www.elsevier.com/locate/yviro

Virology 338 (20

Generation and properties of a human immunodeficiency virus

type 1 isolate resistant to the small molecule CCR5 inhibitor,

SCH-417690 (SCH-D)

Andre J. Marozsana, Shawn E. Kuhmanna, Thomas Morgana, Carolina Herreraa,

Enid Rivera-Trochea, Serena Xub, Bahige M. Baroudyb,1, Julie Strizkib, John P. Moorea,*

aDepartment of Microbiology and Immunology, Joan and Sanford I. Weill Medical College of Cornell University,

1300 York Avenue, W-805, New York, NY 10021, USAbSchering Plough Research Institute, Kenilworth, NJ 07033, USA

Received 19 April 2005; accepted 21 April 2005

Available online 1 June 2005

Abstract

We describe the generation of two genetically related human immunodeficiency virus type 1 (HIV-1) isolates highly (>20,000-fold)

resistant to the small molecule CCR5 inhibitor, SCH-417690 (formerly SCH-D). Both viruses were cross-resistant to other small molecules

targeting entry via CCR5, but they were inhibited by some MAbs against the same coreceptor on primary CD4+ T-cells. The resistant isolates

remained sensitive to inhibitors of other stages of virus entry, and to replication inhibitors acting post-entry. Neither escape mutant could

replicate detectably in peripheral blood mononuclear cells (PBMC) from two donors homozygous for the CCR5-D32 allele and both were

insensitive to the CXCR4-specific inhibitor, AMD3100. Hence, the SCH-D escape mutants retained the R5 phenotype. One of the resistant

isolates was, however, capable of replication in U87.CD4.CXCR4 cells and, after expansion in those cells, was sensitive to AMD3100 in

primary CD4+ T-cells. Hence, some X4 variants may be present in this escape mutant swarm. A notable observation was that the SCH-D

escape mutants were also cross-resistant to PSC-RANTES and AOP-RANTES, chemokine derivatives that are reported to down-regulate cell

surface CCR5 almost completely. However, the extent to which CCR5 is down-regulated was dependent upon the detection MAb. Hence, the

escape mutants may be using a CCR5 configuration that is only detected by some anti-CCR5 MAbs. Finally, two SCH-D-resistant clonal

viruses revealed no amino acid changes in the gp120 V3 region relative to the parental viruses, in marked contrast to clones resistant to the

AD101 small molecule CCR5 inhibitor that possess 4 such sequence changes. Several sequence changes elsewhere in gp120 (V2, C3 and

V4) were present in the SCH-D-resistant clones. Their influence on the resistant phenotype remains to be determined.

D 2005 Elsevier Inc. All rights reserved.

Keywords: HIV-1; CCR5; CCR5 inhibitors; SCH-D; Drug resistance; Escape mutants

Introduction

Entry inhibitors constitute a new class of drugs to treat

infection by human immunodeficiency virus type 1 (HIV-1)

(Baldwin et al., 2003; De Clercq, 2003; Kazmierski et al.,

2003; Kilby and Eron, 2003; Moore and Doms, 2003;

0042-6822/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.virol.2005.04.035

* Corresponding author. Fax: +1 212 746 8340.

E-mail address: jpm2003@med.cornell.edu (J.P. Moore).1 Present affiliation: Avance Pharma Inc., Laval, Quebec, Canada

H7V 5B7.

Moore and Stevenson, 2000; O’Hara and Olson, 2002). The

first member of this class, enfuvirtide, previously known as

T-20, has now been licensed for therapeutic use (Lalezari et

al., 2003; Lazzarin et al., 2003). Several other entry

inhibitors are in various stages of pre-clinical or clinical

development. They target any of several stages in the multi-

stage process by which the HIV-1 envelope glycoprotein

(Env) complex attaches to receptors via the gp120 glyco-

protein, then undergoes the conformational changes that

drive the fusion of the viral and cellular membranes (Kilby

and Eron, 2003; Moore and Doms, 2003). One particular

05) 182 – 199

A.J. Marozsan et al. / Virology 338 (2005) 182–199 183

step that is being actively pursued as a drug target is the

CD4-dependent binding of gp120 to the CCR5 coreceptor

(Trkola et al., 1996; Wu et al., 1996).

Various CCR5-reactive inhibitors of HIV-1 Env-medi-

ated membrane fusion have now been described, including

small molecules, chemokines and their derivatives, and

monoclonal antibodies (MAbs) (Kazmierski et al., 2003;

Olson et al., 1999; Schwarz and Wells, 2002; Seibert and

Sakmar, 2004). The small molecule CCR5 inhibitor SCH-D

(SCH-417690) is a receptor antagonist that has improved

potency and better pharmacology properties compared to its

predecessor, SCH-C (Strizki et al., 2001; Tagat et al., 2004).

SCH-D is in clinical development, and both it and SCH-C

have antiviral activity in HIV-1-infected humans (Maeda et

al., 2004b; Reynes et al., 2002; Schurmann et al., 2004).

It is an unfortunate but well-recognized aspect of HIV-1

drug development that resistance to any single inhibitor

frequently develops in vivo, and often occurs with multiple

inhibitors used in combination (Hammer, 2002; Larder,

2001). Entry inhibitors are highly unlikely, however, to share

resistance pathways with protease inhibitors (PI) or reverse

transcriptase inhibitors (RTI). In addition, inhibitors that

block different stages of the fusion process will probably not

suffer from cross-resistance, as their molecular targets are

different (Moore and Doms, 2003; Wei et al., 2002). Indeed,

countering otherwise drug-resistant strains is an attractive

rationale for the clinical use of entry inhibitors (Kilby and

Eron, 2003; Moore and Doms, 2003). We have, therefore,

been studying the development of resistance to CCR5

inhibitors in vitro, to gain insights into what might happen

if and when these compounds are used extensively in vivo.

We previously described how a primary R5 isolate, CC1/85,

became resistant to AD101, a prototypic CCR5 inhibitor,

when cultured in primary human T-cells (Kuhmann et al.,

2004; Trkola et al., 2002). We concluded that the resistance

of CC1/85 to AD101 involved the retention of CCR5 use in

an inhibitor-insensitive manner, rather than switching to the

use of the CXCR4 coreceptor. The fully AD101-resistant

virus contained 4 amino acid changes in the V3 region of

gp120, which were necessary and sufficient to confer

resistance when engineered into clonal viruses derived from

CC1/85 (Kuhmann et al., 2004). Escape of a different R5

primary virus, the subtype G strain JV1083, from SCH-C

also involves the sequential evolution of multiple amino acid

changes in V3 that create an inhibitor-insensitive R5 variant

(Riley et al., submitted for publication).

We now describe the generation of two different CC1/85

escape mutants under the selection pressure of SCH-D in

vitro. The input virus for one of the escape cultures was a

variant of CC1/85 that had been selected for limited

resistance to AD101; the other was CC1/85 itself. The fully

SCH-D-resistant (>20,000-fold) viruses emerged over 12

and 16 passages, respectively. Like the fully AD101-

resistant virus, the SCH-D escape mutant remained depend-

ent upon CCR5 for entry into primary CD4+ T-cells, rather

than switching to CXCR4 usage, although some variants

capable of CXCR4 use in T-cell lines were detected at low

frequency. However, unlike what was seen in the AD101

escape study, the development of resistance to SCH-D was

not associated with amino acid substitutions in V3. We also

show that the SCH-D escape mutants are cross-resistant to

several other CCR5 inhibitors, including the modified

chemokine PSC-RANTES. In contrast, their sensitivity to

other entry inhibitors, and to PI and RTI, is unchanged.

Results

Generation of an HIV-1 escape mutant to SCH-D

SCH-D is a piperazino-piperidine-based compound that

binds to the CCR5 receptor with high affinity and is an

antagonist of the natural CC-chemokine ligand, RANTES,

MIP-a and MIP-h. The derivation of SCH-D, its chemical

synthesis and its antiviral properties are all described

elsewhere (Tagat et al., 2001, 2004). In brief, SCH-D inhibits

a wide range of HIV-1 strains from various genetic subtypes,

with inhibitory concentrations in the low nM range (Tagat et

al., 2001, 2004). In comparative studies, SCH-D is typically

10-fold more potent than the earlier generation CCR5

inhibitor, SCH-C (Tagat et al., 2004). SCH-D has comparable

potency in vitro to the AD101 compound used in one of our

earlier escape studies although its pharmaceutic and pharma-

cokinetic properties are markedly superior (Strizki et al.,

submitted for publication; Tagat et al., 2004). SCH-D is now

entering Phase 3 clinical trials after demonstrating significant

antiviral activity in earlier trials (Maeda et al., 2004b).

To generate HIV-1 variants resistant to SCH-D, we used

mitogen-stimulated primary human CD4+ T-cells as the

target for virus replication. Two separate, but related, escape

cultures were initiated. In the first study, the input virus was

HIV-1 CC1/85, a subtype B, R5 primary isolate of USA

origin (Connor et al., 1997). When CC1/85 was isolated

from patient Case C in January 1985, he had a CD4+ T-cell

count of 1038 cells/mm3 and only viruses able to use CCR5

were detectable in his blood. However, R5X4 viruses had

evolved within Case C by February 1986 (Connor et al.,

1997). CC1/85 was the input isolate in our earlier CCR5

inhibitor-resistance selection experiment using AD101

(Kuhmann et al., 2004; Trkola et al., 2002).

The second, simultaneously performed escape experi-

ment used CC101.6 as the input virus. CC101.6 had been

isolated from the passage 6 culture of CC1/85 with AD101

(Kuhmann et al., 2004; Trkola et al., 2002). It is partially (3-

fold) resistant to AD101 (and also to SCH-D, see below).

CC101.6 is also much more genetically homogenous than

CC1/85, reflecting a rapid selection for naturally AD101-

resistant variants from among the initial swarm present in

the CC1/85 isolate (Kuhmann et al., 2004). In particular,

CC101.6 contains a single amino acid change in the V3

region, H308P, which was selected for very early in the

CC1/85 escape culture. We considered it worthwhile to

A.J. Marozsan et al. / Virology 338 (2005) 182–199184

investigate whether the escape process would differ for

these two genetically related input viruses.

The CC1/85 and CC101.6 escape cultures were initiated

by culturing the input viruses in the presence of 500 nM of

SCH-D, a concentration that was approximately 500 and

150 times the previously determined IC50 values of 0.95 nM

and 3.0 nM respectively. The PC cultures were maintained

in exactly the same way as the SCH-D-treated cultures,

except that no inhibitor was present. The use of PC cultures

allows the identification and analysis of genotypic and

phenotypic changes that can occur upon prolonged passage

of HIV-1 in vitro, even in PBMC (Kuhmann et al., 2004;

Pugach et al., 2004). The culture supernatants and cells were

passaged weekly and monitored for HIV-1 replication by

measuring p24 antigen production. The SCH-D concen-

tration was increased whenever virus replication in an

inhibitor-treated culture approached that in the correspond-

ing PC culture (Fig. 1). After 16 and 12 weeks, respectively,

the replication of CC1/85 and CC101.6 was completely

insensitive to the highest tested SCH-D concentration (25

AM), suggesting that highly resistant escape mutants had

evolved in both cultures (Figs. 1A, B). The resulting isolates

Fig. 1. Generation of SCH-D escape mutants. (A) The CC1/85 and (B) the

CC101.6 R5 HIV-1 isolates were passaged weekly in freshly activated

primary CD4+ T-cells with or without the SCH-D concentrations indicated

by the solid line. No inhibitor was present for the first passage. Viral

replication was monitored by p24 antigen production at each passage.

were designated D1/85.16 (CC1/85, week 16 escape) and

D101.12 (CC101.6, week 12 escape), respectively (Table 1).

The corresponding PC viruses were designated D1/

85.PC.16 and D101.PC.12.

Sensitivity of SCH-D escape mutants to CCR5 inhibitors

The D1/85.16 and D101.12 variants were tested in CD4+

T-cells for their sensitivity to a range of inhibitors that target

CCR5, including small molecules, MAbs and CC-chemo-

kines, as well to inhibitors of other stages in the virus–cell

fusion process (Figs. 2 and 3; Table 2). We first confirmed

that both D1/85.16 and D101.12 were >20,000-fold and

>8000-fold resistant to SCH-D, compared to the corre-

sponding input viruses, CC1/85 and CC101.6 (Fig. 2). This

degree of resistance to a small molecule CCR5 inhibitor is

comparable to what we observed in studies of the AD101

and SCH-C escape mutants (Kuhmann et al., 2004; Riley et

al., submitted for publication). We have found that the IC50

values for SCH-D for the input and PC viruses vary across

an ¨10-fold range when tested using CD4+ T-cells from

different donors (data not shown). Variation of this

magnitude has been reported for other small molecule

CCR5 inhibitors (Maeda et al., 2004a). We are presently

investigating its cause. However, the degree of SCH-D

resistance shown by both escape mutants exceeds the

natural variation in sensitivity to this inhibitor by at least

1000-fold, and both D1/85.16 and D101.12 were fully SCH-

D-resistant in CD4+ T-cells from every donor we have tested

to date (data not shown).

Both the escape viruses replicated efficiently whether

SCH-D was included in the CD4+ T-cell cultures or not,

indicating that neither had become inhibitor-dependent (data

not shown). In contrast, the PC viruses retained the SCH-D

sensitivity of the parental strains (Fig. 2, Table 2).

Furthermore, unless otherwise indicated, the properties of

the PC viruses were indistinguishable from those of the

input viruses in all the experiments described below. Thus,

the SCH-D-resistance of the escape mutants was a

consequence of their prolonged exposure to the selecting

compound; it was not attributable to in vitro passage per se.

We next investigated whether the SCH-D escape mutants

had acquired cross-resistance in CD4+ T-cells to other small

molecule CCR5 inhibitors, exemplified by AD101, SCH-C

and CMPD 167. The D1/85.16 isolate was >150,000-fold,

>650-fold and >50,000-fold resistant to these three com-

pounds, respectively, compared to CC1/85 (Fig. 3A, Table

2). The corresponding degrees of resistance for the D101.12

escape mutant were >67,000-fold, >2,900-fold and

>100,000-fold, with respect to the CC101.6 isolate.

The D1/85.16 and D101.12 isolates have also acquired

significant resistance (>38-fold and >90-fold) to a natural

ligand of CCR5, the CC-chemokine RANTES (Table 2).

Furthermore, both viruses were also resistant to the more

potent, chemically modified chemokine derivatives PSC-

RANTES and AOP-RANTES (Fig. 3B, Table 2). A more

Table 1

Virus nomenclature

Virus Classification Lineage Phenotype

CC1/85 Parental/input Patient

isolate

R5

D1/85.16 Escape CC1/85 R5

D1/85.PC.16 Passage control CC1/85 R5

D1/85.X4 U87.CD4.CXCR4

expanded escape

D1/85.16 R5/X4 or

R5+X4

CC101.6 Input CC1/85 R5

D101.12 Escape CC101.6 R5

D101.PC.12 Passage control CC101.6 R5

Fig. 2. The escape mutants are resistant to SCH-D. Viruses CC1/85, D1/

85.PC, D1/85.16, CC101.6, D101.PC and D101.12 were tested for their

sensitivity to SCH-D in primary CD4+ T-cells. The extent of HIV-1

replication is represented relative to p24 antigen production in the absence

of any inhibitor, which is defined as 100%. Individual data points were the

averages of values derived from 4 separate experiments, each performed

using duplicate wells.

A.J. Marozsan et al. / Virology 338 (2005) 182–199 185

detailed analysis of the resistance to these compounds is

described below.

In contrast to what was observed with the chemokines,

both SCH-D escape mutants were inhibited by the anti-

CCR5 MAbs PA14 and 2D7 that bind to epitopes formed by

the extracellular domains of CCR5 (Figs. 3C, D, Table 2).

Indeed, D101.12 was substantially more sensitive than the

other isolates to 2D7, including the input and PC viruses. It

was completely (100%) inhibited by 2D7 concentrations

above ¨10 Ag/ml (Fig. 3D). The D1/85.16 escape mutant

was only partially inhibited by this concentration of 2D7.

Thus although 50% inhibition was achieved (Table 2), the

extent of inhibition did not exceed ¨60% at the highest 2D7

concentration tested (20 Ag/ml) (Fig. 3D). The reasons for

the hypersensitivity of D101.12 to 2D7, the partial

sensitivity of D1/85.16, and the differences in the inhibitory

activities of PA14 and 2D7 are all now under investigation.

Neither MAb 45523 or 45531 had inhibitory activity against

any of the parental, PC and escape mutant isolates in CD4+

T-cells (data not shown).

Overall, when replicating in CD4+ T-cells, the SCH-D

escape mutants are resistant to some, but not all, agents that

target CCR5.

The SCH-D escape mutants remain sensitive to other classes

of entry and post-entry inhibitors

The SCH-D-resistant viruses also remained sensitive to

compounds targeting other stages of the virus–cell fusion

process. Thus, the IC50 values of D1/85.16 and D101.12 for

CD4-IgG2, a compound that blocks gp120-CD4 binding,

were not significantly different for the escape mutants and

the PC viruses (Fig. 3E, Table 2). A similar observation was

made using T20 and T1249, peptide inhibitors of gp41-

mediated conformational changes that take place after

receptor binding has occurred (Table 2).

Both the SCH-D escape mutants, as well as the

corresponding PC viruses, were comparable to the parental

isolates in their sensitivities to the NRTI zidovudine, the

NNRTI nevirapine, the PI indinavir and the integrase

inhibitor 118-D-24 (Table 2). This implies, not unexpect-

edly, that adaptation to the presence of SCH-D has primarily

involved the HIV-1 env gene, and not other elements of the

viral genome that influence sensitivity to drugs that act post-

entry. Moreover, if any compensatory mutations have arisen

in the pol gene as a result of SCH-D resistance, they have

not overtly affected sensitivity to the RT, PI or integrase

inhibitors that target the pol-encoded enzymes.

The SCH-D escape mutants continue to use CCR5 for entry

into human CD4+ T-cells, with no evidence of coreceptor

switching

The sensitivity of D1/85.16 and D101.12 to the anti-

CCR5 MAb PA14 suggested that both viruses still required

CCR5 to enter CD4+ T-cells, despite their acquired

resistance to small molecule CCR5 inhibitors. To confirm

this supposition, we evaluated their ability to replicate in

PBMC from two donors who are homozygous for the CCR5

32 allele and hence lack functional CCR5 expression. We

also investigated whether the CXCR4-specific small mole-

cule inhibitor, AMD3100, affected replication of the escape

variants.

The two SCH-D escape mutants were unable to replicate

to significant levels in PBMC from either CCR5-D32

homozygous donor under conditions in which they did so

efficiently in PBMC from a normal (i.e., CCR5 wild-type)

donor. In contrast, the reference viruses NL4-3 (X4), DH123

(R5X4) and CC2/86 (R5X4) all replicated to high levels in

cells from both the CCR5 wild type and CCR5-D32

homozygous donors (Fig. 4A). The escape mutants and

the input and PC viruses were all insensitive to AMD3100

in normal donor PBMC (Fig. 4B). Furthermore, the addition

of a fixed, saturating concentration of AMD3100 (1 AM) did

not significantly affect the replication of D1/85.16 or

D101.12 to PA14 in normal donor PBMC (data not shown).

Under the same experimental conditions, the various X4 and

Fig. 3. Sensitivity of the SCH-D escape mutants to other entry inhibitors. Viruses CC1/85, D1/85.PC, D1/85.16, CC101.6, D101.PC and D101.12 were tested

for their sensitivity to (A) SCH-C, (B) PSC-RANTES, (C) PA14, (D) 2D7, (E) CD4-IgG2 in primary CD4+ T-cells. The extent of viral replication is represented

relative to p24 antigen production in the absence of any inhibitor, which is defined as 100%. Individual data points were the averages of values derived from 4

separate experiments, each performed using duplicate wells.

A.J. Marozsan et al. / Virology 338 (2005) 182–199186

R5X4 reference viruses were all inhibited in the expected

manner by AMD3100 in PBMC from both CCR5 wild-type

and CCR5-D32 donors (Figs. 4A, B).

Taken together, these various experiments show that the

SCH-D escape mutants are both completely dependent upon

CCR5 for entry into human PBMC, and that neither virus

has acquired the ability to use CXCR4 (or any other

coreceptor) to enter these cells.

Replication of SCH-D escape mutants in

coreceptor-expressing cell lines

The coreceptor usage profiles of the SCH-D escape

mutants were further analyzed using GHOST cell lines that

each expressed CD4 and a different putative HIV-1

coreceptor. The D101.12 and D1/85.16 escape mutants,

the CC101.6 input isolate and the D101.PC.12 virus all

Table 2

Sensitivity of test isolates to entry and post-entry inhibitors (IC50 values)

Class Compound CC1/85 D1/85.PC.16 D1/85.16 CC101.6 D101.PC.12 D101.12

CCR5 SCH-D 0.95 0.76 >25,000 3.0 0.96 >25,000

Small moleculesa SCH-C 38 60 >25,000 8.5 19 >25,000

AD101 0.13 0.1 >20,000 0.3 0.3 >20,000

CMPD167 0.02 0.02 >1,000 0.01 0.01 >1000

CXCR4 small moleculea AMD3100 >10,000 >10,000 >10,000 >10,000 >10,000 >10,000

Chemokines and derivativesa RANTES 33 56 >1300 14 6 >1300

AOP-RANTES 2.21 3.41 >100 1.74 2.41 >100

PSC-RANTES 0.20 0.25 >100 0.18 0.33 >100

CCR5 MAbsb 2D7 20 20 6.8 2.0 3.3 0.06

PA14 3.6 2.1 2.3 1.1 1.3 9.6

CD4-gp120 attachmentb CD4-IgG2 4.5 4.0 4.2 16 6.0 2.2

gp41 peptidesb T20 0.41 0.30 0.10 0.70 1.37 0.92

T1249 0.16 0.55 0.11 0.03 0.07 0.07

NRTIa Zidovudine 3.8 10 3.1 10 10 3.0

NNRTIa Nevirapine 7.8 11 7.2 11 18 5.3

PIa Indinavir 3.3 1.3 5.2 3.5 8.6 1.4

Integrase inhibitorc 118-D-24 2.9 2.6 16 1.4 6.1 4.7

a Values in nM.b Values in Ag/ml.c Values in AM (NRTI) nucleoside reverse transcriptase inhibitor, (NNRTI) non-nucleoside reverse transcriptase inhibitor, (PI) protease inhibitor.

A.J. Marozsan et al. / Virology 338 (2005) 182–199 187

failed to replicate in any of the GHOST cell lines, including

GHOST.CD4.CCR5 cells (Table 3). In contrast, the other

PC virus D1/85.PC.16, like the parental CC1/85 virus, could

replicate in GHOST.CD4.CCR5 cells, but not in

GHOST.CD4.CXCR4 cells (Table 3). Hence, the D1/

85.16 escape mutant has lost the ability to use CCR5 to

enter GHOST.CD4 cells, while the CC101.6-based viruses

never had this capacity. None of the viruses have instead

acquired the ability to use CXCR4 or any other coreceptor

to enter these cells.

The escape mutants and control viruses were also tested

using U87.CD4 cells expressing either CCR5 or CXCR4.

The parental virus CC1/85 replicated strongly in

U87.CD4.CCR5 cells, while both PC viruses and the

CC101.6 input virus could do so a little less robustly. The

D101.12 escape mutant replicated detectably but weakly in

U87.CD4.CCR5 cells, but not at all in the U87.CD4.

CXCR4 cells (Table 3). However, an unexpected observa-

tion was made with the D1/85.16 escape mutant. This virus

was unable to infect U87.CD4.CCR5 cells, but it was

unique among the test panel in that it could replicate and

form syncytia in U87.CD4.CXCR4 cells in a manner that

was fully susceptible to the CXCR4 inhibitor, AMD3100

(Table 3, and data not shown). The replication of D1/85.16

in the U87.CD4.CXCR4 cells was as efficient as the

reference X4 and R5X4 viruses NL4-3, DH123 and CC2/

86 (data not shown). Hence, some of the viruses present in

the uncloned D1/85.16 isolate have acquired the ability to

use CXCR4 efficiently, at least when this coreceptor is

expressed in the U87.CD4 cell line. As noted above,

however, D1/85.16 was unable to replicate in

GHOST.CD4.CXCR4 cells and, in common with all the

other test viruses, it could neither replicate nor form

syncytia in MT4 cells (Table 3). Studies with clonal viruses

derived from the D1/85.16 and other isolates described

above have now been initiated.

To further investigate the ability of D1/85.16 to utilize

CXCR4, we propagated it for a short period in

U87.CD4.CXCR4 cells. The properties of the resulting

isolate, D1/85.X4, were then evaluated using primary CD4+

T-cells. Compared with D1/85.16, D1/85.X4 had increased

sensitivity to AMD3100. However, 50% inhibition occurred

only at a high AMD3100 concentration of ¨1 AM, which is

¨50-fold greater than was required to inhibit the NL4-3 X4

reference virus (Fig. 5A). D1/85.X4 also remained sensitive

to the CCR5 MAb, PA14, the inhibition curve for this virus

being comparable to those for the CC1/85, D1/85.PC.16 and

D1/85.16 isolates (Fig. 5B). When a 1-AM concentration of

AMD3100 was also present, the AMD3100-insensitive

component of D1/85.X4 became significantly more vulner-

able to inhibition by PA14 (Fig. 5B). In contrast, the control

viruses, CC1/85, D1/85.PC.16 and D1/85.16 were unaf-

fected when AMD3100 was added to PA14 (data not

shown). Thus, the cell line-expanded D1/85.X4 virus

behaves like a mixture of R5 and X4 variants, or possibly

a genuinely dual-tropic R5X4 strain. In other words, this

isolate retains the ability to use CCR5 for entry into primary

CD4+ T-cells, but at least some viruses are present that can

now instead, or additionally, use CXCR4 to enter these

cells.

The SCH-D escape mutants are resistant to RANTES

derivatives that can down-regulate CCR5

To gain further insights into their mechanism of

resistance, we studied the responses of the SCH-D escape

mutants to the modified chemokines AOP-RANTES and

PSC-RANTES. These compounds are reported to down-

Fig. 4. The SCH-D escape mutants do not use CXCR4 for entry into

PBMC. (A) The escape mutants and the PC viruses were assessed for their

abilities to replicate in PBMC from CCR5-D32 homozygous and CCR5

wild-type donors, in the presence and absence of the CXCR4 inhibitor

AMD3100 (1 AM). The reference viruses NL4-3 (X4), CC1/85 (parental

strain, R5), DH123 (R5X4) and CC2/86 (R5X4) were also tested. The

extent of HIV-1 replication was determined by the amount of p24 antigen

production on day 14. Similar results were obtained using PBMC from a

second CCR5-D32 homozygous donor (not shown). (B) The SCH-D escape

mutants, the PC viruses, the parental virus CC1/85 and the X4 reference

virus NL4-3 were tested for their abilities to replicate in normal donor

primary CD4+ T-cells in the presence of the AMD3100 concentrations

indicated. The extent of HIV-1 replication is expressed relative to what

occurred in the absence of AMD3100 (=100%).

A.J. Marozsan et al. / Virology 338 (2005) 182–199188

regulate a substantial fraction of cell surface CCR5, as well

as binding the receptor with greater affinity than RANTES.

Hence, the chemokine derivatives are significantly more

Table 3

Replication of SCH-D escape mutants and control isolates in coreceptor-expressi

U87.CD4 GHOST.CD4

Virus CCR5 CXCR4 CCR5 CXCR4 BOB BONZO

CC1/85 +++ – ++ – – –

D1/85.PC ++ – + – – –

D1/85.16 – ++* – – – –

CC101.6 ++ – – – – –

D101.PC ++ – – – – –

D101.12 + – – – – –

+ Weak replication.

++ Strong replication.

* Replication blocked by 1 AM AMD3100.

effective than the natural RANTES molecule at inhibiting

R5 virus replication (Hartley et al., 2004; Kawamura et al.,

2004; Simmons et al., 1997; Torre et al., 2000). However,

we noted above that both the D1/85.16 and D101.12 escape

mutants were substantially resistant to both AOP-RANTES

and PSC-RANTES when replicating in primary CD4+ T-

cells (Fig. 3B, Table 2). Thus, D1/85.16 replication was not

affected detectably by PSC-RANTES, whereas D101.12

was only partially inhibited at the highest PSC-RANTES

concentration (Fig. 3B). The resistance displayed by the two

SCH-D escape mutants stands in contrast to the PSC-

RANTES sensitivity of not just the parental and PC viruses,

but also of the AD101 escape isolate, CC101.19 (Fig. 3B,

Table 2 and data not shown). Hence, the SCH-D and AD101

escape mutants differ significantly in their sensitivities to the

modified chemokines, despite their shared genetic origin

and otherwise similar phenotypes, and despite the general

chemical similarity of SCH-D and AD101. Again, we could

find no evidence that the SCH-D escape mutants were using

CXCR4 to enter CD4+ T-cells after PSC-RANTES treat-

ment because AMD3100 (1 AM) did not inhibit their

replication even in the presence of 100 nM PSC-RANTES

(data not shown).

We concluded above that D1/85.16 and D101.12 were

still predominantly or exclusively using CCR5 for repli-

cation in primary CD4+ T-cells. Yet, they are resistant to

compounds that apparently down-regulate a very substantial

fraction of CCR5 from the cell surface (Hartley et al., 2004;

Kawamura et al., 2004; Simmons et al., 1997; Torre et al.,

2000). To investigate this apparent paradox, we studied the

effect of RANTES and its derivatives on the cell surface

exposure of CCR5 on normal donor PBMC, using flow

cytometry and several MAbs previously described as being

CCR5-specific. Nine days after mitogen activation, the

staining of CCR5 on PBMC by the 2D7 MAb was reduced

after 1 h of incubation at 3 -C with 640 nM of RANTES,

100 nM of AOP-RANTES or PSC-RANTES (Fig. 6). In the

case of PSC-RANTES, the level of 2D7 staining was almost

undetectable (<2% positive cells).

We then repeated the above experiment but using two

other anti-CCR5 MAbs, 45531 or 45523, in place of 2D7.

MAb 45523 differs from 45531 and 2D7 in that it only

ng cell lines

MT4

US28 GPR1 CCR1 CCR2 CCR3 CCR8 CXCR4

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

Fig. 5. The U87.CD4.CXCR4 propagated D1/85.16 escape isolate, D1/

85.X4, still uses CCR5 in CD4+ T-cells and is only partially inhibited by

AMD3100. (A) Replication of the viruses D1/85.16 and D1/85.X4, and the

NL4-3 reference strain (X4), in primary CD4+ T-cells was measured in the

presence of the AMD3100 concentrations indicated. The extent of

replication is expressed relative to what occurred in the absence of

AMD3100 (=100%). (B) The viruses CC1/85, D1/85.PC, D1/85.16 and

D1/85.X4 were tested for their sensitivity to PA14 in primary CD4+ T-cells.

D1/85.X4 was tested both with and without the addition of AMD3100 (1

AM). The extent of HIV-1 replication is expressed relative to what occurred

in the absence of PA14 and AMD3100 (=100%).

Fig. 6. The effect of RANTES and its derivatives on the staining of CCR5

on PBMC by different anti-CCR5 MAbs. PBMC from CCR5 wild type

individuals were treated for 1h with 640 nM RANTES, 100 nM AOP-

RANTES or 100 nM PSC-RANTES, then fixed with paraformaldehyde,

stained with MAb 2D7, 45531 or 45523 and analyzed by flow cytometry.

A.J. Marozsan et al. / Virology 338 (2005) 182–199 189

stains PBMC efficiently when the cells are first fixed (data

not shown). Thus, CCR5-D32 homozygous PBMC were not

detectably stained by MAbs 2D7, 45531 and 45523 under

conditions in which the same MAbs stained, on average,

24%, 26% and 35% of the PBMC from a CCR5 wild-type

donor and in which the anti-CXCR4 MAb 12G5 stained

56% of the cells from the CCR5-D32 homozygous donor.

When MAbs 45531 or 45523 were used to probe normal

donor PBMC that had been treated with RANTES, AOP-

RANTES or PSC-RANTES, and then fixed, a significant

amount of CCR5 was still detected on the cell surface (Fig.

6). Hence, whether CCR5 epitopes are removed from the

surface of PBMC upon treatment with RANTES and its

derivatives depends upon which anti-CCR5 MAb is used for

detection. These observations could help explain why the

SCH-D escape mutants are still able to replicate in the

presence of PSC-RANTES (Fig. 3B, Table 2).

CCR5 can be present not just on the cell surface but also

on intracellular membranes (Mack et al., 1998). We were

concerned that the fixed-cell staining procedure we had

used, in particular with MAb 45523, might have permea-

bilized the cells sufficiently to allow the detection of

intracellular forms of CCR5, as well as those present on

the plasma membrane. We therefore probed the fixed PBMC

with a rabbit polyclonal antibody to tubulin, followed by

flow cytometry. No specific tubulin staining was observed

in PBMC fixed with paraformaldehyde alone, whereas

permeabilization with 0.05% saponin allowed efficient

staining by the anti-tubulin antibody (data not shown).

Hence, it seems unlikely that the CCR5 staining observed

with MAbs 45523 and 45531 under the same experimental

conditions of paraformaldehyde fixation represents binding

to intracellular forms of CCR5. Instead, we favor the

explanation that paraformaldehyde treatment has exposed

previously cryptic epitopes on some configurations of cell

surface CCR5.

Env sequences of the SCH-D escape mutants

Our study of the AD101-resistant variant CC101.19

showed that the four most critical amino acid changes

compared to CC1/85 were located within the V3 region of

gp120 (Fig. 7A) (Kuhmann et al., 2004). To see whether a

A.J. Marozsan et al. / Virology 338 (2005) 182–199190

similar pattern of genetic changes had occurred during the

development of SCH-D resistance by the same virus, we

made clonal, SCH-D-resistant viruses by cassetting env

genes from the resistant isolates into the NL4-3 background

(Kuhmann et al., 2004; Marozsan and Arts, 2003; Marozsan

et al., 2005). The same procedure was used to make clonal,

chimeric viruses from the PC isolates (Marozsan and Arts,

2003; Marozsan et al., 2005). The V3 sequences of the

A.J. Marozsan et al. / Virology 338 (2005) 182–199 191

chimeric viruses were determined, as were IC50 values for

SCH-D inhibition of the replication of some of the chimeras

in primary CD4+ T-cells (Fig. 7A).

The SCH-D resistant clone of D1/85.16 had an identical

V3 sequence to the parental CC1/85 virus, but was

essentially completely resistant to SCH-D (Fig. 7A). A

single change, T319A, was present in the V3 region of the

clonal PC virus, D1/85.PC.16. This same change was

observed in 2/12 clones derived from the control virus

from passage 20 in the AD101 escape experiment, and may

represent an adaptation to long-term culture in PBMC

(Kuhmann et al., 2004). The H308P change in the CC101.6

virus that conferred partial resistance to AD101 was retained

in the SCH-D-resistant clone D101.12, but the additional 3

substitutions that conferred full AD101 resistance, shown in

the CC101.19 cl.7 sequence, had not arisen (Fig. 7A). The

AD101-resistant clone was found to also be resistant to

SCH-D (Fig. 7A), presumably due to these 4 changes in V3.

The H308P substitution had reverted back to the wild-type

sequence in the passage control clone from D101.PC.12,

suggesting that the original change was not always stable

during prolonged culture (Fig. 7A).

The clonal viruses derived from both D1/85.16 and

D101.12 were fully resistant to SCH-D (IC50 > 25 AM) in

primary CD4+ T-cells (Fig. 7A). In contrast, the clone

derived from CC1/85 was unusually sensitive to SCH-D,

something we have previously observed in studies of the

AD101 sensitivity of clonal viruses (Kuhmann et al., 2004).

Replication of the presently available D1/85.PC.16 and

CC101.6 clones was too limited for determination of IC50

values (Fig. 7A).

To obtain a fuller understanding of why the D1/85.16 and

D101.12 env clones were resistant to SCH-D, we sequenced

the region from the C1 domain of gp120 through end of the

ectodomain of gp41 (Fig. 7B). It is reasonable to assume

that resistance to SCH-D maps to this surface-exposed

region of Env.

In total, there were 17 amino acid differences between

the depicted CC1/85 and D1/85.16 clones (Fig. 7B). Of

these, 9 occur in gp120 and are shown in Fig. 7C. Position

146 in the V1 region is variable in wild-type CC1/85 clones,

2/8 clones having a glutamic acid residue at this position, 6/

Fig. 7. Sequences and SCH-D sensitivity of clonal viruses. (A) The amino acid sequ

previously sequenced from the CC1/85 isolate, CC1/85 cl.6 (GenBank accession n

this study. Of the CC1/85 clones sequenced in our previous study, 7/8 had identica

the CC101.6 isolate, CC101.6 cl.2 (GenBank accession number AY357371) is th

CC101.6 clones sequenced in our previous study, 9/11 had identical V3 sequence

derived from the AD101-resistant isolate CC101.19. The sequences are recorded

sequence identity. The IC50 values in CD4+ T-cells for SCH-D inhibition of clonal

the right of the sequences. ND=not done. (B) The amino acid sequence of gp120 a

As in (A), CC1/85 cl.6 is used as a reference, dashes indicate identity in the aligne

beginning and end of gp120, the beginning of gp41, and the beginning and end of

numbers indicate the amino acid number, which by convention is based on that o

amino acid to which it refers. The sequences for the D1/85.16 and D101.12

recombination in the yeast system used to create these clones. The alignment ends i

from (B) within the gp120 subunit that either differ between CC1/85 cl.6 and the D

along with the amino acid number within Env (again, based on HXBc2 numberi

8 a lysine (Kuhmann et al., 2004). The selection of K146 in

the D1/85.16 clone may then represent a founder effect.

Glutamic acid at position 146 became dominant at early

passages in the generation of the AD101 escape mutant, and

so the CC101.6 cl.2 and D101.12 clones contain this residue

(Figs. 7B, C) (Kuhmann et al., 2004). Likewise, position

167 in V2 is variable in the CC1/85 clones. 2/8 CC1/85

clones have asparagine at amino acid 167, the remainder

have aspartic acid, and asparagine was found to be present

in the D1/85.16 clone. The remaining substitutions in V2,

V169M and K171R, were not found in the wild-type or PC

viruses in this or our previous study (Figs. 7B, C)

(Kuhmann et al., 2004). Residue 351 in C3 is also variable

in CC1/85 clones. Indeed, CC1/85 cl.6 is the only clone

containing lysine, the other 7 having glutamic acid. The

other changes in C3, G354P and N355D (deletion of N355),

were not seen in any of the sequenced CC1/85 clones (Figs.

7B, C) (Kuhmann et al., 2004). The remaining 8 sequence

differences between D1/85.16 and CC1/85 cl.6 are in the

ectodomain of gp41 (Fig. 7B). Three of these are

conservative substitutions of hydrophobic residues in the

gp41 fusion peptide, the remaining 5 are at locations that are

variable in CC1/85 and represent alternative amino acids

found at those positions in other CC1/85 clones (Fig. 7B)

(Kuhmann et al., 2004).

The CC101.6 and D101.12 clones differ by only 2 amino

acids in gp120 and 4 in gp41 (Figs. 7B,C). The N406K

substitution in the V4 loop results in the loss of a potential

N-linked glycosylation site. The G474D substitution in

D101.12 relative to CC101.6 cl.2 occurs because CC101.6

cl.2 contains glycine at amino acid 474, whereas all the

other sequenced CC101.6 clones, as well as all CC1/85

clones, have aspartic acid (Fig. 7B) (Kuhmann et al., 2004).

Two hydrophobic substitutions are found in the gp41 fusion

peptide and two additional changes are at amino acids 534

and 535. Serine is found at position 534 in the D101.12

clone and 7/8 CC1/85 clones but is alanine is all of the

available CC101.6 clones. The V535M substitution is not

found in any of the CC1/85 or CC101.6 clones (Fig. 7B)

(Kuhmann et al., 2004). Which of the amino acid changes in

the SCH-D-resistant clones account for the escape mutant

phenotype has not yet been determined.

ences of the V3 region from 7 env clones are shown. Among the env clones

umber AY357338) is the most closely related to the D1/85.16 clone used in

l V3 sequences. Likewise, among the env clones previously sequenced from

e one most closely related to the D101.12 clone used in this study. Of the

s. CC101.19 cl.7 (GenBank accession number AY357465) is an env clone

relative to that of CC1/85 cl.6 (top line) with dashes indicating amino acid

viruses containing these env genes in the NL4-3 background are recorded on

nd the ectodomain of gp41 from four of the env clones from (A) are aligned.

d sequences and a dot indicates a gap in the alignment. The positions of the

each conserved and variable region are indicated above the alignment. The

f the prototypic strain HXBc2; the last digit of the number aligns with the

clones begin at amino acid 42 because this is the point of homologous

mmediately before the transmembrane domain in gp41. (C) The amino acids

1/85.16 clone or differ between CC101.6 and the D101.12 clone are shown,

ng) and the region in which that amino acid is found.

A.J. Marozsan et al. / Virology 338 (2005) 182–199192

Discussion

In our previous reports on how HIV-1 escapes from the

selection pressure exerted by small molecule CCR5

inhibitors in vitro, we concluded that the predominant

pathway involved the generation of viral variants that

retained the ability to use CCR5, but in an inhibitor-

insensitive manner (Kuhmann et al., 2004; Riley et al.,

submitted for publication; Trkola et al., 2002). Thus, the

AD101-resistant CC1/85 variants were unable to replicate in

PBMC from CCR5-D32 homozygous donors; inhibited by

CCR5-specific MAbs; insensitive to CXCR4 inhibitors; and

unable to use any coreceptor other than CCR5 to enter

indicator cell lines (Kuhmann et al., 2004; Trkola et al.,

2002). We have since made similar findings using another

primary isolate, JV1083 and SCH-C, albeit with CCR5-

expressing PM1 cells substituting for PBMC (Strizki et al.,

submitted for publication).

The present study, again performed using the CC1/85

primary isolate and PBMC, but with a different inhibitor,

SCH-D, has generated some findings that are consistent

with the previous experiments, but also some unexpected

ones. Two escape mutants were made, one from CC1/85, the

second from a partially (¨3-fold) resistant variant,

CC101.6. The latter virus is more genetically homogenous

than CC1/85, reflecting a rapid selection for naturally

resistant variants during the initial stages of the AD101

resistance-selection process (Kuhmann et al., 2004; Trkola

et al., 2002). SCH-D-resistant viruses were generated from

both input isolates over a comparable number of passages,

16 for CC1/85, 12 for CC101.6. The escape mutants were

completely resistant to SCH-D and also to other small

molecule CCR5 inhibitors: SCH-C, AD101 and CMPD 167.

These compounds are from different chemical families

(Shen et al., 2004). However, all the small molecule

inhibitors probably bind to a broadly similar site within

CCR5’s transmembrane helices (Tsamis et al., 2003). Cross-

resistance within the drug class would be the expected

outcome if these compounds were used clinically, although

it is not necessarily inevitable. In contrast, the SCH-D

escape mutants retained the sensitivity of the parental

isolates to other entry inhibitors, CD4-IgG2 (PRO 542), T-

20 and T-1249, with different mechanisms of action. There

was also no change in their susceptibility to RT, protease or

integrase inhibitors. Hence, as expected, the small molecule

CCR5 inhibitors are unlikely to cause cross-resistance to

other classes of licensed drugs (Moore and Doms, 2003).

Both the SCH-D escape mutants are R5 viruses. Thus,

they lack any ability to replicate in PBMC from CCR5-D32

homozygous individuals; they are inhibited by CCR5-

specific MAbs in CD4+ T-cells from normal donors; and

they have no detectable sensitivity to the CXCR4-specific

inhibitor AMD3100 in the same cells. In this regard, they

resemble the AD101- and SCH-C-escape mutants (Kuh-

mann et al., 2004; Riley et al., submitted for publication;

Trkola et al., 2002). However, the SCH-D-resistant variants

did possess some unusual properties when evaluated in

cell lines expressing various putative coreceptors. For

example, D1/85.16 and D101.12 could not replicate in

GHOST.CD4.CCR5 cells, nor could the PC virus,

D101.PC.12. In contrast, the other PC virus, D1/85.PC.16,

resembled the input CC1/85 isolate by retaining the ability

to replicate in GHOST.CD4.CCR5 cells. The reasons

underlying this pattern of viral tropism remain uncertain

pending additional studies.

Of particular note is that none of the escape mutants or PC

viruses could replicate in any other GHOST.CD4 cell line

that we tested, including GHOST.CD4.CXCR4 cells. The

D101.12 escape mutant was also unable to replicate in

U87.CD4.CXCR4 cells, although it could do so in

U87.CD4.CCR5 cells, albeit only at low levels. However,

the D1/85.16 escape mutant behaved unexpectedly, in that it

could replicate and form syncytia in U87.CD4.CXCR4 cells,

but not in U87.CD4.CCR5 cells. Replication in the

U87.CD4.CXCR4 cells was blocked by AMD3100, suggest-

ing that D1/85.16 was using CXCR4 to enter these cells, and

not any other coreceptor. When the D1/85.16 variant was

propagated for a short period in U87.CD4.CXCR4 cells, then

tested on primary CD4+ T-cells, its properties resembled an

R5X4 virus or a mixture of R5 and X4 variants.

It seems likely then that the swarm of viruses in the D1/

85.16 isolate includes at least some with atypical coreceptor

usage properties compared to the input and PC viruses. The

presence of these variants allows low-level replication in

U87.CD4.CXCR4 cells, but not in GHOST.CD4.CXCR4

cells. Furthermore, the D1/85.16 isolate could not replicate

or form syncytia in MT4 cells, so by the classical definition,

it would not be a syncytium-inducing or SI virus (Moore et

al., 2004; Schuitemaker et al., 1991) Presumably, the

amount or configuration of CXCR4 on the GHOST cells

differs critically from what is present on the U87 cells. The

replication of primary isolates, whether of the R5 or X4

phenotype, in coreceptor-transfected cell lines is frequently

discordant with their usage of CCR5, CXCR4 or other

coreceptors in primary cells (McKnight et al., 1997; Trkola

et al., 1999; Yi et al., 2005; Zhang et al., 1998, 2000). Such

replication restrictions may also apply to different cell lines.

Without deliberate propagation of the D1/85.16 isolate

on U87.CD4.CXCR4 cells, the CXCR4-using variants were

either present at too low a frequency to permit a detectable

level of replication in primary CD4+ T-cells, or they were

unable to use CXCR4 in the configuration or quantity

present on these cells. Genetic sequencing revealed an

unusual amino acid change, G314R, in the crown of the V3

region from some of the clones from the U87.CD4.CXCR4

cell expanded stock (D1/85.X4) (data not shown). The

functional significance of this change is now under

investigation. The selection pressure of SCH-D was, then,

able to drive the emergence of at least some viruses able to

use CXCR4 under some conditions, in vitro, an outcome

that needs to be considered in relation to ongoing clinical

trials of CR5 inhibitors for HIV-1 therapy. However, we did

A.J. Marozsan et al. / Virology 338 (2005) 182–199 193

not observe any usage of CXCR4 by viruses generated in

the other SCH-D resistance culture or by the AD101 or

SCH-C escape mutants, suggesting that switching to

CXCR4 use is not a common pathway for escape from

CCR5 inhibitors (Kuhmann et al., 2004; Riley et al.,

submitted for publication; Trkola et al., 2002).

The paramount property of both SCH-D escape mutants

was, however, their retention of CCR5 usage in primary

CD4+ T-cells. It still remains uncertain exactly how this was

achieved. We initially hypothesized that a CCR5-dependent

escape mutant could either have acquired an increased

affinity for CCR5 that enabled it to out-compete the inhibitor,

or else the ability to use the coreceptor despite the presence of

the inhibitor (Trkola et al., 2002). We obtained evidence that

an increased affinity for CCR5 could occur in the early stages

of the AD101 escape process, but this was associated with

only a modest degree of resistance (¨3-fold). Intuitively, it

seems unlikely that resistance as profound as we have

observed (i.e., >20,000-fold) could arise by this mechanism;

the affinity gain required would be colossal.

Recognition of the inhibitor-bound form of CCR5

remains a plausible, but unconfirmed, escape mechanism.

We have, however, shown that SCH-C can interact with the

macaque form of CCR5 without inhibiting SIV or SHIV

entry, an event explained by a single amino acid difference,

I198M, between the human and macaque coreceptors that is

located outside the inhibitor-binding site (Billick et al.,

2004). On the basis of this and related structure–function

studies, we suggested that small molecule CCR5 inhibitors

either prevent the coreceptor from adopting a configuration

that can be recognized by HIV-1, or stabilize a CCR5

conformation that the virus cannot use (Billick et al., 2004;

Tsamis et al., 2003). Perhaps, the SCH-D escape mutants

have acquired the ability to use the latter configuration.

An alternative escape process is that the mutant virus has

evolved to use a particular configuration of CCR5 to which

SCH-D does not bind (Kuhmann et al., 2004). This CCR5

isoform would either not be recognized by the parental

virus or else it is present at only very low levels because

otherwise SCH-D would have no activity against the

parental virus (or R5 viruses in general). However, some

rare R5 HIV-1 isolates are intrinsically relatively resistant to

small molecule CCR5 inhibitors (Strizki et al., 2001), and

there is variation among SIV-infected macaques in their

response to CCR5 inhibitor therapy (Veazey et al., 2003;

Wolinsky et al., 2004). Viruses, and perhaps also hosts, that

are naturally resistant to CCR5 inhibitors might utilize a

mechanism for Env-CCR5 interaction that is analogous to

the one(s) acquired by the in vitro-selected SCH-D escape

mutants. There is, in fact, evidence based on MAb-

reactivity patterns for the presence of multiple CCR5

configurations on primary cells (Blanpain et al., 2002; Hill

et al., 1998; Lee et al., 1999; Olson et al., 1999). This

evidence also applies to CXCR4 (Baribaud et al., 2001;

Carnec et al., 2005). Studies of G-protein-coupled receptors

in general provide strong support for the existence of these

proteins in alternative allosteric conformations (Kenakin,

2004; Seibert and Sakmar, 2004). Soluble gp120 from the

AD101 escape mutant was unable to bind to human CCR5

expressed on transfected murine fibroblasts, an observation

that might be attributable to the presence of different CCR5

configurations on different cell types, although other

possibilities exist (Kuhmann et al., 2004). We have not

yet performed such studies with gp120 from the SCH-D

escape variants.

The resistance to AOP-RANTES and PSC-RANTES

displayed by the SCH-D escape mutants, but curiously not

by the AD101-resistant virus, might also be consistent with

the hypothesis that cell surface CCR5 exists in multiple

antigenic forms that are recognized differently not just by

various anti-CCR5 MAbs, but also by HIV-1 phenotypic

isolates. PSC-RANTES treatment of PBMC reduced the

staining of cell surface CCR5 proteins by MAb 2D7 by

over 98% within 1 h, yet this treatment did not prevent the

replication of the SCH-D escape mutants. However, MAbs

45523 and 45531 still bound significantly to the surface of

PSC-RANTES-treated cells, at ¨40–60% of the baseline

level. All three MAbs are specific for CCR5; they did not

stain PBMC from CCR5-D32 homozygous individuals.

Perhaps, then, the escape mutants recognize a CCR5

configuration that is also bound by MAbs 45523 and

45531 and which is not affected by PSC-RANTES? At

present, we do not know what such a CCR5 conformation

might be, but one clue for future studies is provided by the

observation that MAb 45523 only binds to CCR5 wild

type PBMC after fixation of the cells. Hence, the 45523

epitope may normally be cryptic, perhaps occluded by

another cell surface molecule or buried within a CCR5

hetero- or homo-dimer. This epitope is formed from

multiple CCR5 domains (Lee et al., 1999). MAb 2D7,

on the other hand, binds to live cells and fixation does not

affect its epitope, which is located principally within the

second extracellular (2-A) loop of CCR5 (Lee et al., 1999;

Olson et al., 1999). The 45531 epitope is also located in

this loop, but in a different area of it (2-B) (Lee et al.,

1999). Thus, the epitopes for the above three MAbs may

partially overlap but they are not identical, allowing the

possibility that they can each recognize different CCR5

configurations to various extents.

The above scenario needs to be reconciled with the

inhibitory effect of MAb 2D7 on the SCH-D escape

mutants. While 2D7 only partially inhibits D1/85.16

(¨60%), the D101.12 isolate is unusually sensitive to this

MAb and is completely blocked by it. How, then, can 2D7

inhibit the replication of D101.12 when the same virus is

little affected by the essentially complete removal of the

2D7 epitope from the cell-surface by PSC-RANTES treat-

ment? AOP-RANTES has been reported to efficiently

down-regulate CCR5 (i.e., physically remove it) from the

surface of various CCR5-transfected cell lines, but it also

occludes the 2D7 epitope (Sabbe et al., 2001; Signoret et al.,

2000). The same is true for PSC-RANTES (data not shown).

A.J. Marozsan et al. / Virology 338 (2005) 182–199194

The inability of 2D7 to stain CCR5 on PSC-RANTES-

treated PBMC does not, therefore, necessarily mean that

CCR5 is completely absent from the PBMC surface. It

could mean that some CCR5 proteins remain on the surface,

but with their 2D7 epitope obscured by the modified

chemokine, but with their epitopes for 45523 and 45531

unaffected. D101.12 might be able to utilize these chemo-

kine-bound CCR5 proteins. However, the efficiency of this

interaction may be diminished compared to the unbound

form of CCR5 because D101.12 was partially sensitive to

the highest concentration of PSC-RANTES.

As noted above, the potency of CCR5 inhibitors in vitro

and in vivo does vary between individuals (Kuhmann et al.,

2004; Moore and Doms, 2003). Additional studies may

reveal whether HIV-1 variants use a CCR5 configuration(s)

that is expressed to different extents between individuals.

Donor-dependent variation in the secretion of chemokines

by PBMC, either basally or in response to treatment with a

chemokine antagonist like SCH-D, could be another

relevant factor. Yet another might be SCH-D-induced up-

regulation of CCR5 over the duration of the 14-day

replication assays, a phenomenon that has been reported

in studies using SCH-C (Xu et al., 2002). An isoform(s) of

CCR5 that is unaffected by PSC-RANTES but recognized

by the SCH-D escape mutants might be upregulated

disproportionately under these conditions.

In genetic studies, we found that escape of CC1/85 from

AD101 was caused by four sequential amino acid changes

in the gp120 V3 region (Kuhmann et al., 2004). A different

R5 primary isolate also escaped from SCH-C in PM1 cells

via V3 sequence changes (Strizki et al., submitted for

publication). In contrast, the V3 sequences of two clones

fully resistant to SCH-D were identical to those of the

corresponding input isolates (Fig. 7). Hence, viruses

resistant to small molecule CCR5 inhibitors can emerge

without any changes in V3. Presumably, genetic changes

elsewhere in the envelope glycoproteins have the same

phenotypic effect.

In the SCH-D-resistant D1/85.16 clone, amino acid

changes in two regions stand out as the ones most likely

to influence SCH-D sensitivity. These changes, V169M and

K171R in the V2 region and G354P and N355D in C3, were

present in the D1/85.16 clone, but in none of the clones

from the parental CC1/85 isolate. The two V2 changes are

proximal to another alteration, D167N, that was a minor

polymorphism in the CC1/85 parental isolate. The D167N

change was previously found to be slowly selected for in the

PC viruses from the AD101 escape cultures, and it is

enriched in the later passages of the AD101-resistant viruses

(Kuhmann et al., 2004; Pugach et al., 2004). The D167N

polymorphism, along with two other changes in V2, I165K

and the deletion of residues 188 and 189, are associated with

increased sensitivity to neutralization by soluble CD4 and

the CD4 binding site MAb b12, and probably cause an

increase in the affinity of Env for cell surface CD4 (Pugach

et al., 2004). Other changes in V2 have also been implicated

in affecting CD4 and coreceptor binding (Platt et al., 2005;

Walter et al., 2005), and the V1/V2 structure is thought to

functionally interact with V3 (Carrillo and Ratner, 1996). In

addition, the V1/V2 structure from one gp120 subunit in the

native Env trimer may physically interact with V3 from a

neighboring subunit (Chen et al., 2005). Thus, it seems

reasonable to assume that the V2 changes in the SCH-D-

resistant virus could have a significant impact on the

receptor interactions of Env, a hypothesis that is now under

investigation.

The C3 changes G354P and N355D are also unique to

the SCH-D-resistant clone D1/85.16 compared to CC1/85

clones (Fig. 7). However, these changes are present in 2/9

clones from the CC2/86 isolate and 1/12 clones from the

CC12/86 isolate, which are R5X4 viruses generated from

the same individual as CC1/85, but at later time points

(Kuhmann et al., 2004). The region containing these

changes is present in the structures of the CD4-liganded

HIV-1 gp120 core and the unliganded SIV gp120 core

(Chen et al., 2005; Kwong et al., 1998). This region is

located in the surface exposed loop E, proximal to the V4

loop and the virion surface (Chen et al., 2005; Kwong et al.,

1998). Whether changes in this region could affect

coreceptor binding, either CCR5 vs. CXCR4 or different

usage of CCR5, is not clear, but it is now under

investigation.

The changes outside of V3 that might influence the SCH-

D resistance of the D101.12 clone are striking by their

scarcity. Within gp120, only one difference between the

SCH-D-resistant D101.12 clone and the clone from the

input CC101.6 virus was unique to this study (i.e., had not

previously been observed). This change, N406K, is found in

the V4 loop and results in the deletion of a potential N-

linked glycosylation site. Of note is that a similar loss of a

glycosylation site in the C-terminal portion of the V4 loop in

the JR-CSF isolate has recently been shown to be associated

with the more efficient use of a CCR5 mutant from which

the N-terminus was deleted (Platt et al., 2005). Relative to

the parental CC1/85 virus, CC101.6 has an H308P

substitution in V3, a change maintained in the D101.12

SCH-D resistant clone. This substitution has been shown to

confer partial resistance to AD101 (Kuhmann et al., 2004),

and the CC101.6 isolate has a similar level of resistance to

SCH-D (Fig. 2 and Table 2). It is possible that the loss of the

V4 glycan acts in concert with P308 to confer the high level

SCH-D resistance of the D101.12 isolate and clone, perhaps

by altering how the resistant virus interacts with CCR5

(Platt et al., 2005). Again, this hypothesis is now under

investigation.

The finding that V3 changes are not involved in the

resistance of the tested D1/85.16 and D101.12 clones was

unexpected given that we had used the same primary isolate,

CC1/85, in both the AD101 and SCH-D resistance experi-

ments, and that both these small molecule CCR5 inhibitors

are of comparable size and potency. However, in each

escape study, we used PBMC from different donors for each

A.J. Marozsan et al. / Virology 338 (2005) 182–199 195

passage, which creates a source of variation even when the

same input virus and similar inhibitors are studied. This

factor might account for why we observed the emergence of

some viruses able to use CXCR4 in some cell lines in the

present study, but not in those where the selection pressure

was from AD101 or SCH-C.

It appears, then, that HIV-1 can follow multiple genetic

pathways to escape from small molecule CCR5 inhibitors.

As we have noted previously, the plasticity of the HIV-1

envelope glycoproteins adds an extra dimension to the drug-

resistance process, compared to what has been experienced

with RT and protease inhibitors (Kuhmann et al., 2004;

Moore and Doms, 2003). Another important variable may

be donor-dependent differences in the expression of CCR5

isoforms. Quasispecies variation in HIV-1-infected individ-

uals creates the potential for even more complexity. An

additional factor to consider in vivo is the presence of

humoral immunity, which also acts against the viral

envelope glycoproteins. Any virus resistant to CCR5

inhibitors, indeed to any entry inhibitor, that emerges in

vivo must also retain its resistance to neutralizing anti-

bodies. Hence, some escape pathways that are favored in

vitro may be disfavored in vivo. We should expect the

unexpected when studying escape from CCR5 inhibitors

under clinical conditions.

Methods

Reagents

The small molecule CCR5 inhibitors, AD101, SCH-C

and SCH-D, have all been described previously and were

synthesized by Schering-Plough Research Institute (Kenil-

worth, NJ) (Strizki et al., 2001; Tagat et al., 2004; Trkola et

al., 2002). SCH-D has is also known as SCH-417690, but

we have retained the SCH-D nomenclature in this paper, for

clarity and convenience. The small molecule CCR5

inhibitor CMPD 167 was a gift from Dr. Marty Springer

(Merck Research Labs, Rahway, NJ) (Veazey et al., 2003;

Wolinsky et al., 2004). The CC-chemokine RANTES was

purchased from RD Systems (Minneapolis, MN). The

chemically modified chemokines PSC-RANTES and

AOP-RANTES were donated by Dr. Oliver Hartley (Centre

Medicale Universitaire, Geneva) (Hartley et al., 2004;

Kawamura et al., 2004; Simmons et al., 1997). The CD4-

IgG2 (PRO 542) molecule (Aarons et al., 2001), the murine

anti-CCR5 MAb PA14 (Olson et al., 1999) and the gp41-

based peptides T20 and T1249 were all gifts from Dr.

William Olson (Progenics Pharmaceuticals Inc, Tarrytown,

NY). The murine anti-CCR5 MAbs 2D7, and 45523 and

45531, were purchased from Pharmingen (San Diego, CA)

and from RD Systems Inc., respectively. The following

reagents were obtained through the NIH AIDS Research and

Reference Reagent Program, Division of AIDS, NIAID,

NIH: the PI, indinavir sulfate; the non-nucleoside reverse

transcriptase inhibitor (NNRTI) nevirapine; the nucleoside

reverse transcriptase inhibitor (NRTI) zidovudine; the

integrase inhibitor 118-D-24; the CXCR4 inhibitor

AMD3100.

Selection of escape mutants in PBMC and assessment of

HIV-1 replication in primary cells

The procedure used to generate SCH-D escape mutants

was closely based on that used to make a virus resistant to

AD101, an earlier generation small molecule CCR5

inhibitor (Kuhmann et al., 2004; Trkola et al., 2002). Two

separate cultures were initiated and carried out simulta-

neously, the input viruses being either CC1/85, an R5

primary isolate (Connor et al., 1993; Trkola et al., 2002), or

CC101.6, a virus isolated from the passage 6 culture of

CC1/85 with AD101 (Kuhmann et al., 2004; Trkola et al.,

2002). CC101.6 is partially (3-fold) resistant to AD101 (and

also to SCH-D, see below). In both cases, passage control

cultures were also established in which the cells and viruses

were treated exactly as in the cultures containing SCH-D,

but without the addition of any inhibitor. Our nomenclature

for isolates and env clones is based on the parental virus, the

selecting drug used and the passage history. Thus, CC1/85

and CC101.6 are the parental isolates, and D1/85.XX or

D101.XX refer to the isolates made after XX passages under

selection pressure from SCH-D. Similarly, D1/85.PC.XX

and D101.PC.XX refer to the control viruses that were

cultured for XX passages in the same PBMC, but in the

absence of SCH-D (Table 1).

In brief, CC1/85 or CC101.6 (1000 TCID50/ml) was

added to 20 ml of PBMC (2 � 106/ml), prepared from

leukopacks pooled from the blood of 4 healthy volunteers,

with sufficient SCH-D (500 nM) to cause >90% inhibition.

The pooled PBMC were depleted of CD8+ T-cells,

fractionated on a Ficoll density gradient, then mitogen-

activated for 3 days, as described previously (Kuhmann et

al., 2004; Trkola et al., 2002). The use of a PBMC pool

minimizes the influence of donor-dependent variation in

CCR5 expression (or in other parameters) on the rate of

HIV-1 replication and, conceivably, on how escape mutants

develop. The CD8 + T-cell-depleted PBMC preparations

are referred to hereafter as primary CD4+ T-cells. The

cultures were passaged weekly, maintaining a constant

density of cells throughout the experiment. The production

of p24 antigen was monitored to ensure that virus

replication was occurring and to determine the extent of

inhibition by SCH-D. Because PBMC from different

donors were used at each passage, p24 production varies

in both the SCH-D-treated and control cultures. SCH-D

(500 nM) was first added after passage 1, then its

concentration was escalated over sequential passages when

viral replication began to increase in the inhibitor-treated

cultures (Trkola et al., 2002). Culture supernatants were

frozen periodically for later inhibitor-sensitivity and cor-

eceptor-utilization studies.

A.J. Marozsan et al. / Virology 338 (2005) 182–199196

The assessment of HIV-1 sensitivity to SCH-D and other

replication inhibitors was performed using primary CD4+ T-

cells as described previously, the assay endpoint being the

production of p24 antigen after 7 and 14 days of culture

(Kuhmann et al., 2004; Pugach et al., 2004). HIV-1

replication in PBMC from CCR5-D32 homozygous donors

(or from CCR5 wild type donors in comparative studies)

was measured as described elsewhere (Kuhmann et al.,

2004; Trkola et al., 2002). Note that these preparations were

not depleted of CD8+ T-cells.

HIV-1 replication in cell lines

The U87.CD4.CCR5 and U87.CD4.CXCR4 cell lines

were obtained through the AIDS Research and Reference

Reagent Program, Division of AIDS, NIAID, NIH: from Dr.

HongKui Deng and Dr. Dan R. Littman (Bjorndal et al.,

1997). The U87 cells were cultured in Dulbecco’s Mod-

ification of Eagle’s Medium (DMEM) with 10% fetal

bovine serum (FBS), 300 Ag/ml G418, 1 Ag/ml puromycin

and penicillin/streptomycin.

GHOST cell lines were obtained through the AIDS

Research and Reference Reagent Program, Division of

AIDS, NIAID, NIH: from Dr. Vineet N. KewalRamani and

Dr. Dan R. Littman (Morner et al., 1999). GHOST cells

were cultured in DMEM with 10% FBS, 500 Ag/ml G418, 1

Ag/ml Puromycin, 100 Ag/ml hygromycin and penicillin/

streptomycin.

Cells from the T-lymphoblast line MT4 were a gift from

Dr. Robert Doms (University of Pennsylvania) and were

cultured in RPMI 1640 medium supplemented with l-

glutamine, 10% FBS and penicillin/streptomycin (Larder

et al., 1989).

HIV-1 replication in the various cell lines was measured

essentially as described previously (Kuhmann et al., 2004;

Trkola et al., 2002). The assay endpoint was the measure-

ment of p24 antigen concentration using an in-house

immunoassay (Kuhmann et al., 2004; Trkola et al., 2002).

Flow cytometry analysis of CCR5 expression

Primary CD4+ T-cells were prepared as described above;

cells from the 4 donors were then pooled and incubated for

6 more days. The cells were harvested, washed with PBS

and incubated for 1 h at 37-C with or without RANTES

(640 nM), AOP-RANTES (100 nM) or PSC-RANTES

(100 nM). After washing with ice-cold PBS, the cells were

fixed in a PBS solution containing 2% paraformaldehyde,

60 mM sucrose, pH 7.4, for 15 min at room temperature.

Fixation is necessary for MAb 45523 to bind to cell surface

CCR5 on primary CD4+ T-cells. In some experiments, cells

were fixed and permeabilized simultaneously by the

addition of 0.05% saponin during the preceding incubation.

The fixed cells were washed twice with PBS containing 20

mM glycine (buffer A), then incubated for 15 min at room

temperature in buffer A supplemented with 1% BSA and

0.05% NaN3 (buffer B). A 50-Al volume of buffer B

containing an anti-CCR5 MAb, either APC-conjugated

2D7 (10 Al/106 cells), FITC-labeled 45531 (20 Ag/ml) or

FITC-labeled 45523 (20 Ag/ml), was then added for 1 h at

room temperature. In some experiments, the cells were

incubated for 45 min in 50 Al buffer B containing rabbit

anti-tubulin (AB935, Chemicon International, Temecula,

CA), washed three times in buffer B and incubated a

further 45 min in 50 Al buffer B containing FITC-

conjugated anti-rabbit IgG (Jackson ImmunoResearch,

West Grove, PA). After 3 washes in buffer B, the cells

were resuspended for analysis using a BD LSR II flow

cytometer (BD Biosciences; San Jose, CA). The parameters

used to select cell populations for analysis were forward-

and side-light scatter, with a total of 10,000 events being

collected for analysis.

Env chimeras and V3 sequence analysis

The cloning of full-length D1/85.PC.16, D1/85.16,

D101.PC.12 and D101.12 env genes from infected PBMC,

and a partial analysis of their sequences, were accomplished

by using a newly developed yeast recombination system that

offers several advantages over previous methods (Marozsan

and Arts, 2003). Briefly, the env gene was PCR-amplified

from the respective virus isolates using the Env-Start and

Env-End primers (Marozsan and Arts, 2003). The product

was then recombined into the pRec D env vector as described

previously (Marozsan and Arts, 2003; Marozsan et al.,

2005). The V3 regions of env clones of the parental, passage

control and SCH-D-resistant viruses were sequenced at the

Cornell Biotechnology Resource Center, making use of

primers CC01 (TTAACCCCACTCTGTGTTAC), CC02

(TAAACCTACCAAGCCTCC), CC03 (CAGTCTAGCA-

GAAGAAGAGG) and CC04 (TGCGCCCATAGTGCTTC).

Acknowledgments

We thank Dr. Robert Doms for providing MT4 cells, Dr.

Ruth Connor for providing the CC1/85 and CC2/86 HIV-1

isolates, Drs. Jim Hoxie and Nelson Michael for blood from

CCR5D32/D 32 homozygous donors, and Dr. William

Olson for providing PA14, T20 and T1249. We are also

grateful to Dr. Oliver Hartley for AOP- and PSC-RANTES.

We thank Tom Ketas, Ashley Palmer, Christopher Kibler,

Janice Riley and Kudzai Chikwava for technical support.

Finally, we thank Dr. Per Johan Klasse for his helpful

critical comments. This work was supported by NIH grant

R01 AI41420 (JPM), NIH Immunology training grant T32

AI07621 (AJM), NIH Fellowship grant F32 AI062664

(SEK), and by Schering Plough Research Institute. JPM is a

Stavros S. Niarchos Scholar. The Department of Micro-

biology and Immunology at the Weill Medical College

gratefully acknowledges the support of the William Ran-

dolph Hearst Foundation.

A.J. Marozsan et al. / Virology 338 (2005) 182–199 197

References

Aarons, E.J., Beddows, S., Willingham, T., Wu, L., Koup, R.A., 2001.

Adaptation to blockade of human immunodeficiency virus type 1 entry

imposed by the anti-CCR5 monoclonal antibody 2D7. Virology 287 (2),

382–390.

Baldwin, C.E., Sanders, R.W., Berkhout, B., 2003. Inhibiting HIV-1 entry

with fusion inhibitors. Curr. Med. Chem. 10 (17), 1633–1642.

Baribaud, F., Edwards, T.G., Sharron, M., Brelot, A., Heveker, N., Price,

K., Mortari, F., Alizon, M., Tsang, M., Doms, R.W., 2001. Antigeni-

cally distinct conformations of CXCR4. J. Virol. 75 (19), 8957–8967.

Billick, E., Seibert, C., Pugach, P., Ketas, T., Trkola, A., Endres, M.J.,

Murgolo, N.J., Coates, E., Reyes, G.R., Baroudy, B.M., Sakmar, T.P.,

Moore, J.P., Kuhmann, S.E., 2004. The differential sensitivity of human

and rhesus macaque CCR5 to small-molecule inhibitors of human

immunodeficiency virus type 1 entry is explained by a single amino

acid difference and suggests a mechanism of action for these inhibitors.

J. Virol. 78 (8), 4134–4144.

Bjorndal, A., Deng, H., Jansson, M., Fiore, J.R., Colognesi, C., Karlsson,

A., Albert, J., Scarlatti, G., Littman, D.R., Fenyo, E.M., 1997.

Coreceptor usage of primary human immunodeficiency virus type 1

isolates varies according to biological phenotype. J. Virol. 71 (10),

7478–7487.

Blanpain, C., Vanderwinden, J.M., Cihak, J., Wittamer, V., Le Poul, E.,

Issafras, H., Stangassinger, M., Vassart, G., Marullo, S., Schlndorff, D.,

Parmentier, M., Mack, M., 2002. Multiple active states and oligome-

rization of CCR5 revealed by functional properties of monoclonal

antibodies. Mol. Biol. Cell 13 (2), 723–737.

Carnec, X., Quan, L., Olson, W.C., Hazan, U., Dragic, T., 2005. Anti-

CXCR4 monoclonal antibodies recognizing overlapping epitopes differ

significantly in their ability to inhibit entry of human immunodeficiency

virus type 1. J. Virol. 79 (3), 1930–1933.

Carrillo, A., Ratner, L., 1996. Cooperative effects of the human

immunodeficiency virus type 1 envelope variable loops V1 and V3 in

mediating infectivity for T cells. J. Virol. 70 (2), 1310–1316.

Chen, B., Vogan, E.M., Gong, H., Skehel, J.J., Wiley, D.C., Harrison, S.C.,

2005. Structure of an unliganded simian immunodeficiency virus gp120

core. Nature 433 (7028), 834–841.

Connor, R.I., Mohri, H., Cao, Y., Ho, D.D., 1993. Increased viral burden

and cytopathicity correlate temporally with CD4+ T-lymphocyte decline

and clinical progression in human immunodeficiency virus type 1-

infected individuals. J. Virol. 67 (4), 1772–1777.

Connor, R.I., Sheridan, K.E., Ceradini, D., Choe, S., Landau, N.R., 1997.

Change in coreceptor use coreceptor use correlates with disease

progression in HIV-1-infected individuals. J. Exp. Med. 185 (4),

621–628.

De Clercq, E., 2003. The bicyclam AMD3100 story. Nat. Rev., Drug

Discov. 2 (7), 581–587.

Hammer, S.M., 2002. HIV drug resistance: implications for management.

Top. HIV Med. 10 (5), 10–15.

Hartley, O., Gaertner, H., Wilken, J., Thompson, D., Fish, R., Ramos, A.,

Pastore, C., Dufour, B., Cerini, F., Melotti, A., Heveker, N., Picard, L.,

Alizon, M., Mosier, D., Kent, S., Offord, R., 2004. Medicinal chemistry

applied to a synthetic protein: development of highly potent HIV entry

inhibitors. Proc. Natl. Acad. Sci. U.S.A. 101 (47), 16460–16465.

Hill, C.M., Kwon, D., Jones, M., Davis, C.B., Marmon, S., Daugherty,

B.L., DeMartino, J.A., Springer, M.S., Unutmaz, D., Littman, D.R.,

1998. The amino terminus of human CCR5 is required for its function

as a receptor for diverse human and simian immunodeficiency virus

envelope glycoproteins. Virology 248 (2), 357–371.

Kawamura, T., Bruse, S.E., Abraha, A., Sugaya, M., Hartley, O., Offord,

R.E., Arts, E.J., Zimmerman, P.A., Blauvelt, A., 2004. PSC-RANTES

blocks R5 human immunodeficiency virus infection of Langerhans cells

isolated from individuals with a variety of CCR5 diplotypes. J. Virol. 78

(14), 7602–7609.

Kazmierski, W., Bifulco, N., Yang, H., Boone, L., DeAnda, F., Watson, C.,

Kenakin, T., 2003. Recent progress in discovery of small-molecule

CCR5 chemokine receptor ligands as HIV-1 inhibitors. Bioorg. Med.

Chem. 11 (13), 2663–2676.

Kenakin, T., 2004. G-protein coupled receptors as allosteric machines.

Recept. Channels 10 (2), 51–60.

Kilby, J.M., Eron, J.J., 2003. Novel therapies based on mechanisms of HIV-

1 cell entry. N. Engl. J. Med. 348 (22), 2228–2238.

Kuhmann, S.E., Pugach, P., Kunstman, K.J., Taylor, J., Stanfield, R.L.,

Snyder, A., Strizki, J.M., Riley, J., Baroudy, B.M., Wilson, I.A., Korber,

B.T., Wolinsky, S.M., Moore, J.P., 2004. Genetic and phenotypic

analyses of human immunodeficiency virus type 1 escape from a small-

molecule CCR5 inhibitor. J. Virol. 78 (6), 2790–2807.

Kwong, P.D., Wyatt, R., Robinson, J., Sweet, R.W., Sodroski, J.,

Hendrickson, W.A., 1998. Structure of an HIV gp120 envelope

glycoprotein in complex with the CD4 receptor and a neutralizing

human antibody. Nature 393 (6686), 648–659.

Lalezari, J.P., Henry, K., O’Hearn, M., Montaner, J.S., Piliero, P.J.,

Trottier, B., Walmsley, S., Cohen, C., Kuritzkes, D.R., Eron Jr., J.J.,

Chung, J., DeMasi, R., Donatacci, L., Drobnes, C., Delehanty, J.,

Salgo, M., 2003. Enfuvirtide, an HIV-1 fusion inhibitor, for drug-

resistant HIV infection in North and South America. N. Engl. J. Med.

348 (22), 2175–2185.

Larder, B., 2001. Mechanisms of HIV-1 drug resistance. Aids 15 (Suppl. 5),

S27–S34.

Larder, B.A., Darby, G., Richman, D.D., 1989. HIV with reduced

sensitivity to zidovudine (AZT) isolated during prolonged therapy.

Science 243 (4899), 1731–1734.

Lazzarin, A., Clotet, B., Cooper, D., Reynes, J., Arasteh, K., Nelson, M.,

Katlama, C., Stellbrink, H.J., Delfraissy, J.F., Lange, J., Huson, L.,

DeMasi, R., Wat, C., Delehanty, J., Drobnes, C., Salgo, M., 2003.

Efficacy of enfuvirtide in patients infected with drug-resistant HIV-1 in

Europe and Australia. N. Engl. J. Med. 348 (22), 2186–2195.

Lee, B., Sharron, M., Blanpain, C., Doranz, B.J., Vakili, J., Setoh, P., Berg,

E., Liu, G., Guy, H.R., Durell, S.R., Parmentier, M., Chang, C.N., Price,

K., Tsang, M., Doms, R.W., 1999. Epitope mapping of CCR5 reveals

multiple conformational states and distinct but overlapping structures

involved in chemokine and coreceptor function. J. Biol. Chem. 274

(14), 9617–9626.

Mack, M., Luckow, B., Nelson, P.J., Cihak, J., Simmons, G., Clapham,

P.R., Signoret, N., Marsh, M., Stangassinger, M., Borlat, F., Wells,

T.N., Schlondorff, D., Proudfoot, A.E., 1998. Aminooxypentane-

RANTES induces CCR5 internalization but inhibits recycling: a novel

inhibitory mechanism of HIV infectivity. J. Exp. Med. 187 (8),

1215–1224.

Maeda, K., Nakata, H., Koh, Y., Miyakawa, T., Ogata, H., Takaoka, Y.,

Shibayama, S., Sagawa, K., Fukushima, D., Moravek, J., Koyanagi, Y.,

Mitsuya, H., 2004a. Spirodiketopiperazine-based CCR5 inhibitor which

preserves CC- chemokine/CCR5 interactions and exerts potent activity

against R5 human immunodeficiency virus type 1 in vitro. J. Virol. 78

(16), 8654–8662.

Maeda, K., Nakata, H., Ogata, H., Koh, Y., Miyakawa, T., Mitsuya, H.,

2004b. The current status of, and challenges in, the development of

CCR5 inhibitors as therapeutics for HIV-1 infection. Curr. Opin.

Pharmacol. 4 (5), 447–452.

Marozsan, A.J., Arts, E.J., 2003. Development of a yeast-based recombi-

nation cloning/system for the analysis of gene products from diverse

human immunodeficiency virus type 1 isolates. J. Virol. Methods 111

(2), 111–120.

Marozsan, A.J., Moore, D., Lobritz, M., Fraundorf, E., Abraha, A., Reeves,

J.D., Arts, E.J., 2005. Differences in the fitness of two diverse wild-type

human immunodeficiency virus type 1 isolates are related to the

efficiency of cell binding and entry. J. Virol. 79, 7121–7134.

McKnight, A., Wilkinson, D., Simmons, G., Talbot, S., Picard, L., Ahuja,

M., Marsh, M., Hoxie, J.A., Clapham, P.R., 1997. Inhibition of human

immunodeficiency virus fusion by a monoclonal antibody to a coreceptor

(CXCR4) is both cell type and virus strain dependent. J. Virol. 71 (2),

1692–1696.

Moore, J.P., Doms, R.W., 2003. The entry of entry inhibitors: a fusion

A.J. Marozsan et al. / Virology 338 (2005) 182–199198

of science and medicine. Proc. Natl. Acad. Sci. U.S.A. 100 (19),

10598–10602.

Moore, J.P., Stevenson, M., 2000. New targets for inhibitors of HIV-1

replication. Nat. Rev., Mol. Cell Biol. 1 (1), 40–49.

Moore, J.P., Kitchen, S.G., Pugach, P., Zack, J.A., 2004. The CCR5 and

CXCR4 coreceptors-central to understanding the transmission and

pathogenesis of human immunodeficiency virus type 1 infection. AIDS

Res. Hum. Retroviruses 20 (1), 111–126.

Morner, A., Bjorndal, A., Albert, J., Kewalramani, V.N., Littman, D.R.,

Inoue, R., Thorstensson, R., Fenyo, E.M., Bjorling, E., 1999. Primary

human immunodeficiency virus type 2 (HIV-2) isolates, like HIV-1

isolates, frequently use CCR5 but show promiscuity in coreceptor

usage. J. Virol. 73 (3), 2343–2349.

O’Hara, B.M., Olson, W.C., 2002. HIV entry inhibitors in clinical

development. Curr. Opin. Pharmacol. 2 (5), 523–528.

Olson, W.C., Rabut, G.E., Nagashima, K.A., Tran, D.N., Anselma, D.J.,

Monard, S.P., Segal, J.P., Thompson, D.A., Kajumo, F., Guo, Y., Moore,

J.P., Maddon, P.J., Dragic, T., 1999. Differential inhibition of human

immunodeficiency virus type 1 fusion, gp120 binding, and CC-

chemokine activity by monoclonal antibodies to CCR5. J. Virol. 73

(5), 4145–4155.

Platt, E.J., Shea, D.M., Rose, P.P., Kabat, D., 2005. Variants of human

immunodeficiency virus type 1 that efficiently use CCR5 lacking the

tyrosine-sulfated amino terminus have adaptive mutations in gp120,

including loss of a functional N-glycan. J. Virol. 79 (7), 4357–4368.

Pugach, P., Kuhmann, S.E., Taylor, J., Marozsan, A.J., Snyder, A., Ketas,

T., Wolinsky, S.M., Korber, B.T., Moore, J.P., 2004. The prolonged

culture of human immunodeficiency virus type 1 in primary lympho-

cytes increases its sensitivity to neutralization by soluble CD4. Virology

321 (1), 8–22.

Reynes, J., Rouzier, R., Kanouni, T., Baillat, V., Baroudy, B., Keung, A.,

Hogan, C., Markowitz, M., Laughlin, M., 2002. 9th Conference on

Retroviruses and Opportunistic Infections, Seattle, WA.

Riley, J., Wojcik, L., Wei, H., Xu, S., Kuhmann, S., Marozsan, A.J.,

Whitcomb, J.M., Petropoulos, C.J., Moore, J.P., Strizki, J.M., submitted

for publication. Genotypic and phenotypic analysis of Human Immu-

nodeficiency Virus Type 1 escape variants generated under the selection

pressure of the CCR5 antagonist SCH-C (SCH-351125). In Vitro.

Sabbe, R., Picchio, G.R., Pastore, C., Chaloin, O., Hartley, O., Offord, R.,

Mosier, D.E., 2001. Donor- and ligand-dependent differences in C-C

chemokine receptor 5 reexpression. J. Virol. 75 (2), 661–671.

Schuitemaker, H., Kootstra, N.A., de Goede, R.E., de Wolf, F., Miedema,

F., Tersmette, M., 1991. Monocytotropic human immunodeficiency

virus type 1 (HIV-1) variants detectable in all stages of HIV-1 infection

lack T-cell line tropism and syncytium-inducing ability in primary T-cell

culture. J. Virol. 65 (1), 356–363.

Schurmann, D., Rouzier, R., Nougarede, R., Reynes, J., Fatkenheuer, G.,

Raffi, F., Michelet, C., Tarral, A., Hoffmann, C., Kiunke, J., Sprenger,

H., vanLier, J., Sansone, A., Jackson, M., Laughlin, M., 2004. 11th

Conference on Retroviruses and Opportunistic Infections, San

Francisco, CA.

Schwarz, M.K., Wells, T.N., 2002. New therapeutics that modulate

chemokine networks. Nat. Rev., Drug Discov. 1 (5), 347–358.

Seibert, C., Sakmar, T.P., 2004. Small-molecule antagonists of CCR5 and

CXCR4: a promising new class of anti-HIV-1 drugs. Curr. Pharm. Des.

10 (17), 2041–2062.

Shen, D.M., Shu, M., Willoughby, C.A., Shah, S., Lynch, C.L., Hale, J.J.,

Mills, S.G., Chapman, K.T., Malkowitz, L., Springer, M.S., Gould,

S.L., DeMartino, J.A., Siciliano, S.J., Lyons, K., Pivnichny, J.V., Kwei,

G.Y., Carella, A., Carver, G., Holmes, K., Schleif, W.A., Danzeisen, R.,

Hazuda, D., Kessler, J., Lineberger, J., Miller, M.D., Emini, E.A., 2004.

Antagonists of human CCR5 receptor containing 4-(pyrazolyl)piper-

idine side chains: Part 2. discovery of potent, selective, and orally

bioavailable compounds. Bioorg. Med. Chem. Lett. 14 (4), 941–945.

Signoret, N., Pelchen-Matthews, A., Mack, M., Proudfoot, A.E., Marsh, M.,

2000. Endocytosis and recycling of the HIV coreceptor CCR5. J. Cell

Biol. 151 (6), 1281–1294.

Simmons, G., Clapham, P.R., Picard, L., Offord, R.E., Rosenkilde, M.M.,

Schwartz, T.W., Buser, R., Wells, T.N., Proudfoot, A.E., 1997. Potent

inhibition of HIV-1 infectivity in macrophages and lymphocytes by a

novel CCR5 antagonist. Science 276 (5310), 276–279.

Strizki, J.M., Xu, S., Wagner, N.E., Wojcik, L., Liu, J., Hou, Y., Endres, M.,

Palani, A., Shapiro, S., Clader, J.W., Greenlee, W.J., Tagat, J.R.,

McCombie, S., Cox, K., Fawzi, A.B., Chou, C.C., Pugliese-Sivo, C.,

Davies, L., Moreno, M.E., Ho, D.D., Trkola, A., Stoddart, C.A., Moore,

J.P., Reyes, G.R., Baroudy, B.M., 2001. SCH-C (SCH 351125), an

orally bioavailable, small molecule antagonist of the chemokine

receptor CCR5, is a potent inhibitor of HIV-1 infection in vitro and in

vivo. Proc. Natl. Acad. Sci. U.S.A. 98 (22), 12718–12723.

Strizki, J.M., Tremblay, C., Xu, S., Wojcik, L., Wagner, N., Gonsiorek,

W., Hipkin, R.W., Chou, C.C., Pugliese-Sivo, C., Xiao, Y., Tagat, J.,

Cox, K., Priestly, T., Sorota, S., Huang, W., Hirsch, M.S., Reyes,

G.R., Baroudy, B.M., submitted for publication. Discovery and

characterization of SCH-417690 (SCH-D), a second generation

CCR5 antagonist.

Tagat, J.R., McCombie, S.W., Steensma, R.W., Lin, S., Nazareno, D.V.,

Baroudy, B., Vantuno, N., Xu, S., Liu, J., 2001. Piperazine-based

CCR5 antagonists as HIV-1 inhibitors: I. 2 (S)-methyl piperazine as a

key pharmacophore element. Bioorg. Med. Chem. Lett. 11 (16),

2143–2146.

Tagat, J.R., McCombie, S.W., Nazareno, D., Labroli, M.A., Xiao, Y.,

Steensma, R.W., Strizki, J.M., Baroudy, B.M., Cox, K., Lachowicz, J.,

Varty, G., Watkins, R., 2004. Piperazine-based CCR5 antagonists as

HIV-1 inhibitors: IV. Discovery of 1-[(4,6-dimethyl-5-pyrimidinyl)

carbonyl]-4-[4-[2-methoxy-1(R)-4-(trifluoromethyl)phenyl]ethyl-3(S)-

methyl-1-piperaz inyl]-4-methylpiperidine (Sch-417690/Sch-D), a

potent, highly selective, and orally bioavailable CCR5 antagonist.

J. Med. Chem. 47 (10), 2405–2408.

Torre, V.S., Marozsan, A.J., Albright, J.L., Collins, K.R., Hartley, O.,

Offord, R.E., Quinones-Mateu, M.E., Arts, E.J., 2000. Variable

sensitivity of CCR5—tropic human immunodeficiency virus type 1

isolates to inhibition by RANTES analogs. J. Virol. 74 (10),

4868–4876.

Trkola, A., Dragic, T., Arthos, J., Binley, J.M., Olson, W.C., Allaway, G.P.,

Cheng- Mayer, C., Robinson, J., Maddon, P.J., Moore, J.P., 1996. CD4-

dependent, antibody-sensitive interactions between HIV-1 and its co-

receptor CCR-5. Nature 384 (6605), 184–187.

Trkola, A., Matthews, J., Gordon, C., Ketas, T., Moore, J.P., 1999. A cell

line-based neutralization assay for primary human immunodeficiency

virus type 1 isolates that use either the CCR5 or the CXCR4 coreceptor.

J. Virol. 73 (11), 8966–8974.

Trkola, A., Kuhmann, S.E., Strizki, J.M., Maxwell, E., Ketas, T., Morgan,

T., Pugach, P., Xu, S., Wojcik, L., Tagat, J., Palani, A., Shapiro, S.,

Clader, J.W., McCombie, S., Reyes, G.R., Baroudy, B.M., Moore, J.P.,

2002. HIV-1 escape from a small molecule, CCR5-specific entry

inhibitor does not involve CXCR4 use. Proc. Natl. Acad. Sci. U.S.A.

99 (1), 395–400.

Tsamis, F., Gavrilov, S., Kajumo, F., Seibert, C., Kuhmann, S., Ketas, T.,

Trkola, A., Palani, A., Clader, J.W., Tagat, J.R., McCombie, S.,

Baroudy, B., Moore, J.P., Sakmar, T.P., Dragic, T., 2003. Analysis of

the mechanism by which the small-molecule CCR5 antagonists SCH-

351125 and SCH-350581 inhibit human immunodeficiency virus type 1

entry. J. Virol. 77 (9), 5201–5208.

Veazey, R.S., Klasse, P.J., Ketas, T.J., Reeves, J.D., Piatak Jr., M.,

Kunstman, K., Kuhmann, S.E., Marx, P.A., Lifson, J.D., Dufour, J.,

Mefford, M., Pandrea, I., Wolinsky, S.M., Doms, R.W., DeMartino,

J.A., Siciliano, S.J., Lyons, K., Springer, M.S., Moore, J.P., 2003. Use

of a small molecule CCR5 inhibitor in macaques to treat simian

immunodeficiency virus infection or prevent simian-human immuno-

deficiency virus infection. J. Exp. Med. 198 (10), 1551–1562.

Walter, B.L., Wehrly, K., Swanstrom, R., Platt, E., Kabat, D., Chesebro, B.,

2005. Role of low CD4 levels in the influence of human immunode-

ficiency virus type 1 envelope v1 and v2 regions on entry and spread in

macrophages. J. Virol. 79 (8), 4828–4837.

A.J. Marozsan et al. / Virology 338 (2005) 182–199 199

Wei, X., Decker, J.M., Liu, H., Zhang, Z., Arani, R.B., Kilby, J.M., Saag,

M.S., Wu, X., Shaw, G.M., Kappes, J.C., 2002. Emergence of resistant

human immunodeficiency virus type 1 in patients receiving fusion

inhibitor (T-20) monotherapy. Antimicrob. Agents Chemother. 46 (6),

1896–1905.

Wolinsky, S.M., Veazey, R.S., Kunstman, K.J., Klasse, P.J., Dufour, J.,

Marozsan, A.J., Springer, M.S., Moore, J.P., 2004. Effect of a CCR5

inhibitor on viral loads in macaques dual-infected with R5 and X4

primate immunodeficiency viruses. Virology 328 (1), 19–29.

Wu, L., Gerard, N.P., Wyatt, R., Choe, H., Parolin, C., Ruffing, N.,

Borsetti, A., Cardoso, A.A., Desjardin, E., Newman, W., Gerard, C.,

Sodroski, J., 1996. CD4-induced interaction of primary HIV-1 gp120

glycoproteins with the chemokine receptor CCR-5. Nature 384 (6605),

179–183.

Xu, S., Wojcik, L., Strizki, J., 2002. 9th Conference on Retroviruses and

Opportunistic Infection, Seattle, WA.

Yi, Y., Shaheen, F., Collman, R.G., 2005. Preferential use of CXCR4 by

R5X4 human immunodeficiency virus type 1 isolates for infection of

primary lymphocytes. J. Virol. 79 (3), 1480–14806.

Zhang, Y.J., Dragic, T., Cao, Y., Kostrikis, L., Kwon, D.S., Littman, D.R.,

KewalRamani, V.N., Moore, J.P., 1998. Use of coreceptors other than

CCR5 by non-syncytium-inducing adult and pediatric isolates of

human immunodeficiency virus type 1 is rare in vitro. J. Virol. 72

(11), 9337–9344.

Zhang, Y., Lou, B., Lal, R.B., Gettie, A., Marx, P.A., Moore, J.P., 2000. Use

of inhibitors to evaluate coreceptor usage by simian and simian/human

immunodeficiency viruses and human immunodeficiency virus type 2

in primary cells. J. Virol. 74 (15), 6893–6910.