Silke Meister, 1 Claas Otto, 1 Armin Papkalla, 1 Mandy Krumbiegel, Stefan Pöhlmann, 2 and Frank Kirchhoff 3

Transcription

1 Virology 284, (2001) doi: /viro , available online at on Basic Amino Acid Residues in the V3 Loop of Simian Immunodeficiency Virus Envelope Alter Viral Coreceptor Tropism and Infectivity but Do Not Allow Efficient Utilization of CXCR4 as Entry Cofactor Silke Meister, 1 Claas Otto, 1 Armin Papkalla, 1 Mandy Krumbiegel, Stefan Pöhlmann, 2 and Frank Kirchhoff 3 Institute for Clinical and Molecular Virology, University of Erlangen-Nürnberg, Schlossgarten 4, Erlangen, Germany Received October 11, 2000; returned to author for revision December 19, 2000; accepted January 30, 2001 In contrast to human immunodeficiency viruses type 1 and type 2 (HIV-1 and HIV-2, respectively), simian immunodeficiency virus (SIVmac) rarely uses CXCR4 (X4) for efficient entry into target cells. Basic amino acid residues in the V3 loop of HIV Env allow efficient coreceptor utilization of X4. Therefore, we investigated if similar changes in the SIVmac Env protein also mediate a coreceptor switch from CCR5 (R5) to X4. Functional analysis revealed that none of eight SIVmac variants, containing V3 regions with an overall charge between 4 and 10, efficiently utilized X4 as entry cofactor. Nonetheless, these alterations had differential effects on SIV coreceptor tropism and on Env expression levels. A single amino acid substitution of L328R, located near the tip of the V3 loop, resulted in grossly reduced Env expression levels and impaired viral infectivity. Notably, additional basic residues restored efficient Env expression and virion incorporation but not infectivity. In comparison to the L328R mutation, changes of P334K and D337K had little disruptive effects on SIVmac entry and replication. Interestingly, mutation of L320K and P321R disrupted coreceptor usage of GPR15 but not R5. These changes also impaired SIVmac replication in peripheral blood mononuclear cells (PBMC) derived from a 32/ 32 donor but not in R5-expressing human or simian PBMC. Our results show that positively charged amino acid residues in the V3 loop affect SIVmac coreceptor tropism and infectivity but do not allow efficient utilization of X Academic Press INTRODUCTION Primate lentiviruses enter target cells by binding the external envelope glycoprotein (gp120) to the cellular CD4 receptor and subsequent interactions with chemokine receptors that mediate fusion between the viral and host cell membrane (reviewed by Doms and Peiper, 1997; Moore et al., 1997; Bieniasz and Cullen, 1998; Unutmaz et al., 1998; Berger et al., 1999). With the unique exception of SIVrcm from red-capped mangabeys, which uses CCR2B (Chen et al., 1998b), primary human and simian immunodeficiency viruses (HIV and SIV, respectively) frequently utilize the CC chemokine receptor CCR5 (R5) as entry cofactor (Alkhatib et al., 1996; Choe et al., 1996; Deng et al., 1996; Doranz et al., 1996; Dragic et al., 1996; Chen et al., 1997, 1998a; Hill et al., 1997; Kirchhoff et al., 1997; Marcon et al., 1997). Some differences between the different groups of primate lentiviruses exist, however, in the ability to use additional coreceptors. CXCR4 (X4) is the second major coreceptor of HIV-1, the 1 These authors contributed equally to this study. 2 Current address: Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA To whom correspondence and reprint requests should be addressed at present address: Abteilung Virologie, Universitätsklinikum, Albert-Einstein-Allee 11, Ulm, Germany. Fax: frank.kirchhoff@medizin.uni-ulm.de. main causative agent of AIDS (Feng et al., 1996). HIV-1 variants that efficiently use X4 evolve in about 50% of all AIDS patients (Van t Wout et al., 1998; Xiao et al., 1998). The appearance of X4-tropic variants is associated with a poor prognosis and might play an important role in AIDS pathogenesis. A number of alternative coreceptors that mediate entry or Env-mediated cell cell fusion have been described (reviewed by Unutmaz et al., 1998). However, efficient utilization of these alternative receptors is rare and replication of HIV-1 in primary cells seems to be mainly mediated by R5 or X4 (Zhang et al., 1998a,b, 1999, 2000). In comparison to HIV-1, the second type of human immunodeficiency virus (HIV-2) is less restricted to CCR5 and CXCR4 and frequently uses additional coreceptors, such as GPR15, STRL33, CCR1, CCR2b, CCR3, CCR8, CCR9, ChemR23, US28, V28, APJ, RDC-1, and BLR1/ CXCR5 for entry into target cells (Bron et al., 1997; Deng et al., 1997; Edinger et al., 1997, 1998; Sol et al., 1997; Owen et al., 1998; Mörner et al., 1999). In contrast to both HIV-1 and HIV-2, even laboratoryadapted strains of SIV derived from infected macaques (SIVmac) use R5, and frequently the orphan receptors GPR15 and STRL33, but rarely X4 for efficient entry (Chen et al., 1997, 1998a; Hill et al., 1997; Kirchhoff et al., 1997; Marcon et al., 1997). The exact role of these orphan receptors in the pathogenesis of SIV remains to be clarified but recent findings suggest that GPR15 and STRL33 usage, in addition to R5, is of limited relevance for SIV /01 $35.00 Copyright 2001 by Academic Press All rights of reproduction in any form reserved.

2 288 MEISTER ET AL. FIG. 1. Amino acid substitutions in the SIVmac V3 region investigated in this study. Dashes indicate identity to the V3 sequence of SIVmac239. Positions corresponding to basic residues (shaded) and an acidic D residue (boxed) in the wild-type V3 region are indicated. The numbers above the alignment specify the amino acid position in the 239 Env. The number of charged amino acid residues and the net charge of the V3 loop sequences are given in the right panel. mac replication in vivo (Pöhlmann et al., 1999, 2000). It is striking that SIVmac rarely uses X4, because (i) macaque- and mangabey-derived X4 is fully functional for HIV entry (Chen et al., 1998a; Marx and Chen, 1998); (ii) X4-tropic SHIVs can replicate efficiently and induce immunodeficiency in infected macaques (Harouse et al., 1999); and (iii) SIVmac is closely related to HIV-2, which frequently utilizes X4, and both originate from naturally infected sooty mangabeys (Gao et al., 1992). Elucidation of biological differences between HIV and SIV is important, because infection of rhesus macaques with SIVmac represents the most commonly used animal model for studies of AIDS pathogenesis and vaccine development (Desrosiers and Ringler, 1989; Johnson and Hirsch, 1992). For HIV-1 it has been well established that isolates containing a V3 loop with a relatively low positive net charge ( 3 to 6) commonly utilize R5, whereas isolates with a more basic V3 region (usually 7 to 9) use X4 with high efficiency (DeJong et al., 1992; ; Shioda et al., 1992; Choe et al., 1996; Cocchi et al., 1996; Bieniasz et al., 1997; Hoffman et al., 1998). Also for HIV-2 it has been shown that an increase in the net positive charge of the V3 region allows efficient utilization of X4 (Isaka et al., 1999). The V3 region is also a determinant of SIVmac cell tropism (Kirchhoff et al., 1994), but the requirements for efficient utilization of X4 are currently unknown. In the present study, we investigated if a more basic V3 loop region in SIVmac Env mediates a coreceptor switch from R5 to X4. We demonstrate that additional basic residues in the V3 loop of SIVmac affect coreceptor tropism and infectivity, but do not allow efficient utilization of X4 for entry or replication. RESULTS SIVmac239 (239wt) is a well-characterized molecular clone that causes AIDS in infected rhesus macaques (Kestler et al., 1990). It replicates well in rhesus peripheral blood mononuclear cells (rpbmc) and in immortalized T-cell lines, but only poorly in rhesus macaque alveolar macrophages (Mori et al., 1992). SIVmac239 efficiently utilizes R5, GPR15, and human-derived STRL33, but not X4 as entry cofactor (Edinger et al., 1997; Chen et al., 1997; Hill et al., 1997; Farzan et al., 1997; Kirchhoff et al., 1997; Marcon et al., 1997; Pöhlmann et al., 2000). Alterations in the V3 loop region of 239wt Env affect viral cell tropism (Kirchhoff et al., 1994). However, changes that mediated efficient replication in macrophages mapped to different regions in Env (Mori et al., 1992). The V3 loop of 239wt encompasses 34 amino acids and has a net charge of 4, resembling that of R5-tropic HIV-1 isolates. We introduced basic amino acid residues at positions 10, 11, 18, 24, and 27 in the V3 region of 239wt gp120 (Fig. 1) to investigate if an increase in basic charge affects SIV coreceptor tropism. These substitutions were introduced at positions similar to those that were previously shown to be associated with efficient utilization of X4 by HIV-1 and HIV-2 strains (Shioda et al., 1992; Choe et al., 1996; Cocchi et al., 1996; Bieniasz et al., 1997; Isaka et al., 1999) and resulted in a net charge between 5 and 10 (Fig. 1). First, we investigated the expression levels of the mutated glycoproteins in 293T cells transfected with proviral 239wt constructs differing only in the env gene. Western blot analysis revealed that most V3 Env variants were efficiently expressed and processed (Fig. 2A, Table 1). A single amino acid substitution of L328R located close to the tip of the loop, however, resulted in strongly reduced gp120 and moderately reduced gp42 expression levels, compared to 239wt transfected cells (Fig. 2A, lane 3). Notably, additional mutations of L320K-P321R (KR) or P334K-D337K (KK) both individually or in combination restored efficient expression of 239Env carrying the L328R substitution (Fig. 2A, lanes 5, 7, 8). The amounts of the 239KR, 239KK, 239KRR, 239KRKK, 239RKK, and 239KRRKK gp120/160 proteins detected in the cellular extracts were similar or only slightly reduced compared to those detected in 239wt transfected cells (Table 1). Thus, with the exception of L328R-Env, all mutated Env proteins were efficiently expressed in transfected 293T cells. To investigate if the mutated Envs were also found in the cell-free virus stocks used for infection, virus con-

3 DETERMINANTS OF SIV CORECEPTOR TROPISM 289 FIG. 2. Detection of viral proteins in 293T cells transfected with proviral SIVmac239 constructs containing wild-type or mutated V3 sequences. Proteins in the cellular extracts (A) or in the cell-free culture supernatant (B) of transiently transfected 293T cells were separated through 10% (top) or 5% (bottom) SDS PAGE and detected by ECL immunoblot using pooled sera derived from SIV-infected rhesus macaques or an antigag serum derived from SIVmac p27 hybridoma cells (B, middle). Virus was harvested and cell lysates were prepared and used for Western blot analysis as described under Materials and Methods. Similar results were obtained in an independent experiment. taining a normalized amount of p27 antigen was harvested from the supernatant of the transfected cells, pelleted, and used for Western blot analysis. Comparable amounts of p27, Nef, p55gag, and p66pol were detected in all cell culture supernatants (Fig. 2B, Table 1). No gp120 was found, however, in the supernatant of 293T cells transfected with the 239L328R construct, even after long overexposure (Fig. 2B, bottom, lane 3). In comparison, the gp42 levels were only about threefold reduced relative to 239wt. The amount of gp120 detected in the supernatant of 239KRKK-transfected cells was also strongly reduced (Fig. 2B, lane 6), despite relatively normal expression levels in the transfected cells (Fig. 2A, lane 6). In contrast, the gp120 levels were comparable to 239wt in the 239KR, 239KRR, and 239RKK virus stocks, Mutant TABLE 1 Effect of Mutations in the SIVmac239 V3 Region on Protein Expression Levels Cellular extract a Supernatant a p27 Nef TM gp120 p27 Nef TM gp KR L328R KK KRR KRKK RKK KRRKK a Viral proteins were detected by ECL immunoblot in the cellular extract of transfected 293T cells or in the culture supernatant (Fig. 2). Signals were analyzed as described under Materials and Methods. Expression levels are given in percentage relative to the amounts of 239wt protein detected. and only slightly reduced in supernatant of cells transfected with the proviral 239KK and 239KRRKK constructs (Table 1). A variety of indicator cells was infected with the 239wt Env variants to evaluate the effect of basic residues in the V3 loop on viral coreceptor tropism and infectivity. As shown in Fig. 3, the 239KR variant entered efficiently into MAGI-R5, U373-R5, and P4R5 cells, which coexpress CD4 and CCR5, but not into GPR15-expressing MAGI cells. Thus, substitution of L320K-P321R (KR) selectively impaired utilization of GPR15. In agreement to absence of detectable gp120 in the virus supernatant (Fig. 2B, lane 3), the 239L328R variant was noninfectious. In contrast, the 239KK mutant, containing changes of P334K and D337K in the C-terminal half of the V3 loop, consistently infected all indicator cells with 239wt-like efficiency. The 239KR and 239KK variants also entered with 239wt-like efficiency into smagi cells that express GPR1, GPR15, and as-yet-unknown entry cofactors (S. Pöhlmann and F. Kirchhoff, unpublished observations) (Fig. 3C). The 239KRR, 239RKK, and 239KRRKK variants showed little if any infectivity, despite efficient Env expression level. The combined changes present in the 239KRKK variant greatly reduced viral infectivity, although the L320K-P321R (KR) and P334K-D337K (KK) substitutions individually had little disruptive effects on R5 usage. Inefficient infection with the 239KRKK mutant is consistent with the low amounts of gp120 detected in the viral stocks (Fig. 2B, lane 6). Nonetheless, significant entry into U373-R5 cells was observed (Fig. 3E), suggesting that the residual KRKK-Env protein was able to mediate infection. Importantly, all functional SIVmac V3 variants entered X4-expressing U373 cells much less efficiently than R5-expressing U373 cells (Fig. 3F).

4 290 MEISTER ET AL. FIG. 3. Basic amino acid residues in the V3 loop affect SIVmac239 infectivity. The indicated reporter cell lines were infected with virus stocks containing 50 ng p27 antigen. Cells were plated on 48-well dishes and infected the next day. After 4 h incubation, the cell culture medium was changed and -galactosidase activities were measured 2 days later as described under Materials and Methods. Viral infectivity is shown relative to 239wt (%). In F entry of 239wt into R5 expressing U373 cells is shown for comparison. U373-X4 cells could readily be infected with X4-tropic HIV isolates (not shown). Values were derived from six independent infections performed with two different virus stocks. The average -galactosidase reporter activities obtained for wild-type infected cells are indicated. c.p.s., counts per second. Coreceptor expression levels are highly important for SIV and HIV entry (Kozak et al., 1997; Rucker et al., 1997; Edinger et al., 1998; Pöhlmann et al., 1999; Sharron et al., 2000). We infected transiently transfected 293T cells coexpressing CD4 and a variety of G-protein coupled receptors (GPCRs) with 239wt variants carrying the luciferase reporter gene to investigate the effect of the V3 changes on viral entry via a broad panel of GPCRs at high expression levels. None of the SIVmac variants, including 239wt, showed appreciable entry into cells coexpressing human or rhesus-derived CD4 and hucxcr4, rhcxcr4, rhccr4, smccr3, rhccr3, smccr8, or rhccr8 (hu, human; rh, rhesus; sm, sooty mangabey; data not shown). Similarly, huapj was used only very inefficiently as SIV entry cofactor (Fig. 4). Only 239wt could utilize rhgpr1 as entry cofactor, albeit with about 10- to 20-fold lower efficiency than hustrl33, rhccr5, or rhgpr15 (Fig. 4). The results obtained with the CCR5- and GPR15-expressing 293T cells were similar to those obtained with the stable transfected indicator cell lines shown in Fig. 3. The 239KK variant entered with high efficiency, compared to the remaining V3 variants, into R5-, GPR15-, and STRL33-expressing cells. The 239KR mutant also utilized R5 and STRL33 but not GPR15 FIG. 4. Entry of SIVmac239 V3 variants into transiently transfected 293T cells coexpressing CD4 and various GPCRs. 293T cells were transfected with plasmids expressing human CD4 and the entry cofactors indicated. At 1 day posttransfection, the cells were detached from the plates, seeded in 48-well dishes, and infected in triplicate with replication-competent SIV carrying the luciferase gene in place of the nef gene. Virus stocks containing normalized amounts of p27 antigen (100 ng) were used for infection. Luciferase activities in the cellular extracts were measured at 3 days postinfection. Results were confirmed in an independent experiment. No significant entry was observed in cells cotransfected with plasmids expressing CD4 and hucxcr4, rhcxcr4, smccr3, rhccr3, smccr8, or rhccr8.

5 DETERMINANTS OF SIV CORECEPTOR TROPISM 291 FIG. 5. Replication of SIVmac239 V3 variants in PM1 and CEMx174 cells. The cells were infected with 5 ng p27 containing virus derived from transiently transfected 293T cells. Virus production was monitored by assays of reverse transcriptase (RT) activity in the culture supernatants at the indicated days postinfection. Similar results were obtained in two independent experiments. PSL, photograph stimulated light emission. for efficient entry. Of the remaining five SIVmac239 V3 variants, only 239KRKK showed appreciable levels of entry into R5-expressing cells (Fig. 4). Thus, all forms containing the L328R changes close to the tip of the loop were defective for infectivity. Next, we investigated the effect of the V3 mutations on 239wt replication in PM1 cells that express both R5 and X4, and in CEMx174 cells that express GPR15 and X4 but not CCR5. Overall, the replication data correlated with the infectivity results shown in Figs. 3 and 4. Only the 239KR and 239KK variants that entered efficiently into R5-expressing cells replicated above background levels in PM1 cells (Fig. 5). However, viral replication was less efficient and delayed compared to 239wt. In agreement with the grossly impaired infectivity for GPR15-expressing MAGI or 293T cells, the 239KR variant did not replicate in CEMx174 cells (Fig. 5). In contrast, the 239KK variant was infectious for CEMx174 cells, although the replication kinetics were significantly delayed compared to 239wt. None of the remaining V3 mutants, 239L328K, 239KRR, 239KRKK, 239RKK, and 239KRRKK were replication-competent in PM1 or CEMx174 cells, even when very high viral doses (up to 200 ng p27 antigen) were used for infection (Fig. 5, and data not shown). Furthermore, no replication was observed in Jurkat cells, which express CD4 and X4 and allow efficient propagation of X4-tropic HIV-1 isolates (data not shown). These results suggest that additional basic amino acid residues in the V3 region have a differential effect on 239wt coreceptor tropism and infectivity but, unlike HIV-1 and HIV-2 Env, do not allow utilization of X4 for efficient entry or replication. To assess the infectivity of the various 239wt mutants for primary cells, human and rhesus macaque PBMC were isolated from whole blood and infected with the different SIVmac239 V3 variants. SIVmac 239wt replicated efficiently in both human- and rhesus-derived cells (Fig. 6). In agreement with the relatively efficient utilization of CCR5 (Figs. 3 and 4), the 239KR and 239KK variants replicated in the PBMC cultures, albeit with delayed kinetics compared to 239wt. Concordant with the inefficient infection of the cell lines, and the inability to replicate in PM1 or CEMx174 cells, none of the re- FIG. 6. Replication of SIVmac239 V3 variants in human and rhesus PBMC. The cells were infected with virus containing 5 ng of p27 antigen. Infections, cell culture, and quantitation of infectivity were performed as described under Materials and Methods. Similar results were obtained in three independent experiments using different virus stocks and PBMC from different donors.

6 292 MEISTER ET AL. FIG. 7. Replication of SIVmac239 V3 variants in PBMC derived from a 32/ 32 donor. Infections were performed with virus stock containing 1, 10, and 100 ng of p27 antigen as indicated in the figure. maining V3 variants showed significant levels of replication. Thus, replication in human and rhesus PBMC correlated with the ability of the SIVmac V3 variants to utilize CCR5 as entry cofactor. In the final set of experiments we investigated the ability of the SIVmac239 V3 variants to replicate in PBMC derived from a homozygous 32/ 32 donor. Similar to the results obtained with PBMC derived from normal donors (Fig. 6), the 239KK variant replicated efficiently, but with delayed kinetics compared to 239wt, in 32/ 32 PBMC (Fig. 7). In contrast, the 239KR variant, which utilized R5 but not GPR15 for efficient entry, did not show significant levels of replication, even when high doses of virus were used for infection (Fig. 7). These findings resemble the results obtained after infection of CEMx174 cells (Fig. 5), which express GPR15 but not R5, and suggest that the orphan receptor GPR15 allows efficient replication of SIVmac in human PBMC. DISCUSSION In this study, we show that an increase in the net positive charge of the SIVmac239 V3 region was not associated with an efficient utilization of X4 as entry cofactor. It has been shown previously that a few amino acid changes in the C-terminal half of the HIV-2 V3 loop determine coreceptor usage of R5 and X4 (Isaka et al., 1999). The SIVmac V3 variants analyzed in our study contain basic residues at positions in the V3 region which are similar to those found to be associated with efficient utilization of X4 by HIV-2. It remains possible that other changes in the V3 region of SIVmac might shift the coreceptor usage. Nonetheless, our results suggest that the requirements for effective coreceptor usage of X4 might differ between SIVmac and HIV-1 or HIV-2 Env proteins. A relatively high number of changes in SIVmac Env might be required for a coreceptor switch from R5 to X4. Thus, the probability that such SIVmac variants emerge in vivo could be low, compared to HIV infection of human individuals. Recently, it has been described that some primary SIV isolates from sooty mangabeys (SIVsm) utilize X4 as a coreceptor for entry (Owen et al., 2000). In contrast to HIV-2 strains, however, all SIVsm isolates analyzed in this study entered X4-expressing indicator cell lines with much lower efficiency than R5-expressing cells. Furthermore, X4-utilizing SIV strains were frequently blocked less effectively than HIV isolates by X4 antagonists (Owen et al., 2000). CXCR4 was not described as a functional coreceptor for primary or laboratory-adapted SIVmac or SIVsm strains in a number of previous studies (Chen et al., 1997; Edinger et al., 1997; Hill et al., 1997; Kirchhoff et al., 1997; Marcon et al., 1997; Schols and De Clerk, 1998). We found that virus reisolated from wt infected macaques at late stages of infection was unable to utilize X4 for replication (S. Pöhlmann, C. Stahl-Hennig, and F. Kirchhoff, unpublished observations). This is consistent with reports that SIV variants, which evolve during the course of infection, show increased replicative capacity and cytopathic properties but not a switch in coreceptor usage from R5 to X4 (Rudensy et al., 1998; Kimata et al., 1999). In agreement with the results of the present study, these findings suggest that the SIVmac Env might interact with specific coreceptors in a manner somewhat different from the HIV-1 and HIV-2 Env proteins. Changes of L320K and P321R in the N-terminal half of the SIVmac V3 loop severely reduced utilization of GPR15 for entry and replication but had much less effect on coreceptor usage of CCR5 and STRL33. In an early study on the function of SIVmac V3 loop, it has been shown that a change of P321S impairs viral replication in CEMx174 cell, which express GPR15, but not in R5-expressing macrophages or rpbmcs (Kirchhoff et al., 1994). Thus, amino acid residues at positions 320 and 321 in SIVmac Env are important for the coreceptor usage of GPR15 but less critical for the interaction with CCR5 and STRL33. Changes of P334K and D337K in the C-terminal half of the SIVmac V3 region did not allow utilization of X4 or

7 DETERMINANTS OF SIV CORECEPTOR TROPISM 293 disrupt coreceptor usage of R5. Similarly, the introduction of positively charged amino acids at positions 333Q 3 K and 336N 3 K of the Env protein of the M-tropic SIVmac316 strain did not impair efficient viral replication in rhesus alveolar macrophages (Kirchhoff et al., 1994). In comparison, it has been described that the C-terminal half of the HIV-2 V3 region determines the coreceptor specificity for X4 and R5 (Isaka et al., 1999). In the case of SIVmac Env, changes in the C-terminal half of the V3 region frequently have little effect on coreceptor tropism. We found, however, that even highly homologous amino acid substitutions in the N-terminal half of the SIVmac V3 loop have differential effects on R5 and GPR15 coreceptor usage (Pöhlmann et al., 1999; S. Pöhlmann and F. Kirchhoff, unpublished observations). Thus, residues located between the potential N-linked glycosylation site and the conserved tip of the SIVmac V3 loop seem to be important for the interaction with specific coreceptors, like GPR15. An unexpected finding of our study was that a single L328R change located at the crown of the V3 loop had differential effects on Env expression levels, depending on the presence or absence of other amino acid substitutions in this region. When the L328R change was introduced into the 239wt Env, it resulted in severely reduced gp120 expression levels. It had only little effect on Env expression levels, however, when additional basic amino acid residues were introduced elsewhere in the SIVmac V3 loop. Strikingly, the L328R substitution seemed to have restorative effects on gp120 virion association when it was present in the 239KRKK background. All SIVmac239 variants containing the L328R substitution were impaired in replication and infectivity. This is consistent with previous observations with both HIV-1 and SIVmac showing that the conserved tip of the V3 loop is critical for infection (Berger et al., 1992; Page et al., 1992; Kirchhoff et al., 1994). In conclusion, we show that changes in the SIVmac V3 region can affect gp120 expression levels as well as viral infectivity and coreceptor tropism. Amino acid residues in the N-terminal half of the loop are critical for the ability to utilize GPR15 as entry cofactors. The exact requirements for the interaction of SIVmac Env with CXCR4 need further clarification. The results suggest, however, that the SIVmac and HIV Env proteins interact with X4 in a somewhat different manner and that basic amino acid residues in the SIVmac V3 region are insufficient to mediate a coreceptor switch from R5 to X4. Mutants MATERIALS AND METHODS The SIVmac239 V3 mutants were generated by splice overlap extension PCR essentially as described previously (Pöhlmann et al., 1999). Briefly, the 239wt env region was amplified using p5outv3 (5 -GAGTCTTGTGA- CAAACA-3 ) and p3outv3 (5 -TCCTCTGCAATTTGTCCA- 3 ) as outer primers, and three pairs of inner primers: (i) L320K-P321R: pk320r53 (5 -GACAGTTAAACGAGTCAC- CATTATG-3 ) and pk320r35 (5 -GAC-TCGTTGAACTGTC TTATTTCCTG-3 ); (ii) L328R: pr328r53 (5 -GTCTGGAC- GAGTT-TTCCACTCACAA-3 ) and pr328r35 (5 -AAA- CTCGTCCAGACATAATGGTG-3 ); and (iii) P334K-D337K: pk334k53 (5 -AAAAGATCAATAAGAGGCCAAAGCAGGC- AT-3 ) and pk334k35 (5 -CTTATTGATCTTTTGTGAGTG- GAAAAC-3 ). Changes compared to the 239wt sequence are shown in bold letters. PCR products were gel purified, mixed, and subjected to a second round of PCR with the outer primers p5outv3 and p3outv3. The unique XhoI and MroI restriction sites flanking the V3 region were used for cloning into a modified full-length 239wt provirus as described previously (Pöhlmann et al., 1999). The 239KR, 239L328R, and 239KK variants were used as templates for additional rounds of SOE-PCR with the primers listed above to generate the 239KRR, 239KRKK, 239RKK, and 239KRRKK mutants (Fig. 1). All PCR-derived inserts were sequenced to ensure that only the intended changes were present. Using standard cloning techniques, the deleted env gene of pbr239 envluc (Kirchhoff et al., 1997) was replaced with the intact 239wt and V3 mutant env genes to generate replication-competent luciferase reporter viruses. Cells CEMx174, PM1, and Jurkat cells were maintained in RPMI 1640 medium supplemented with 10% FCS. smagi cells (Chakerian et al., 1997), MAGI-CCR5, MAGI-GPR15, and 293T cells as well as U373-MAGI-CCR5E and U373- MAGI-CXCR4CEM cells (Vodicka et al., 1997) were grown in DMEM medium supplemented with 10% FCS and antibiotics. Rhesus and human peripheral blood mononuclear cells were isolated using lymphocyte separation medium (Organon Teknika Corporation, Durham, NC) stimulated for 3 days with 2 g/ml phytohemagglutinin and cultured in RPMI 1640 medium with 20% FCS and 100 U/ml IL-2. Supernatants were collected at 3 or 4 day intervals and virus production was measured by reverse transcriptase assay as described (Potts, 1990). Virus stocks For the generation of virus stocks, 293T cells were transfected by the calcium phosphate method with 5 g of the full-length proviral SIVmac239 constructs differing only in the env gene as described previously (Deng et al., 1996; Pöhlmann et al., 1999). After overnight incubation, the medium was changed and virus was harvested 24 h later. Viral stocks were aliquoted and frozen at 80 C. The p27 antigen concentrations of viral stocks were quantitated with an SIV p27 antigen capture ELISA obtained through the National Institutes of Health AIDS Research and Reference Reagent Program.

8 294 MEISTER ET AL. Infections MAGI, U373, and P4R5 reporter cell lines were seeded in 48-well dishes and infected with viral stocks containing 50 ng p27 antigen in a total volume of 0.5 ml medium. After 4 h incubation the cell culture medium was changed. The cells were lysed and the -galactosidase activities were measured using a commercially available kit (Tropix) 2 to 3 days after infection. To determine entry in 293T cells that overexpress CD4 and the various coreceptors, 293T cells were transiently cotransfected with 5 g (each) of the CD4 and the various coreceptor expression vectors. pbabe-puro vectors expressing human and rhesus macaque CCR5 were provided by Nathaniel Landau and Preston Marx through the AIDS Research and Reference Program, Division of AIDS, NIAID, NIH. Expression vectors for human GPR15 and STRL33 were kindly provided by Dan Littman (Skirball Institute for Molecular Medicine, New York, NY). Vectors expressing rhesus macaque GPR15 and STRL33 were generated as described previously (Pöhlmann et al., 2000). After overnight incubation, the medium was changed and the cells were seeded in 48-well dishes. The following day cells were infected with luciferasereporterviruses differing only in the V3 coding region of the env gene. Virus stocks containing 100 ng p27 antigen were used for infection. At 3 days after infection, cells were lysed and the luciferase activities in 20- l cell lysates were determined using a commercially available kit (Promega). Western blot analysis 293T cells were transfected with 10 g DNA of the proviral constructs. After overnight incubation, the medium was replaced by fresh DMEM. At 48 h after transfection, the viral supernatant was removed and the cells were harvested and lysed with lysis buffer (0.5% Nonidet- P40, 0.15 M NaCl, and 50 mm HEPES buffer [ph 7.5]) containing 10 mm NaF and 1 mm PMSF (Sigma Chemicals, St. Louis, MO). Virus stocks containing 460 ng p27 antigen in a total volume of 1 ml medium were used for the detection of viral antigens in the cell-free culture supernatants. Virions were pelleted in a microcentrifuge (14,000 rpm, 1 h, 4 C), washed with PBS, and resuspended in 15 l lysis buffer. Viral proteins were separated through 5 or 10% SDS PAGE and detected by ECL immunoblot using pooled sera from six SIVmac-infected rhesus macaques. For detection of p27 core protein, an antigag serum derived from SIVmac p27 hybridoma cells (55 2F12) was used (Higgins et al., 1992). For ECL detection, horseradish peroxidase-conjugated secondary antibodies were used. Signals were analyzed using the GelDoc system (Bio-Rad), following the protocols provided by the manufacturer. ACKNOWLEDGMENTS A number of reagents were obtained through the AIDS Research and Reference Program, Division of AIDS, NIAID, NIH. We thank Nathaly Finze for expert technical assistance and Bernhard Fleckenstein for constant support and encouragement. We also thank Bryce Chackerian and July Overbaugh for smagi cells. We are also indebted to the people who donated blood for this study, particularly Andrea Höger. This work was supported by grants from the Deutsche Forschungsgemeinschaft, the Johannes and Frieda Marohn Foundation, and the Wilhelm Sander Foundation. REFERENCES Alkhatib, G., Combadiere, C., Broder, C. C., Feng, Y., Kennedy, P. E., Murphy, P. M., and Berger, E. A. (1996). CC-CKR5: A RANTES, MIP- 1alpha, MIP-1beta receptor as a fusion cofactor for macrophagetropic HIV-1. Science 272, Berger, E. A., Sisler, J. R., and Earl, P. L. (1992). Human immunodeficiency virus type 1 envelope glycoprotein molecules containing membrane fusion-impairing mutations in the V3 region efficiently undergo soluble CD4-stimulated gp120 release. J. Virol. 66, Berger, E. A., Murphy, P. M., and Farber, J. M. (1999). Chemokine receptors as HIV-1 coreceptors: Roles in viral entry, tropism, and disease. Annu. Rev. Immunol. 17, Bieniasz, P. D., Fridell, R. A., Aramori, I., Ferguson, S. S. G., Caron, M. G., and Cullen, B. R. (1997). HIV-1 induced cell fusion is mediated by multiple regions within both the viral envelope and the CCR5 coreceptor. EMBO J. 16, Bieniasz, P. D., and Cullen, B. R. (1998). Chemokine receptors and human immunodeficiency virus infection. Front. Biosci. 3, Bron, R., Klasse, P. J., Wilkinson, D., Clapham, P. R., Pelchen-Matthews, A., Power, C., Wells, T. N. C., Kim, J., Peiper, S. C., Hoxie, J. A., and Marsh, M. (1997). Promiscuous use of CC and CXC chemokine receptors in cell-to-cell fusion mediated by a human immunodeficiency virus type 2 envelope protein. J. Virol. 71, Chen, Z., Zhou, P., Ho, D. D., Landau, N. R., and Marx, P. A. (1997). Genetically divergent strains of simian immunodeficiency virus use CCR5 as a coreceptor for entry. J. Virol. 71, Chen, Z., Gettie, A., Ho, D. D., and Marx, P. A. (1998a). Primary SIVsm isolates use the CCR5 coreceptor from sooty mangabeys naturally infected in west Africa: A comparison of coreceptor usage of primary SIVsm, HIV-2, and SIVmac. Virology 246, Chen, Z., Kwon, D., Jin, Z., Monard, S., Telfer, P., Jones, M. S., Lu, C. Y., Aguilar, R. F., Ho, D. D., and Marx, P. A. (1998b). Natural infection of a homozygous delta 24 CCR5 red-capped mangabey with an R2btropic simian immunodeficiency virus. J. Exp. Med. 188, Choe, H., Farzan, M., Sun, Y., Sullivan, N., Rollins, B., Ponath, P. D., Wu, L., Mackay, C. R., LaRosa, G., Newman, W., Gerard, N., Gerard, C., and Sodroski, J. (1996). The beta-chemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates. Cell 85, Cocchi, F., DeVico, A. L., Garzino-Demo, A., Cara, A., Gallo, R. C., and Lusso, P. (1996). The V3 domain of the HIV-1 gp120 envelop glycoprotein is critical for chemokine-mediated blockade of infection. Nat. Med. 2, De Jong, J. J., De Ronde, A., Keulen, W., Tersmette, M., and Goudsmit, J. (1992). Minimal requirements for the human immunodeficiency virus type 1 V3 domain to support the syncytium-inducing phenotype: analysis by single amino acid substitution. J. Virol. 66, Deng, H., Liu, R., Ellmeier, W., Choe, S., Unutmaz, D., Burkhart, M., Di Marzio, P., Marmon, S., Sutton, R. E., Hill, C., Davis, M., Peiper, C. B., Schall, S. C., Littman, D. R., and Landau, N. R. (1996). Identification of a major co-receptor for primary isolates of HIV-1. Nature 381, Deng, H. K., Unutmaz, D., KewalRamani, V. N., and Littman, D. R. (1997).

9 DETERMINANTS OF SIV CORECEPTOR TROPISM 295 Expression cloning of new receptors used by simian and human immunodeficiency viruses. Nature 388, Desrosiers, R. C., and Ringler, D. J. (1989). Use of SIV for AIDS research. Intervirology 30, Doms, R. W., and Peiper, S. C. (1997). Unwelcome guests with master keys: How HIV uses chemokine receptors for cellular entry. Virology 235, Doranz, B. J., Rucker, J., Yi, Y., Smyth, R. J., Samson, M., Peiper, S. C., Parmentier, M., Collman, R. G., and Doms, R. W. (1996). A dual-tropic primary HIV-1 isolate that uses fusin and the beta-chemokine receptors CKR-5, CKR-3, and CKR-2b as fusion cofactors. Cell 85, Dragic, T., Litwin, V., Allaway, G. P., Martin, S. R., Huang, Y., Nagashima, K. A., Cayanan, C., Maddon, P. J., Koup, R. A., Moore, J. P., and Paxton, W. A. (1996). HIV-1 entry into CD4 cells is mediated by the chemokine receptor CC-CKR-5. Nature 381, Edinger, A. L., Amedee, A., Miller, K., Doranz, B. J., Endres, M., Sharron, M., Samson, M., Lu, Z. H., Clements, J. E., Murphey-Corb, M., Peiper, S. C., Parmentier, M., Broder, C. C., and. Doms, R. W. (1997). Differential utilization of CCR5 by macrophage and T cell tropic simian immunodeficiency virus strains. Proc. Natl. Acad. Sci. USA 94, Edinger, A. L., Hoffman, T. L., Sharron, M., Lee, B., O Dowd, B., and Doms, R. W. (1998). Use of GPR1, GPR15, and STRL33 as coreceptors by diverse human immunodeficiency virus type 1 and simian immunodeficiency virus envelope proteins. Virology 249, Farzan, M., Choe, H., Martin, K., Marcon, L., Hofmann, W., Karlsson, G., Sun, Y., Barrett, P., Marchand, N., Sullivan, N., Gerard, N., Gerard, C., and Sodroski, J. (1997). Two orphan seven-transmembrane segment receptors which are expressed in CD4-positive cells support simian immunodeficiency virus infection. J. Exp. Med. 186, Feng, Y., Broder, C. C., Kennedy, P. E., and Berger, E. A. (1996). HIV-1 entry cofactor: Functional cdna cloning of a seven-transmembrane, G protein-coupled receptor. Science 272, Gao, F., Yue, L., White, A. T., Pappas, P. G., Barchue, J., Hanson, A. P., Greene, B. M., Sharp, P. M., Shaw, G. M., and Hahn, B. H. (1992). Human infection by genetically diverse SIVSM-related HIV-2 in west Africa. Nature 358, Harouse, J. M., Gettie, A., Tan, R. C., Blanchard, J., and Cheng-Mayer, C. (1999). Distinct pathogenic sequela in rhesus macaques infected with CCR5 or CXCR4 utilizing SHIVs. Science 284, Higgins, J. R., Sutjipto, S., Marx, P. A., and Pedersen, N. C. (1992). Shared antigenic epitopes of the major core proteins of human and simian immunodeficiency virus isolates. J. Med. Primatol. 21, Hill, C. M., Deng, H., Unutmaz, D., Kewalramani, V. N., Bastiani, L., Gorny, M. K., Zolla-Pazner, S., and Littman, D. R. (1997). Envelope glycoproteins from human immunodeficiency virus types 1 and 2 and simian immunodeficiency virus can use human CCR5 as a coreceptor for viral entry and make direct CD4-dependent interactions with this chemokine receptor. J. Virol. 71, Hoffman, T. L., Stephens, E. B., Narayan, O., and Doms, R. W. (1998). HIV type I envelope determinants for use of the CCR2b, CCR3, STRL33, and APJ coreceptors. Proc. Natl. Acad. Sci. USA 95, Isaka, Y., Sato, A., Miki, S., Kawauchi, S., Sakaida, H., Hori, T., Uchiyama, T., Adachi, A., Hayami, M., Fujiwara, T., and Yoshie, O. (1999). Small amino acid changes in the V3 loop of human immunodeficiency virus type 2 determines the coreceptor usage for CXCR4 and CCR5. Virology 264, Johnson, P. R., and Hirsch, V. M. (1992). SIV infection of macaques as a model for AIDS pathogenesis. Int. Rev. Immunol. 8, Kestler, H. W., Kodama, T., Ringler, D., Marthas, M., Pedersen, N., Lackner, A., Regier, D., Sehgal, P., Daniel, M. D., and Desrosiers, R. C. (1990). Induction of AIDS in rhesus monkeys by molecularly cloned simian immunodeficiency virus. Science 248, Kimata, J. T., Gosink, J. J., KewalRamani, V. N., Rudensey, L. M., Littman, D. R., and Overbaugh, J. (1999). Coreceptor specificity of temporal variants of simian immunodeficiency virus Mne. J. Virol. 73, Kirchhoff, F., Mori, K., and Desrosiers, R. C. (1994). The V3 domain is a determinant of simian immunodeficiency virus cell tropism. J. Virol. 68, Kirchhoff, F., Pöhlmann, S., Hamacher, M., Means, R. E., Kraus, T., Überla, K., and Di Marzio, P. (1997). SIVmac variants with differential T-cell and macrophage tropism use CCR5, and an unidentified cofactor expressed in CEMx174 cells for efficient entry. J. Virol. 71, Kozak, S. L., Platt, E. J., Madani, N., Ferro, F. E., Peden, K., and Kabat. D. (1997). CD4, CXCR-4, and CCR-5 dependencies for infections by primary patient and laboratory-adapted isolates of human immunodeficiency virus type 1. J. Virol. 71, Marcon, L., Choe, H., Martin, K., Farzan, M., Ponath, P. D., Wu, L., Newman, W., Gerard, N., Gerard, C., and Sodroski, J. (1997). Utilization of C-C chemokine receptor 5 by the envelope glycoproteins of a pathogenic simian immunodeficiency virus, SIVmac239. J. Virol. 71, Marx, P. A., and Chen, Z. (1998). The function of simian chemokine receptors in the replication of SIV. Semin. Immunol. 10, Moore, J. P., Trkola, A., and Dragic, T. (1997). Co-receptors for HIV-1 entry. Curr. Opin. Immunol. 9, Mori, K., Ringler, D. J., Kodama, T., and Desrosiers, R. C. (1992). Complex determinants of macrophage tropism in env of simian immunodeficiency virus. J. Virol. 66, Mörner, A., Björndal, A., Albert, J., Kewalramani, V. N., Littman, D. R., Inoue, R., Thorstensson, R., Fenyo, E. M., and 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, Owen, S. M., Ellenberger, D., Rayfield, M., Wiktor, S., Michel, P., Grieco, M. H., Gao, F., Hahn, B. H., and Lal, R. B. (1998). Genetically divergent strains of human immunodeficiency virus type 2 use multiple coreceptors for viral entry. J. Virol. 72, Owen, S. M., Masciotra, S., Novembre, F., Yee, J., Switzer, W. M., Ostyula, M., and Lal, R. B. (2000). Simian immunodeficiency viruses of diverse origin can use CXCR4 as a coreceptor for entry into human cells. J. Virol. 74, Page, K. A., Stearns, S. M., and Littman, D. R. (1992). Analysis of mutations in the V3 domain of gp160 that affect fusion and infectivity. J. Virol. 66, Pöhlmann, S., Stolte, N., Münch, J., Ten Haaft, P., Heeney, J. L., Stahl- Hennig, C., and Kirchhoff, F. (1999). BOB/GPR15 coreceptor usage has no significant effect on replication of simian immunodeficiency virus in vivo. J. Infect. Dis. 180, Pöhlmann, S., Benhur, L., Meister, S., Krumbiegel, M., Leslie, G., Doms, R. W., and Kirchhoff, F. (2000). Simian immunodeficiency virus utilizes human and sooty mangabey but not rhesus macaque STRL33 for efficient entry. J. Virol. 74, Potts, B. J. (1990). Mini reverse transcriptase (RT) assay. In Techniques in HIV Research (Aldovini, A., and Walker, B. D., Eds.), pp Stockton, New York. Rucker, J., Edinger, A. L., Sharron, M., Samson, M., Lee, B., Berson, J. F., Yi, Y., Margulies, B., Collman, R. G., Doranz, B. J., Parmentier, M., and Doms, R. W. (1997). Utilization of chemokine receptors, orphan receptors, and herpesvirus-encoded receptors by diverse human and simian immunodeficiency viruses. J. Virol. 71, Rudensey, L. M., Kimata, J. T., Long, E. M., Chackerian, B., and Overbaugh, J. (1998). Changes in the extracellular envelope glycoprotein of variants that evolve during the course of simian immunodeficiency virus SIVMne infection affect neutralizing antibody recognition, syncytium formation, and macrophage tropism but not replication, cytopathicity, or CCR-5 coreceptor recognition. J. Virol. 72, Schols, D., and De Clercq, E. (1998). The simian immunodeficiency virus mnd(gb-1) strain uses CXCR4, not CCR5, as coreceptor for entry in human cells. J. Gen. Virol. 79,

10 296 MEISTER ET AL. Sharron, M., Pöhlmann, S., Price, K., Lolis, E., Tsang, M., Kirchhoff, F., Doms, R. W., and Lee, B. (2000). Expression and coreceptor activity of STRL33/Bonzo on primary peripheral blood lymphocytes. Blood 96, Shioda, T., Levy, J. A., and Cheng-Meyer, C. (1992). Small amino acid changes in the V3 hypervariable region of gp120 can affect the T-cell-line and macrophage tropism of human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 89, Sol, N., Ferchal, F., Braun, J., Pleskoff, O., Treboute, C., Ansart, I., and Alizon, M. (1997). Usage of the coreceptors CCR-5, CCR-3, and CXCR-4 by primary and cell line-adapted human immunodeficiency virus type 2. J. Virol. 71, Unutmaz, D., KewalRamani, V. N., and Littman, D. R. (1998). G proteincoupled receptors in HIV and SIV entry: New perspectives on lentivirus-host interactions and on the utility of animal models. Semin. Immunol. 10, Van t Wout, A. B., Blaak, H., Ran, L. J., Brouwer, M., Kuiken, C., and Schuitemaker, H. (1998). Evolution of syncytium-inducing and nonsyncytium-inducing biological virus clones in relation to replication kinetics during the course of human immunodeficiency virus type 1 infection. J. Virol. 72, Vodicka, M. A., Goh, W. C., Wu, L. I., Rogel, M. E., Bartz, S.R., Schweickart, V.L., Raport, C. J., and Emerman, M. (1997). Indicator cell lines for detection of primary strains of human and simian immunodeficiency viruses. Virology 233, Xiao, L., Rudolph, D. L., Owen, S. M., Spira, T. J., and Lal, R. B. (1998). Adaptation to promiscuous usage of CC and CXC-chemokine coreceptors in vivo correlates with HIV-1 disease progression. AIDS 12, Zhang, L., He, T., Huang, Y., Chen, Z., Guo, Y., Wu, S., Kunstman, K. J., Brown, R. C., Phair, J. P., Neuman, A. U., Ho, D. D., and Wolinsky, S. M. (1998a). Chemokine coreceptor usage by diverse primary isolates of human immunodeficiency virus type 1. J. Virol. 72, Zhang, Y.-J., Dragic, T., Cao, Y., Kostrikis, L., Kwon, D. S., Littman, D. R., Kewalramani, V. N., and Moore, J. P. (1998b). 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, Zhang, Y.-J., and Moore, J. P. (1999). Will multiple coreceptors need to be targeted by inhibitors of human immunodeficiency virus type 1 entry? J. Virol. 73, Zhang, Y., Lou, B., Lal, R. B., Gettie, A., Marx, P. A., and 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,