R5 to X4 coreceptor switch of human immunodeficiency virus type 1 B’ and Bʼ/C recombinant subtype isolates in China

The chemokine receptors CCR5 and CXCR4 play an important role as coreceptors for human immunodeficiency virus type 1 (HIV-1) entring into cells. HIV-1 isolates can be distinguished by the chemokine coreceptors. Nonsyncytium inducing (NSI), macrophage tropic viruses utilizing CCR5, are called R5 viruses; syncytium inducing (SI) isolates use CXCR4 and known as X4 viruses. R5 viruses generally are associated with latent stage of infection and X4 viruses with later, more severe stages of infection. This phenotypic switch from R5 to X4 correlated with the decreased CD4+ T cell count and the increased viral load (VL).1 Mutations in the V3 region of gp120 of HIV env, specifically at positions 11 and 25, correlated with coreceptor usage and other biological properties.2,3 From the latest two major molecular epidemiological studies of HIV in China, the percentage of people infected with HIV-1 B'/C recombinant subtype increased from 29% in 1996 - 1998 to 47% in 2003, and those infected with HIV-1 subtype B’ decreased from 50% to 33%.4,5 To investigate the difference between two major subtype coreceptor usage, 71 B’ or B'/C recombinant subtype isolates were analyzed by CD4+ T cell count, VL, and V3 sequence variability obtained from HIV-1 seropositive individuals from Shanxi, Anhui and Henan provinces and Xinjiang Vygur Autonomous Region. METHODS Subjects Seventy-one viruses were isolated from 71 HIV-1 seropositive individuals (47 men and 24 women) came from high HIV incidence regions in China (Shanxi n=8, Anhui n=24, Henan n=14, and Xinjiang n=25) between 2004 and 2005. The average age of subjects was 37 years (ranged 27–50 years). Blood specimens were collected and fractionated by centrifugation. CD4+ count and VL were measured and viruses were isolated and cultured. This study was approved by the Institutional Research Ethics Community and all subjects signed the written informed consents before blood collection. Cell lines GHOST cells were derived from the human osteosarcoma cell line, which had been engineered to stably express CD4 and one of the chemokine receptors CCR3, CCR5, CXCR4, Bonzo, or the orphan receptor BOB. The GHOST-R5 and GHOST-X4 cell lines were supplied by the National Institutes of Health AIDS Research and Reference Reagent Program. The cells were maintained for up to 24 passages and then replaced with fresh cells from the cryopreserved stock which had been frozen after the second or third passage.6 Virus isolates Viruses were isolated from patients’ peripheral blood mononuclear cells (PBMCs) by Ficoll-Paque gradient centrifugation (Amersham Biosciences, Sweden), and cocultured with phytohaemagglutinin-stimulated PBMCs from two HIV-1-negative human donors. The cells were maintained in RPMI 1640 medium (HyClone, USA) containing 20 U/ml recombinant interleukin-2 (IL-2, National Institutes of Health, USA). HIV-1 p24 was quantified using a commercial enzyme linked immunosorbent assay kit (Vironostika HIV-1 Microelisa system, bioMérieux, France) once each week, and the cultures were maintained for 4 weeks. Virus culture supernatants with p24>2 ng/ml were stored as separate aliquots in liquid nitrogen until used.7 Viral load and CD4+ count Plasma VL was measured using a HIV-1 nucleotide fluorescence quantitative assay kit (BD Biosciences, USA) with the detection lower limit of 500 copies/ml. TriTEST CD3/CD4/CD45 and CD3/CD8/CD45 three colour direct immunofluorescent reagents with TruCOUNT Tubes (BD Biosciences, USA) followed by flow cytometric analysis were used to establish CD4+ cell counts.8 Coreceptor assay in GHOST cells GHOST cells were seeded in 24-well plates (Corning Inc, Spain) at the density of 6 × 104 cells · well-1 · 0.5 ml-1. On the following day, the monolayers, about 70% confluent, were infected with virus stocks (200 μl/well) in the presence of 8 μg/ml DEAE-dxtran to enhance the infective efficiency. After 4–5 days, cells were harvested and analyzed with a flow cytometer (Elite ESP, Beckman Coulter, Germany) and a total of 10 000 to 15 000 events were scored. We expected an approximately 10 fold shift in mean GFP fluorescence of infected cells over noninfected cell.9 The Ghost-R5 and -X4 cells with SF33 were positive control and the cells without HIV-1 infection were negative control. Sequence analysis DNA was harvested from whole blood of HIV infected patients using a DNA Blood Mini Kit (QIAGEN, Germany). Nested polymerase chain reaction was used to sequence the V3 region using the external primers ED5/ED12 (5′-ATGGGATC AAAGCCTAAAGCC ATGT G-3′/5′-AGTGCTTCCTGCTGCTCCCA-3′) and the internal primers Env-7/Env-8 (5′-CTGTTAAATGGCAG TCTAGC-3′/5′-CACTTCTCCAATTGTCCCTCA-3′). The responsive and cycling parameters were set according to the Takara Ex Taq PCR kit's specification.10 Purified products (Gel Extraction Kit, QIAGEN, USA) were sequenced on an ABI 377 Sequencer (Applied Biosciences) and analyzed using GCG sequence analysis software. Statistical analysis Coreceptor usage difference between subtypes B’ and B'C rececombinant viruses was compared using the Fisher's exact probability. Positive charges of V3 loop results were expressed as mean ± standard deviation (SD) and analyzed using Student's t test. P value less than 0.05 was considered statistically significant. All statistical calculations were completed by the statistical software SPSS11.0. RESULTS CCR5 and CXCR4 coreceptor usage of HIV-1 subtype B’ and B'/C recombinant subtype isolates The env and gag regions of viral sequence from 71 subjects were analyzed and showed that 46 samples were the subtype B’ and 25 were the B'/C recombinant viruses. The cultured GHOST cell lines with and without HIV infection were analyzed by flow cytometry. Among 46 subtype B’ isolates, thirty-one (67.4%) used CCR5 coreceptor for entry, whereas 15 (32.6%) used both CCR5 and CXCR4. All 25 (100%) B'/C recombinant viruses used CCR5. There was a significant difference of coreceptor usage between two groups of B'/C recombinant viruses and subtype B’ viruses (χ2 =10.336, P<0.001). Relationship between virus coreceptor usage and CD4+ count Among 46 subtype B’ isolates, 20 utilized CCR5 and 3 utilized R5/X4 in CD4+ counts >200/μl group (n=23), 5 utilized CCR5 and 4 utilized R5/X4 in CD4+ counts 100–200/μl group (n=9), 7 utilized CCR5 and 7 utilized R5/X4 in CD4+ counts <100/μl group (n=14). The progressive decrease of CD4+ count followed the increase in the proportion of dual tropic in subtype B’ viruses. This did not appear in the B'/C recombinant viral pool; these viruses used CCR5 coreceptor regardless of CD4+ count (Fig.).Fig: . Relationship of CD4+ count and HIV-1 coreceptor usage separated by subtype. The number of dual tropic viruses increase with the decreased CD4+ count in patients infected with subtype B’. Despite a range of CD4+ counts in patients infected with B'/C subtype viruses, coreceptor usage is restricted to CCR5.Relationship between virus coreceptor usage and VL All 11 subtype B’ isolates with VL <103 copies/ml used CCR5 coreceptor for entry into the cell. Of the 11 subtype B’ isolates with VL 103-105 copies/ml, 8 used CCR5 and 3 were dual tropic. Of the 24 subtype B’ isolates with VL >105 copies/ml, 13 used CCR5 and 11 were dual tropic. VL dispersed in different ranges: viruses’ coreceptor switched from R5 to R5/X4 with the increase in VL followed the increase in the proportion of dual tropic subtype B’ viruses. However, the B'/C recombinant viruses used CCR5 coreceptor regardless of VL (n=5 in VL<103 copies/ml group; n=5 in 103-105 copies/ml group; n=15 in >105 copies/ml group, Table).Table: The relationship between virus coreceptor usage and viral load (n(%))Relationship of V3 positive charge and V3 loop tip between HIV-1 subtype B’ and B'/C recombinant viruses Positive charge was calculated by subtracting the number of negatively charged amino acids (aspartic acid (D) and glutamic acid (E)) from the number of positively charged amino acids (arginine (R) and lysine (K)). The formula was (R+K)-(D+E). B'/C recombinant isolates (9) all had the GPGQ sequence at the V3 tip without the mutations increasing positive charge at positions 11 and 25, and overall positive charges ranged from 2 to 4. In 19 subtype B’ isolates, the sequences GPGK (n=4), GPGQ (n=3) and GPGR (n=12) were at the V3 tip, 14 had mutations at positions 11 and 25, and 5 had mutations at position 25. Overall positive charges ranged from 3 to 6. The difference of positive charges between two groups (B'/C recombinant virus 3.2 ± 0.6, subtype B’ virus 4.5± 0.8, P<0.05) is significant. DISCUSSION HIV-1 enters target cells through fusion of the virus envelope with the target cell membrane, which involves sequential interactions between the external envelope glycoprotein gp120, the primary receptor CD4, and one of seven transmembrane chemokine coreceptors.1,2 The chemokine receptor CCR5 acts as a coreceptor for most primary NSI viruses that are transmitted between individuals and predominate early in the natural infection. SI viruses later emerge in infected patients that can utilize the chemokine receptor CXCR4.11 Our research showed 31 were R5 and 15 were dual tropic (R5/X4) among the 46 HIV-1 subtype B’ isolates. The 25 HIV B'/C recombinant isolates collected from Xinjiang were all R5 viruses. In 40%-50% of patients with HIV-1 subtype B’ infection, a phenotypic switch from R5 to X4 appears in the progression to AIDS. This switch is associated with the decrease of CD4+ count and the increase of VL and disease severity.12 Thus, VL and CD4+ count may be used as an indicator of disease progression and the effectiveness of antiretroviral therapies. The present study showed the proportion of dual tropic virus increased with the increased VL and the decreased CD4+ count in patients infected with subtype B’ viruses. However, a stable range of VL and CD4+ counts occurred in patients infected with B'/C subtype viruses. In all 25 B'/C recombinant virus of this study, coreceptor usage was restricted to CCR5. The positively charged amino acids at positions 11 and 25 of V3 loop are believed to play an important role in the switch of viral coreceptors from R5 to X4 virus. The B'/C recombinant subtype viruses, rarely had mutation at positions 11 and 25 of V3, positive charges from 2 to 4 and coreceptor usage was restricted to CCR5. Of the subtype B’ isolates explored, mutations at positions 11 and 25 of V3 seen in most isolates, positive charge ranged from 3 to 6, and dual tropic were seen. The significant difference of positive charges between two groups may imply a difference in the biological characteristics between B'/C recombinant and subtypes B’. Interestingly, the phenotypic switch is rare in HIV-1 subtype C viruses. In a Zimbabwe sample, only 3 out of 22 viral isolates were X4 viruses.13 Out of 31 subtype C infected individuals in Malawi, no X4 viruses were seen.14 A previous study has shown that R5 activates resting CD4+ T lymphocytes more than X4, thus R5 virus was more easily spread than X4.15 The results of this study showed lower V3 sequence variability and lack of coreceptor usage switch in B'/C recombinant subtype; this may lead to virus evading the immune system and spreading faster. In summary, this study showed the frequency of R5 to X4 switch increases with the decrease of CD4+ counts and the increase of VL in subtype B’ isolates. Whereas in B'/C subtype, all isolates used CCR5 coreceptor for viral entry regardless of CD4+ count and VL. The B'/C V3 sequence is highly conserved compared with B’ subtype. This imply B'/C recombinant virus may be fitter and transmit more efficiently than subtype B’ virus. Acknowledgments: We are grateful to the AIDS Research and Reference Reagent Program, NIAID, NIH, for providing GHOST cell lines, and to Dr. Kavin Fang for modifying the manuscript.