25 44) 50 V3 loop D 0.1 ( ) 6.9 (4 12) 6.1 ( ) 50 V3 loop 782-D 22.5 f ( ) V3 loop 838-D 0.3 ( ) ND CD4bd 559/64-D 2 (0.4 10) CD4bd 654-D 2 (0.2 8) CD4bd 728/29-D ND CD4bd 830-D 2.7 (1 7.2) C5 450-D C5 670-D ND e 50 gp41 I 246-D (6 20) gp41 I gp41 II gp41 II 1281-D (2 2.5) gp41 II (5 15) 2.3 (0.5 10) 3.6 (0.7 18) 50 a mab concentration (µg/ml) required to reduce p24 production by 50% relative to cultures with no mab. b Each virus isolate was used at TCID 50 /well. c No ID 50 was achieved at the highest mab concentration tested. d Geometric mean and range of ID 50 from two to five experiments e Not determined. f Data from one representative experiment. was diluted appropriately in the assay medium and mixed the cells were lyzed by adding 50 µl of 5% Triton X-100 in with the virus, and the mixture was incubated for 1 h at 37 C PBS per well and the p24 concentration in each well was in the CO 2 incubator. At the end of incubation, 50 µl of each measured using the previously described p24 ELISA (46). antibody/virus mixture was added to the cells from which The percent neutralization was calculated as [1 (mean p24 medium had been aspirated. Each antibody/virus condition in the presence of antibody/mean p24 in the absence of was tested in four or six replicate wells. The cells were antibody)] 100. The 50% neutralization titers (ID 50 ) were incubated for 18 h, then the culture supernatant (containing calculated by interpolation between points above and below the antibody/virus mixture) was removed by careful aspiration 50% respectively on the neutralization curves. and replaced with 200 µl of the assay medium containing human ril-2 (20 U/ml) (Boehringer Mannheim, Indianapolis, IN) and PHA (0.8 µg/ml) (Difco, Detroit, MI). For experiments Results with HIV-1 serum and plasma samples, the culture supernatsubjects Neutralizing activity of polyclonal antibodies from HIV-1 ant was replaced two or three more times (180 µl each) at against primary isolates. days 1 and 2 post-infection to remove anti-p24 antibodies Four primary isolates were examined for sensitivity to neutralization which were present in the samples. Six days after infection, by five serum/plasma samples. The data derived

4 1284 Antibody neutralization of HIV-1 primary isolates Fig. 1. Neutralization of primary isolates BZ167, 91US056, 92HT593 and US4 by HIV-1 plasma and serum samples. These virus isolates were tested in the neutralization assays against serum F (A), serum L (B), FDA-2 serum (C), plasma pool 1 (D) and serum pool 2 (E). from testing each combination of virus and serum/plasma Neutralization of primary isolates by a panel of human mab specimen in two to three experiments are presented in Table specific for HIV-1 1 as geometric mean titers for ID 50. Representative titration To determine the specificity of human antibodies which can curves for each specimen against the four viruses are shown inhibit infection of HIV-1 primary isolates, we tested a panel in Fig. 1(A E). All of the five specimens had neutralizing of mab with well-defined antigenic specificities (39 45 and activity, as demonstrated by the ability to inhibit at least one M. K. Gorny et al., unpublished data) against the same four primary isolate. One individual s serum (FDA-2) neutralized primary isolates tested above. The mab panel consisted of all four viruses tested, with reciprocal ID 50 titers ranging from six anti-v3, four anti-cd4 binding domains (CD4bd), two anti- 82 up to 185. Sera F and L effectively inhibited one isolate C5 and five anti-gp41 mab (Table 2). These individual mab (BZ167), but had low or no activity against the other three were tested at concentrations ranging from 0.04 to 50 µg/ml, viruses. The plasma and serum pools were more broadly levels which are attainable in vivo following passive antibody reactive, neutralizing three and four isolates respectively. In transfer (47). The mean ID 50 values in µg/ml (derived from each experiment, no detectable inhibition was observed with two to five experiments) are presented in Table 2, and a negative control serum (data not shown). representative titration curves for anti-v3, anti-cd4bd and Each virus isolate displayed distinct sensitivity to the poly- anti-gp41 mab against each isolate are shown in Fig. 2(A C). clonal antibody samples. Isolate BZ167 was sensitive to Each virus displayed a distinct pattern of neutralization by neutralization by all of the sera and plasma tested with ID 50 the mab. BZ167 was inhibited by eight of 17 mab, in particular titers ranging from 100 to 158, while isolate US4 was inhibited by mab to the V3 loop (419-D, D, 782-D and 838-D), reproducibly by two serum samples (FDA-2 and serum pool the CD4bd (559/64-D, 654-D and 830-D) and cluster II of 2) with ID 50 titers of 155 and 90 respectively. Both 91US056 gp41 (98-6), whereas US4 was not neutralized by any of the and 92HT593 isolates were repeatedly neutralized by FDA-2 mab tested. Isolates 92HT593 and 91US056 were affected serum, plasma pool 1 and serum pool 2, but the ID 50 values by mab to the V3 loop (419-D and D) and mab to were 2- to 7-fold lower for 92HT593 than for 91US056. In both cluster I (246-D) and cluster II (1281-D and 98-6) of sum, while BZ167 was neutralized with greatest frequency by gp41, but not by the anti-cd4bd mab tested here. Isolate the polyclonal specimens tested (five out of five), the other 92HT593 had previously been shown to be neutralized by an viruses, when neutralized, were inhibited by similar dilutions anti-gp41 mab (2F5), as well as by IgG1b12, a recombinant of sera or plasma. mab to CD4bd. In contrast, 91US056 was neutralized by

5 Antibody neutralization of HIV-1 primary isolates 1285 Fig. 2. Neutralizing activity of mab to V3 loop (A), CD4bd (B) and gp41 (C) against primary isolates BZ167, 91US056, 92HT593 and US4. These mab were tested for dose-dependent neutralization within a concentration range of µg/ml, according to the neutralization assay protocol described in Methods. For each mab and virus combination, the neutralization curve from one experiment, out of two to five repeated experiments, is shown. IgG1b12, but not by 2F5 (32). US4 was refractory to both of mab to two previously described epitope clusters on gp41 these mab (data not shown). The two anti-c5 mab (670-D were tested (Table 2 and Fig. 2C). One mab to cluster I (246- and 450-D) did not neutralize any of the tested isolates. The D), which spans the immunodominant disulfide loop of gp41 results suggest that while some neutralizing epitopes are at positions (40), had weak neutralizing activity shared by two or three of the four viruses, each of these four against primary isolate 91US056. Recently, Cotropia et al. (52) clade B viruses is antigenically distinct and expresses a also reported a weakly neutralizing human mab recognizing a different profile of neutralizing epitopes. linear epitope within the cluster I site of gp41. In addition, Among the six anti-v3 mab tested, four (419-D, D, two of three mab specific for the cluster II epitope (1281-D 782-D and 838-D) had neutralizing activities against one or and 98-6), residing near the transmembrane region of gp41 more isolates (Table 2 and Fig. 2A). The specificities of the (amino acids ) (40), showed significant neutralization. reactive anti-v3 mab include epitopes at the tip of the loop mab 1281-D was effective against 91US056, while 98-6 (419-D and D) and on the N-terminal flanking region neutralized BZ167, 91US056 and 92HT593. The mab 98-6 is (838-D and 782-D) (42 and M. K. Gorny et al., manuscript in directed to a conformational epitope close to or spanning the preparation). However, not all mab to these regions showed same region for which the broadly neutralizing anti-gp41 mab neutralizing activity (257-D and 268-D). Neutralization of 2F5 is specific (29,53). However, unlike 98-6, mab 2F5 was 91US056 isolate by two anti-v3 mab (419-D and D) shown to neutralize 92HT593, but not 91US056 (32). confirms the ability of anti-v3 mab to neutralize NSI isolates, Of the nine mab with ID 50 of 25 µg/ml, three mab (98-6, as we (28,48,49) and others (50,51) have previously shown. 419-D and D) were cross-reactive with two or more In this study, three CD4bd mab (559/64-D, 654-D and viruses (Table 2 and Fig. 2). These cross-neutralizing mab 830-D) showed neutralizing activities; all three inhibited BZ167 are directed to the cluster II epitope of gp41 (98-6), as well (Fig. 2B). The other three primary isolates were not affected as to the V3 loop of gp120 (419-D and D). While the by any of the CD4bd mab tested. Previously published studies V3 loop is, by definition, highly variable, these latter two mab presented similar data for poor activity of other anti-cd4bd are specific to relatively conserved regions at the crown of mab (26), with the exception of the recombinant mab the loop of clade B viruses. IgG1b12, which has broad neutralizing activity for primary We examined 15 mab for neutralization against isolate US4 isolates, including 92HT593 and 91US056 (31,32). and none showed any effect (Table 2) even though this virus

6 1286 Antibody neutralization of HIV-1 primary isolates was used at a low virus input (10 TCID 50 /well). This isolate was also resistant to neutralization by mab 2F5 and IgG1b12 (data not shown), which were known to have broad reactivities against primary isolates (29,31,32). Based on the neutralization study with polyclonal antibodies, US4 is not resistant biologically to antibody-mediated neutralization (Table 1 and Fig. 1; also ref. 54). One possible explanation for the failure of the mab to neutralize this virus was that a single mab was not sufficient and that synergistic activity of two or more mab was necessary for effective neutralization (55). To examine this possibility, we mixed 10 mab (419-D, D, 782-D, 838-D, 559/64-D, 654-D, 450-D, 670-D, 1281-D and 98-6) which, except for 450-D and 670-D, neutralized at least one of the viruses tested here. The cocktail was evaluated for activity against US4 in the neutralization assay. The mab mixture was titrated starting with a total concentration of 50 µg/ml, but no significant neutralization was observed (Fig. 3A). This contrasts to FDA-2 serum (Fig. 3B and also Fig. 1C) and serum pool 2 (Fig. 1E), which achieved 50% neutralization of US4 with geometric mean titers of 155 and 90 (Table 1) Fig. 3. Neutralization of isolate US4 by a mab cocktail (A) as compared to FDA-2 serum (B). A cocktail of 10 mab was used at a total concentration of µg/ml, whereas FDA-2 serum was tested at 2-fold serial dilutions starting at 1:10. The cocktail was prepared by mixing equal amounts of 10 mab described in Methods. distinct set of neutralizing epitopes, these viruses also share some neutralizing epitopes which permit cross-neutralization respectively. These findings indicate that neutralizing antibod- to occur. ies for the US4 isolate are present in certain serum samples, Shared as well as distinct neutralizing epitopes may be but are not represented by any of the mab tested in this study. displayed by laboratory-adapted strains and primary isolates. Thus, certain mab to the V3 loop and the CD4bd are capable Discussion of neutralizing both laboratory strains and primary isolates (26,28,31,32,41,42,56 58). Similarly, an anti-gp41 mab 2F5 Antibody-mediated virus neutralization is a function of the was effective against both laboratory and primary strains presence of neutralizing antibodies directed to critical viral (29,32). However, some mab have been found to neutralize epitopes and the expression of these epitopes on the surface primary isolates only. Notably, an anti-gp41 mab, 98-6, which of the virions being tested. Thus, detection of neutralizing did not affect laboratory strains (SF-2, IIIB and RF) (39,44,59) activity by HIV-1 sera or anti-hiv-1 mab derived from the showed neutralizing activity against three of the primary cells of HIV-1 subjects is a multifactorial phenomenon. It isolates tested herein (Table 2). This mab is directed to a reflects both the prevalence with which particular neutralizing conformational epitope, cluster II of gp41 (encompassing antibodies are produced and the potency of these antibodies, amino acids ) (40) close to the linear epitope (amino as well as the existence and exposure of the epitopes for acids ) recognized by mab 2F5. In addition, an anti- which these antibodies are specific on the viruses being V2 mab, 697-D, was found previously to be effective against studied. To dissect these aspects of HIV-1 neutralization by certain primary isolates but not against laboratory strains of antibodies from infected human subjects, we have examined HIV-1 (30). the neutralization of four primary HIV-1 isolates by five HIV-1 The V3 loop was previously defined as a key neutralizing serum or plasma specimens and by 19 human mab. The epitope for laboratory strains (reviewed in 60). Based on the results show that each primary isolate tested could be neutralized, results with our mab panel, four anti-v3 mab were also shown but differed in its susceptibility to various polyclonal to neutralize the SI primary isolate BZ167 and two of these and mab preparations, indicating that each primary isolate mab neutralized SI primary isolate 92HT593. These four anti- displays a distinct array of neutralizing epitopes. However, V3 mab collectively recognized core epitopes spanning nine among these four primary isolates, cross-neutralization was amino acids at the tip and on the N-terminal flanking sequence observed; the mab with the most frequent and most extensive of the V3 loop (42 and M. K. Gorny et al., manuscript in cross-neutralizing activity were specific for the V3 loop of preparation). As suggested by a number of investigators (61 gp120 and for an epitope near the transmembrane domain 63), the V3 loop of SI viruses may be readily exposed and of gp41. reactive with anti-v3 antibodies. The potency of these mab As shown in Tables 1 and 2, each of the four primary for SI viruses, relative to NSI viruses, may reflect the structural isolates tested exhibited a unique pattern of sensitivity and and/or charge differences in this region of SI and NSI viruses reactivity with the polyclonal and mab in the panel, confirming (64 66). The activity of the anti-v3 mab with BZ167 and the extensive antigenic variation among these clade B primary 92HT593 is also likely due to the fact that these mab were isolates. Similar finding were reported previously with laborat- selected for reactivity with the V3 peptide of laboratoryadapted ory-adapted strains (42,44 and M. K. Gorny, manuscript SI viruses (42 and M. K. Gorny et al., manuscript in in preparation). Nonetheless, cross-neutralization of primary preparation). Nonetheless, two anti-v3 mab also reproducibly isolates was noted above and in previous studies (25,26,32 neutralized the NSI virus 91US056, indicating that anti-v3 34) with polyclonal and mab reagents, and also was reported mab can neutralize both SI and NSI primary isolates. The previously for laboratory strains (29). Thus, while each virus, molecular basis of neutralization by these anti-v3 antibodies whether a primary isolate or a laboratory strain, displays a is not yet understood. It has been suggested that they interfere

7 Antibody neutralization of HIV-1 primary isolates 1287 at one of the post-binding steps, possibly by blocking the V3 any more sensitive to neutralization than the other three loop cleavage required for fusion of the virion with the cell isolates tested here. Taken together, these data suggest that membrane (67,68). The V3 loop has also been implicated as BZ167 may bear neutralization epitopes which are highly a major determinant of tissue tropism for HIV-1 viruses (69), immunogenic and frequently expressed by other clade B and most recently the V3 sequence was found to influence primary isolates. Thus, terms describing viruses as neutralization the utilization of both the CXCR4 and CCR5 co-receptors sensitive or neutralization resistant are misleading withthe (49,70,71). The results summarized herein suggest that what- out careful definition and without reference to issues that ever the mechanism by which the V3 loop contributes to virus include definition of antibody prevalence, antibody potency infectivity, it is involved with the infectious process of both SI and virion antigenic structures. and NSI viruses. The degree of neutralizability of a virus may also be related The US4 primary isolate stands in contrast with the other to the method used to quantify neutralizing activity. For three viruses examined in this study. It was neutralized example, the in vitro neutralization assay used in the studies effectively by only two of the polyclonal specimens and by described above employed unstimulated PBMC as target none of the mab tested. These data suggest that production cells for infection. This assay had been shown to improve the of neutralizing antibodies to epitopes displayed by this virus is detection of HIV-1 neutralization by antibodies compared to a less frequent occurrence and that immunogenic neutralizing the conventional assay utilizing PHA-stimulated PBMC as epitopes present on common clade B viruses are absent or target cells (35). It was noted previously that the resting cell unexposed on US4. Nonetheless, this isolate is not resistant assay was significantly more sensitive when used to detect to neutralization since, when the relevant antibodies are neutralizing activities of an anti-v3 mab (35) and sera from present, it can be neutralized. Previous reports have also immunized chimpanzees (36) and humans (37). In this study, described primary isolates which are neutralized poorly by similar findings were observed, e.g. 3- to 80-fold lower most HIV-1 sera and mab (26). Similarly, Nyambi et al. (34) concentrations of mab D and 98-6 were required to have described primary isolates which are neutralized by obtain 50% neutralization of the 92HT593 isolate in the resting only a minority of sera tested and have suggested that these cell assay than in the PHA-blast assay (data not shown). Such isolates, like US4, express neutralizing epitopes which are quantitative differences were also found with mab 419-D, restricted in their distribution amongst primary isolates. Such D (also reported in 35) and 838-D against BZ167 findings suggest that neutralizing epitopes other than those isolate. More dramatically, for a number of mab and virus defined to date with human anti-hiv-1 mab derived from the combinations, such as 559/64-D or 654-D mab against BZ167 cells of HIV-1 individuals remain to be identified. These virus, and 419-D, 1281-D or 98-6 mab against 91US056 virus, neutralizing antibodies may be specific for epitopes present neutralization was readily detectable in the resting cell assay, on native or oligomeric forms of gp120 (72), but absent from but not in the PHA-blast assay, even at mab concentrations recombinant monomeric forms of gp120 or gp160 or from up to 100 µg/ml (data not shown). The differences obtained peptides, the molecules most frequently used for selection of with the two assays were apparent when the same virus input anti-hiv mab. (measured as TCID 50 ) was used in both types of assays and The data discussed above also help to clarify an ambiguity when the p24 measurements were performed at days 4 8 in the published HIV literature. Reference has frequently been post-infection. made to neutralization-resistant isolates, yet this term has Recently, a paper was published by Zhou and Montefiori not been carefully defined and, indeed, seems to mean (73) that questions the greater sensitivity of the resting cell different things to different authors. Most literally defined, a assay compared to the PHA-blast assay. Essentially every neutralization-resistant isolate might be one which cannot be point raised by these authors, including the differential growth neutralized. However, a virus which is not neutralized by a kinetics, the reduced infectivity of virus on quiescent versus given panel of polyclonal and/or mab may be neutralizable activated cells, the effect of the level of virus input, the time with the appropriate antibody. Possibly, this virus bears of assay in the two types of assays and the use of monoclonal neutralization epitopes which are not carried by most primary versus polyclonal antibodies, has been addressed previously isolates within the same cladal grouping. US4 is an example (35 37,74 and data above). Discrepancies between the of an isolate which appears, on the basis of its lack of reaction results of Zhou and Montefiori and ours concerning the relative with 17 mab (15 mab in our panel plus 2F5 and IgG1b12), sensitivities of the two neutralization assays can be attributed to be neutralization resistant. Nonetheless, it is clearly capable primarily to two factors. First, technical differences in the of being neutralized when appropriate antibodies are present, method of conducting the resting cell assay may have contributed as with serum FDA-2 which neutralized this isolate with a 50% to the inability of Zhou and Montefiori to grow NSI viruses. neutralizing titer of 1:155. Conversely, neutralization-sensitive One obvious example is their use of viruses as 1:5 diluted strains have been designated, but it is unclear whether the stocks, regardless of the titers of the different virus stocks. definition of this term refers to viruses which are neutralized The data reported above and by Zolla-Pazner and Sharpe by most sera and mab or whether the infectivity of such (35) clearly indicate that NSI viruses such as 91US056, US4 strains can be blocked with small amounts of antibody. By and JR-CSF grow under the conditions of the resting cell the former definition, a strain like BZ167 might be termed a assay with p24 levels ranging from 10 to 60 ng/ml by day 6. neutralization-sensitive isolate since it was neutralized by five Second, in the experiments where the two neutralization of five polyclonal specimens and by eight of 17 mab. However, assays were directly compared, Zhou and Montefiori were evaluated on the basis of the titer and potency of reagents inconsistent in the time points used for harvesting cultures, such as FDA-2 serum and mab 98-6, BZ167 is not intrinsically variously measuring neutralization in the linear phase of

8 1288 Antibody neutralization of HIV-1 primary isolates virus production, at the peak of the virus production or on virus load and phenotype switch during primary infection with human immunodeficiency virus type 1. J. Infect. Dis. 175:231. appropriate days. According to their own data (73) and ours 6 Poignard, P., Klasse, P. J. and Sattentau, Q. J Antibody (74), the resting cell assay gave consistent neutralization neutralization of HIV-1. Immunol. Today 17:239. results on several days, whereas results with the blast assay 7 Loofbourow J. C., Cabasso, V. J., Roby, R. E. and Anuskiewicz, were highly dependent upon the day of the assay. Thus, by W Rabies immune globulin (human) clinical trials and dose varying the assay time rather than defining and standardizing determination. J. Am. Med. Ass. 217: Iwarson, S., Ahlmen, J., Eriksson, E., Hermodsson, S., Kjellman, their assay condition, Zhou and Montefiori (73) were not able H., Ljunggren, C. and Selander, D Hepatitis B immune to recognize and report the greater sensitivity of the resting globulin in prevention of hepatitis B among hospital staff members. cell assay. Moreover, they did not provide data directly J. Infect. Dis. 135:473. comparing neutralization results where equivalent virus inputs 9 Groothuis, J. R., Levin, M. J., Rodriguez, W., Hall, C. B., Long, were used in the two assays. Their work, therefore, fails to C. E., Kim, H. W., Lauer, B. A. and Hemming, V. G Use of intravenous gamma globulin to passively immunize high risk refute the data reported here as well as in previous reports children against respiratory syncytial virus: safety and (35 37,74) that the resting cell assay is equally or more pharmacokinetics. Antimicrob. Ag. Chemother. 35:1469. sensitive than the PHA-blast assay depending on the virus/ 10 Prince, A. M., Reesink, H., Pascual, D., Horowitz, B., Hewlett, I., antibody pair tested. Murthy, K. K., Cobb, K. E. and Eichberg, J. W Prevention of HIV infection by passive immunization with HIV immunoglobulin. In conclusion, we have demonstrated the neutralization of AIDS Res. Hum. Retroviruses 7:971. four HIV-1 clade B primary isolates by polyclonal antibodies 11 Emini E. A., Schleif, W. A., Nunberg, J. H., Conley, A. J., Eda, Y., and by mab specific for the V3 loop and CD4bd of gp120, Tokiyoshi, S., Putney, S. D., Matsushita, S., Cobb, K. E., Jett, as well as by mab directed to gp41. Our data also suggest C. M., Eichberg, J. W. and Murthy, K. K Prevention of that other functional but unidentified antibodies that are HIV-1 infection in chimpanzees by gp120 V3 domain-specific monoclonal antibody. Nature 355:728. distinct from the mab tested here can be elicited in HIV-1 12 Safrit, J. T., Fung, M. S. C., Andrews, C. A., Braun, D. G., Sun, infected humans. W. N. C., Chang, T. W. and Koup, R. A hu-pbl-scid mice can be protected from HIV-1 infection by passive transfer of monoclonal antibody to the principal neutralizing determinant of Acknowledgements envelope gp120. AIDS 7: Gauduin, M.-C., Safrit, J. T., Weir, R., Fung, M. S. C. and Koup, This work was supported in part by a Merit Review grant and the R. A Pre- and post-exposure protection against human Research Center for AIDS and HIV Infection (RCAHI) of the Departimmunodeficiency virus type 1 infection mediated by a monoclonal ment of Veterans Affairs and by NIH grants AI32424 and AI antibody. J. Infect. Dis. 171:1203. C. E. H. was supported by NIH AIDS Institutional Training Grant 14 Ugen, K. E., Goedert, J. J., Boyer, J., Refaeli, Y., Frank, I., Williams, AI The authors would like to thank Dr Suman Laal, Dr Juan W. V., Willoughby, A., Landesman, S., Mendez, H., Rubinstein, A., Bandres, Lisa Bastiani and Karen Samanich for helpful discussions Kieber-Emmons, T. and Weiner, D. B Vertical transmission and comments on the manuscript. of human immunodeficiency virus (HIV) infection. Reactivity of maternal sera with glycoprotein 120 and 41 peptides from HIV Abbreviations type 1. J. Clin. Invest. 89: Scarlatti, G., Albert, J., Rossi, P., Hodara, V., Biraghi, P., CD4bd CD4-binding domain Muggiasca, L. and Fenyö, E. M Mother to child transmission ID of human immunodeficiency virus type 1: correlation with 50 50% inhibiting dose NSI non-syncytium inducing neutralizing antibodies against primary isolates. J. Infect. Dis. PBMC peripheral blood mononuclear cell 168:207. PHA phytohemagglutinin 16 Cao, Y., Qin, L., Zhang, L., Safrit, J. and Ho, D. D Virologic SI syncytium inducing and immunologic characterization of long-term survivors of human TCID immunodeficiency virus type 1 infection. N. Engl. J. Med. 332: % tissue culture infectious dose 17 Arendrup, M., Nielsen, C. M., Hansen, J.-E. S., Mathiesen, L., Lindhardt, B., Scheibel, E. and Nielsen, J. O Neutralizing antibodies against two HIV-1 strains in consecutively collected References serum samples: cross neutralization and association to HIV-1 1 Robey, W. G., Safai, B., Oroszlan, S., Arthur, L. O., Gonda, M. A., related disease. Scand. J. Infect. Dis. 24:21. Gallo, R. C. and Fischinger, P. J Characterization of 18 Prince A. M., Pascual, D., Kosolapov, L. B., Kurokawa, D., Baker, envelope and core structural gene products of HTLV-III with sera L. and Rubinstein, P Prevalence, clinical significance, from AIDS patients. Science 228:593. and strain specificity of neutralizing antibody to the human 2 Gaines, H., von Sydow, M., Sönnerborg, A., Albert, J., Czajkowski, immunodeficiency virus. J. Infect. Dis. 156:268. J., Pehrson, P. O., Chiodi, F., Moberg, L., Fenyö, E.-M., Åsjö, B. 19 Berman, P. W., Gregory, T. J., Riddle, L., Nakamura, G. R., and Forsgren, M Antibody response in primary human Champe, M. A., Porter, J. P., Wurm, F. M., Hershberg, R. D., immunodeficiency virus infection. Lancet i:1249. Cobb, K. E. and Eichberg, J Protection of chimpanzees 3 Horsburgh, C. R., Jr, Ou, C. Y., Jason, J., Holmberg, S. D., from infection by HIV-1 vaccination with recombinant glycoprotein Longini, I. M., Jr, Schable, C., Mayer, K. H., Lifson, A. R., 120 but not gp160. Nature 345:622. Schochetman, G., Ward, J. W., Rutherford, G. W., Evatt, B. L., 20 Girard, M., Kieny, M. P., Pinter, A., Barre-Sinoussi, F., Nara, P., Seage, G. R., III and Jaffe, H. W Duration of human Kulbe, H., Kusumi, K., Chaput, A., Reinhart, T., Muchmore, E., immunodeficiency virus infection before detection of antibody. Ronco, J., Kaczorek, M., Gomard, E., Gluckman, J.-C. and Fultz, Lancet ii:637. P. N Immunization of chimpanzees confers protection 4 Albert, J., Abrahamsson, B., Nagy, K., Aurelius, E., Gaines, H., against challenge with human immunodeficiency virus. Proc. Natl Nyström, G. and Fenyö, E. M Rapid development of isolate- Acad. Sci. USA 88:542. specific neutralizing antibodies after primary HIV-1 infection and 21 el-amad, Z., Murthy, K. K., Higgins, K., Cobb, E. K., Haigwood, N. consequent emergence of virus variants which resist neutralization L., Levy, J. A. and Steimer, K. S Resistance of chimpanzees by autologous sera. AIDS 4:107. immunized with recombinant gp120 SF2 to challenge by HIV-1 SF2. 5 Lathey, J. L., Pratt, R. D. and Spector, S. A Appearance AIDS 9:1313. of autologous neutralizing antibody correlates with reduction in 22 Hanson, C. V Measuring vaccine-induced HIV

9 Antibody neutralization of HIV-1 primary isolates 1289 neutralization: report of a Workshop. AIDS Res. Hum. virus-uninfected recipients of candidate AIDS vaccines. J. Infect. Retroviruses 10:645. Dis. 175: Matthew, T. J The dilemma of neutralization resistance of 38 Johnson, V. A. and Byington, R. E Quantitative assays for HIV-1 field isolates and vaccine development. AIDS Res. Hum. virus infectivity. In Aldovini, A. and Walker, B. D., eds, Techniques Retroviruses 10:631. in HIV Research, p. 71. Stockton Press, New York. 24 Daar, E. S., Li, X. L., Moudgil, T. and Ho, D. D High 39 Tyler, D. S., Stanley, S. D., Zolla-Pazner, S., Gorny, M. K., concentrations of recombinant soluble CD4 are required to Shadduck, P. P., Langlois, A. J., Matthews, T. J., Bolognesi, D. P., neutralize primary human immunodeficiency virus type 1 isolates. Palker, T. J. and Weinhold, K. J Identification of sites Proc. Natl Acad. Sci. USA 87:6574. within gp41 that serve as targets for antibody-dependent cellular 25 Moore, J. P., Cao, Y., Leu, J., Qin, L., Korber, B. and Ho, D. D. cytotoxicity by using human monoclonal antibodies. J. Immunol Inter- and intraclade neutralization of human 145:3276. immunodeficiency virus type 1: genetic clades do not correspond 40 Xu, J.-Y., Gorny, M. K., Palker, T., Karwowska, S. and Zolla-Pazner, to neutralization serotypes but partially correspond to gp120 S Epitope mapping of two immunodominant domains of antigenic serotypes. J. Virol. 70:427. gp41, the transmembrane protein of human immunodeficiency 26 Moore, J. P., Cao, Y., Qing, L., Sattentau, Q. J., Pyati, J., Koduri, virus type 1, using ten human monoclonal antibodies. J. Virol. R., Robinson, J., Barbas, C. F., III, Burton, D. R. and Ho, D. D. 65: Primary isolates of human immunodeficiency virus type 1 41 Karwowska, S., Gorny, M. K., Buchbinder, A., Gianakakos, V., are relatively resistant to neutralization by monoclonal antibodies Williams, C., Fuerst, T. and Zolla-Pazner, S Production of to gp120, and their neutralization is not predicted by studies with human monoclonal antibodies specific for conformation and linear monomeric gp120. J. Virol. 69:101. non-v3 epitopes of gp120. AIDS Res. Hum. Retroviruses 8: Hansen, J.-E. S., Nielsen, C., Arendrup, M., Olofsson, S., 42 Gorny, M. K., Xu, J.-Y., Karwowska, S., Buchbinder, A. and Zolla- Mathiesen, L., Nielsen, J. O. and Clausen, H Broadly Pazner, S Repertoire of neutralizing human monoclonal neutralizing antibodies targeted to mucin-type carbohydrate antibodies specific for the V3 domain of HIV-1 gp120. J. epitopes of human immunodeficiency virus. J. Virol. 65:6461. Immunol. 150: Conley, A. J., Gorny, M. K., Kessler, J. A., II, Boots, L. J., 43 McKeating, J. A., Bennett, J., Zolla-Pazner, S., Schutten, M., Lineberger, D., Emini, E. A., Ossorio, M., Koenig, S., Williams, Ashelford, S., Brown, A. L. and Balfe, P Resistance of a C. and Zolla-Pazner, S Neutralization of primary human human serum-selected human immunodeficiency virus type 1 immunodeficiency virus type 1 isolates by the broadly reactive escape mutant to neutralization by CD4 binding site monoclonal anti-v3 monoclonal antibody, D. J. Virol. 68:6994. antibodies is conferred by a single amino acid change in gp Conley, A. J., Kessler II, J. A., Boots, L. J., Tung, J.-S., Arnold, B. J. Virol 67:5216. A., Keller, P. M., Shaw, A. R. and Emini, E. A Neutralization 44 Forthal, D. N., Landucci, G., Gorny, M. K., Zolla-Pazner, S. and of divergent human immunodeficiency virus type 1 variants and Robinson, W. E., Jr Functional activities of 20 human primary isolates by IAM-41-2F5, an anti-gp41 human monoclonal immunodeficiency virus type 1 (HIV-1)-specific human monoclonal antibody. Proc. Natl Acad. Sci. USA 91:3348. antibodies. AIDS Res. Hum. Retroviruses 11: Gorny M. K., Moore, J. P., Conley, A. J., Karwowska, S., Sodroski, 45 Zolla-Pazner, S., O Leary, J., Burda, S., Gorny, M. K., Kim, M., J., Williams, C., Burda, S., Boots, L. J. and Zolla-Pazner, S Mascola, J. and McCutchan, F Serotyping of primary Human anti-v2 monoclonal antibody that neutralizes primary but human immunodeficiency virus type 1 isolates from diverse not laboratory isolates of human immunodeficiency virus type 1. geographic locations by flow cytometry. J. Virol. 69:3807. J. Virol. 68: Laal, S., Burda, S., Sharpe, S. and Zolla-Pazner, S A rapid, 31 Burton, D. R., Pyati, J., Koduri, R., Sharp, S. J., Thornton, G. B., automated, microtitre assay for measuring neutralization for HIV- Parren, P. W. H. I., Sawyer, L. S. W., Hendry, R. M., Dunlop, N., 1. AIDS Res. Hum. Retroviruses 9:781. Nara, P. L., Lamacchia, M., Garratty, E., Stiehm, E. R., Bryson, Y. 47 Günthard, H. F., Gowland, P. L., Schüpbach, J., Fung, M. S. C., J., Cao, Y., Moore, J. P., Ho, D. D. and Barbas, C. F., III Böni, J., Liou, R.-S., Chang, N. T., Grob, P., Graepel, P., Braun, Efficient neutralization of primary isolates of HIV-1 by a D. G. and Lüthy, R A phase I/IIA clinical study with a recombinant human monoclonal antibody. Science 266:1024. chimeric mouse human monoclonal antibody to the V3 loop of 32 Trkola, A., Pomales, A. B., Yuan, H., Korber, B., Maddon, P. J., human immunodeficiency virus type 1 gp120. J. Infect. Dis. Allaway, G. P., Katinger, H., Barbas, C. F., III, Burton, D. R., Ho, 170:1384. D. D. and Moore, J. P Cross-clade neutralization of primary 48 Stomatatos, L., Zolla-Pazner, S., Gorny, M. K. and Cheng-Mayer, isolates of human immunodeficiency virus type 1 by human C Binding of antibodies to virion-associated gp120 monoclonal antibodies and tetrameric CD4 IgG. J. Virol. 69:6609. molecules of primary-like human immunodeficiency virus type 1 33 Wrin, T., Crawford, L., Sawyer, L., Weber, P., Sheppard, H. W. and (HIV-1) isolates: effect on HIV-1 infection of macrophages and Hanson, C. V Neutralizing antibody response to autologous peripheral blood mononuclear cells. Virology 229:360. and heterologous isolates of human immunodeficiency virus. J. 49 Hill, C. M., Deng, H. K., Unutmaz, D., Kewal-Ramani, V. N., Acquired Immune Defic. Syndr. 7:211. Bastiani, L., Gorny, M. K., Zolla-Pazner, S. and Littman, D. R. 34 Nyambi, P. N., Nkengasong, J., Lewi, P., Andries, K., Janssens, Envelope glycoproteins from HIV-1, HIV-2, and SIV can use W., Fransen, K., Heyndrickx, L., Piot, P. and van der Groen, G. human CCR5 as co-receptor for viral entry and make direct CD Multivariate analysis of human immunodeficiency virus type dependent interactions with this chemokine receptor. J. Virol., 1 neutralization data. J. Virol. 70:6235. in press. 35 Zolla-Pazner, S. and Sharpe, S A resting cell assay for 50 Fouts, T. R., Binley, J. M., Trkola, A., Robinson, J. E. and Moore, improved detection of antibody-mediated neutralization of HIV J. P Neutralization of the human immunodeficiency virus type I primary isolates. AIDS Res. Hum. Retroviruses 11:1449. type 1 primary isolate JR-FL by human monoclonal antibodies 36 Lubeck, M. D., Natuk, R. N., Myaghikh, M., Kaylan, N., Aldrich, correlates with antibody binding to the oligomeric form of the K., Sinangil, F., Alipanah, S., Murthy, S. C. S., Chanda, P. K., envelope glycoprotein complex. J. Virol. 71:2779. Nigida, S., Markham, P. D., Zolla-Pazner, S., Steimer, K, Wade, 51 Moore, J. P., Trkola, A., Korber, B., Boots, L. J., Kessler II, J. A., M., Reitz, M. S., Jr, Arthur, L. O., Mizutani, S., Davis, A., Hung, McCutchan, F. E., Mascola, J., Ho, D. D., Robinson, J. and Conley, P., Gallo, R. C., Eichberg, J. and Robert-Guroff, M Longterm A. J A human monoclonal antibody to a complex epitope protection of chimpanzees against high-dose HIV-1 in the V3 region of gp120 of human immunodeficiency virus type challenge induced by immunization. Nature Med. 3: has broad reactivity within and outside clade B. J. Virol. 69: Zolla-Pazner, S., Alving, C., Belshe, R., Berman, P., Burda, S., 52 Cotropia, J., Ugen, K. E., Kliks, S., Broliden, K., Broliden, P.-A., Chigurupati, P., Clements, M. L., Duliege, A.-M., Excler, J.-L., Hoxie, J. A., Srikantan, V., Williams, W. V. and Weiner, D. B Hioe, C., Kahn, J., McElrath, M. J., Sharpe, S., Sinangil, F., Steimer, A human monoclonal antibody to HIV-1 gp41 with neutralizing K., Walker, M. C., Wassef, N. and Xu, S Neutralization of activity against diverse laboratory isolates. J. Acquired Immune a clade B primary isolate by sera from human immunodeficiency Defic. Syndr. and Hum. Retrovirol. 12:221.