SI phenotyping sufficient - CiteSeerX

Co-receptor usage was more predictive than NSI/SI phenotype for HIV replication in macrophages: is NSI/SI phenotyping sufficient? Janet L. Lathey,* Donald Brambilla,† Maureen M. Goodenow,‡ Mostafa Nokta,§ Suraiya Rasheed,|| Edward B. Siwak,¶ James W. Bremer,** Diana D. Huang,** Yanjie Yi,†† Patricia S. Reichelderfer,‡‡ and Ronald G. Collman†,1 *Department of Pediatrics, University of California San Diego, La Jolla, California; †New England Research Institutes, Watertown, Massachusetts; ‡Department of Pathology, University of Florida, Gainsville, Florida; § Department of Internal Medicine, University of Texas, Medical Branch, Galveston, Texas; ||Department of Pathology, University of Southern California, Los Angeles, California; ¶Department of Otorhinolaryngology and Communicative Sciences, Baylor College of Medicine, Houston, Texas; **Department of Immunology/Microbiology, Rush Medical College, Chicago, Illinois; ††Pulmonary, Allergy, and Critical Care Division, University of Pennsylvania, Philadelphia, Pennsylvania; and ‡‡National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland

Abstract: A monocyte-derived macrophage (MDM) culture assay was used to define the replication kinetics of HIV isolates. Ten-day-old MDMs were infected with HIV. Supernatants were collected and assayed for HIV p24 on days 3, 7, 10, and 14 post-infection (PI). In this assay, SF162 (macrophage tropic, NSI) produced increasing amounts of HIV p24 antigen with increasing time in culture. BRU (nonmacrophage tropic, SI) infection resulted in low levels of HIV p24 antigen with no increase in production during the culture period. A panel of 12 clinical isolates was evaluated. All isolates produced detectable levels of HIV p24 antigen in MDMs. However, the NSI viruses had significantly higher log10 HIV p24 antigen values at all times PI (P < 0.01). Co-receptor usage was determined for all 12 isolates (8 NSI and 4 SI). All SI isolates used CXCR4 for entry; two used CXCR4 only, one used CXCR4, CCR5, and CCR3, and one was a mixture of two isolates using CXCR4 and CCR5. None of the NSI viruses used CXCR4 for entry. All used CCR5 as their predominant coreceptor. Of the eight NSI isolates, three used CCR5 only, two used CCR5 and CCR2b, one used CCR5 and CCR3, and one used CCR5, CCR3, and CCR2b. Log10 HIV p24 antigen production on day 14 PI for viruses that used CCR5ⴙCCR3 (3.79 ⴙ 1.40) was greater than for viruses that used CCR5ⴙCCR2b (3.22 ⴙ 1.55) or CCR5 (3.32 ⴙ 1.49), and all were greater than those that used CXCR4 only (1.69 ⴙ 0.28), regardless of SI phenotype (P < 0.05). Thus, in these primary isolates, macrophage tropism and replication kinetics were closely linked to CCR5 utilization, whereas SI capacity was closely linked to CXCR4 utilization. Furthermore, viruses, which could use CCR5 and CCR3 for entry, had a replication advantage in 324

Journal of Leukocyte Biology Volume 68, September 2000

macrophages, regardless of SI phenotype. J. Leukoc. Biol. 68: 324 –330; 2000. Key Words: macrophage tropism co-receptors 䡠 HIV replication kinetics

䡠

viral phenotype

䡠

HIV

INTRODUCTION Host and viral factors during HIV infection play a role in determining disease progression and the way in which an individual responds to anti-retroviral therapy. One viral factor, which has been evaluated, is viral tropism. Historically, HIV isolates have been classified as either syncytium-inducing (SI), T-tropic, or nonsyncytium-inducing (NSI), macrophage tropic. Viruses with SI phenotype generally demonstrated rapid growth kinetics, cytopathicity for T cells, and lacked the ability to grow in macrophages. In contrast, viruses of the NSI phenotype usually grow more slowly, are not cytopathic for T cells, and replicate in macrophages. The presence of an SI phenotype has been associated with disease progression even during anti-retroviral therapy. Conversely, the presence and maintenance of the NSI phenotype has been associated with reduced disease progression and asymptomatic disease (1, 2). However, 50% of adult individuals that progress to AIDS have circulating virus with an NSI phenotype (3, 4), and rapid disease progression can occur in infants and children from whom only NSI viruses were isolated (5). In addition, the assumption has been that a switch from SI back to NSI during antiretroviral treatment would result in a positive response to therapy (i.e., Correspondence: Janet L. Lathey, Ph.D., Pediatric Infectious Diseases, University of California, San Diego, 9500 Gilman Dr. #0672, La Jolla, CA 92093-0672. E-mail: [email protected] 1 For the Macrophage Tropic Viral Kinetic Team (MTVK), Pediatric AIDS Clinical Trials Group (PACTG), NIAID.

http://www.jleukbio.org

reduced chance of disease progression). However, this was not true for ACTG 175. There was no significant difference in disease progression between individuals who maintained the SI phenotype and those that switched from SI back to NSI (6). Thus, it is possible that losing the SI phenotype does not necessarily indicate a switch back to a macrophage-tropic less-cytopathic phenotype. This was demonstrated in PACTG 138, which included four detectable conversions from SI to NSI between baseline and week 56 –132 (7). Of those four NSI isolates, only two demonstrated macrophage tropism. For the remaining two cases, neither the baseline SI nor the posttreatment NSI isolates were able to replicate in macrophages. Thus, a virus conversion back to NSI does not automatically indicate a change in tropism. As a further complication to defining tropism, dual tropic viruses have been isolated. The dual tropic activity of one such virus, 89.6, has been attributed to expanded co-receptor usage (8). Taken together, these data suggest more stringent criteria than SI/NSI may be necessary to define HIV tropism. Here we investigate replication kinetics in macrophages and viral co-receptor usage for their abilities to define viral tropism.

prototype viral strains. At this time, the cells were 95–100% esterase positive. Supernatants were harvested on days 3, 7, 10, and 14 from duplicate wells and assayed for HIV-1 p24 antigen by ELISA. Macrophage cultures were set up in three independent laboratories for SF162/BRU comparison and four laboratories for all other comparisons. The cloned prototype viral strains (SF162 and BRU) were included as controls in all experiments.

Viral phenotype assay Syncytium induction was determined by use of MT-2 cells (9). Viral isolates that replicated in PBMC, as determined by p24 antigen production, and formed syncytia on MT-2 cells were considered to be SI. Viral isolates that replicated in PBMC and did not form syncytia on MT-2 cells were considered NSI.

Co-receptor assay

MATERIALS AND METHODS

Co-receptor usage was determined on the basis of entry into quail QT6 cells that were cotransfected with CD4 and CCR5, CXCR4, CCR2b or CCR3. One day after transfection, cells were infected with 3000 TCID50 of Dnase-treated virus and then lysed 2 days later. Cell lysates were amplified by PCR using primers directed at conserved LTR sequences that detect early viral reverse transcription products, followed by Southern blot with an oligonucleotide probe. Details of this assay, primers, and probe have been described previously (8). Controls included cells transfected with CD4 alone, heat inactivated virus, and prototype HIV-1 strains with established patterns of co-receptor use. Definite co-receptor use was considered viral entry clearly detected in three of three replicate experiments, whereas detection in one or two of three replicates was considered intermediate use.

Viral stocks

Genotype analysis

The prototypic viral strains, HIV-1 SF162 and HIV-1 BRU, were obtained from the AIDS Research and Reference Reagent Program (Ogden BioServices, Rockville, MD). They were biologically cloned by limiting dilution culture: HIV-1 SF162 in macrophages and HIV-1 BRU in lymphocytes by Robert Buckheit at the Frederick Research Center (Frederick, MD). The 12 clinical isolates were selected randomly from a panel of 25 isolates generated during the course of ACTG virology studies using standard methods. The cloned viruses, as well as the 12 clinical isolates, were grown, titered, and phenotyped by the NIAID Virology Quality Assurance Laboratory at Rush Medical College (Chicago, IL).

The V3 loop of the ENV gene was sequenced using either DNA from infected PBMC culture pellets or virion RNA from culture supernatants. DNA from pellets was prepared by cell lysis as described in the DAIDS Virology Manual (9). Sequencing templates were then generated by PCR amplification using nested primers described by Simmonds et al. (10). The outer primers ⫹6957 and ⫺7381 and the inner primers ⫹7009 and ⫺7331 were used. Viral RNA was extracted using the Nuclisens Isolation kit (Organon-Teknika, Durham, NC). The RNA was reversed-transcribed using AMV-RT and random hexamers (Promega, Madison, WI). The cDNA was then amplified using the PE Biosystems (Foster City, CA) XL PCR Kit by kit instructions with 250 pM primers MSF12 and MSR5 (11). A 1.5-kb product was made under the conditions described above for ⫹6957/⫺7381 from these amplicons but using primers env 1 (5⬘TCACAGTHTATTATGGGGTACCTGT) and env 2 (5⬘ATAATTGTCTGGCCTGTACCGTCA). Sequencing templates were then generated using the ⫹6957/⫺7381 and ⫹7009/⫺7331 protocols described above. Sequencing reactions were performed using the ABI Prism Big Dye Terminator Cycle Sequencing Ready Reaction Kit (PE Biosystems) according to kit instructions using ⫹7009 and ⫺7331 as overlapping sequencing primers. Data were analyzed using ABI Prism software, DNA Sequencing Analysis 3.0, Factura 2.0, and Auto Assembler 2.0. Nucleic acid sequences were translated and amino acid sequences aligned and compared.

Viral culture Viral stocks of cloned viruses and viral isolates were grown and titered in peripheral blood mononuclear cells (PBMC). PBMC were separated by FicollHypaque density centrifugation and cultured for 1–3 days in medium (RPMI1640 with glutamine, penicillin [100 U/ml]/streptomycin [100 g/ml], and 20% FBS) containing PHA (2.5 ␮g/ml) and 3% interleukin-2 (IL-2). On day of infection, media were removed and 1 ml of supernatant viral samples were added to 10 million PBMC. After 1 h, 9 ml of medium, which contained 5% IL-2, was added to each sample. Half of the media were replaced on days 3, 10, and 17 PI. On days 7 and 14, 10 million PHA-stimulated PBMC were added and medium volume was doubled (concentration maintained at 1 million cells/ml). Supernatant harvested on day 21 was titered by limiting dilution culture on PHA-stimulated PBMC. TCID50 was calculated by the method of Spearman-Karber.

Macrophage isolation and culture Macrophage cultures were performed according to the consensus protocol of the Pediatric AIDS Clinical Trials Group. PBMC were separated by FicollHypaque density centrifugation. After two washes with PBS, PBMC were suspended at a concentration of 2.5 ⫻ 106/ml in RPMI-1640 with glutamine and penicillin (100 U/ml)/streptomycin (100 g/ml) (medium) and plated at 1 ml per well in 24-well plates. After 1 h, 1 ml of medium containing 20% FBS was added (10% FBS final concentration). After 3 days, half of the medium with 10% FBS was replaced in each well. After 7 days of culture, cells (macrophages) were vigorously washed three times with PBS. Macrophages were cultured and additional 3 days, then infected with either HIV-1 isolates or

Statistical analysis The HIV p24 antigen values were transformed to base 10 logarithms and the transformed duplicate values were averaged before data analysis. Values that were ⬍10 pg/ml were set at 1 pg/ml before log transformation. Two-way analysis of variance (ANOVA) was used to compare average HIV p24 antigen levels at each time point among isolates, controlling for laboratory, and to compare changes from day 3 among isolates, again controlling for laboratory. This approach was also used to compare HIV p24 antigen levels at each time point and changes since day 3 in SI and NSI viruses and in groups of isolates defined by co-receptor use. Interactions between laboratory and isolate, between laboratory and type (SI vs. NSI), and between laboratory and co-receptor class were included in the initial models to determine if results varied over laboratories. No statistically significant variation among laboratories was identified, so the interactions were removed form the final models.

Lathey et al. NSI/SI does not predict macrophage tropism

325

Fig. 1. Replication patterns in macrophages of prototype strains and clinical isolates. A) SF162 vs. BRU, n ⫽ 3 assays. B) NSI clinical isolates (8) vs. SI clinical isolates (4), n ⫽ 4 assays of each isolate. Left, Mean ⫾ SD log10 HIV p24, P value is given in parenthesis for difference between SF162 and BRU or NSI and SI for day 3, 7, 10, and 14 PI. Right, Mean ⫾ SD log10 change from baseline for days 7, 10, 14 (log10 HIV p24 antigen for day 7, 10, or 14 PI log10 HIV p24 antigen for 3). P value is given in parenthesis for difference between SF162 and BRU or NSI and SI for day 7, 10, and 14.

RESULTS Replication of prototypic NSI and SI viral strains in macrophages To determine the replication pattern for typical NSI and SI viruses in macrophages, 12,500 TCID50 of HIV-1 SF162 (NSI) and HIV-1 BRU (SI) were used to infect macrophages in three independent laboratories. Although there was variation among the laboratories in actual log10 HIV p24 antigen values, which resulted in lack of significance, the mean values were higher for SF162 at all time points (Fig. 1). To normalize for the difference in levels of HIV p24 antigen measured among the labs, the change from baseline (day 3) in log10 HIV p24 antigen was calculated for each viral strain (Fig. 1). The change from base line was constant at day 7, 10, and 14 PI for HIV-1 BRU. Conversely, the change from baseline was increasing at each time point for SF162. The difference between the two reached significance by day 14 (P ⫽ 0.05). Thus, the NSI, but not the SI, virus demonstrated increasing replication kinetics in macrophages.

NSI) were evaluated for replication kinetics in macrophages in four independent laboratories. Macrophages were infected with 7500 TCID50 of each isolate and the prototype SI and NSI viruses. Log10 HIV p24 antigen production was measured at 3, 7, 10, and 14 days PI. At all time points, the mean log10 HIV p24 antigen values for the NSI isolates were greater than those for the SI isolates (P ⬍ 0.01) (Fig.1). However, the pattern of log10 HIV p24 antigen production was similar for the SI and NSI viruses with both increasing over time. This was demonstrated by the lack of difference in change from baseline (day 3 PI) for all days PI between the NSI and SI isolates (Fig. 1). In addition, when the mean log10 HIV p24 antigen values for day 14 PI were evaluated for the individual clinical isolates, there was a continuum of HIV p24 antigen levels. All isolates retained some ability to replicate in macrophages with no distinct cutoff between SI and NSI isolates (Table 1). Thus, although there was a difference in the total log10 HIV p24 antigen production between NSI and SI clinical isolates, there was no difference in the growth kinetics in macrophages, nor was there a HIV p24 antigen cutoff that separated NSI from SI isolates.

Replication of NSI versus SI isolates in macrophages

Genetic analysis of isolates

To determine whether clinical isolates behaved like the prototype HIV strains, twelve clinical isolates (four SI and eight

Because the clinical isolates were not cloned, it was possible that the viral stocks contained multiple instead of single iso-

326



TABLE 1.

Virus

BRUe 016 018 004 051 050 008 014 017 005 242 251 245 SF162e

Genotype and Phenotype of Viruses in MTVK Panel

V3 amino acid sequence

Net ⫹ chargea

CTRPNNNTRKSIRIQRGPGRAFVTIGK IGNMRQAHC -I----K-TRR-H- ----S-Y-T EGL--I-K-Y----HK-IKQR-H- ------Y-TKNIR-DI--------S-------N- ------Y-T-QI--DI-----I--- I---RVHM ----YAYATKAI--DI----------------P- ------YAT-DI--DI--------------R-TM ----VYY-T-QI--DI-R--------------N- -------AT-DI--DI--------S-------N- -----IYAT-DI--DI--------S----E--H- ------Y-T-EV--DI----------------H- Q-----YAT-EI--DI----------------HL ------YAT-DI--DI--------S----- -H ------Y G- I--DI----------------HM ---K--Y-T-EV--DI----------------T- ------YAT-DI--DI-----

8 6 6 4 6 3 6 3 3 1 3 3 4 3 3

SI/NSIb

Co-receptor usagec

log10 p24 in MDMd

SI SI SI NSI SI

CXCR4 CXCR4 CXCR4 CCR5 CCR5 [CXCR4]

2.90 1.49 1.90 2.67 2.88

SI NSI NSI NSI NSI NSI NSI NSI NSI

CXCR4 [CCR5, CCR3] CCR5 [CCR2b] CCR5 [CCR2b] CCR5 CCR5 CCR5 [CCR3] CCR5 CCR5 [CCR3, CCR2b] CCR5 [CCR3]

3.06 3.13 3.31 3.34 3.39 3.85 4.06 4.46 5.05

Net positive charge of V3 loops was determined by [K ⫹ R] ⫺ [D ⫹ E]. b Syncytium formation was measured in MT2 cells. NSI, non-syncytium inducing; SI, syncytium inducing. c Co-receptor usage was determined by infection of QT6 cells transfected with CD4 and individual coreceptor cDNAs. d Virus replication in MDM cultures as measured by HIV p24 antigen by ELISA in supernatants on day 14 PI. e Prototype viruses. a

lates. Multiple isolates might explain the inability to differentiate SI from NSI isolates. The envelope regions of the isolates and the prototype viruses were sequenced by PCR (Table1). Of the 12 isolates, only one (#051) was a mixture of two isolates. NSI/SI phenotype was confirmed by net positive charge analysis of V3 loops. All SI viruses had a charge of ⱖ6. The sample with two isolates contained one SI (charge ⫽ ⫹6) and one NSI (charge ⫽ ⫹3). Thus, the lack of difference in replication kinetics between NSI and SI isolates could not totally be explained by the presence of mixed isolates.

Co-receptor usage of isolates

other co-receptor usage (n ⫽ 3; 1 CCR5⫹CCR3, NSI; 1 CCR5⫹CCR2b⫹CCR3, NSI; 1 CCR5⫹CXCR4⫹CCR3, SI). Group 2 included isolates tht use CCR2b and CCR5 (n ⫽ 2; 2 NSI). Group 3 included isolates that used only CCR5 (n ⫽ 4; 4 NSI). Group 4 included isolates that used only CXCR4 (n ⫽ 2; 2 SI). The sample that was a mixture of two isolates was excluded. Isolates that used CCR3 with any combination of co-receptors produced the highest levels of HIV p24 antigen at all time points. There was little to no difference in the levels of HIV p24 antigen produced by isolates that used CCR5 with or without CCR2b. The lowest level of HIV p24 antigen produc-

Multiple co-receptor usage by the isolates might also help explain the lack of difference for replication in macrophages. Co-receptor usage was determined for all 12 isolates and the 2 prototype viruses (Table 1). All NSI viruses used at least CCR5. No NSI isolates used CXCR4 for entry. Use of CXCR4 was exclusively associated with SI viruses. Two SI isolates could also use CCR5 in addition to CXCR4. One of these was the mixed isolate. The other could also use CCR3. The two SI viruses with multiple co- receptor usage did have a higher level of HIV p24 antigen production than the two that used only CXCR4. Thus, multiple co-receptor usage could be associated with the lack of difference in replication kinetics in macrophages between NSI and SI isolates.

Association of co-receptor usage with replication in macrophages It is possible that co-receptor usage is more predictive of viral replication in macrophages than NSI/SI phenotype. To determine whether co-receptor usage was related to replication kinetics in macrophages, the 12 clinical isolates (TCID50 ⫽ 7,500) were used to infect macrophages in four independent laboratories. Supernatants were collected and assayed for HIV p24 antigen on days 3, 7, 10, and 14. Isolates were divided into four groups according to co-receptor usage (Fig. 2). Group 1 included all isolates that use CCR3 regardless of

Fig. 2. Replication patterns in macrophages of co-receptor groups, CCR3 (3) includes CCR3 plus any other receptor, CCR2b (2) includes CCR2b plus CCR5, CCR5 (4) includes only CCR5, and CXCR4 (2) includes only CXCR4, means ⫾ SD are plotted, n ⫽ 4 assays of each isolate. For all time points, the mean log10 HIV p24 antigen values for CXCR4 were significantly different from all other groups, P ⬍ 0.05.

Lathey et al. NSI/SI does not predict macrophage tropism

327

tion was observed from group 4, the isolates that only used CXCR4. The level of p24 production in the CXCR4-only group was significantly lower at all time points than all other groups (P ⬍ 0.05). Thus, isolates that used CCR4 had the least ability to replicate in macrophages, whereas isolates that used CCR3 and CCR5, including the isolate that was SI and used CXCR4, replicated to the highest level.

when clinical isolates of the NSI and SI phenotypes were evaluated. Although NSI isolates had a greater level of production of HIV p24 antigen, isolates of both the NSI and SI phenotype had increasing replication kinetics. Because 11/12 isolates were a single virus population, this did not appear to be related to multiple population viral stocks. Many of the isolates, however, did have the ability to use multiple coreceptors for entry. CXCR4 usage was exclusively associated with SI viruses. However, SI viruses, which could use CCR5 and CCR3 in addition to CXCR4, had a replication advantage in macrophages over viruses that used CXCR4 alone. CCR3using CCR5 viruses also appeared to have a replication advantage over CCR5 viruses that could not use CCR3. Several aspects of the primary isolate macrophage-replication patterns were notable. In contrast to the T-tropic X4 prototype strain BRU, both CXCR4-restricted primary isolates replicated in macrophages, based on HIV p24 antigen levels that increased over time (Table 1 and Figs. 1, 2). This result is consistent with recent studies showing that CXCR4 is a functional co-receptor on macrophages for some primary isolates (12–15). However, the peak antigen levels reached by the CXCR4-restricted strains were modest compared with those produced by isolates that used CCR5, which emphasizes that for these strains CCR5 is a more efficient pathway for macrophage infection than CXCR4. Primary HIV-1 isolates are typically grouped as M-tropic/ NSI versus T-tropic/SI. Among these isolates, however, CXCR4 use was tightly associated with the SI phenotype regardless of other co-receptors used, and CCR5 use was associated with greater replication in macrophages regardless of CXCR4 use. Thus, whereas M-tropic and NSI phenotypes are frequently co-associated, these results show that the phenotypes of SI versus NSI, and efficient versus inefficient macrophage infection, are independently governed by the use of distinct co-receptors and should not be considered reciprocal features of the same phenotype. In contrast, replication in transformed cell lines (the T-tropic phenotype) and syncytia formation in MT-2 cells (the SI phenotype) are both determined by CXCR4 utilization, so these two biological characteristics are both phenotypically and mechanistically linked. A previ-

Effect of CCR3 at varying infectious doses To determine if input viral concentration determined the effect observed for isolates able to use CCR3, macrophages were infected with various TCID50 ranging from 500 to 7,500 in four independent laboratories. Supernatants were collected and assayed for HIV p24 antigen on days 3, 7, 10, and 14. Isolates were divided into four groups depending on TCID50: group 1, 501–748; group 2, 1879 –2007; group 3, 3007– 4765; group 4, 7,500. Groups 1–3 contained one CCR3-using isolate each. HIV p24 antigen levels are shown for day 10 PI (Table 2). This is the point in Figure 2 where the replication curve started to flatten. Data from day 10 PI in Figure 2 is shown for comparison. In each case, the isolates that could use CCR3 produced higher levels of HIV p24 than any isolates in its own group. In addition, the isolates that used only CXCR4 produced the least HIV p24 antigen. Thus, use of CCR3 with CCR5 could predict a high level of HIV p24 antigen production, whereas use of only CXCR4 predicted a low level of HIV p24 antigen production in macrophages.

DISCUSSION The macrophage culture system developed by the Macrophage Tropism Viral Kinetics Team for the Pediatric AIDS Clinical Trials Group was used to examine the relationship of NSI/SI phenotype and replication in macrophages. A prototypic NSI strain HIV-1 SF162 replicates with increasing replication kinetics (increased HIV p24 antigen production over time). Conversely, a prototypic SI strain HIV-1 BRU replicated with a flat kinetic replication curve (HIV p24 antigen did not increase after day 3 PI). The separation, however, was not observed TABLE 2.

Log10 HIV p24 Antigen Production Sorted by TCID

Isolate

Co-receptor

SI/NSI

Log p24a 7,500 TCID

Log 24b Various TCID

TCID50b

51 17 245 16 4 14 50 8 5 242 18 251

CXCR4/CCR5 CCR5 CCR5(2b, 3) CXCR4 CCR5 CCR5(2b) CXCR4(5, 3) CCR5(2b) CCR5 CCR5(3) CXCR4 CCR5

SI NSI NSI SI NSI NSI SI NSI NSI NSI SI NSI

2.69 (1.30) 3.19 (1.65) 4.20 (1.08) 1.37 (1.59) 2.23 (1.87) 2.72 (1.69) 2.97 (1.68) 2.92 (1.61) 3.04 (1.58) 3.50 (1.03) 1.81 (1.36) 3.79 (1.20)

1.04 (1.43) 1.68 (1.52) 2.41 (0.73) 0.55 (1.11) 1.89 (1.33) 1.98 (0.83) 2.59 (0.59) 1.17 (1.09) 1.96 (1.34) 2.55 (0.79) 0.96 (1.66) 1.22 (1.23)

748 501 501 2007 1897 1897 1879 4738 4765 3007 7500 7500

a Log10 HIV p24 antigen of day 10 PI, isolates from Fig. 2, TCID50 ⫽ 7,500. column).

328


b

Log10 HIV p24 antigen of day 10 PI with varying input TCID50 (accompanying


ous report (16) correlates number of basic amino acid substitutions in V3 with SI/NSI phenotype. In our study, SI viruses display a higher net positive charge than NSI viruses. Furthermore, a high V3 charge was closely associated with CXCR4 utilization rather than with lack of CCR5 use, because the mean V3 charge was 6.3 for isolates restricted to CXCR4, 6.0 for CXCR4/CCR5 isolates, and 3 for strains that used CCR5 with or without other non-CXCR4 co-receptors. Our data, then, suggest that the V3 charge is related directly to CXCR4 utilization and that factors in addition to V3 charge distribution may impact CCR5 usage. We found that isolates that used CCR3 in addition to CCR5 had the highest levels of replication in macrophages. However, it is unlikely that these strains’ replication advantage is a result of entry into macrophages through CCR3 in addition to CCR5, because little if any CCR3 is expressed by macrophages (17, 18), and macrophages that lack CCR5 cannot be infected by HIV-1 isolates that use CCR3 (19). In addition, we did not observe any CCR3 expression on the macrophages grown using this protocol (data not shown). Instead, it is more likely that co-utilization of CCR3 is a marker for some other factor such as more efficient CCR5 utilization. Use of CCR3 in addition to CCR5 indicates that a viral envelope can use a range of related co-receptor structures. This might reflect a more “fusogenic” envelope glycoprotein or might be associated with more efficient use of CCR5 if there are conformational variations or post-translational modifications of CCR5 in primary cells that limit use by isolates that are more restricted to specific structure. In this study we observed that NSI/SI phenotype was not an adequate predictor of HIV replication in macrophages or complete co-receptor usage. The range in viral replication patterns we have observed is supported by a previous report by Simmons et al. (20). They described a continuum of relative levels of HIV p24 antigen production from clinical isolates grown in macrophages. In addition, there are clinical situations that have required additional evaluations beyond NSI/SI phenotyping. These include the areas of transmission and disease progression. In one such study by Yu et al. (21) of 12 seroconverters, 3 were thought to be infected by SI viruses based on NSI/SI phenotyping. However, after genotypic and macrophage culture analysis, two of three viruses were found to be dual tropic and the third a mixed population of SI and NSI variants. Thus, NSI/SI phenotyping had given an incomplete picture. Transmission of HIV from mother to infant is an area complicated by viral phenotype issues. Infants are most often infected with NSI viruses. However, a low percentage of mothers with NSI viruses actually transmit HIV to their infants (22). In addition, mothers with SI viruses who transmit usually transmit NSI viruses to their infants. Thus, phenotyping alone is not a good predictor of transmission. In studies that looked at only qualitative growth (replication positive or negative) in macrophages, there were no differences in the ability of viruses from transmitters or nontransmitters, whether SI or NSI, to replicate in macrophages. Most were capable of replication in macrophages (23–25). This is similar to our results observed with the panel of isolates described in this report; all were capable of some level of replication in macrophages. However, when the macrophage replication kinetics of isolates from

transmitters and nontransmitters were evaluated differences were observed. Lathey et al. (23) showed that isolates taken at delivery from transmitters compared wth nontransmitters had increasing replication kinetics in macrophages (0.65 ⫾ 0.21 vs. 0.07 ⫾ 0.11 log10 HIV p24 antigen increase between days 11 and 15 PI, respectively). Of these 80%, four of five of the transmitter isolates had increases of log10 HIV p24 antigen ⱖ 0.5 log10 compared with only 17% (1/12) of nontransmitters. Thus, for mother-to-infant transmission, NSI phenotype or qualitative growth in macrophages could not differentiate transmitters from nontransmitters. However, the replication kinetics in macrophages were predictive of transmission. Viral phenotype issues also complicate disease progression studies. The SI phenotype has been associated with progression to AIDS, however, 50% of individuals can progress to AIDS with only NSI isolates (3, 4). In a study by Blaak et al. (26) isolates from long-term survivors (LTS) and progressors were all NSI isolates, however, they could be differentiated when they were examined for replication kinetics in vitro. Early in the course of infection, viruses from 5/7 LTS and 3/3 progressors had a low replicative rate. Late in infection, only viruses from 4/7 LTS remained with a low replicative rate. However, all viruses, including the high-replicating viruses, remained NSI. Thus, individuals with NSI viruses early in disease retained NSI viruses late in disease, but the replication rate of the NSI viruses had increased. In addition, at the time of isolation of viruses with high replicative rate in vitro, high levels of HIV RNA in serum were observed in vivo. This suggests that changes in replication kinetics of NSI isolates that are observed in vitro may be associated with viral load in vivo. This association could not have been made from simply NSI/SI phenotyping, because all viruses were NSI. In summary, we have shown that NSI/SI phenotyping is not sufficient for the prediction of replication of HIV isolates in macrophages or predicting co-receptor usage of dual tropic isolates. Co-receptor usage is predictive of growth in macrophages with viruses using only CXCR4 having the lowest level of replication. However, there is a range of levels of replication by viruses using CCR5, with the addition of CCR3-increasing viral replication in macrophages. For clinical evaluations, it may be necessary to do some combination of viral growth kinetics and co-receptor analyses in addition to NSI/SI phenotyping.

ACKNOWLEDGMENTS We would like to thank the Virology Committee of the Pediatric AIDS Clinical Trials Group for supporting this project. We would like to acknowledge David Polstra and Tom Giesler of the Virology Quality Assurance Program, Rush Medical College, Susanna Lamers of Gene Genie, and Rodney Trout of UCSD for technical assistance. The following grant support was used in part for this project: SSS-97PVCL01 and SSS-97PICL04 (J. L. L.), NO-AI-35712 and NO-AI-85354 (D. B., J. W. B., D. D. H.), HD-32259, HL-58005, AI-39015 (M. M. G.), Rasheed Research Fund at USC School of Medicine (S. R.), AACTG Immunology Support Lathey et al. NSI/SI does not predict macrophage tropism

329

Laboratory at UTMB (M. N.), Eugene B. Casey Foundation at Baylor College of Medicine (E. B. S.), and AI-35502 (R. G. C.).

REFERENCES 1. Schuitemaker, H., Koot, M., Kootstra, N.A., Dercksen, M. W., de Goode, R. E. Y., Van Strenwijk, R. P., Lange, J. M. A., Eeftink Schattenkerk, J. K. M., Miedema, F., Tersmette, M. (1992) Biological phenotype of human immunodeficiency virus type 1 clones at different stages of infection: progression of disease is associated with a shift from monocytotropic to T-cell-tropic populations. J. Virol. 66, 1354 –1360. 2. Lathey, J. L., Hughes, M. D., Fiscus, S. A., Pi, T., Jackson, J. B., Rasheed, S., Elbeik, T., Reichman, R., Japour, A., D’Aquila, R. T., Scott, W., Griffith, B. P., Hammer, S. M., Katzenstein, D. A., for the AIDS Clinical Trails Group Protocol 175 Team (1998) Variability and prognostic values of virologic and CD4 cell measures in human immunodeficiency virus type 1-infected patients with 200-500 CD4 cells/mm3 (ACTG 175). J. Infect. Dis. 177, 617– 624. 3. Tersmette, M., Lange, J. M. A., de Goode, R. E. Y., de Wolf, F., Eeftink-Schattenkerk, J. K. M., Schellekens, P. Th. A., Coutinho, R. A., Huisman, J. G., Goutsmit, J., Miedema, F. (1989) Association between biological properties of human immunodeficiency virus variants and risk for AIDS and AIDS mortality. Lancet 1, 983–985. . 4. Richman, D. D., Bozzette, S. A. (1994) The impact of syncytium-inducing phenotype of human immunodeficiency virus on disease progression. J. Infect. Dis. 169, 968 –974 . 5. Spencer, L. T., Ogino, M. T., Dankner, W. M., Spector, S. A. (1994) Clinical significance of human immunodeficiency virus type 1 phenotypes in infected children. J. Infect. Dis. 169, 491– 495 . 6. Fiscus, S., Hughes, M. D., Lathey, J. L., Pi, T., Jackson, J. B., Rasheed, S., Elbeik, T., Reichman, R., Japour, A., Byington, R., Scott, W., Griffith, B. P., Katzenstein, D. A., Hammer, S. M. for the AIDS Clinical Trails Group Protocol 175 Team (1998) Changes in virologic markers as predictors of CD4 cell decline and progression of disease in human immunodeficiency virus type 1-infected adults treated with nucleosides. J. Infect. Dis. 177, 625– 633 . 7. Viani, R. M., Smith, I. L., Spector, S. A. (1998) Human immunodeficiency virus type 1 phenotypes in children with advanced disease treated with long-term zalcitabine. J. Infect. Dis. 177, 565–570 . 8. Doranz, B. J., Rucker, J., Yi, Y., Smyth, R. J., Samson, M., Peiper, S. C., Parmentier, M., Collman, R. G., 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, 1149 –1158 . 9. US Department of Health and Human Services (1994) ACTG virology manual for HIV laboratories. NIH Publication 94-3828 SIA-1-SIA-3, Bethesda, MD. 10. Simmonds, P., Balfe, P., Ludlum, C.A., Bishop, J. O., Brown, A. J. L. (1990) Analysis of sequence diversity in hypervariable regions of the external glycoprotein of human immunodeficiency virus type 1. J. Virol. 64, 5840 –5850 . 11. Salminen, M. O., Koch, C., Sanders-Buel, E., Ehrenberg, P. K., Michael, N. L., Carr, J. K., Burke, D. S., McCutchan, F. E. (1995) Recovery of virtually full-length HIV-1 provirus of diverse subtypes from primary virus cultures using the polymerase chain reaction. Virol. 213, 80 – 86 . 12. Yi, Y., Rana, S., Turner, J. D., Gaddis, N., Collman, R. G. (1998) CXCR-4 expressed by primary macrophages and supports CCR5-independent infection by dual-tropic but not T-tropic isolates of human immunodeficiency virus type 1. J. Virol. 72, 772–777 .

330


13. Yi, Y., Isaacs, S. N., Williams, D. A., Frank, I., Schols, D., DeClercq, E., Kolson, D. L., Collman, R. G. (1999) Role of CXCR4 in cell-cell fusion and infection of monocyte-derived macrophages by primary human immunodeficiency virus type 1 (HIV-1) strains: two distinct mechanisms of HIV-1 dual tropism. J. Virol. 73, 7117–7125 . 14. Simmons, G., Reeves, J. D., McKnight, A., Dejucq, N., Hibbitts, S., Power, C. A. Aarons, E., Schols, D., DeClercq, E., Proudfoot, A. E., Clapham, P. R. (1998) CXCR4 as a functional coreceptor for human immunodeficiency virus type 1 infection of primary macrophages. J. Virol. 72, 8453– 8457 . 15. Verani, A.; Pesenti, E.; Polo, S.; Tresoldi, E.; Scarlatti, G.; Lusso, P.; Siccardi, A.G. and Vercelli, D. (1998) CXCR4 is a functional coreceptor for infection of human macrophages by CXCR4-dependent primary HIV-1 isolates. J. Immunol. 161, 2084 –2088 . 16. Milich, L., Margolin, B., Swanstorm, R. (1993) V3 loop of the human immunodeficiency virus type 1 env protein: Interpreting sequence variability. J. Virol. 67, 5623–5634 . 17. Sica, A. Saccani, A., Borsatti, A., Power, C. A., Wells, T. N., Luini, W., Sozzani, S., Mantovani, A. (1997) Bacterial lipopolysaccharide rapidly inhibits expression of C-C chemokine receptors in human monocytes. J. Exp. Med. 185, 969 –974 . 18. Naif, H. M., Li, S., Alali, M., Sloane, A., Wu, L., Kelly, M., Lynch, G., Lloyd, A., Cunningham, A. L (1998) CCR5 expression correlates with susceptibility of maturing monocytes to human immunodeficiency virus type 1 infection. J. Virol. 72, 830 – 836 . 19. Rana, S., Besson, G., Cook, D. G., Rucker, J., Smyth, R. J., Yi, Y., Turner, J. D., Guo, H. H., Du, J. G., Peiper, S. C., Lavi, E., Samson, M., Libert, F., Liesnard, C., Vassart, G., Doms, R. W., Parmentier, M., Collman, R. G. (1997) Role of CCR5 in infection of primary macrophages and lymphocytes by macrophage-tropic strains of human immunodeficiency virus: resistance to patient-derived and prototype isolates resulting from the dccr5 mutation. J. Virol. 71, 3219 –3227 . 20. Simmons, G., Wilkinson, D., Reeves, J. D., Dittmar, M. T., Beddows, S., Weber, J., Carnegie, G., Desselberger, U., Gray, P. W., Weiss, R. A., Clapham, P. R. (1998) Primary, syncytium-inducing human immunodeficiency virus type 1 isolates are dual-tropic and most can use either Lestr or CCR5 as coreceptors for virus entry. J. Virol. 70, 8355– 8360 . 21. Yu, X., Wang, Z., Vlahov, D., Markham, R. B., Farzadegan, H., Margolick, J. B. (1998) Infection with dual-tropic human immunodeficiency virus type 1 variants associated with rapid total T cell decline and disease progression in injection drug users. J. Infect. Dis. 178, 388 –396 . 22. Bryson, Y. J. (1996) Perinatal HIV-1 transmission: recent advances and therapeutic interventions. AIDS 10(S3), S33–S42 . 23. Lathey, J. L., Tsou, J., Brinker, K., Hsia, K., Meyer, W. A., Spector, S. A. (1999) Lack of autologous neutralizing antibody to human immunodeficiency virus type-1 (HIV-1) and macrophage tropism are associated with mother-to-infant transmission. J. Infect. Dis. 180, 344 –350 . 24. Ometto, L., Zanotto, C., Maccabruni, A., Caselli, D., Truscia, D., Giaquinto, C., Ruga, E., Chieco-Bianchi, L., De Rossi, A. (1995) Viral phenotype and host-cell susceptibility to HIV-1 infection as risk factors for mother-to-child HIV-1 transmission. AIDS 9, 427– 434. 25. Kliks, S., Wara, D. W., Landers, D.V., Levy, J. A. (1994) Features of HIV-1 that could influence maternal-child transmission. JAMA 272, 467– 474 . 26. Blaak, H., Brouwer, M., Ran, L. J., de Wolf, F., Schuitemaker, H.(1998) In vitro replication kinetics of human immunodeficiency virus type 1 (HIV-1) variants in relation to virus load in long-term survivors of HIV-1 infection. J. Infect. Dis. 177, 600 – 610.
