Time dependence of protective post-exposure prophylaxis with human monoclonal antibodies against pathogenic SHIV challenge in newborn macaques

Virology 358 (2007) 69 – 78 www.elsevier.com/locate/yviro

Time dependence of protective post-exposure prophylaxis with human monoclonal antibodies against pathogenic SHIV challenge in newborn macaques Flavia Ferrantelli a,b , Kathleen A. Buckley a , Robert A. Rasmussen a,b , Alistair Chalmers a , Tao Wang a,b , Pei-Lin Li a,b , Alison L. Williams a , Regina Hofmann-Lehmann a,b , David C. Montefiori c , Lisa A. Cavacini b,e , Hermann Katinger f,g , Gabriela Stiegler f,g , Daniel C. Anderson d , Harold M. McClure d , Ruth M. Ruprecht a,b,⁎ a

Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Boston, MA, USA b Department of Medicine, Harvard Medical School, Boston, MA, USA c Department of Surgery, Duke University Medical Center, Durham, NC, USA d Yerkes National Primate Research Center, Emory University, Atlanta, GA, USA e Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA, USA f Institute of Applied Microbiology, University of Agriculture, Vienna, Austria g Polymun Scientific Immunobiological Forschung GmbH, Vienna, Austria Received 14 April 2006; returned to author for revision 18 May 2006; accepted 28 July 2006 Available online 25 September 2006

Abstract In a primate model of postnatal virus transmission, we have previously shown that 1 h post-exposure prophylaxis (PEP) with a triple combination of neutralizing monoclonal antibodies (nmAbs) conferred sterilizing protection to neonatal macaques against oral challenge with pathogenic simian–human immunodeficiency virus (SHIV). Here, we show that nmAbs can also partially protect SHIV-exposed newborn macaques against infection or disease, when given as 12 or 24 h PEP, respectively. This work delineates the potential and the limits of passive immunoprophylaxis with nmAbs. Even though 24 h PEP with nmAbs did not provide sterilizing immunity to neonatal monkeys, it contained viremia and protected infants from acute disease. Taken together with our results from other PEP studies, these data show that the success of passive immunization depends on the nmAb potency/dose and the time window between virus exposure and start of immunotherapy. © 2006 Elsevier Inc. All rights reserved. Keywords: Post-exposure prophylaxis; Passive immunization; Neutralizing monoclonal antibodies; Pathogenic simian–human immunodeficiency virus (SHIV); Rhesus macaques; Oral infection

Introduction Passive immunoprophylaxis is a very useful tool to investigate the role of antibodies against HIV infection and, as such, can be used in primate models of simian–human immunodeficiency virus (SHIV) infection, as the latter viruses encode ⁎ Corresponding author. Dana-Farber Cancer Institute, 44 Binney Street, JFB809, Boston, MA 02115, USA. Fax: +1 617 632 3112. E-mail address: [email protected] (R.M. Ruprecht). 0042-6822/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2006.07.056

HIV-1 env genes in the context of the simian immunodeficiency virus (SIV) backbone. Because the use of anti-HIV IgG in preclinical studies in primates led to inconsistent results (reviewed in Ferrantelli and Ruprecht, 2002), we limited our studies to combinations of human neutralizing monoclonal antibodies (nmAbs) known to target conserved envelope epitopes. Several such anti-HIV nmAbs are currently available, which are known to neutralize primary HIV isolates of different clades and groups in vitro (Binley et al., 2004; Ferrantelli et al., 2004a; Kitabwalla et al., 2003; Stiegler et al., 2001; Trkola et

70

F. Ferrantelli et al. / Virology 358 (2007) 69–78

al., 1995; Verrier et al., 2001; Xu et al., 2001; Zwick et al., 2001a, 2001b), either when used singly or in combinations. We and others have shown previously (Baba et al., 2000; Ferrantelli et al., 2003, 2004b; Hofmann-Lehmann et al., 2001, 2002; Mascola et al., 1999, 2000; Parren et al., 2001; Veazey et al., 2003) that passive administration with anti-HIV envelope (Env) nmAbs can protect monkeys against intravenous (I.V.) or mucosal challenge with SHIVs. In our past studies in newborn macaques, we achieved complete protection against oral challenge with the pathogenic SHIV89.6P either with a triple or a quadruple combination of nmAbs administered as 1 h post-virus exposure prophylaxis (PEP) (Ferrantelli et al., 2004b). Here, we present two comparative studies that tested the efficacy of PEP with nmAbs given at 1 h, 12 h or 24 h after virus challenge in order to define the time window of opportunity for effective antibody treatment against SHIV infection. Results Intravenous PEP with four anti-HIV nmAbs against SHIV89.6P oral challenge in rhesus newborns Eleven newborn macaques were challenged orally with the highly pathogenic SHIV89.6P. Whereas a group of four control animals was left untreated (Fig. 1A), two treatment groups of four and three neonatal macaques were treated intravenously with the combination of IgG1b12, 2G12, 2F5, and 4E10 either 1 h, or 12 h post-virus exposure, respectively; a second treatment-dose was administered 8 days later (Figs. 1B, C). All controls displayed high plasma viremia levels (Fig. 1D) and had steep CD4+ T-cell losses within 14 days (Fig. 1G); all four rhesus infants were euthanized within 7 weeks of challenge because of AIDS. In contrast, all treated animals were protected from either SHIV89.6P infection (Figs. 1E, F) or virus-induced acute disease (Figs. 1H, I). Three of the four rhesus newborns that were given 1 h PEP were protected from persistent systemic infection, with two animals having undetectable plasma viremia throughout a 47month follow-up and one animal, monkey RWq-8, showing a single very low viremia blip at week 20 post-challenge (not followed by seroconversion) and again low plasma viremia between weeks 199 and 206 (Fig. 1E). The fourth nmAb-treated animal (RYq-8) presented with a delayed-peak plasma viremia that was almost 100-fold lower than that of the controls; this animal was protected from CD4+ T-cell depletion (Fig. 1H). Three of the monkey infants that were treated 1 h post-virus exposure remained healthy during a 47-month follow-up (Fig. 1H); RAq-8 remained virus negative until it was euthanized at week 7 of age due to Staphylococcus aureus aspiration pneumonia unrelated to the virus challenge or nmAb treatment (Fig. 1H). Persistent systemic infection was prevented in one out of three neonatal macaques in the 12 h PEP group (animal RQs-8); this animal only displayed two blips of transient, low plasma viral RNA (vRNA) at weeks 16 and 97 post-challenge (Fig. 1F and data not shown). The other two animals had approximately

100-fold lower levels of plasma vRNA than the controls (Fig. 1F); both of these infants were protected from SHIV89.6Pinduced acute disease (Fig. 1I). The macaque with transient vRNA blips (RQs-8) and RPs-8 were also protected from disease over a follow-up of nearly 4 years (Fig. 1I), whereas RYr-8 had severely depleted CD4+ T cells from week 20 following viral exposure onward (Fig. 1I) and was euthanized at week 69. Cocultivation experiments of peripheral blood mononuclear cells (PBMC) of all four monkeys that were protected from persistent systemic infection, RAq-8, RQp-8, RWq-8, and RQs-8, were virus-negative (not shown). PBMC DNA isolated from the same animals was also negative by PCR for viral sequences. Virus status of lymph nodes of protected animals To determine whether treated animals that displayed undetectable plasma viremia throughout the study might be nonetheless latently infected, we tested for the presence of viral sequences in lymph nodes (LNs). Inguinal LN biopsies were obtained from SHIV-seronegative animals RQp-8, RWq-8, and RQs-8 (it was not possible to obtain LNs from RAq-8 because the animal was no longer in the study at the time biopsies were performed). LN cells from these animals were negative for both viral DNA and RNA sequences, except for RWq-8, which resulted in low positivity for vRNA but not DNA (not shown). This result is compatible with the fact that RWq-8 was exposed to the virus and had occasional blips of low-level plasma vRNA (Fig. 1E). Intramuscular PEP with three anti-HIV Env nmAbs against SHIV89.6P oral challenge in neonatal macaques In a separate study, eight macaque newborns were also exposed orally to SHIV89.6P. Four control animals were left untreated (Fig. 2A), whereas four infants were treated intramuscularly (I.M.) with three nmAbs (2F5, 2G12, and 4E10) 24 h post-exposure and, again, 8 days later (Fig. 2B). All four controls displayed high levels of plasma vRNA, dropped their CD4 counts irreversibly within 2 weeks after challenge, and died of AIDS (Figs. 2C, E). All treated animals became viremic, but with delayed and lower peaks of viremia than the controls. None of the infants given immunoprophylaxis showed acute disease. In fact, differences between CD4+ T-cell counts in control versus treated animals were significant (pweek 5 = 0.03). Three of the treated animals remained healthy throughout a follow-up of 24 weeks (Fig. 2F), whereas the nmAb-treated infant that displayed the highest viremia set-point (ROp-9) presented with a slow decline in CD4+ T-cell counts starting a few weeks postchallenge (Figs. 2D, F). Humoral immune responses Control animals remained seronegative as assessed by Western blot assay (Fig. 3A) and/or ELISA (data not shown), a result that can be explained by early progression to AIDS,

F. Ferrantelli et al. / Virology 358 (2007) 69–78 71

Fig. 1. (A–C) Experimental design of the first neonatal immunoprophylaxis study (1 h or 12 h post-exposure immunization with 4 nmAbs). (D–I) Plasma viremia and T-helper cell counts. (A) Four newborn macaques served as untreated controls and were given only oral virus. (B) Seven newborn macaques were challenged orally with 15 AID50 of SHIV89.6P (yielding 99% probability of infection) within 5 days of birth (open arrows) and infused twice I.V. with human nmAbs IgG1b12, 2G12, 2F5, and 4E10 (30 mg/kg each—solid black arrows). The first mAb infusion started either 1 h or 12 h post-virus exposure; identical treatment was administered on day 8 or 8.5 post-challenge, respectively. Animals were observed for at least 47 months. Plasma vRNA load (D–F) and peripheral absolute CD4+ T-cell counts (G–I) of neonates challenged orally with SHIV89.6P. The sensitivity of the RT-PCR assay was 50 copies/ml (D–F), dotted line. †All control animals were sacrificed within 7 weeks post-exposure because of AIDS-related conditions; nmAb-treated infant RAq-8 was sacrificed at week 7 for causes non-related to the virus challenge, whereas nmAb-treated macaque RYr-8 was euthanized at week 69 because of AIDS-associated clinical conditions.

72

F. Ferrantelli et al. / Virology 358 (2007) 69–78

Fig. 2. (A–B) Experimental design of the second neonatal immunoprophylaxis study (24 h post-exposure prophylaxis (PEP) with 3 nmAbs). (C–F) Plasma viremia and T-helper cell counts. (A) Four newborn macaques served as untreated controls and were challenged orally with 15 AID50 of SHIV89.6P (yielding 99% probability of infection) within 5 days of birth (open arrows). (B) Four newborn macaques were also challenged orally with 15 AID50 of SHIV89.6P (open arrows) and then treated twice I.M. with human nmAbs 2G12, 2F5, and 4E10 (40 mg/kg each—solid black arrows). The first mAb administration started 24 h post-virus exposure; identical treatment was administered on day 9 post-challenge. Animals were observed for at least 8 months. Plasma vRNA load (C–D) and peripheral absolute CD4+ T-cell counts (E–F) of neonates challenged orally with SHIV89.6P. The sensitivity of the RT-PCR assay was 50 copies/ml (C and D, dotted line). †All control animals were sacrificed within 14 weeks post-exposure.

which did not permit these animals to mount virus-specific humoral immune responses. In both studies, only nmAb-treated animals that were overtly infected (sustained vRNA levels > 104 copies/ml) developed anti-SIV and/or HIV antibodies, whereas monkeys with negative plasma and cellular viremia did not seroconvert (Fig. 3). As such, animal RWq-8, which had lowlevel vRNA blips at some time points, was negative by Western blot analysis at 58 weeks post-inoculation (Fig. 3A). The fact that monkey ROp-9, given PEP at 24 h, had no anti-SIV Gag reactivity at week 54 (Fig. 3A) is due to its severe depletion of CD4 T cells, starting from week 28 onward. At this time point, no anti-SIV Gag reactivity was seen (Fig. 3B), although the

animal had mounted anti-HIV-1 Env antibody responses (Fig. 3C). This observation is consistent with the notion that selective loss of anti-Gag antibody responses is a poor prognostic sign that heralds immunodeficiency (Binley et al., 1997). Serum or plasma nAb titers were also assessed in vitro against SHIV89.6P. Treated animals of both studies displayed early (weeks 1–5) neutralizing activity as a direct consequence of the passive administration of human nmAbs, whereas only animals that were not protected from infection mounted virus-specific neutralizing antibody (nAb) responses, which were detectable at later time points (Table 1). The five control animals that could be tested for the presence of nAb showed borderline- or low-

F. Ferrantelli et al. / Virology 358 (2007) 69–78

73

Fig. 3. Plasma anti-SIV (A and B) or anti-HIV (C) antibody assessment by Western blot for the 1 h or 12 h PEP and 24 h PEP studies. In addition to the controls provided by the manufacturer, pre- and post-challenge plasma from animals infected with a SHIV clade C were use as negative and positive controls of the assay, respectively.

positive titers at one or two time points post-challenge. Baseline low titers found in several animals at time zero were likely due to non-specific antiviral activity of individual serum/plasma samples. Cellular immune responses Cellular immune responses of treated animals in the 1 h vs. 12 h PEP study were assessed by ELISPOT assay between weeks 29 and 33 and by proliferation assays at different time points after virus challenge. PBMC of monkeys that had undetectable plasma viremia at all times (RQp-8) or low-level vRNA blips (RWq-8 and RQs-8) did not respond to in vitro stimulation with Gag, Pol, or Nef peptides as assessed by ELISPOT (not shown); these three animals had also been consistently negative for proviral

DNA sequences in their PBMC DNA within the limits of our assay sensitivity. In contrast, infected, nmAb-treated macaques RYq-8 and RPs-8 showed positive ELISPOTS to at least two viral antigens (not shown). Conversely, animal RYr-8, which was also virus-positive, did not display positive ELISPOTS, consistent with its concomitant immunodeficiency at the time of testing (ELISPOT analysis performed at week 31; week 32, CD4+ T-cell counts: 17 cells/μl). All nmAb-treated animals, except for RAq-8 (which was no longer in the study after the first 7 weeks) and RPs-8, exhibited low-positive, yet significant, viralantigen-specific proliferative responses to either Gag, Nef, or Env proteins at least once or twice between weeks 12 and 40 post-challenge (SI ≥ 4) (Table 2). This finding can be explained by considering that even nmAb-treated animals that were protected from persistent systemic infection had

74

F. Ferrantelli et al. / Virology 358 (2007) 69–78

Table 1 Serum or plasma neutralizing antibody titers to SHIV89.6P a Group

Animal # b

Outcome of virus challenge

Weeks post-exposure 0

Control 1 h PEP 1 h PEP 1 h PEP 1 h PEP 12 h PEP 12 h PEP 12 h PEP Control Control Control Control PEP 24 h PEP 24 h PEP 24 h PEP 24 h

RPo-8 RAq-8 RQp-8 e RWq-8 RYq-8 RPs-8 RQs-8 RYr-8 h RMp-9 RPp-9 RWn-9 RZn-9 RHu-9 RIt-9 RLp-9 ROp-9

AIDS, death Protected from infection Protected from infection Transient vRNA blips; protected from disease Protected from disease Protected from disease Transient vRNA blips; protected from disease Delayed disease AIDS, death AIDS, death AIDS, death AIDS, death Protected from disease Protected from disease Protected from disease Protected from acute disease

1

<20 <20 nt <20 <20 <20 <20 <20 41 <20 <20 <20 <20 20 <20 25

51 62 118 nt 228 64 nt 212 nt nt nt nt 105 21 139 104

2 <20 213 224 147 184 194 310 215 46 28 22 <20 871 352 340 510

5 <20 30 <20 43 85 36 101 53 <20 nt 22 34 59 236 50 176

8

12 c

<20 <20 c <20 nt nt 50 <20 938 <20 45 <20 j 74 60 827 6337 1438

d

nt nt <20 nt 1792 193 <20 >540 <20 i <20 < 20 <20 k 41 1695 22,706 20,361

20

40

nt nt <20 nt nt nt nt 3168 nt nt nt nt 2527 797 24,347 22,946

nt nt <20 <20 f 315 77 g <20 g 455 nt nt nt nt nt nt nt nt

a nAb titers are the reciprocal serum or plasma dilution at which 50% of cells were protected from virus-induced killing as measured by neutral red uptake. This corresponds to a 90% reduction of the viral Gag synthesis (Bures et al., 2000). b No data available for control animals REq-8, RFq-8, or RSl-8 because of the lack of samples. c Sample collected at week 7. d nt = not tested. e RQp-8 also had titers <20 at weeks 16 and 24. f Sample collected at week 45. g Sample collected at week 41. h RYr-8 also had a titer of 1515 at week 16. i Sample collected at week 14. j Sample collected at week 7. k Sample collected at week 11.

been exposed to live virus and might have therefore mounted some antiviral cellular immune responses. Discussion In the first study reported here, two groups of neonatal macaques were challenged orally with SHIV89.6P and treated

I.V. with a quadruple combination of nmAbs as either 1 h or 12 h PEP, whereas the control group was exposed to oral virus but left untreated. All four controls became highly viremic and had to be euthanized within 7 weeks because of AIDS. In the 1 h PEP group, three out of four newborns were protected from persistent systemic infection against SHIV89.6P challenge, and the fourth monkey was protected from disease. When given

Table 2 I.V. 1 h vs. 12 h PEP: T-cell proliferative responses after SHIV89.6P challenge a Group (PEP at)

Animal #

Weeks 12

Weeks 28–33

Weeks 40

Gag

Nef

Env

Con A

Gag

Nef

Env

Con A

Gag

Nef

Env

Con A

1 h PEP

RQp-8 b

2c

3c

7c

67 c

28

nt nt nt 16 nt

nt nt nt 3 nt

nt nt nt 24 nt

24 58 64 98 34 169 18

≤1

nt e nt nt ≤1 nt

≤1 3 2 4 2 2 9

≤1

RWq-8 d RYq-8 g RPs-8 h RQs-8 i RYr-8

2 2 2 ≤1 ≤1 ≤1 ≤1

≤1

1 h PEP 1 h PEP 12 h PEP 12 h PEP 12 h PEP

≤1 2 8 3 ≤1 ≤1 ≤1

≤1 f ≤1 f nt nt ≤1

≤1 f ≤1 f nt nt 3

≤1 f ≤1 f nt nt 5

22 f 7f nt nt 7

a Proliferation stimulation indices for PBMC in the presence of soluble proteins ≥4 assayed at various time points after SHIV89.6P challenge. Stimulation indices ≥4 are in bold and underlined and indicate positive responses. b RQp-8 was tested also at weeks 36 and 44: no antigen-specific responses were detected. c Sample from week 16. d RWq-8 was tested also at week 37: no antigen-specific responses were detected. e nt = not tested. f Sample from week 41. g RYq-8 was tested also at week 37: no antigen-specific responses were detected. h RPs-8 was tested also at weeks 26 and 37: no antigen-specific responses were detected. i RQs-8 was tested also at weeks 26 and 37: no antigen-specific responses were detected.

F. Ferrantelli et al. / Virology 358 (2007) 69–78

12 h PEP, persistent systemic infection was prevented in one out of three neonates, and disease was either prevented or delayed in the two other infant monkeys. This study provided the proof of concept that 12 h PEP with nmAbs given I.V. can be protective against oral challenge with the highly pathogenic SHIV. These results were further extended by another study in macaque newborns (Ferrantelli et al., 2004b), in which a triple combination of anti-HIV nmAbs (2G12, 2F5, and 4E10) given I.M. as 1 h PEP protected all treated infants from oral challenge with SHIV89.6P. Taken together, these and the previous data led to the second study reported here, in which the same triple combination of nmAbs was administered I.M. 24 h post-virus exposure. All four treated neonatal macaques were protected from acute SHIV89.6P-caused disease; three of the infants have been healthy throughout a follow-up of 8 months, whereas the fourth animal had a decline in CD4 counts by week 20 after virus exposure. All four control infants were infected and had severe losses of CD4+ T cells within 2 weeks. The present work delineates the potential and the limits of passive immunoprophylaxis with human nmAbs. Taken together with our results from other PEP studies (Ferrantelli et al., 2003, 2004b), our data show that the success of passive immunization depends on the nmAb potency/dose and the time window between virus exposure and the start of immunoprophylaxis. Given that PEP with human nmAbs completely prevented systemic infection in some of the experimental animals, potential protective mechanisms need to be considered. Clearly, the nmAbs' direct neutralizing ability blocks virus entry, but this mechanism alone does not explain why virus did not reappear in the completely protected monkeys after all passively administered human nmAbs were cleared. It is likely that, during the initial time period when no nmAbs were present, virions penetrated the mucosa of the oral cavity or gastrointestinal (GI) tract. After nmAb treatment was administered, cell-free virus was then covered by human antibodies and eliminated through phagocytosis of immune complexes by cells bearing either complement or Fc-receptors, including antigen presenting cells (reviewed by Mc Cann et al., 2005). In addition, we believe it quite likely that a few target cells were productively infected during the time window between virus challenge and nmAb treatment. Such cells would then be destroyed via different, though not mutually exclusive mechanisms, including antibody-dependent cell-mediated cytotoxicity (ADCC) or complement activation. Consistent with this hypothesis, three of the human nmAbs, which share the same IgG1 backbone, have been found previously to activate complement and to possess ADCC activity in vitro (Hezareh et al., 2001; Trkola et al., 1996; Wolbank et al., 2003). Lastly, it is also possible that productive infection of early target cells occurring before nmAb treatment led to presentation of viral peptides on MHC class I molecules. This in turn would generate MHC class I-restricted cytotoxic lymphocyte (CTL) responses that could work in concert with ADCC to eliminate the virusproducing early target cells. We did not have sufficient numbers of cells to conduct studies to assess the generation of MHC class

75

I-restricted cellular immune responses during the first month of life, when such responses were most likely detectable. When we tested for ELISPOT responses later on, the virus-exposed, uninfected animals were negative. However, we have detected positive, but low-level, virus-specific proliferative responses in some of the protected infants. It will be important to assess the longest time interval postvirus exposure at which nmAb prophylaxis can still provide complete protection. Such a time window might reasonably be between 6 and 12 h post-exposure. In fact, in our 12 h PEP study, I.V. administration of a quadruple nmAb combination protected one of three newborns against persistent infection after oral exposure to pathogenic SHIV89.6P. This result suggests that both the time window post-exposure and treatment potency in that study arm were, overall, borderline protective against infection. It is possible that the triple combination of 2G12, 2F5, and 4E10, which was given at the higher dose of 40 mg/kg and which protected 4 out of 4 treated animals either from infection when administered I.M. as 1 h PEP (Ferrantelli et al., 2004b) or from disease when given as 24 h PEP, might still provide sterilizing immunity when given within 6–12 h postexposure. In line with this hypothesis, a recent study showed that antiviral neutralizing polyclonal IgG which did not protect 2 out of 2 pig-tailed macaques from infection when administered 24 h after I.V. SHIV challenge indeed conferred sterilizing immunity to 3 of 4 macaques when given 6 h post-virus exposure (Nishimura et al., 2003). It should be noted that, even though 24 h PEP with human nmAbs was not sterilizing in our preclinical SHIV/primate model, the same 24 h PEP regimen might still be completely protective in a clinical setting. In fact, virus inocula associated with natural exposure are much lower than those given to monkeys, which correspond to 99% probability of infection (Spouge, 1992). Also of note, the nmAbs used in our PEP studies in neonatal monkeys have shown potent cross-clade and cross-group neutralizing activity in vitro against primary HIV isolates. Several independent studies attested that such nmAbs, when used in various combinations or even as single neutralizing agents, neutralized most primary isolates tested and belonging to HIV clade A, B, C, D, E, and F (Binley et al., 2004; Kitabwalla et al., 2003; Stiegler et al., 2001; Trkola et al., 1995; Verrier et al., 2001; Xu et al., 2001; Zwick et al., 2001a, 2001b), or group O (Ferrantelli et al., 2004a). Moreover, among the nmAbs that were protective in our neonatal preclinical studies, 2G12, 2F5, and 4E10 have already been proven safe and shown to have relatively long half-lives in phase I clinical trials involving HIV-infected human adults (Armbruster et al., 2002, 2004; Cavacini et al., 1998; Wolfe et al., 1996). The combination of nmAbs 2G12 and 2F5 also had antiviral activity in asymptomatic HIV-infected individuals, as evaluated in a phase I study (Stiegler et al., 2002). In a more recent study (Trkola et al., 2005), the triple combination of 2G12, 2F5, and 4E10 was evaluated in adult individuals with either acute or chronic, stable HIV-1 infection that was completely suppressed by highly active antiviral therapy (HAART). These individuals received

76

F. Ferrantelli et al. / Virology 358 (2007) 69–78

passive immunization after HAART interruption, and the ability of the human nmAb combination to prevent or delay virus rebound was assessed. While the nmAb treatment was well tolerated in all individuals, efficacy was more pronounced in the acutely infected cohort. Analysis of viral neutralization escape mutants revealed that only nmAb 2G12 appeared to exert selective pressure. Detailed pharmacokinetic analysis demonstrated that this nmAb had a significantly longer half-life and higher peak and trough levels than the two anti-gp41 nmAbs, 2F5 and 4E10. More recently, these same two nmAbs were reported to crossreact with cardiolipin (Haynes et al., 2005). It is possible that this autoreactivity may account for the more rapid elimination of 2F5 and 4E10 compared to 2G12, as all three of these nmAbs share the same IgG1 backbone. The fourth nmAb we used in our neonatal primate study, IgG1b12, was found to recognize ribonucleoprotein, double-stranded DNA, and other autoantigens (Haynes et al., 2005). Our neonatal primate model was originally designed to allow evaluation of strategies aimed at preventing mother-tochild transmission of HIV-1 (MTCT), which can occur in utero, during birth, or via breast-feeding. The problem of milkborne HIV-1 transmission is most prevalent in sub-Saharan Africa, where many women continue to breast-feed either because of lack of access to clean water, social pressure, and/or fear of stigmatization. In these areas, even when perinatal antiviral drug treatments are available, the treatment benefits might be overcome by the practice of breast-feeding (Petra Study Team, 2002). The latter is both difficult to avoid because of economic and cultural reasons, and necessary, as a source of maternal immunoglobulins (Ig) to guarantee immune protection against several pathogens to children (reviewed in Brandtzaeg, 2003). Since intrapartum MTCT is thought to occur predominantly via mucosal exposure of the fetus, we feel that our oral virus challenge model in neonatal monkeys mimics important aspects of intrapartum MTCT, in addition to serving as a model for milk-borne virus transmission. We have used this model to develop passive immunization with combinations of broadly reactive human nmAbs to prevent MTCT via breast feeding. At the same time, passive immunization given soon after delivery may serve as PEP after intrapartum virus exposure. If successful as a true prophylactic regimen and if a sufficiently large time window between virus exposure and onset of nmAb therapy can be demonstrated, passive immunization with broadly reactive human nmAbs may be clinically beneficial as PEP after intrapartum HIV-1 exposure and as prevention of milk-borne HIV-1 transmission. Given the sobering data by Haynes et al. (2005) that the majority of the human nmAbs we used in our neonatal primate studies exhibit anti-self reactivity, we believe that such autoreactivity may potentially complicate the use of the human nmAbs in passive immunization. In the future, it will be important to perform more extensive safety studies with the currently available broadly reactive human nmAbs, especially with 4E10, which has been reported to possess lupus anticoagulant activity (Haynes et al., 2005). Equally important

will be the identification of new, potent nmAbs with broad antiviral reactivity but hopefully lacking autoreactivity. This approach could lead to the identification of novel, conserved epitopes that could have a significant impact on the development of nAb-responses-based AIDS vaccines. Overall, passive immunization with human nmAbs in the SHIV/ nonhuman primate models is a very useful tool to assess the importance of Abs against HIV infection. Materials and methods Animals, SHIV89.6P challenge and human nmAb administration Rhesus macaques were housed at the Yerkes National Primate Research Center, Emory University, a facility fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. All procedures were approved by the Animal Care and Use Committees of Emory University and the Dana-Farber Cancer Institute. We are committed to comply with the Principles for Use of Animals, the Guide for the Care and Use of Laboratory Animals, the Provisions of the Animal Welfare Act, and other applicable laws and regulations. The SHIV89.6P stock was prepared and titrated as described previously (Hofmann-Lehmann et al., 2001). Within 5 days of birth, newborn macaques were challenged orally with fifteen 50% animal infectious doses (AID50) as reported previously (Ferrantelli et al., 2003). Either triple or quadruple combinations of human IgG1 nmAbs 2G12, 2F5, 4E10, and IgG1b12 were used in these studies. 2G12 recognizes a conformational epitope consisting of mannose residues on the gp120 subunit of HIV Env (Scanlan et al., 2002); 2F5 and 4E10 bind the conserved HIV gp41 sequences ELDKWA (Muster et al., 1993; Parker et al., 2001) and NWFDIT (Stiegler et al., 2001; Zwick et al., 2001a), respectively. 2G12, 2F5, and 4E10 were provided by Polymun Scientific, Vienna, Austria. IgG1b12 targets the CD4 binding site of gp120 (Roben et al., 1994) and was produced as described (Hofmann-Lehmann et al., 2001). nmAbs were of clinical grade purity and endotoxin-free. The first part of the present study included 11 neonatal rhesus monkeys, all of which were exposed orally to the pathogenic SHIV89.6P. One control group of 4 animals was left untreated, whereas a group of 4 newborn macaques was infused I.V. with human nmAbs IgG1b12, 2G12, 2F5, and 4E10 (30 mg/kg of body weight for each antibody) 1 h post-inoculation. Another treatment group of 3 neonates received the four nmAbs starting at 12 h post-challenge. The identical nmAb regimen was given to all 7 treated infants on day 8 post-inoculation. nmAbs were administered I.V. through a 0.2 μm syringe filter previously flushed with physiological solution. In the second experiment, 8 newborn macaques were challenged orally with SHIV89.P. Four animals were left untreated as controls, whereas the other 4 neonates were treated twice I.M. with the three human nmAbs 2G12, 2F5, and 4E10 at a dose of 40 mg/kg each, 24 h and 9 days post-challenge.

F. Ferrantelli et al. / Virology 358 (2007) 69–78

Cocultivation of PBMC, measurement of plasma vRNA and cellular viral DNA levels PBMC were co-cultured with CEM×174/GFP cells as described previously (Rasmussen et al., 2002). Plasma vRNA loads were assessed by quantitative RT-PCR for SIV gag sequences (QiaAmp RNA Blood Mini-Kit—Qiagen) (Hofmann-Lehmann et al., 2000), and the presence of viral DNA was similarly evaluated in PBMC by PCR for SIV gag sequences. Lymph node cell sample preparation and vRNA/DNA detection Lymph node cells were isolated from 1 to 2 inguinal lymph nodes per animal by enzymatic digestion to ensure the preservation of follicular dendritic cells amid other cell types, as reported (Ferrantelli et al., 2004b). Cell samples were therefore assayed by RT-PCR (QiaAmp RNA Blood Mini-Kit—Qiagen) for SIV-RNA gag sequences or SIV gag DNA (sensitivity, 1 copy of viral DNA per μg of genomic DNA) (HofmannLehmann et al., 2000). Serological assays Triplicate plasma samples were tested for the presence of anti-SIV or anti-HIV antibodies by Western blot assay (ZeptoMetrix) according to the manufacturer's instructions and/or by ELISA using recombinant SIV p27 (Baba et al., 2000). Serum or plasma nAb titers against a SHIV89.6P produced in human PBMC were measured by an MT-2 cellbased killing assay (Bures et al., 2000; Crawford et al., 1999). Cellular immunity PBMC were tested in triplicate wells for antigen-specific proliferative responses against SIVmac239 Gag, Nef (AIDS Research Reference Reagent Program, NIAID, NIH), or HIV89.6 Env (kindly provided by P. Earl, NIAID, NIH) as described (Rasmussen et al., 2002); stimulation indices (SI) were calculated as reported previously (Ferrantelli et al., 2004b) and considered positive when ≥ 4. ELISPOT assays were also performed as described (Larsson et al., 1999). Acknowledgments We thank Dr. Dennis Burton (Scripps Research Institute, La Jolla, CA, USA) for the gift of CHO cells producing mAb IgG1b12 and Susan E. Sharp for editorial assistance. Sponsorship: This work was supported in part by National Institutes of Health grants RO1 AI34266 and RO1 DE12937 to R.M.R.; PO1 AI48240 to R.M.R., R.A.R., and R.H.-L.; AI30034 to D.C.M.; Center for AIDS Research Core grant IP30 28691 awarded to the Dana-Farber Cancer Institute as support for the Institute's AIDS research efforts and RR00165, providing base grant support to the Yerkes National Primate Research Center. R.H.-L. is the recipient of a professorship by the Swiss National Science Foundation.

77

References Armbruster, C., Stiegler, G.M., Vcelar, B.A, Jager, W., Michael, N.L., Vetter, N., Katinger, H.W., 2002. A phase I trial with two human monoclonal antibodies (hMAb2F5, 2G12) against HIV-1. AIDS 16, 227–233. Armbruster, C., Stiegler, G.M., Vcelar, B.A., Jager, W., Koller, U., Jilch, R., Ammann, C.G., Pruenster, M., Stoiber, H., Katinger, H.W., 2004. Passive immunization with the anti-HIV-1 human monoclonal antibody (hMAb) 4E10 and the hMAb combination 4E10/2F5/2G12. J. Antimicrob. Chemother. 54, 915–920. Baba, T.W., Liska, V., Hofmann-Lehmann, R., Vlasak, J., Xu, W., Ayehunie, S., Cavacini, L.A, Posner, M.R., Katinger, H., Stiegler, G., Bernacky, B.J., Rizvi, T.A., Schmidt, R., Hill, L.R., Keeling, M.E., Lu, Y., Wright, J.E., Chou, T.C., Ruprecht, R.M., 2000. Human neutralizing monoclonal antibodies of the IgG1 subtype protect against mucosal simian–human immunodeficiency virus infection. Nat. Med. 6, 200–206. Binley, J.M., Klasse, P.J., Cao, Y., Jones, I., Markowitz, M., Ho, D.D., Moore, J. P., 1997. Differential regulation of the antibody responses to Gag and Env proteins of human immunodeficiency virus type 1. J. Virol. 71, 2799–2809. Binley, J.M., Wrin, T., Korber, B., Zwick, M.B., Wang, M., Chappey, C., Stiegler, G., Kunert, R., Zolla-Pazner, S., Katinger, H., Petropoulos, C.J., Burton, D.R., 2004. Comprehensive cross-clade neutralization analysis of a panel of anti-human immunodeficiency virus type 1 monoclonal antibodies. J. Virol. 78, 13232–13252. Brandtzaeg, P., 2003. Mucosal immunity: integration between mother and the breast-fed infant. Vaccine 21, 3382–3388. Bures, R., Gaitan, A., Zhu, T., Graziosi, C., McGrath, K.M., Tartaglia, J., Caudrelier, P., El Habib, R., Klein, M., Lazzarin, A., Stablein, D.M., Deers, M., Corey, L., Greenberg, M.L., Schwartz, D.H., Montefiori, D.C., 2000. Immunization with recombinant canarypox vectors expressing membrane-anchored glycoprotein 120 followed by glycoprotein 160 boosting fails to generate antibodies that neutralize R5 primary isolates of human immunodeficiency virus type 1. AIDS Res. Hum. Retroviruses 16, 2019–2035. Cavacini, L.A., Samore, M.H., Gambertoglio, J., Jackson, B., Duval, M., Wisnewski, A., Hammer, S., Koziel, C., Trapnell, C., Posner, M.R., 1998. Phase I study of a human monoclonal antibody directed against the CD4binding site of HIV type 1 glycoprotein 120. AIDS Res. Hum. Retroviruses 14, 545–550. Crawford, J.M., Earl, P.L., Moss, B., Reimann, K.A., Wyand, M.S., Manson, K.H., Bilska, M., Zhou, J.T., Pauza, C.D., Parren, P.W., Burton, D.R., Sodroski, J.G., Letvin, N.L., Montefiori, D.C., 1999. Characterization of primary isolate-like variants of simian–human immunodeficiency virus. J. Virol. 73, 10199–10207. Ferrantelli, F., Ruprecht, R.M., 2002. Neutralizing antibodies against HIV— Back in the major leagues again. Curr. Opin. Immunol. 14, 495–502. Ferrantelli, F., Hofmann-Lehmann, R., Rasmussen, R.A., Wang, T., Xu, W., Li, P.L., Montefiori, D.C., Cavacini, L.A., Katinger, H., Stiegler, G., Anderson, D.C., McClure, H.M., Ruprecht, R.M., 2003. Post-exposure prophylaxis with human monoclonal antibodies prevented SHIV89.6P infection or disease in neonatal macaques. AIDS 17, 301–309. Ferrantelli, F., Kitabwalla, M., Rasmussen, R.A., Cao, C., Chou, T.C., Katinger, H., Stiegler, G., Cavacini, L.A., Bai, Y., Cotropia, J., Ugen, K.E., Ruprecht, R.M., 2004a. Potent cross-group neutralization of primary human immunodeficiency virus isolates with monoclonal antibodies-implications for acquired immunodeficiency syndrome vaccine. J. Infect. Dis. 189, 71–74. Ferrantelli, F., Rasmussen, R.A., Buckley, K.A., Li, P.L., Wang, T., Montefiori, D.C., Katinger, H., Stiegler, G., Anderson, D.C., McClure, H.M., Ruprecht, R.M., 2004b. Complete protection of neonatal rhesus macaques against oral exposure to pathogenic simian–human immunodeficiency virus by human anti-HIV monoclonal antibodies. J. Infect. Dis. 189, 2167–2173. Haynes, B.F., Fleming, J., St. Clair, E.W., Katinger, H., Stiegler, G., Kunert, G., Robinson, J., Scearce, R.M., Plonk, K., Staats, H.F., Ortel, T.L., Liao, H.X., Alam, S.M., 2005. Cardiolipin polyspecific autoreactivity in two broadly neutralizing HIV-1 antibodies. Science 308, 1906–1908. Hezareh, M., Hessell, A.J., Jensen, R.C., van de Winkel, J.G., Parren, P.W., 2001. Effector function activities of a panel of mutants of a broadly

78

F. Ferrantelli et al. / Virology 358 (2007) 69–78

neutralizing antibody against human immunodeficiency virus type 1. J. Virol. 75, 12161–12168. Hofmann-Lehmann, R., Swenerton, R.K., Liska, V., Leutenegger, C.M., Lutz, H., McClure, H.M., Ruprecht, R.M., 2000. Sensitive and robust one-tube real-time reverse transcriptase-polymerase chain reaction to quantify SIV RNA load: comparison of one- versus two-enzyme systems. AIDS Res. Hum. Retroviruses 16, 1247–1257. Hofmann-Lehmann, R., Vlasak, J., Rasmussen, R.A., Smith, B.A., Baba, T.W., Liska, V., Ferrantelli, F., Montefiori, D.C., McClure, H.M., Anderson, D.C., Bernacky, B.J., Rizvi, T.A., Schmidt, R., Hill, L.R., Keeling, M.E., Katinger, H., Stiegler, G., Cavacini, L.A., Posner, M.R., Chou, T.C., Andersen, J., Ruprecht, R.M., 2001. Postnatal passive immunization of neonatal macaques with a triple combination of human monoclonal antibodies against oral simian–human immunodeficiency virus challenge. J. Virol. 75, 7470–7480. Hofmann-Lehmann, R., Vlasak, J., Rasmussen, R.A., Jiang, S., Li, P.L., Baba, T.W., Montefiori, D.C., Bernacky, B.J., Rizvi, T.A., Schmidt, R., Hill, L.R., Keeling, M.E., Katinger, H., Stiegler, G., Cavacini, L.A., Posner, M.R., Ruprecht, R.M., 2002. Postnatal pre- and postexposure passive immunization strategies: protection of neonatal macaques against oral simian–human immunodeficiency virus challenge. J. Med. Primatol. 31, 109–119. Kitabwalla, M., Ferrantelli, F., Wang, T., Chalmers, A., Katinger, H., Stiegler, G., Cavacini, L.A., Chou, T.C., Ruprecht, R.M., 2003. Primary African HIV clade A and D isolates: effective cross-clade neutralization with a quadruple combination of human monoclonal antibodies raised against clade B. AIDS Res. Hum. Retroviruses 19, 125–131. Larsson, M., Jin, X., Ramratnam, B., Ogg, G.S., Engelmayer, J., Demoitie, M.A., McMichael, A.J., Cox, W.I., Steinman, R.M., Nixon, D., Bhardwaj, N., 1999. A recombinant vaccinia virus based ELISPOT assay detects high frequencies of Pol-specific CD8 T cells in HIV-1-positive individuals. AIDS 13, 767–777. Mascola, J.R., Lewis, M.G., Stiegler, G., Harris, D., VanCott, T.C., Hayes, D., Louder, M.K., Brown, C.R., Sapan, C.V., Frankel, S.S., Lu, Y., Robb, M.L., Katinger, H., Birx, D.L., 1999. Protection of macaques against pathogenic simian/human immunodeficiency virus 89.6PD by passive transfer of neutralizing antibodies. J. Virol. 73, 4009–4018. Mascola, J.R., Stiegler, G., VanCott, T.C., Katinger, H., Carpenter, C.B., Hanson, C.E., Beary, H., Hayes, D., Frankel, S.S., Birx, D.L., Lewis, M.G., 2000. Protection of macaques against vaginal transmission of a pathogenic HIV-1/SIV chimeric virus by passive infusion of neutralizing antibodies. Nat. Med. 6, 207–210. Mc Cann, C.M., Song, R., Ruprecht, R.M., 2005. Antibodies: can they protect against HIV infection? Curr. Drug Targets Infect. Disord. 5, 95–111. Muster, T., Steindl, F., Purtscher, M., Trkola, A., Klima, A., Himmler, G., Ruker, F., Katinger, H., 1993. A conserved neutralizing epitope on gp41 of human immunodeficiency virus type 1. J. Virol. 67, 6642–6647. Nishimura, Y., Igarashi, T., Haigwood, N.L., Sadjadpour, R., Donau, O.K., Buckler, C., Plishka, R.J., Buckler-White, A., Martin, M.A., 2003. Transfer of neutralizing IgG to macaques 6 h but not 24 h after SHIV infection confers sterilizing protection: implications for HIV-1 vaccine development. Proc. Natl. Acad. Sci. U. S. A. 100, 15131–15136. Parker, C.E., Deterding, L.J., Hager-Braun, C., Binley, J.M., Schulke, N., Katinger, H., Moore, J.P., Tomer, K.B., 2001. Fine definition of the epitope on the gp41 glycoprotein of human immunodeficiency virus type 1 for the neutralizing monoclonal antibody 2F5. J. Virol. 75, 10906–10911. Parren, P.W., Marx, P.A., Hessell, A.J., Luckay, A., Harouse, J., Cheng-Mayer, C., Moore, J.P., Burton, D.R., 2001. Antibody protects macaques against vaginal challenge with a pathogenic R5 simian/human immunodeficiency virus at serum levels giving complete neutralization in vitro. J. Virol. 75, 8340–8347. Petra Study Team, 2002. Efficacy of three short-course regimens of zidovudine and lamivudine in preventing early and late transmission of HIV-1 from mother to child in Tanzania, South Africa, and Uganda (Petra study): a randomised, double-blind, placebo-controlled trial. Lancet 359, 1178–1186. Rasmussen, R.A., Hofmann-Lehmann, R., Montefiori, D.C., Li, P.L., Liska, V., Vlasak, J., Baba, T.W., Schmitz, J.E., Kuroda, M.J., Robinson, H.L., McClure, H.M., Lu, S., Hu, S.L., Rizvi, T.A., Ruprecht, R.M., 2002. DNA

prime/protein boost vaccine strategy in neonatal macaques against simian human immunodeficiency virus. J. Med. Primatol. 31, 40–60. Roben, P., Moore, J.P., Thali, M., Sodroski, J., Barbas III, C.F., Burton, D.R., 1994. Recognition properties of a panel of human recombinant Fab fragments to the CD4 binding site of gp120 that show differing abilities to neutralize human immunodeficiency virus type 1. J. Virol. 68, 4821–4828. Scanlan, C.N., Pantophlet, R., Wormald, M.R., Ollmann Saphire, E., Stanfield, R., Wilson, I.A., Katinger, H., Dwek, R.A., Rudd, P.M., Burton, D.R., 2002. The broadly neutralizing anti-human immunodeficiency virus type 1 antibody 2G12 recognizes a cluster of alpha1 → 2 mannose residues on the outer face of gp120. J. Virol. 76, 7306–7321. Spouge, J.L., 1992. Statistical analysis of sparse infection data and its implications for retroviral treatment trials in primates. Proc. Natl. Acad. Sci. U. S. A. 89, 7581–7585. Stiegler, G., Kunert, R., Purtscher, M., Wolbank, S., Voglauer, R., Steindl, F., Katinger, H., 2001. A potent cross-clade neutralizing human monoclonal antibody against a novel epitope on gp41 of human immunodeficiency virus type 1. AIDS Res. Hum. Retroviruses 17, 1757–1765. Stiegler, G., Armbruster, C., Vcelar, B., Stoiber, H., Kunert, R., Michael, N.L., Jagodzinski, L.L., Ammann, C., Jager, W., Jacobson, J., Vetter, N., Katinger, H., 2002. Antiviral activity of the neutralizing antibodies 2F5 and 2G12 in asymptomatic HIV-1-infected humans: a phase I evaluation. AIDS 16, 2019–2025. Trkola, A., Pomales, A.B., Yuan, H., Korber, B., Maddon, P.J., Allaway, G.P., Katinger, H., Barbas III, C.F., Burton, D.R., Ho, D.D., Moore, J.P., 1995. Cross-clade neutralization of primary isolates of human immunodeficiency virus type 1 by human monoclonal antibodies and tetrameric QD4-lgG. J. Virol. 69, 6609–6617. Trkola, A., Purtscher, M., Muster, T., et al., 1996. Human monoclonal antibody 2G12 defines a distinctive neutralization epitope on the gp120 glycoprotein of human immunodeficiency virus type 1. J. Virol. 70, 1100–1108. Trkola, A., Kuster, H., Rusert, P., Joos, B., Fischer, M., Leemann, C., Manrique, A., Huber, M., Rehr, M., Oxenius, A., Weber, R., Stiegler, G., Vcelar, B., Katinger, H., Aceto, L., Gunthard, H.F., 2005. Delay of HIV-1 rebound after cessation of antiretroviral therapy through passive transfer of human neutralizing antibodies. Nat. Med. 11, 615–622. Veazey, R.S., Shattock, R.J., Pope, M., Kirijan, J.C., Jones, J., Hu, Q., Ketas, T., Marx, P.A., Klasse, P.J., Burton, D.R, Moore, J.P., 2003. Prevention of virus transmission to macaque monkeys by a vaginally applied monoclonal antibody to HIV-1 gp120. Nat. Med. 9, 343–346. Verrier, F., Nadas, A., Gorny, M.K., Zolla-Pazner, S., 2001. Additive effects characterize the interaction of antibodies involved in neutralization of the primary dualtropic human immunodeficiency virus type 1 isolate 89.6. J. Virol. 75, 9177–9186. Wolbank, S., Kunert, R., Stiegler, G., Katinger, H., 2003. Characterization of human class-switched polymeric (immunoglobulin M [IgM] and IgA) antihuman immunodeficiency virus type 1 antibodies 2F5 and 2G12. J. Virol. 77, 4095–4103. Wolfe, E.J., Cavacini, L.A., Samore, M.H., Posner, M.R., Kozial, C., Spino, C., Trapnell, C.B., Ketter, N., Hammer, S., Gambertoglio, J.G., 1996. Pharmacokinetics of F105, a human monoclonal antibody, in persons infected with human immunodeficiency virus type 1. Clin. Pharmacol. Ther. 59, 662–667. Xu, W., Smith-Franklin, B.A., Li, P.L., Wood, C., He, J., Du, Q., Bhat, G.J., Kankasa, C., Katinger, H., Cavacini, L.A., Posner, M.R., Burton, D.R., Chou, T.C., Ruprecht, R.M., 2001. Potent neutralization of primary human immunodeficiency virus clade C isolates with a synergistic combination of human monoclonal antibodies raised against clade B. J. Hum. Virol. 4, 55–61. Zwick, M.B., Labrijn, A.F., Wang, M., Spenlehauer, C., Saphire, E.O., Binley, J.M., Moore, J.P., Stiegler, G., Katinger, H., Burton, D.R., Parren, P.W., 2001a. Broadly neutralizing antibodies targeted to the membraneproximal external region of human immunodeficiency virus type 1 glycoprotein gp41. J. Virol. 75, 10892–10905. Zwick, M.B., Wang, M., Poignard, P., Stiegler, G., Katinger, H., Burton, D.R., Parren, P.W., 2001b. Neutralization synergy of human immunodeficiency virus type 1 primary isolates by cocktails of broadly neutralizing antibodies. J. Virol. 75, 12198–12208.