Most rhesus macaques infected with the CCR5-tropicSHIVAD8 generate cross-reactive antibodies thatneutralize multiple HIV-1 strainsMasashi Shingaia,1, Olivia K. Donaua,1, Stephen D. Schmidtb,1, Rajeev Gautama, Ronald J. Plishkaa,Alicia Buckler-Whitea, Reza Sadjadpoura, Wendy R. Leea, Celia C. LaBranchec, David C. Montefioric,John R. Mascolab, Yoshiaki Nishimuraa, and Malcolm A. Martina,2

aLaboratory of Molecular Microbiology, bVaccine Research Center, National Institute of Allergy and Infectious Diseases, National Institutes of Health,Bethesda, MD 20892; and cDepartment of Surgery, Duke University Medical Center, Durham, NC 27710

Contributed by Malcolm A. Martin, October 8, 2012 (sent for review September 5, 2012)

The induction of broadly reacting neutralizing antibodies has beena major goal of HIV vaccine research. Characterization of a patho-genic CCR5 (R5)-tropic SIV/HIV chimeric virus (SHIV) molecular clone(SHIVAD8-EO) revealed that eight of eight infected animals developedcross-reactive neutralizing antibodies (NAbs) directed against an en-velope glycoprotein derived from the heterologous HIV-1DH12 strain.A panel of plasmas, collected from monkeys inoculated with eithermolecularly cloned or uncloned SHIVAD8 stocks, exhibited cross-neu-tralization againstmultiple tier 1 and tier 2HIV-1 clade B isolates. OneSHIVAD8-infected animal also developed NAbs against clades A and CHIV-1 strains. In this particular infected macaque, the cross-reactinganti–HIV-1 NAbs produced between weeks 7 and 13 were directedagainst a neutralization-sensitive virus strain, whereas neutralizingactivities emerging at weeks 41–51 targeted more neutralization-resistant HIV-1 isolates. These results indicate that the SHIVAD8 ma-caque model represents a potentially valuable experimental systemfor investigating B-cell maturation and the induction of cross-reac-tive NAbs directed against multiple HIV-1 strains.

Amajor challenge in HIV vaccine research has been the de-velopment of immunogens capable of eliciting potent, broadly

acting, neutralizing antibodies (NAbs). It is now appreciated that10–30% of HIV-1–infected individuals produce cross-reactiveNAbs of significant breadth (1–6). Less than 1% of such persons,so-called “elite neutralizers,” produce potent cross-clade–neu-tralizing activity, but only 2–3 y after virus acquisition (6). Al-though the emergence of broadly reacting anti–HIV-1 NAbs inelite neutralizers has been associated with multiple rounds of so-matic hypermutation (7), little is known about vaccine strategiesable to elicit such antibodies. Longitudinal studies of HIV-1–infected persons have suggested that set-point plasma virus loads,CD4+ T-cell levels, duration of the infection, or antibody-bindingavidity may contribute to the development of cross-reacting NAbs(5, 8, 9). It is also possible that the induction of such antibodiesdepends on unique gp120 epitopes associated with specific HIV-1strains and individual B-cell repertoires or is simply a randomprocess (10, 11). A nonhuman primate model capable of gener-ating cross-reactive anti–HIV-1–neutralizing activity could pro-vide answers to some of these questions and contribute to thedevelopment of an effective prophylactic vaccine. In this regard,we recently reported that one rhesus monkey, inoculated with anuncloned preparation of the R5-tropic SHIVAD8, developedbroad, potent, and glycan-specific NAbs with cross-reacting ac-tivity against virus isolates from different HIV-1 clades similar tothat described for “elite” HIV-1 neutralizers (12).In this study, we describe the construction of a pathogenic

SHIVAD8 molecular clone (SHIVAD8-EO). During its character-ization, we discovered that eight of eight SHIVAD8-EO–infectedanimals generated cross-reactive NAbs directed against a SHIVcarrying an envelope glycoprotein derived from a different HIV-1isolate (namely HIV-1DH12). Because this result suggested thata majority of monkeys infected with the SHIVAD8-EO family of

viruses might be able to generate NAbs against heterologous HIV-1 isolates, plasma samples collected from a cohort of 11 rhesusmacaques, infected with either uncloned SHIVAD8 swarm stocksor SHIVAD8 molecular clones, were tested for their capacity toneutralize a panel of clade A, B, and C HIV-1 isolates. All of theplasmas from this group of 11 animals neutralized several tier 1Aand 1B clade B HIV-1 strains. Three of the 11 macaques alsogenerated >1:100 IC50 neutralization titers against some tier 2clade B HIV-1 isolates, and one of the three produced significantNAb titers against some clade A and C isolates. Taken together,these findings indicate that SHIVAD8 is uniquely immunogenicduring infections of rhesus monkeys and may be a particularlyuseful reagent for identifying viral determinants that drive B-cellmaturation resulting in cross-reactive anti–HIV-1 NAbs.

ResultsInfection of Rhesus Monkeys with the SHIVAD8-EO Molecular Clone.Although we had previously reported the construction and char-acterization of the R5-tropic SHIVAD8 and had prepared uncl-oned SHIVAD8 swarm stocks as challenge viruses for vaccineexperiments (13, 14), we had not obtained a pathogenic SHIVAD8molecular clone, capable of durably maintaining chronic virusinfection and inducing clinical immunodeficiency in inoculatedanimals. This was accomplished, as described in Materials andMethods, by amplifying SHIVAD8 vpu and env genes from theplasmas of several SHIVAD8–infected monkeys and inserting in-dividual amplicons into the genetic background of SIVmac239(Figs. S1 and S2). One of the resulting clones (SHIVAD8-EO) ex-hibiting robust replication kinetics in cultured rhesus peripheralblood mononuclear cells (PBMC) was inoculated into 11 animals:4 by the i.v. (5,000 or 500 TCID50) and 7 by the intrarectal (5,000or 1,000 TCID50) routes. Similar to uncloned SHIVAD8 derivatives(13, 14), the levels of set point viremia in macaques infected withthe SHIVAD8-EO molecular clone varied widely (102 to >105 RNAcopies/mL) (Fig. 1A). Memory CD4+ T cells at an effector site[recovered by bronchoalveolar lavage (BAL)] declined markedlyin all of the monkeys during the acute phase of the SHIVAD8-EOinfection (Fig. 1B), and a gradual loss of total circulating CD4+ Tcells occurred in most animals (Fig. 1C). Monkey DC0L was eu-thanized at week 95 post infection (PI) with clinical symptoms ofimmunodeficiency following protracted anorexia and diarrhea that

Author contributions: M.S., R.G., Y.N., and M.A.M. designed research; M.S., O.K.D., S.D.S.,R.G., R.J.P., A.B.-W., R.S., W.R.L., C.C.L., and Y.N. performed research; M.S. contributednew reagents/analytic tools; M.S., O.K.D., S.D.S., R.G., R.J.P., A.B.-W., R.S., C.C.L., D.C.M.,J.R.M., Y.N., and M.A.M. analyzed data; and M.S., Y.N., and M.A.M. wrote the paper.

The authors declare no conflict of interest.1M.S., O.K.D., and S.D.S. contributed equally to this article.2To whom correspondence should be addressed. E-mail: mmartin@niaid.nih.gov.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1217443109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1217443109 PNAS | November 27, 2012 | vol. 109 | no. 48 | 19769–19774

MICRO

BIOLO

GY

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

was accompanied by marked weight loss. Macaque DC6W alsoexperienced marked weight loss and was euthanized at week 111with a gastric lymphoma. Macaque CD8T was euthanized at week117 with multiple intra-abdominal lymphomas.

Neutralizing Antibodies Generated in SHIVAD8-Infected Macaques.Werecently reported that only 3 of 19 macaques inoculated withSHIVAD8 swarm viruses developed sustained levels of autologousNabs (13). In the current study, autologous NAbs, developing ineight of the macaques infected with the SHIVAD8-EO molecularclone were initially assessed using plasma samples diluted 1:20. Asshown in Fig. 2A, five of the eight animals produced autologousneutralizing activity (>50% neutralization). In four of thesemacaques (DC6W, DCF1, FZH, and JG7), autologous NAbsbecame measurable between weeks 20 and 40 PI, but were delayeduntil week 74 in the fifth (DC8T) animal. Individual samples fromthree of these monkeys (DC6W, DCF1, and JG7) had autologousneutralization IC50 titers ranging from 1:148 to 1:161.In an earlier study, we reported that one rhesus monkey

inoculated with uncloned SHIVAD8 developed extraordinarilybroad, cross-clade, and high-titered NAbs similar to that describedfor HIV-1 “elite neutralizers” (12). This potent neutralizing ac-tivity targeted the gp120 N332 glycan. Because we were curiouswhether animals infected with the SHIVAD8-EO molecular clonemight also produce antibodies able to neutralize a heterologousHIV-1 isolate, plasma samples (1:20 dilution) from the sameeight SHIVAD8-EO–infected monkeys were tested for their ca-pacity to neutralize pseudovirions bearing the CXCR4 (X4)-tropicSHIVDH12-CL7 envelope glycoprotein, originally derived from theHIV-1DH12 isolate (15–17). The HIV-1–derived env genes presentin SHIVAD8-EO and SHIVDH12-CL7 are 90% and 84% identical atthe level of nucleotide and amino acid sequences, respectively. Asshown in Fig. 2B, all eight plasmas, including samples from ani-mals (DC0L, DC7W, and DCV9) with no or extremely low au-tologous NAbs, exhibited neutralizing activity against SHIVDH12.We next determined the cross-reactive IC50 NAb titers directed

against SHIVDH12 present in the plasmas of a different cohort of11 macaques, which had been inoculated with molecularly clonedSHIVAD8 or recently described SHIVAD8 swarm stocks (13) (Fig.3A). Only 4 (macaques DC6W, DCF1, DA55, DA70) of these 11SHIVAD8–infected animals had developed autologous NAbs.

Nonetheless, the plasmas from all 11 SHIVAD8–infected animalsgenerated anti-SHIVDH12 neutralizing IC50 titers (ranging from1:236 to 1:3,040).The production of cross-reacting NAbs against SHIVDH12 by

SHIVAD8–infected monkeys raised the possibility that a similaractivity might have also been generated against other HIV-1

Fig. 1. Virus replication and CD4+ T-cell dynamicsin rhesus macaques inoculated i.v. or intrarectallywith the molecularly cloned SHIVAD8-EO. The levelsof plasma viremia (A), percentage of BAL fluid CD4+

T cells (B), and absolute numbers of circulating CD4+

T cells (C) are shown.

Fig. 2. Autologous and cross-reacting NAbs produced by animals infectedwith the molecularly cloned SHIVAD8-EO. Plasma samples from the indicatedSHIVAD8-EO–infected macaques were collected at different times post in-fection, diluted 1:20, and assayed for neutralizing activity against theautologous SHIVAD8 (A) or the heterologous SHIVDH12 (B). The dashed linein each panel at 50% neutralization represents the threshold of virussuppression in the TZM-bl cell assay. The IC50 neutralization titers de-termined for plasmas collected from macaques DC6W (week 52), DCF1(week 40), and JG7 (week32) were determined by reciprocal dilutionand assay in TZM-bl cells. Asterisks indicate NAbs detected at week 95. nd:not done.

19770 | www.pnas.org/cgi/doi/10.1073/pnas.1217443109 Shingai et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

isolates. Plasma samples from the same cohort of 11 animals,infected with the molecularly cloned or uncloned SHIVAD8 viru-ses, were assessed for their capacity to neutralize a panel of cladeA, B, and C HIV-1 strains. As shown in Fig. 3 B and C, plasmasfrom all of the monkeys possessed neutralizing activity against tier1A (very high sensitivity) and tier 1B (above-average sensitivity)clade B HIV-1 isolates (18). In addition, three of the macaques(CL5E, DCF1, and DA55) generated >1:100 IC50 NAb titersagainst some tier 2 (moderate sensitivity) clade B HIV-1 strains(HIV-1JR-FL, HIV-176515, HIV-1TRO.11, and HIV-1CAAN.A2). Theplasma from monkey CL5E exhibited the widest breadth, in-cluding neutralization activity against cladeA andCHIV-1 strains.A neutralization profile for the previously reported (12) “eliteneutralizer” macaque CE8J at the time of its euthanasia at week100 PI is also shown at the bottom of Fig. 3.In the context of the neutralization phenotypes shown in Fig.

3, SHIVDH12 ranks with several highly neutralization-sensitiveclade B HIV-1 strains. However, when its sensitivity was evalu-ated against a panel of standardized plasma pools from HIV-1infected individuals (18), SHIVDH12 exhibited a tier 2 neutrali-zation phenotype (Table S1).To ascertain whether the temporal appearance of cross-react-

ing NAbs directed against the more difficult-to-neutralize tier 2HIV-1 strains occurred at the same time as neutralizing activityagainst tier 1 isolates in macaque CL5E (Fig. 3), the IC50 neu-tralization titers present in plasma samples collected at varioustimes following SHIVAD8 inoculation of this animal were deter-mined. Surprisingly, significant levels of neutralizing activity be-came detectable only against the tier 1B HIV-1SS1196.1 strain by

week 13 PI (Fig. 4A). NAbs directed against three tier 2 clade BHIV-1 isolates or clade A and C HIV-1 strains all emerged be-tween weeks 41 and 51 PI. In an independent experiment, cross-reacting NAbs directed against the highly neutralization-sensitiveSHIVDH12 appeared between weeks 7–10 PI in two representativeSHIVAD8–infected monkeys (Fig. 4B). Taken together, theseresults suggest that SHIVAD8–infected macaques may producecross-reactive neutralization activity in one or two phases of theinfection: (i) at early times (weeks 7–13 PI), a majority of ani-mals may generate high titers against relatively neutralization-sensitive tier 1A and tier 1B HIV-1 strains; and (ii) severalmonths later, some monkeys also develop NAbs able to suppressthe more difficult-to-neutralize tier 2 clade B or clade A and CHIV-1 isolates.

SHIVs Bearing the HIV-1AD8 Envelope Glycoprotein Are More Efficientin Generating Cross-Reacting NAbs than a SHIV Expressing the HIVDH12

env Gene During Infections of Rhesus Macaques. Two earlier studieshave reported cross-reactive anti–HIV-1–neutralizing activitydeveloping in a small fraction of SHIV–inoculated animalsduring the chronic phase of their infections (19, 20). The avail-ability of preserved plasma samples from large cohorts of theX4-tropic SHIVDH12–infected monkeys (36 animals) or the R5-tropic SHIVAD8–infected monkeys (34 animals not evaluated inFig. 3) provided the opportunity to directly compare the levels ofcross-reacting plasma-neutralizing activity generated in macaquesinoculated with SHIVDH12 or SHIVAD8 against a common het-erologous tier 1B clade B HIV-1 isolate (HIV-1SS1196.1) (21). Asshown in Fig. 5, only 1 of 36 SHIVDH12–infected animals generated

Fig. 3. Macaques infected with uncloned or molecularly cloned SHIVAD8 viruses develop cross-reacting NAbs against multiple HIV-1 strains. IC50 neutrali-zation titers against the highly sensitive SHIVDH12 (A), the indicated clade B (B) and clade A and C (C) HIV-1 isolates in plasmas of a cohort of 11 SHIVAD8–

infected animals were determined by reciprocal dilution and assay in TZM-bl cells. Values represent reciprocal serum dilution required to achieve 50%neutralization. The previously reported (12) neutralizing activity of week 100 plasma from animal CE8J is shown at the bottom of each panel. Value between20 and 99, green; value between 100 and 999, yellow; value ≥1,000, red. Blank cell indicates that value was not determined.

Shingai et al. PNAS | November 27, 2012 | vol. 109 | no. 48 | 19771

MICRO

BIOLO

GY

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

IC50 NAb titers >1:100 (a titer of 1:146) against HIV-1SS1196.1compared with 29 of 34 SHIVAD8–infected macaques (titersranging from 1:105 to 1:1,707) against HIV-1SS1196.1. In this singlehead-to-head comparison, the SHIVAD8 envelope glycoprotein,expressed during in vivo infections, was clearly superior in elic-iting cross-reactive anti–HIV-1 NAbs.

DiscussionThe most unique and unexpected property of the SHIVAD8 familyof viruses revealed in this study was its capacity to induce cross-reactive anti–HIV-1 NAbs in a majority of infected animals.Monkey CL5E, which produced NAbs against clade A, B, and CHIV-1 strains, and macaques DCF1 and DA55, which generatedIC50-neutralizing titers >1:100 against the tier 2 HIV-1JR-FL iso-late, represent three SHIVAD8–infected animals that developedcross-reacting NAb responses of significant breadth. Unlike thedevelopment of NAbs in many HIV-1–infected persons (22–24),macaques infected with the SHIVAD8 family of viruses generatedcross-reactive NAbs well before or in lieu of autologous NAbs.There have been at least two earlier studies reporting that some

SHIV–infected macaques develop cross-reactive neutralizing ac-tivity. One showed that the X4-tropic SHIVHXB2 or SHIV89.6 andits derivatives generated NAbs against the exquisitely sensitive T-cell–adapted HIV-1MN and HIV-1SF162 strains (20). The cross-reactive neutralizing activity produced by SHIVHXB2–infectedmonkeys became measurable 76–124 wk PI but not after 21–41 wkof infection. The second study, monitoring anti–HIV-1 NAbs de-veloping in animals infected with the R5-tropic SHIVSF162-P4,reported that autologous neutralization activity appeared withinthe first month of infection in 11 of 12 macaques and that cross-reactive NAbs became detectable in only 3 of 12 monkeys between6 and 10 mo PI, with the highest titers directed against the tier 1HIV-1MN and HIV-1NL4-3 strains (19).The temporal appearance of cross-reacting neutralizing ac-

tivity in SHIVAD8–infected macaque CL5E was quite intriguing.At face value, at least two bursts of NAb production occurred inthis animal, each apparently targeting HIV-1 strains with dif-ferent neutralization sensitivities. The first was directed againstthe sensitive tier 1B HIV-1SS1196.1 (Fig. 4A) and occurred be-tween weeks 5 and 13 PI. Two SHIVAD8–infected macaquesalso generated cross-reactive NAbs against the neutralization-sensitive SHIVDH12 in the same time frame (between weeks 5and 10 PI) (Fig. 4B). In contrast, the second wave of NAbproduction in animal CL5E was delayed until week 41 PI,reaching a peak at week 51, and was directed against moreneutralization-resistant HIV-1 isolates. The nearly identicalemergence times of the second burst of cross-reactive NAbssuggest that the same conserved epitope may be targeted ineach of these viruses. It should be pointed out that the IC50neutralization titers generated against both the sensitive or themore resistant HIV-1 strains generally correlated with levels of

set-point viremia in the SHIVAD8–infected monkeys (Fig. 1 andrefs. 13 and 14). For example, macaques DC6W, BIH, CK5T,and DC4N, with some of the highest neutralization titersagainst the tier 1B SS1196.1 HIV-1 isolate (Figs. 3 and 5), allhad set-point plasma viral loads between 103 and 104 viral RNAcopies/mL, whereas animal JLD, with a lower level (∼200 RNAcopies/mL) of plasma viremia (Fig. 5), had a very low neutral-ization titer against the SS1196.1 strain. All three macaques(CL5E, DCF1, and DA55) producing NAbs against the moreresistant tier 2 HIV-1 isolates had set-point plasma viremialevels between 103 and 104 RNA copies/mL.The delayed appearance of phase 2 neutralization activity in

macaque CL5E is also reminiscent of the emergence of the pre-viously described potent cross-clade NAbs generated by SHIVAD8–

infected monkey CE8J, which first became detectable betweenweeks 32 and 36 PI (12). It is worth noting that a similar two-phasepattern of cross-reacting NAb development has been reported ina longitudinal study of HIV-1–infected individuals (25). Theemergence of the second phase required∼2.5 y during these HIV-1infections whereas SHIVAD8–infected animals developed activityagainst more difficult-to-neutralize HIV-1 strains within 1 y ofvirus inoculation.Although a time course of NAb development was not deter-

mined for each of the 11 SHIVAD8–infected animals listed in Fig.3, it is quite likely that, in addition to monkey CL5E, macaquesDCF1 and DA55, both of which generated neutralization IC50titers >1:100 against the relatively resistant tier 2 HIV-1JR-FLisolate, also produced both early and late-phase NAbs. Thus, ofthe 11 SHIVAD8–infected animals intensively evaluated for thedevelopment of cross-reacting NAbs (Fig. 3), 3 monkeys likelyexhibited a “two-phase” phenotype of NAb development, and 8monkeys were able to neutralize only sensitive HIV-1 strains. It isnot currently known whether or not a common epitope is targetedby the early emerging cross-reactive NAbs in the latter group ofSHIVAD8–infected animals.At present, we do not know which epitopes associated with the

SHIVAD8 envelope glycoprotein are driving the production ofcross-reactive neutralizing activity or which epitopes in heterol-ogous HIV-1 strains are targeted for neutralization. Becausea majority of SHIVAD8–infected monkeys are able to generatecross-reactive neutralizing activity against multiple HIV-1 iso-lates, studies of B-cell maturation and the generation of broadlyreacting NAbs, triggered by a single HIV-1 envelope glycopro-tein (namely HIV-1AD8) are now possible using the macaquemodel. One could envisage experiments monitoring B-cell mat-uration patterns and NAb development in the three SHIVAD8–

infected monkeys (CL5E, DCF1, and DA55 in Fig. 3) able toneutralize the tier 2 HIV-1JR-FL isolate at IC50 titers >1:100. Theresults obtained from such a study could guide the developmentof immunogens designed to elicit late-phase cross-reactive anti–HIV-1 NAbs.

Fig. 4. Rhesus monkeys generate cross-reactiveNAbs during two phases of the SHIVAD8 infection.(A) The IC50 neutralization titers in plasma samplescollected at various times from macaque CL5E againstthe indicated HIV-1 strains were determined by re-ciprocal dilution and assay in TZM-bl cells. Values rep-resent reciprocal serum dilution required to achieve50% neutralization. Value between 20 and 99, green;value between 100 and 999, yellow; value≥1,000, red.(B) Cross-reactive anti-SHIVDH12 NAbs rapidly de-velop in two representative SHIVAD8–infected ani-mals. Plasma samples, collected from SHIVAD8–infectedmacaques DCF1 or CL5E at the indicated times, werediluted 1:20 and assayed for neutralizing activityagainst pseudovirions carrying the SHIVDH12-CL-7 (15)envelope glycoprotein in TZM-bl cells.

19772 | www.pnas.org/cgi/doi/10.1073/pnas.1217443109 Shingai et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

The failure of many SHIVAD8–infected animals to generatepotent and sustained autologous NAbs remains an enigma. Werecently reported that only 3 of 19 monkeys inoculated withSHIVAD8 swarm stocks produced autologous neutralizing activ-ity (13), and in the present study only 5 of 8 macaques inoculatedwith the SHIVAD8-EO molecular clone developed autologousNAbs (Fig. 2A). Perhaps envelope glycoproteins associated withrelatively neutralization-resistant HIV-1 strains like the tier 3HIV-1AD8, rather than envelopes from tissue-culture-adaptedvirus isolates, may present unique epitopes to the immune sys-tem, which elicit an ensemble of cross-reactive NAbs that emergeat different times and possess varying potencies.

Materials and MethodsConstruction and Characterization of a Pathogenic SHIVAD8 Molecular Clone.The strategy used to obtain a potentially pathogenic SHIVAD8 molecularclone was (i) to amplify env gene-containing segments from a previously

described (14) cohort of nine animals, infected with serially passaged SHIVAD8

derivatives (Table S1); (ii) to construct full-length SHIV molecular clones car-rying some of these amplified env genes; and (iii) to assess the infectivity ofreconstructed viruses in rhesus PBMC and in inoculated macaques. The de-tailed construction of SHIVAD8 molecular clones and the characterization ofresulting virus stocks are described in SI Materials and Methods.

Cell Culture. HEK293T (293T) and TZM-bl cells were cultured in Dulbecco’smodified Eagle’s MEM supplemented with 10% (vol/vol) heat-inactivated FBS.Rhesus monkey PBMCs were prepared and cultured as described previously (26).

Animal Experiments. Rhesus macaques (Macaca mulatta) were maintainedin accordance with the guidelines of the Committee on Care and Use ofLaboratory Animals (27) and were housed in a biosafety level 2 NIAID fa-cility. Phlebotomies, euthanasia, and tissue sample collections were per-formed as previously described (28). BAL fluid lymphocytes were preparedas previously described (29). All animals were negative for the MHC class IMamu-A*01 allele.

Fig. 5. SHIVAD8 is far superior to SHIVDH12 in eliciting potent cross-reactive NAbs during infections of rhesus macaques. Anti–HIV-1SS1196.1 IC50 neutralizingtiters in plasma collected from the indicated infected monkeys were determined by reciprocal dilution and assay in TZM-bl cells. Values represent reciprocalserum dilution required to achieve 50% neutralization. Value between 20 and 99, green; value between 100 and 999, yellow; value ≥1,000, red.

Shingai et al. PNAS | November 27, 2012 | vol. 109 | no. 48 | 19773

MICRO

BIOLO

GY

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

Quantitation of Plasma Viral RNA Levels. Viral RNA levels in plasma weredetermined by real-time reverse transcription-PCR (ABI Prism 7900HT se-quence detection system; Applied Biosystems) as previously reported (28).

Lymphocyte Immunophenotyping and Data Analysis. EDTA-treated bloodsamples and BAL fluid lymphocytes were stained for flow cytometric analysisas described previously (30–32).

Neutralization Assays. The neutralization activity present in plasma samplescollected from rhesus macaques infected with molecularly cloned or swarmSHIVAD8 stocks were measured in single-cycle infections of TZM-bl cells byEnv pseudoviruses as previously described (14, 33–36). In preliminary assays,1:20 dilutions of plasma samples were incubated with pseudotyped viruses,expressing the SHIVAD8 CK15-3 amplicon or the env gene derived fromSHIVDH12-Cl7 (15) and prepared by cotransfecting 293T cells with pNLenv1and pCMV vectors expressing the respective envelope proteins, as previouslydescribed (14). Any plasma sample causing a 50% reduction of luciferase ac-tivity compared with that obtained with an uninfected plasma sample wasconsidered positive for NAbs. All experiments were performed in quadrupli-cate and repeated twice.

The 50% neutralization inhibitory dose (IC50) titer was calculated as theplasma dilution causing a 50% reduction in relative luminescence units (RLU)compared with levels in virus control wells after subtraction of cell control

RLU as previously described (18, 33). The neutralization phenotype (tierlevels) of the SHIVAD8 molecular clone was determined by TZM-bl cell assay(18, 33, 36) using plasma samples from a cohort study, which exhibit a widerange of neutralizing activities against subtype B HIV-1 isolates (37).

Neutralization Phenotypes of HIV-1 and SHIV Preparations. Using standardizedplasma pools from HIV-1–infected individuals (18), we determined that themolecularly cloned SHIVAD8-EO exhibited a tier 2 neutralization phenotype(Table S1), the level usually associated with circulating HIV-1 strains. The pa-rental HIV-1AD8 exhibited a tier 3 phenotype in this assay. SHIVDH12CL7 dis-played a tier 2 neutralization phenotype with this pool of plasmas.

ACKNOWLEDGMENTS. We thank Keiko Tomioka, Robin Kruthers, and RanjiniIyengar for determining plasma viral RNA loads and Boris Skopets and RahelPetros for diligently assisting in the maintenance of animals and assisting withprocedures. We are indebted to the National Institutes of Health AIDSResearch and Reference Reagent Program for providing AMD3100 and toJulie Strizki and Shering Plough for providing AD101. Plasma samples fordetermining the neutralization phenotype of virus stocks were provided byÁineMcKnight through the Bill andMelindaGates Foundation’s Collaborationfor AIDS Vaccine Discovery/Comprehensive Antibody Vaccine Immune Moni-toring Consortium (Grant 38619). This work was supported by the IntramuralResearch Program of the National Institute of Allergy and Infectious Diseases,National Institutes of Health.

1. Binley JM, et al. (2008) Profiling the specificity of neutralizing antibodies in a largepanel of plasmas from patients chronically infected with human immunodeficiencyvirus type 1 subtypes B and C. J Virol 82(23):11651–11668.

2. Doria-Rose NA, et al. (2010) Breadth of human immunodeficiency virus-specific neu-tralizing activity in sera: Clustering analysis and association with clinical variables. JVirol 84(3):1631–1636.

3. Doria-Rose NA, et al. (2009) Frequency and phenotype of human immunodeficiencyvirus envelope-specific B cells from patients with broadly cross-neutralizing anti-bodies. J Virol 83(1):188–199.

4. Gray ES, et al. (2009) Antibody specificities associated with neutralization breadth inplasma from human immunodeficiency virus type 1 subtype C-infected blood donors.J Virol 83(17):8925–8937.

5. Sather DN, et al. (2009) Factors associated with the development of cross-reactiveneutralizing antibodies during human immunodeficiency virus type 1 infection. JVirol 83(2):757–769.

6. Simek MD, et al. (2009) Human immunodeficiency virus type 1 elite neutralizers: In-dividuals with broad and potent neutralizing activity identified by using a high-throughput neutralization assay together with an analytical selection algorithm. JVirol 83(14):7337–7348.

7. Scheid JF, et al. (2009) Broad diversity of neutralizing antibodies isolated frommemory B cells in HIV-infected individuals. Nature 458(7238):636–640.

8. Euler Z, et al. (2012) Longitudinal analysis of early HIV-1-specific neutralizing activityin an elite neutralizer and in five patients who developed cross-reactive neutralizingactivity. J Virol 86(4):2045–2055.

9. Gray ES, et al.; and the CAPRISA002 Study Team (2011) The neutralization breadth ofHIV-1 develops incrementally over four years and is associated with CD4+ T cell de-cline and high viral load during acute infection. J Virol 85(10):4828–4840.

10. Mahalanabis M, et al. (2009) Continuous viral escape and selection by autologousneutralizing antibodies in drug-naive human immunodeficiency virus controllers. JVirol 83(2):662–672.

11. Rademeyer C, et al.; HIVNET 028 study team (2007) Genetic characteristics of HIV-1subtype C envelopes inducing cross-neutralizing antibodies. Virology 368(1):172–181.

12. Walker LM, et al. (2011) Rapid development of glycan-specific, broad, and potentanti-HIV-1 gp120 neutralizing antibodies in an R5 SIV/HIV chimeric virus infectedmacaque. Proc Natl Acad Sci USA 108(50):20125–20129.

13. Gautam R, et al. (2012) Pathogenicity and mucosal transmissibility of the R5-tropicsimian/human immunodeficiency virus SHIV(AD8) in rhesus macaques: Implicationsfor use in vaccine studies. J Virol 86(16):8516–8526.

14. Nishimura Y, et al. (2010) Generation of the pathogenic R5-tropic simian/human immu-nodeficiency virus SHIVAD8by serial passaging in rhesusmacaques. JVirol84(9):4769–4781.

15. Sadjadpour R, et al. (2004) Induction of disease by a molecularly cloned highlypathogenic simian immunodeficiency virus/human immunodeficiency virus chimera ismultigenic. J Virol 78(10):5513–5519.

16. Shibata R, et al. (1997) Infection and pathogenicity of chimeric simian-human im-munodeficiency viruses in macaques: Determinants of high virus loads and CD4 cellkilling. J Infect Dis 176(2):362–373.

17. Shibata R, et al. (1995) Isolation and characterization of a syncytium-inducing, mac-rophage/T-cell line-tropic human immunodeficiency virus type 1 isolate that readilyinfects chimpanzee cells in vitro and in vivo. J Virol 69(7):4453–4462.

18. Seaman MS, et al. (2010) Tiered categorization of a diverse panel of HIV-1 Envpseudoviruses for assessment of neutralizing antibodies. J Virol 84(3):1439–1452.

19. Kraft Z, et al. (2007) Macaques infected with a CCR5-tropic simian/human immuno-deficiency virus (SHIV) develop broadly reactive anti-HIV neutralizing antibodies. JVirol 81(12):6402–6411.

20. Montefiori DC, et al. (1998) Neutralizing antibodies in sera from macaques infectedwith chimeric simian-human immunodeficiency virus containing the envelope gly-coproteins of either a laboratory-adapted variant or a primary isolate of humanimmunodeficiency virus type 1. J Virol 72(4):3427–3431.

21. Li M, et al. (2005) Human immunodeficiency virus type 1 env clones from acute andearly subtype B infections for standardized assessments of vaccine-elicited neutral-izing antibodies. J Virol 79(16):10108–10125.

22. Albert J, et al. (1990) Rapid development of isolate-specific neutralizing antibodiesafter primary HIV-1 infection and consequent emergence of virus variants which resistneutralization by autologous sera. AIDS 4(2):107–112.

23. Bunnik EM, Pisas L, van Nuenen AC, Schuitemaker H (2008) Autologous neutralizinghumoral immunity and evolution of the viral envelope in the course of subtype Bhuman immunodeficiency virus type 1 infection. J Virol 82(16):7932–7941.

24. Montefiori DC, et al. (1991) Homotypic antibody responses to fresh clinical isolates ofhuman immunodeficiency virus. Virology 182(2):635–643.

25. Mikell I, et al. (2011) Characteristics of the earliest cross-neutralizing antibody re-sponse to HIV-1. PLoS Pathog 7(1):e1001251.

26. Imamichi H, et al. (2002) Amino acid deletions are introduced into the V2 region ofgp120 during independent pathogenic simian immunodeficiency virus/HIV chimericvirus (SHIV) infections of rhesus monkeys generating variants that are macrophagetropic. Proc Natl Acad Sci USA 99(21):13813–13818.

27. Committee on Care and Use of Laboratory Animals (1985) Guide for the Care and Useof Laboratory Animals (National Institutes of Health, Bethesda, MD) DHHs Publ. No.NIH 85-23, Revised Ed.

28. Endo Y, et al. (2000) Short- and long-term clinical outcomes in rhesus monkeys in-oculated with a highly pathogenic chimeric simian/human immunodeficiency virus. JVirol 74(15):6935–6945.

29. Igarashi T, et al. (2003) Macrophage-tropic simian/human immunodeficiency viruschimeras use CXCR4, not CCR5, for infections of rhesus macaque peripheral bloodmononuclear cells and alveolar macrophages. J Virol 77(24):13042–13052.

30. Nishimura Y, et al. (2005) Resting naive CD4+ T cells are massively infected andeliminated by X4-tropic simian-human immunodeficiency viruses in macaques. ProcNatl Acad Sci USA 102(22):8000–8005.

31. Nishimura Y, et al. (2007) Loss of naïve cells accompanies memory CD4+ T-cell de-pletion during long-term progression to AIDS in Simian immunodeficiency virus-in-fected macaques. J Virol 81(2):893–902.

32. Nishimura Y, et al. (2004) Highly pathogenic SHIVs and SIVs target different CD4+ Tcell subsets in rhesus monkeys, explaining their divergent clinical courses. Proc NatlAcad Sci USA 101(33):12324–12329.

33. Shu Y, et al. (2007) Efficient protein boosting after plasmid DNA or recombinantadenovirus immunization with HIV-1 vaccine constructs. Vaccine 25(8):1398–1408.

34. Wei X, et al. (2002) Emergence of resistant human immunodeficiency virus type 1 inpatients receiving fusion inhibitor (T-20) monotherapy. Antimicrob Agents Chemo-ther 46(6):1896–1905.

35. Willey R, Nason MC, Nishimura Y, Follmann DA, Martin MA (2010) Neutralizing an-tibody titers conferring protection to macaques from a simian/human immunodefi-ciency virus challenge using the TZM-bl assay. AIDS Res Hum Retroviruses 26(1):89–98.

36. Wu X, et al. (2009) Mechanism of human immunodeficiency virus type 1 resistance tomonoclonal antibody B12 that effectively targets the site of CD4 attachment. J Virol83(21):10892–10907.

37. Dreja H, et al. (2010) Neutralization activity in a geographically diverse East Londoncohort of human immunodeficiency virus type 1-infected patients: Clade C infectionresults in a stronger and broader humoral immune response than clade B infection. JGen Virol 91(Pt 11):2794–2803.

19774 | www.pnas.org/cgi/doi/10.1073/pnas.1217443109 Shingai et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0