AIDS RESEARCH AND HUMAN RETROVIRUSESVolume 15, Number 6, 1999, pp. 561± 570Mary Ann Liebert, Inc.

Expression and Characterization of HIV Type 1 EnvelopeProtein Associated with a Broadly Reactive Neutralizing

Antibody Response

GERALD V. QUINNAN, JR.,1 PENG FEI ZHANG,1 DA WEI FU,1,2 MING DONG,1

HARVEY J. ALTER,3 and INTERNATIONAL COLLABORATORS4

ABSTRACT

We have studied envelope protein from a donor with nonprogressive HIV-1 infection whose serum containsbroadly cross-reactive, primary virus NA. DNA was extracted from lymphocytes, which had been collectedapproximately 6 and 12 months prior to the time of collection of the cross-reactive serum, and env genes weresynthesized, cloned, expressed on pseudoviruses, and phenotyped in NA assays. Two clones from each timepoint had identical V3 region nucleotide sequences, utilized CCR5 but not CXCR4 for cell entry, and hadsimilar reactivities with reference sera. Analysis of the full nucleotide sequence of one clone (R2) demonstratedit to be subtype B and have normal predicted glycosylation. R2 pseudovirus was compared with others ex-pressing env genes of various clades for neutralization by sera from U.S. donors (presumed or known sub-type B infections), and from individuals infected with subtypes A, C, D, E, and F viruses. Neutralization bythe U.S. sera of R2 and other clade B pseudoviruses was low to moderate, although R2 was uniquely neu-tralized by all. R2 was neutralized by 3/3, 3/3, 2/5, 5/8, and 3/4 clade A, C, D, E, and F sera, respectively. R2and a clade E pseudovirus were neutralized by largely complementary groups of sera, potentially definingtwo antigenic subgroups of HIV-1. The results suggest that the epitope(s) that induced the cross-clade reac-tive NA in donor 2 may be expressed on the R2 envelope.

561

INTRODUCTION

THE TARGET OF HIV-1 neutralizing antibodies (NAs) is the

envelope glycoprotein complex. This complex is a multi-

meric structure composed of three or four copies each of the

gp120 surface and gp41 transmembrane glycoproteins.1 There

are a number of neutralization domains on each of the three or

four heterodimeric components of the complex.2±9 The amino

acid compositions of the proteins vary substantially from strain

to strain. Some of the neutralization domains are in regions that

tend to vary greatly, while others are in regions that tend to be

highly conserved. The variable neutralization domains include

those in variable (V) regions 1, 2, and 3 of gp120, while the

conserved domains include the primary receptor-binding site,

and other epitopes in gp120 and gp41. Amino acid sequence

variation is undoubtedly the explanation for the variation that

is seen in specificity of neutralization sensitivity among virus

strains. However, it has not been possible to classify antigenic

subtypes of HIV-1 based on genetic analyses, and various re-

gions of the envelope complex even outside of the neutraliza-

tion domains have been shown to contribute to antigenic vari-

ability.10,11 Findings indicate that the neutralization of primary

isolates of HIV may be mediated primarily by antibodies

directed against non-V3 region epitopes expressed on the

1Division of Tropical Public Health, Department of Preventive Medicine and Biometrics, Uniformed Services University of the Health Sci-ences, Bethesda, Maryland 20814.

2Present address: National Vaccine and Serum Institute, Beijing, China.3Department of Transfusion Medicine, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892.4International Collaborators who contributed to this study included the following: J. Mascola, U.S. Military HIV Research Program; HIVNET

Investigators S. Allen (Birmingham, AL), J. Ellner (Cleveland, OH), K. Nelson (Baltimore, MD), and H. Sheppard (Berkeley, CA); and G. Fran-cis, UNAIDS Program, London, England.

oligomeric complex but not on monomeric gp120, while labo-

ratory-adapted strains are more readily neutralized by antibod-

ies directed against V3.12,13 The identity of the non-V3 epitopes

recognized on primary isolates is not established. The presence

of antibodies that have broadly neutralizing activity against pri-

mary isolates of many subtypes of HIV-1 in sera from infected

people is unusual, but the nature of the envelope proteins in in-

dividuals with such antibodies may be of interest for defining

the epitopes that may be broadly immunogenic in vaccines. This

article describes the cloning of envelope genes from such an

individual and the characterization of the envelope protein ex-

pressed.14,15

MATERIALS AND METHODS

Reference serum donor envelope gene cloning

The donor of the HIV-1 Neutralizing Serum (2) (Reference

2), available through the National Institutes of Health (NIH)

AIDS Research and Reference Reagent Program (Bethesda,

MD; Catalog No. 1983; provided by L. Vujcic and G. Quin-

nan), is a participant in a long-term cohort study at the Clini-

cal Center of the NIH (this serum has been referred to as

ª FDA2º ).14 The blood used to prepare Reference 2 had been

collected in the spring of 1989. Peripheral blood mononuclear

cells that had been cryopreserved from donations obtained ap-

proximately 6 months and 1 year prior to the time of Reference

2 collections were used as sources of DNA for env gene cloning.

The cells had not been stored to maintain viability. DNA was

extracted using phenol±chloroform from approximately 1±3 3106 cells from each donation.16 The DNA was used as template

in a nested polymerase chain reaction, similar to that described

previously, except that rTth was used as the DNA polymerase,

following the manufacturer instructions.17,18 The DNA was

cloned into the expression vector pSV7d, as previously de-

scribed.16,19

Other env gene clones and virus pools

The following HIV-1 env clones in the expression vector

pSV3 were obtained from the AIDS Research and Reference

Reagent Program, to which they had been provided by F. Gao

and B. Hahn: 93MW965.26 (clade C), 92RW020.5 (clade A),

92UG975.10 (clade G), and 93TH966.8 (clade E).20 The pro-

duction of env clones from the molecular virus clones NL4-

3, AD8, and SF162 has been previously described.16,21±23 env

gene of the Z2Z6 strain was cloned similarly, using molecu-

lar virus clone plasmid as template in the polymerase chain

reaction, and cloning the genes into the plasmid pSV7d.24

The production of primary isolate env clones from partici-

pants in the Multicenter AIDS Cohort Study, designated here

P9 and P10, has also been previously described.16 P9 and P10

virus pools were prepared by single subpassages of the cell

culture medium from primary cultures in phytohemagglutinin

(PHA)-stimulated blasts.16 The use of molecular virus clones

for preparation of virus pools of NL4-3 in H9 cells, and of

NL(SF162) and AD8 in PHA blasts, has also been previously

described.16

Cell cultures

The H9 cell line was obtained from R. Gallo.25 The Molt 3

cell line was obtained from the American Type Culture Col-

lection (ATCC, Rockville, MD).26 The HOS cell lines ex-

pressing CD4 and various coreceptors for HIV-1 were obtained

from the NIH AIDS Research and Reference Reagent Program

(provided by N. Landau), as was the PM1 cell line (provided

by P. Lusso and M. Reitz).27±29 The 293T cell line was ob-

tained from the ATCC, with permission from the Rockefeller

Institute.30 The H9, Molt 3, and PM1 cell cultures were main-

tained in RPMI 1640 medium supplemented with 10% fetal

bovine serum and antibiotics (GIBCO, Grand Island, NY). The

HOS and 293T cells were maintained in Dulbecco’s minimal

essential medium (GIBCO), with similar supplements, except

that the HOS cell medium was supplemented with puromycin

for maintenance of plasmid stability. Cryopreserved human pe-

ripheral blood lymphocytes were stimulated with PHA and used

for virus infections.16,31 Reverse transcriptase activity was as-

sayed as previously described.32

QUINNAN ET AL.562

TABLE 1. COMPARATIVE NEUTRALIZATION OF PSEUDOVIRUSES

EXPRESSING MULTIPLE ENVELOPE CLONES

Neutralization titer against clonea

Serum 10.1 10.2 3.1 3.2

Reference 1 1:32 1:64 1:32 1:64Reference 2 1:128 1:128 1:128 1:128

aClones 10.1 and 10.2 were obtained from the October 1988sample, and clones 3.1 and 3.2 were obtained from the March1988 sample.

TABLE 2. CORECEPTOR DEPENDENCY OF R2 PSEUDOVIRUS ENTRY INTO HOS-CD4 CELLS

Infectivity titer

In HOS-CD4 cells expressing:In PM1

Pseudovirus CCR1 CCR2b CCR3 CCR4 CCR5 CXCR4 cells

R2 , 1:4 , 1:4 , 1:4 , 1:4 1:64 , 1:4 1:32P9 , 1:4 , 1:4 , 1:4 , 1:4 1:256 , 1:4 1:8NL4-3 , 1:4 , 1:4 , 1:4 , 1:4 1:32 . 1:256 1:8AD8 , 1:4 , 1:4 , 1:4 , 1:4 1:256 , 1:4 1:32

Virus neutralization assays

The virus NL4-3 was used in neutralization assays, which

employed Molt 3 cells as target cells and used giant cell for-

mation for end-point determination, as previously de-

scribed.14 The amounts of virus used were sufficient to re-

sult in the formation of 30±50 giant cells per well.14,33 The

viruses NL(SF162) and AD8, and P9 and P10 were tested for

neutralization in PHA-stimulated human lymphoblasts in the

presence of interleukin 2 (IL-2).16,31 In the latter assays 10%

of the cell suspension was removed each week, 50% of the

medium was changed each week, and medium was sampled

twice weekly from each well for reverse transcriptase assay.

The reverse transcriptase assays were performed on the test

samples from the first sampling date at which the nonneu-

tralized control wells had reverse transcriptase activity about

10±20 times that of background, generally on day 14 or 17

of the assay. The neutralization end point was considered to

be the highest dilution of serum at which reverse transcrip-

tase activity was reduced at least 50%. The References 1 and

2 and the Negative Reference Serum were used as positive

and negative controls (NIH AIDS Reagent Program, provided

by L. Vujcic and G. Quinnan).14

Pseudovirus construction and assays of pseudovirusesfor infectivity and neutralization

Pseudoviruses were constructed and assayed using methods

similar to those described previously.16,27,32 pSV7d-env plas-

mid DNA and pNL4-3.Luc.E2 R 2 were cotransfected into

70±80% confluent 293T cell cultures using the calcium phos-

phate±HEPES buffer technique, following the manufacturer in-

structions (Promega, Madison, WI) in 24-well plastic tissue cul-

ture trays or 25-cm2 flasks.16,27,32 After 24 hr the medium was

replaced with medium containing 1 mM sodium butyrate.16,32

Two days after transfection medium was harvested, passed

through a 45-m m pore size sterile filter (Millipore, Bedford,

MA), supplemented with additional fetal bovine serum to a fi-

nal concentration of 20% and stored at 2 80°C.

Pseudovirus infectivity was assayed in PM1 or HOS-CD4

cells expressing various coreceptors. Transfection supernatants

were serially diluted and inoculated into cells in 96-well plates,

pseudovirus was allowed to adsorb to cells, and then the cul-

tures were incubated for 4 days. Assays were routinely per-

formed in triplicate. The cells were washed twice with phos-

phate-buffered saline, and lysed with 5 m l of cell culture lysing

reagent (Promega) for 15 min. The cells were then triturated

BROADLY NEUTRALIZING HIV ANTIBODIES AND ANTIGEN 563

TABLE 3. INFERRED AMINO ACID SEQUENCE OF THE R2 ENVELOPE CLONE FROM DONOR 2

Amino acid residuea Residue No.

aAmino acid residues are identified by standard single-letter designations. Boldface lettersindicate predicted N-linked glycosylation sites.

into the medium, and 10-m l aliquots of the suspensions were

transferred to wells of 96-well luminometry plates. Substrate

was added in 50-m l volumes automatically, and the lumines-

cence read using a MicroLumatPlus luminometer (EG&G

Berthold, Hercules, CA). Mock pseudovirus controls were used

in each assay, consisting of medium harvested from 293T cell

cultures transfected with pSV7d (without an env insert) and

pNL4-3.Luc.E2 R 2 , and processed in the same way as cultures

for pseudovirus preparations. Infectivity end points were de-

termined by a modified Reed±Muench method; an individual

well was considered positive if the luminescence was at least

10-fold greater than that of the mock control, and the end point

was considered to be the highest dilution at which the calcu-

lated frequency of positivity was $ 50%.16,32,33 Luminescence

resulting from infection with minimally diluted samples was

generally about 10,000-fold greater than background.

Neutralization tests were performed using PM1 or HOS-CD4

cells. Aliquots of 25 m l of two-fold serial serum dilutions were

mixed with equal volumes of diluted pseudovirus in wells of

96-well plates. The pseudovirus dilutions were selected so as

to expect luminescence in the presence of nonneutralizing

serum of about 100 times background. Assays were performed

in triplicate. The virus±serum mixtures were incubated for 60

min at 4°C, after which 150-m l aliquots of PM1 cell suspen-

sions containing 1.5 3 104 cells were added, or the suspensions

were transferred to wells containing HOS-CD4 cells. The as-

says were then processed similarly to the infectivity assays. The

neutralization end points were calculated by a modified Reed-

Muench method in which the end point was considered to be

the highest serum dilution calculated to have a frequency of

$ 50% for reducing luminescence by $ 90% compared with the

nonneutralized control. This method of end-point determination

yields good test-to-test consistency, and results that are com-

parable to results obtained in our virus neutralization assays,

described above.16 Pseudovirus titrations were conducted in du-

plicate in parallel with each neutralization assay.

Human sera

All studies were done in compliance with applicable regu-

lations for the protection of human subjects. The sera used

were References 1 and 2; sera from Thai donors with subtype

B or E infections (provided by J. Mascola, Walter Reed Army

Institute of Research, Rockville, MD); sera from Zambian,

Ugandan, and Thai donors with subtype A, C, and E infec-

tions, respectively (provided by H. Shepard, S. Allen, J. Ell-

ner, and K. Nelson of the HIVNET project); and sera from

Ugandan and Brazilian donors with subtype D and F infec-

tions, respectively (provided by G. Francis of the UNAIDS

program). In addition, sera from HIV-1-infected residents of

the Washington, D.C. and Baltimore, Maryland metropolitan

areas collected in the time period of 1985±1990 (presumed

QUINNAN ET AL.564

FIG. 1. Phylogenetic analyses of the gp120 and gp41 nucleotide coding sequences of clone R2. Alignments were performedusing the Clustal algorithm of Higgins and Sharp in the program DNA Star.34,36,37 The graphs at the bottom of the two figuresindicate the percent similarity distances represented by the dendrograms. GenBank accession numbers for the sequences repre-sented are as follows: MW959, U08453; MW960, U08454; D747, X65638; BR020, U27401; BR029, U27413; RU131, U30312;UG975, U27426; AD8, M60472; HXB, K03455; NDK, M27323; Z2Z6, M22639; UG021, U27399; CM235, L03698; TH022,U09139; TH006, U08810; UG275, L22951; SF1703, M66533; RW020, U08794; RW009, U08793; U455, M62320; and Z321,M15896.

subtype B infections), and available in our laboratory from

previous studies, were used.

Nucleic acid sequencing

Nucleotide sequence analysis was performed using the

dideoxy cycle sequencing technique and AmpliTaq FS DNA

polymerase, according to the manufacturer directions (Applied

Biosystems, Foster City, CA). After the sequencing reaction the

DNA was purified using Centriflex gel-filtration cartridges (Ad-

vanced Genetic Technologies, Gaithersburg, MD). Sequencing

gels were run and analyzed using an Applied Biosystems Prism,

model 377 DNA sequencer. Sequencing was performed on both

strands. Sequence alignment was performed using the Editseq

and Megalign programs in DNA Star according to the method

of Higgins and Sharp.34

RESULTS

Comparability of clones isolated at different time points

From the samples of patient cells from each of the two time

points, env clones were recovered that encoded proteins that

were capable of mediating pseudovirus entry into target cells.

Two such clones from each time point were further character-

ized. As shown in Table 1, the envelopes of all four clones me-

diated infection for PM1 cells and were neutralized compara-

bly by References 1 and 2. Pseudoviruses carrying envelopes

corresponding to each clone were also tested for infectivity for

HOS-CD4 cells expressing either CXCR4 or CCR5, and all four

were infectious only for the cells expressing CCR5, as shown

in Table 2. Nucleotide sequences including the V3 regions were

analyzed for each clone, with more than 300 bases assigned for

each, and no differences between the clones were found (re-

sults not shown). Based on the absence of a demonstration of

differences in these assays, a single clone from the March sam-

ple was selected for use in subsequent assays (clone 3.1 in Table

1), and is designated R2 hereafter.

Clone R2 genotype and host range phenotype

The complete nucleotide sequence of the env gene clone

R2 was determined, and found to have an open reading frame

of 2598 bases (Genbank accession number AF 128126). The

amino acid sequence deduced from this sequence is shown

in Table 3. There are 30 predicted N-linked glycosylation

sites, compared with 29 in the consensus clade B sequence;

4 consensus N-linked glycosylation sites are lacking in R2,

including those at residues 146, 215, 270, and 368 (number-

ing according to the Human Retroviruses and AIDS Database

clade B consensus sequence), in the V2, C2, C2, and V4 re-

BROADLY NEUTRALIZING HIV ANTIBODIES AND ANTIGEN 565

FIG. 2. Neutralization of clade B viruses and pseudoviruses by sera from 10 subjects infected with HIV-1 strains presumed tobe clade B. The P9 and P10 viruses (P9-V and P10-V) are primary isolates from two of the serum donors.16 The virus andpseudovirus neutralization assays were performed as described in Materials and Methods. Each point represents the results ob-tained with an individual serum. The open bars represent the standard deviations about the geometric means, indicated by the midlines. The numbers above the results obtained using pseudoviruses indicate the probabilities obtained from testing the nullhypothesis by paired t testing comparing the individual pseudoviruses to R2. The results obtained with the SF162 pseudoviruswere also significantly greater than those obtained with P10 (p 5 0.02) and NL4-3 (p 5 0.02), but not P9 and AD8 pseudoviruses.

gions of gp120, respectively.35 The consensus glycosylation

sequences at residues 215 and 270 are highly and moderately

variable, respectively.

Genotypic analyses conducted included evaluation of the

gp120 and gp41 nucleotide coding sequences in comparison with

those of a number of strains of clades A through G, as shown in

Fig. 1.36,37 Both coding regions were more closely related to clade

B than non-clade B sequences. Comparative analyses of smaller

regions of the predicted gp120 and gp41 amino acid sequences

were also performed (results not shown). The regions analyzed

included the following: each constant and variable region of

gp120; the proximal gp41 ectodomain, including the fusion do-

main and the leucine zipper region; the part of gp41 extending

from the end of the leucine zipper to the second dysteine; the re-

maining gp41 ectodomain, the transmembrane region; and the cy-

toplasmic region. R2 consistently related more closely with the

clade B sequences than the others.

Comparative sensitivity of R2 and other clade Bviruses and pseudoviruses to neutralization by serafrom individuals with clade B infections

The neutralization of R2 pseudovirus was compared to other

clade B viruses and pseudoviruses as shown in Fig. 2. Of

the five virus/pseudovirus comparisons made (P9, P10, NL4-3,

AD8, and SF162 V and PV), there were no significant differ-

ences in the neutralization of matched viruses and pseudo-

viruses by paired t test (statistical results not shown). Each of

the pseudovirus preparations was neutralized by seven, eight,

or nine of the sera tested, and the geometric mean titers ranged

from 1:13.9 to 1:56, while the R2-PV was neutralized by all 10

of the sera tested, with a geometric mean titer of 1:73.5. Al-

though the neutralization titers of each of the different sera

against R2 and the other pseudoviruses were frequently within

fourfold, the neutralization of R2-PV was significantly greater

by paired t test comparing log-transformed titers than four of

the other pseudovirus preparations.

Comparative neutralization of pseudovirusesexpressing R2 and other envelopes of diverse subtypesby sera from diverse subtype infections

The results of comparative neutralization testing using sera

from individuals infected with HIV-1 strains of subtypes A, C,

D, E, and F, Reference 1 and 2, and one Thai clade B serum

are shown in Table 4. Reference 2 neutralized the pseudovirus

expressing the homologous R2 envelope at the modest titer of

1:64 in the experiment shown and within twofold of this titer

QUINNAN ET AL.566

TABLE 4. NEUTRALIZATION OF PSEUDOVIRUSES EXPRESSING ENVELOPES OF VARIOUS CLADES BY SERA FROM

PEOPLE INFECTED WITH VARIOUS CLADES OF HIV-1

NA titer against pseudovirus (clade)a

R2 P9 P10 BR020 MW965 Z2Z6 TH966 UG975Clade Serumb (B) (B) (B) (A) (C) (D) (E) (G)

B Ref 1 32 16 32 , 10 256 10 , 8 , 10Ref 2 64 32 64 10 128 40 8 10WR465 20 NTc 80 , 10 640 10 , 10 10

A 7570 320 160 20 80 2560 , 10 , 10 , 105374 40 , 10 , 10 , 10 640 , 10 , 10 , 105837 40 20 , 10 80 2560 , 10 , 10 , 10

C ZAM107 40 10 , 10 10 1280 , 10 , 10 , 10ZAM708 10 , 10 , 10 , 10 320 , 10 , 10 , 10ZAM218 80 , 10 , 10 , 10 1280 , 10 , 10 , 10

D UG240 , 10 NT NT NT NT , 10 20 NTUG370 , 10 NT NT NT NT , 10 10 NTUG386 , 10 NT NT NT NT 20 10 NTUG097 10 NT NT NT NT 10 10 NTUG118 10 NT NT NT NT 40 20 NT

E WR659 10 , 10 , 10 , 10 20 , 10 40 , 10WR901 , 10 , 10 , 10 40 320 10 40 10WR177 , 10 , 10 , 10 40 640 10 80 , 10WR657 , 10 , 10 10 10 640 , 10 80 , 10WR593 , 10 , 10 , 10 , 10 160 10 40 , 10008 , 10 , 10 , 10 , 10 10 , 10 , 10 , 10053 20 , 10 , 10 , 10 40 , 10 20 , 10062 20 10 , 10 10 320 , 10 20 , 10

F BR318 , 10 NT NT NT NT , 10 , 10 NTBR019 10 NT NT NT NT , 10 , 10 NTBR020 20 NT NT NT NT , 10 , 10 NTBR029 10 NT NT NT NT 20 , 10 NT

aNeutralization titers are the dilutions at which 90% inhibition of luminescence was observed.bSera were the Reference Neutralizing Human Serum 1 and 2, or were provided by Dr. J. Mascola, HIVNET, or the UNAIDSProgram, as described in the text.cNT, not tested.

in many other experiments. It neutralized the other seven

pseudoviruses tested at low to moderate titers, as well. The R2

pseudovirus was neutralized by 18 of 26 sera. The other two

clade B pseudoviruses were neutralized less often, and were

also neutralized infrequently by the clade E sera. The clade C

pseudovirus was substantially more sensitive, in general, to neu-

tralization than the others tested. Cross-reactivity of R2 with

clade A was suggested by moderate to high titers of all three

clade A sera tested, in spite of minimal neutralization of the

clade A pseudovirus, BR020. Cross-reactivity with clade C was

suggested by similarly efficient neutralization of R2 by clade

C sera and moderate neutralization of the clade C pseudovirus

MW065 by Reference 2. Some cross-reactivity of R2 with clade

D was suggested in that Reference 2 neutralized the clade D

pseudovirus at a moderate dilution; however, the neutralization

of R2 by clade D sera was seen at only the lowest serum dilu-

tion tested and in only three of five cases. Minor cross-reac-

tivity of R2 with clade E was suggested in that Reference 2

neutralized the clade E pseudovirus TH966, at only the lowest

dilution tested, and the three of eight clade sera that neutralized

R2 did so at low dilutions only. No clade F pseudovirus was

tested, but minor cross-reactivity of R2 with clade F was sug-

gested by the finding that three of four clade F sera neutralized

R2 at low dilutions. Reference 2 neutralized the clade G

pseudovirus UG975 at the lowest dilution tested and no clade

G sera were tested. Some of the results obtained using the other

clade B sera against clade A and C pseudoviruses, and clade A

and C sera against clade B pseudoviruses, also suggested cross-

reactivity, but R2 was more cross-reactive than the other two

clade B clones. Cross-reactivities among some of the other

serum±pseudovirus combinations were also observed.

To evaluate further the low cross-reactivity suggested be-

tween clades B and E, the sera used in the tests shown in Fig.

2 were tested for neutralization of the clade E pseudovirus.

None of them neutralized (results not shown).

DISCUSSION

We have cloned, expressed, and characterized envelope

genes from the donor of Reference 2, which is noted for its ca-

pacity to neutralize primary HIV isolates of varied subtypes.15

Clones obtained from samples collected at two different time

points were similar to each other. The envelope gene was sim-

ilar to other clade B envelopes in nucleotide and predicted

amino acid sequence, depended on CCR5 as a coreceptor for

cell entry, and was commonly neutralized by human sera. This

envelope protein was also found to express one or more neu-

tralization epitopes in common with at least some strains of

other clades of HIV-1, and may be useful for study of the po-

tential for induction of broadly cross-reactive immunity by vac-

cination.

The patient from whom the env genes described in this arti-

cle were cloned was selected as donor of the plasma that was

used to prepare Reference 2 on the basis of preliminary screen-

ing of sera from 10 participants in a long-term cohort study of

patients with HIV-1 infection conducted at the NIH Clinical

Center. This donor and the donor of Reference 1 had higher,

and more cross-reactive neutralizing antibodies in their sera

than did the other eight participants when tested against three

laboratory-adapted clade B strains of HIV-1. The donor of Ref-

erence 2 has since been observed to have continued in good

health as recently as approximately 1 year ago, 8 years after the

original serum donation. The nonprogressive nature of this

donor’s HIV infection is of interest in view of evidence that

neutralizing antibody levels correlate to some degree with non-

progressive HIV infection.38 However, in this case there is no

basis for judging whether the neutralizing antibody status is a

consequence of the donor’s nonprogressive infection status or

vice versa, or even if the two are related. The cross-reactivity

of the donor’s neutralizing antibodies could have resulted from

infection with a strain that expressed unusually cross-reactive

epitopes, or from maturation of the neutralizing antibody re-

sponse to a relatively high level of cross-reactivity as a result

of the donor’ s nonprogressive status.

Pseudovirus technology was used for study of the antigenic

characteristics of this envelope protein, since the approach is

not confounded by variables affecting different virus prepara-

tions unrelated to the envelope protein itself.16,39±41 Pseudo-

virus expressing the R2 envelope was modestly sensitive to neu-

tralization, with respect to the titers obtained from individuals

infected with strains of HIV-1 presumed to be clade B, and ho-

mologous sera (Reference 2). Various studies of neutralization

escape mutants of HIV-1 indicate that the sensitivity of virus

to neutralization can vary as a result of mutations that do not

affect the primary structure of the neutralization epitopes rec-

ognized by the antibodies tested.3,10,11,32 Thus, it is not unex-

pected that R2 might express highly cross-reactive neutraliza-

BROADLY NEUTRALIZING HIV ANTIBODIES AND ANTIGEN 567

TABLE 5. COMPARISON OF V3 REGION AMINO ACID

SEQUENCES OF CLONE R2 WITH PHENETIC SUBGROUP

CONSENSUS SEQUENCES 1 THROUGH 13 AND CLADE ATHROUGH E CONSENSUS SEQUENCESa

Clone, subgroupor clade V3 region amino acid sequence

R2 NNTR.KSIPMGPGRAFYTTGQIIGDIRQAHC

Phenetic 1 ----.---HI----------D----------

Phenetic 2 ----.---SI-------A--E----------

Phenetic 3 ----.---SI-------A--K----------

Phenetic 4 ----.---RI---Q---A--D----------

Phenetic 5 ----.---HI-------A--K----------

Phenetic 6 K--RRR-H.I---------K-----------

Phenetic 7 ----.T--TI---QV--R--K----------

Phenetic 8 KKM-.T-ARI----V-HK--D---S-TK-Y-

Phenetic 9 ----.Q-THI---Q-L---.D---K------

Phenetic 10 ----.QGTHI-----Y---.N----------

Phenetic 11 ----.QRTSI-Q-QAL---.E-R------A-

Phenetic 12 D-IKIQRT-I-Q-Q-L---RITGYI.G----

Phenetic 13 Q-K-.QGT-I-L-Q-L---R.-K----K---

Clade A ----.--VHI---Q---A--D----------

Clade B ----.---HI----------E----------

Clade C ----.---RI---QT-YA--D----------

Clade D ----.QRTHI---Q-L---.R----------

Clade E ----.T--TI---QV--R--D------K-Y-

aDashes indicate residues at which the individual sequencesare identical to R2. The periods indicate sites of insertions ordeletions.

tion epitope(s), but not necessarily be highly sensitive to neu-

tralization. Phylogenetic comparison of various regions of R2

and other HIV-1 envelope gene nucleic acid sequences demon-

strated that R2 is descended from a clade B progenitor strain

throughout its full length.

The R2 pseudovirus was neutralized by sera from people

known to be infected with clade A, B, C, D, E, and F strains

of HIV-1. None of the sera tested neutralized all of the

pseudoviruses tested except Reference 2. Comparison with

other envelopes tested indicates that R2 may be more cross-re-

active with clade A and C strains than clade D and E strains.

Notably, the clade E pseudovirus was neutralized by 12 of 13

clade D and E sera and by 1 of 13 clade A, B, C, and F sera,

while R2 was neutralized by only 6 of 13 clade D and E sera,

but 12 of 13 of the clade A, B, C, and F sera. This distinction

between the two envelopes suggests that they may generally

distinguish two antigenic subgroups of HIV-1. That clade B and

E sera can be distinguished in neutralization assays has been

reported.31,42 Overall the results indicate that the R2 envelope

expresses neutralization epitope(s) that are common to varying

degrees among strains of different clades.

The loss of glycosylation sites in the HIV-1 envelope has

been associated with enhanced sensitivity to neutralizing anti-

bodies.43 Conversely, during the course of SIV infection there

is evolution of the V1 and V4 regions of the envelope gene re-

sulting in the extension of these regions with the addition of

glycosylation sites.44,45 These added glycosylation sites are as-

sociated with resistance to neutralization, probably as a result

of masking of neutralization epitopes by the sugar side

chains.44,45 The inferred amino acid sequence of the R2 enve-

lope glycoprotein has an overall number of N-linked glycosy-

lation sites similar to other strains of HIV-1. Moreover, in each

of the variable length regions of the V1 and V4 regions of this

protein corresponding to the segments between residues 132

and 159 and residues 396 and 415, respectively, there are three

predicted N-linked glycosylation sites, while in the vast ma-

jority of strains studied to date there are between one and three

predicted glycosylation sites in each of the corresponding seg-

ments.35

Genetic classification of strains of HIV has been useful in

efforts to understand the epidemiology of the virus, but has not

been highly predictive of antigenic relatedness among strains.46

Korber et al. have described a phenetic classification of strains

of HIV-1 into 13 subgroups based on similarities of V3 region

amino acid sequences, and Plantier et al. have presented evi-

dence that sera from infected patients can be grouped accord-

ing to their reactivity with peptides consisting of amino acid

sequences corresponding to consensus V3 sequences of the var-

ious phenetic subgroups.47,48 Since there are multiple neutral-

ization epitopes on the HIV envelope, it cannot be predicted

whether the antigenic grouping they describe will be reflected

in neutralization phenotypes of different virus strains. However,

to the extent that neutralization phenotype reflects dominance

of anti-V3 antibodies, such a grouping may be seen. The pre-

dicted V3 region amino acid sequence of the R2 envelope dif-

fers from the phenetic group 1 consensus sequence at three

residues, and from group 2, 3, and 5 consensus sequences at

four residues, as shown in Table 5. Two of these distinguish-

ing mutations, in residues 313 and 314, are at residues that were

found by Ghiara et al. to interact with anti-V3 antibody, and to

influence the interaction of the antibody with more apical

residues.49 The presence of a proline at residue 315, rather than

a basic residue, and of a glutamine at residue 329, rather than

an acidic residue, are additional unusual features of the V3 re-

gion of R2.35 Whether the V3 region or other regions of the en-

velope of this virus contributed to induction of the cross-

reactive antibody response seen in this patient and the cross-re-

activity of the R2 envelope will be the subject of additional

study.

ACKNOWLEDGMENTS

This research was supported by NIH Grant AI37438 and

USUHS Grant RO87EZ. The authors are grateful for technical

assistance provided by Dr. Xi Chen.

REFERENCES

1. Luciw PA. In: Fields Virology (Fields BN, Knipe DM, and How-

ley PM, eds.), 3rd Ed. Lippincott-Raven, Philadelphia, 1996, pp.

1881±1952.

2. Thali M, Furman C, Ho DD, Robinson J, Tilley S, Pinter A, and

Sodroski J: Discontinuous, conserved neutralization epitopes

overlapping the CD4-binding region of human immunodefi-

ciency virus type 1 gp120 envelope glycoprotein. J Virol 1992;

66:5635±5641.

3. Zwart G, Langedijk H, Van der Hoek L, de Jong JJ, Wolfs TF, Ra-

mautarsing C, Bakker M, De Ronde A, and Goudsmit J: Immun-

odominance and antigenic variation of the principal neutralization

domain of HIV-1. Virology 1991;181:481±489.

4. Moore JP, Sattentau QJ, Yoshiyama H, Thali M, Charles M, Sul-

livan N, Poon S-W, Fung MS, Traincard F, Pinkus M, Robey G,

Robinson JE, Ho DD, and Sodroski J: Probing the structure of the

V2 domain of human immunodeficiency virus type 1 surface gly-

coprotein gp120 with a panel of eight monoclonal antibodies: hu-

man immune response to the V1 and V2 domains. J Virol

1993;67:6136±6151.

5. Trkola A, Purtscher M, Muster T, Ballaun C, Buchacher A, Sulli-

van N, Srinivassan K, Sodroski J, Moore JP, and Katinger H: Hu-

man monoclonal antibody 2G12 defines a distinctive neutralization

epitope of human immunodeficiency virus type 1. J Virol 1996;70:

1100±1108.

6. Muster T, Stein F, Purtscher M, Trkola A, Klima A, Himmler G,

Ruker F, and Katinger H: A conserved neutralization epitope on

gp41 of human immunodeficiency virus type 1. J Virol 1993;

67:6642±6647.

7. Thali M, Moore JP, Furman C, Charles M, Ho DD, Robinson J,

and Sodroski J: Characterization of conserved human immunode-

ficiency virus type 1 gp120 neutralization epitopes exposed upon

gp120-CD4. J Virol 1993;67:3978±3988.

8. Cotropia J, Ugen KE, Kliks S, Broliden K, Broliden PA, Hoxie JA,

Srikantan V, Williams WV, and Weiner DB: A human monoclonal

antibody to HIV-1 gp41 with neutralizing activity against diverse

laboratory isolates. J Acquir Immune Defic Syndr Hum Retrovirol

1996;12:221±232.

9. Sabri F, Chiodi F, and Fenyö EM: Lack of correlation between V3

amino acid sequence and syncytium-inducing capacity of some

HIV type 1 isolates. AIDS Res Hum Retroviruses 1996;12:

855±858.

10. Thali M, Charles M, Furman C, Cavacini L, Posner M, Robinson

J, and Sodroski J: Resistance to neutralization by broadly reactive

antibodies to the human immunodeficiency virus type 1 gp120 gly-

QUINNAN ET AL.568

coprotein conferred by a gp41 amino acid change. J Virol 1994;

68:674±680.

11. Back NKT, Smit L, Schutten M, Nara PL, Tersmette M, and

Goudsmit J: Mutations in human immunodeficiency virus type 1

gp41 affect sensitivity to neutralization by gp120 antibodies. J Vi-

rol 1993;67:6897±6902.

12. Hioe CE, Xu S, Chigurupati P, Burda S, Williams C, Gorny MK,

and Zolla-Pazner S: Neutralization of HIV-1 primary isolates by

polyclonal and monoclonal human antibodies. Int Immunol

1997;9:1281±1290.

13. VanCott TC, Mascola JR, Kaminski RW, Kalyanaraman V, Hall-

berg PL, Burnett PR, Ulrich JT, Rechtman DJ, and Birx DL: An-

tibodies with specificity to native gp120 and neutralization activ-

ity against primary human immunodeficiency virus type 1 isolates

elicited by immunization with oligomeric gp160. J Virol 1997;

71:4319±4330.

14. Vujcic LK and Quinnan GV Jr: Preparation and characterization

of human HIV type 1 neutralizing reference sera. AIDS Res Hum

Retroviruses 1995;11:783±787.

15. D’ Souza MP, Durda P, Hanson CV, Milman G, and collaborating

investigators: Evaluation of monoclonal antibodies to HIV-1 by

neutralization and serological assays: An international collabora-

tion. AIDS 1991;5:1061±1070.

16. Quinnan G, Zhang P, Fu D, Dong M, and Margolick J: Evolution

of neutralizing antibody response against HIV-1 virions and

pseudovirions in Multicenter AIDS Cohort Study participants.

AIDS Res Hum Retroviruses 1998;14:939±949.

17. Barnes WM: The fidelity of Taq polymerase catalyzing PCR is im-

proved by an N-terminal deletion. Gene 1992;112:29±35.

18. Cariello NF, Swenberg JA, and Skopek TR: Fidelity of Thermo-

coccus litoralis DNA polymerase (Vent) in PCR determined by

denaturing gradient gel electrophoresis. Nucleic Acids Res 1991;

19:4193±4198.

19. Stuve LL, Brown-Shimer S, Pachl C, Naharian R, Diaz D, and

Burke RL: Structure and expression of the herpes simplex virus

type 2 glycoprotein gB gene. J Virol 1987;61:326±335.

20. Gao F, Yue L, Craig S, Thornton CL, Robertson DL, McCutchan

FE, Bradac JA, Sharp PM, and Hahn BH: Genetic variation of HIV

type 1 in four World Health Organization-sponsored vaccine eval-

uation sites: Generation of functional envelope (glycoprotein 160)

clones representative of sequence subtypes A, B, C, and E. WHO

Network for HIV Isolation and Characterization. AIDS Res Hum

Retroviruses 1994;10:1359±1368.

21. Adachi A, Gendelman HE, Koenig S, Folks T, Willey R, Rabson

A, and Martin MA: Production of acquired immunodeficiency

syndrome-associated retrovirus in human and nonhuman cells

transfected with an infectious molecular clone. J Virol 1986;

59:284±291.

22. Theodore TS, Englund G, Buckler-White A, Buckler CE, Martin

MA, and Peden KW: Construction and characterization of a stable

full-length macrophage-tropic HIV type 1 molecular clone that di-

rects the production of high titers of progeny virions. AIDS Res

Hum Retroviruses 1996;12:191±194.

23. Englund G, Theodore TS, Freed EO, Engleman A, and Martin MA:

Integration is required for productive infection of monocyte-de-

rived macrophages by human immunodeficiency virus type 1. J Vi-

rol 1995;69:3216±3219.

24. Seth A, Hodge DR, Thompson DM, Robinson L, Panayiotakis A,

Watson DK, and Papas TS: ETS family proteins activate tran-

scription from HIV-1 long terminal repeat. AIDS Res Hum Retro-

viruses 1993;9:1017±1023.

25. Mann DL, O’ Brien SJ, Gilbert DA, Reid Y, Popovic M, Read-Con-

nole E, Gallo RC, and Gazdar AF: Origin of the HIV-susceptible

human CD4 1 cell line H9. AIDS Res Hum Retroviruses

1989;5:253±255.

26. Daniel MD, Li Y, Naidu YM, Durda PJ, Schmidt DK, Troup CD,

Silva DP, MacKey JJ, Kestler HW III, Sehgal PK, King NW, Ohta

Y, Hayami M, and Desrosiers RC: Simian immunodeficiency virus

from African green monkeys. J Virol 1988;62:4123±4128.

27. Deng H, Liu R, Ellmeier W, Choe S, Unutmaz D, Burkhart M, Di

Marzio P, Marmon S, Sutton RE, Hill CM, Davis CB, Peiper SC,

Schall TJ, Littman DR, and Landau NL: Identification of a major

co-receptor for primary isolates of HIV. Nature (London) 1996;

381:661±666.

28. Landau NR and Littman DR: Packaging system for rapid produc-

tion of murine leukemia virus vectors with variable tropism. J Vi-

rol 1992;66:5110±5113.

29. Lusso P, Cocchi F, Balotta C, Markham PD, Louie A, Farci P, Pal

R, Gallo RC, and Reitz MS Jr: Growth of macrophage-tropic and

primary human immunodeficiency virus type 1 (HIV-1) isolates

in a unique CD4 1 T-cell clone (PM1): Failure to downregulate

CD4 and to interfere with cell-line-tropic HIV-1. J Virol 1995;

69:3712±3720.

30. Liou HC, Sha WC, Scott ML, and Baltimore D: Sequential induc-

tion of NF-kappa B/Rel family proteins during B-cell terminal dif-

ferentiation. Mol Cell Biol 1994;14:5349±5359.

31. Mascola J, Louwagie J, McCutchan FE, Fischer CL, Hegerich PA,

Wagner KF, Fowler AK, McNeil JG, and Burke DS: Two anti-

genically distinct subtypes of human immunodeficiency virus type

1: Viral genotype predicts neutralization serotype. J Infect Dis

1994;169:48±54.

32. Park EJ, Vujcic LJ, Anand R, Theodore TS, and Quinnan GV: Mu-

tations in both gp120 and gp41 are responsible for the broad neu-

tralization resistance of variant HIV-1 MN to antibodies directed

at V3 and non-V3 epitopes. J Virol 1998;72:7099±7107.

33. Lennette EH: General principles underlying laboratory diagnosis

of viral and rickettsial infections. In: Diagnostic Procedures of Vi-

ral and Rickettsial Disease (Lennette EH and Schmidt MJ, eds.).

American Public Health Association, New York, 1964, p. 45.

34. Higgins DG and Sharp PM: Fast and sensitive multiple sequence

alignments on a microcomputer. Comput Appl Biosci 1989;

5:151±153.

35. Myers G, Berzofsky JA, Korber B, Smith RF, and Pavlakis GN:

Human Retroviruses and AIDS 1992. Los Alamos National Labo-

ratory, Los Alamos, New Mexico, 1993.

36. Saitou N and Nei M: The neighbor-joining method: A new method

for reconstructing phylogenetic trees. Mol Biol Evol 1987;

4:406±425.

37. Myers EW and Miller W: Optimal alignments in linear space. Com-

put Appl Biosci 1988;4:11±17.

38. Montefiori DC, Pantaleo G, Fink LM, Zhou JT, Zhou JY, Bilska

M, Miralles GD, and Fauci AS: Neutralizing and infection-en-

hancing antibody responses to human immunodeficiency virus type

1 in long-term nonprogressors. J Infect Dis 1996;173:60±67.

39. Wrin T, Loh TP, Vennari JC, Schuitemaker H, and Nunberg JH:

Adaptation to persistent growth in the H9 cell line renders a pri-

mary isolate of human immunodeficiency virus type 1 sensitive to

neutralization by vaccine sera. J Virol 1995;69:39±48.

40. Fortin JF, Cantin R, and Tremblay MJ: T cells expressing activated

LFA-1 are more susceptible to infection with human immunodefi-

ciency virus type 1 particles bearing host-encoded ICAM. J Virol

1998;72:2105±2112.

41. Zolla-Pazner S and Sharpe S: A resting cell assay for improved de-

tection of antibody-mediated neutralization of HIV type 1 primary

isolates. AIDs Res Hum Retroviruses 1995;11:1449±1458.

42. Porter KR, Mascola JR, Hupudio H, Ewing D, VanCott TC, An-

thony RL, Corwin AL, Widodo S, Ertono S, McCutchan FE, Burke

DS, Hayes CG, Wignall FS, and Graham RR: Genetic, antigenic

and serologic characterization of human immunodeficiency virus

type 1 from Indonesia. J Acquir Immune Defic Syndr Hum Retro-

virol 1997;14:1±6.

43. Schonning K, Jansson B, Olofsson S, Nielsen JO, and Hansen JS:

BROADLY NEUTRALIZING HIV ANTIBODIES AND ANTIGEN 569

Resistance to V3-directed neutralization caused by an N-linked

oligosaccharide depends on the quaternary structure of the HIV-1

envelope oligomer. Virology 1996;218:134±140.

44. Overbaugh J and Rudensey LM: Alterations in potential sites for

glycosylation predominate during evolution of the simian immun-

odeficiency virus envelope gene in macaques. J Virol 1992;

66:5937±5948.

45. Chackerian B, Rudensey LM, and Overbaugh J: Specific N-linked

and O-linked glycosylation modifications in the envelope V1 do-

main of simian immunodeficiency virus variants that evolve in the

host alter recognition by neutralizing antibodies. J Virol 1997;

71:7719±7727.

46. Moore JP, Cao Y, Leu J, Qin L, Korber B, and Ho DD: Inter- and

intraclade neutralization of human immunodeficiency virus type 1:

Genetic clades do not correspond to neutralization serotypes but

partially correspond to gp120 antigenic serotypes. J Virol 1996;

70:427±444.

47. Korber BT, MacInnes K, Smith RF, and Myers G: Mutational

trends in V3 loop protein sequences observed in different genetic

lineages of human immunodeficiency virus type 1. J Virol

1994;68:6730±6744.

48. Plantier JC, Le Pogam S, Poisson F, Buzelay L, Lejeune B, and

Barin F: Extent of antigenic diversity in the V3 region of the sur-

face glycoprotein, gp120, of human immunodeficiency virus type

1 group M and consequences for serotyping. J Virol 1998;72:

677±683.

49. Ghiara JB, Stura EA, Stanfield RL, Profy AT, and Wilson IA: Crys-

tal structure of the principal neutralization site of HIV. Science

1994;264:82±85.

Address reprint requests to:

Gerald V. Quinnan, Jr.

Department of Preventive Medicine and Biometrics

Uniformed Services University of the Health Sciences

4301 Jones Bridge Road

Bethesda, Maryland 20814

QUINNAN ET AL.570