Linkages between HIV-1 specificity for CCR5 or CXCR4 and in vitro usage of alternative coreceptors during progressive HIV-1 subtype C infection

Abstract

Background: Human immunodeficiency virus type 1 (HIV-1) subtype C (C-HIV) is spreading rapidly and is now responsible for >50% of HIV-1 infections worldwide, and >95% of infections in southern Africa and central Asia. These regions are burdened with the overwhelming majority of HIV-1 infections, yet we know very little about the pathogenesis of C-HIV. In addition to CCR5 and CXCR4, the HIV-1 envelope glycoproteins (Env) may engage a variety of alternative coreceptors for entry into transfected cells. Whilst alternative coreceptors do not appear to have a broad role in mediating the entry of HIV-1 into primary cells, characterizing patterns of alternative coreceptor usage in vitro can provide valuable insights into mechanisms of Env-coreceptor engagement that may be important for HIV-1 pathogenesis.

Results: Here, we characterized the ability of luciferase reporter viruses pseudotyped with HIV-1 Envs (n = 300) cloned sequentially from plasma of 21 antiretroviral therapy (ART)-naïve subjects experiencing progression from chronic to advanced C-HIV infection over an approximately 3-year period, who either exclusively maintained CCR5-using (R5) variants (n = 20 subjects) or who experienced a coreceptor switch to CXCR4-using (X4) variants (n = 1 subject), to utilize alternative coreceptors for entry. At a population level, CCR5 usage by R5 C-HIV Envs was strongly linked to usage of FPRL1, CCR3 and CCR8 as alternative coreceptors, with the linkages to FPRL1 and CCR3 usage becoming statistically more robust as infection progressed from chronic to advanced stages of disease. In contrast, acquisition of an X4 Env phenotype at advanced infection was accompanied by a dramatic loss of FPRL1 usage. Env mutagenesis studies confirmed a direct link between CCR5 and FPRL1 usage, and showed that the V3 loop crown, but not other V3 determinants of CCR5-specificity, was the principal Env determinant governing the ability of R5 C-HIV Envs from one particular subject to engage FPRL1.

Conclusions: Our results suggest that, in the absence of coreceptor switching, the ability of R5 C-HIV viruses to engage certain alternative coreceptors in vitro, in particular FPRL1, may reflect an altered use of CCR5 that is selected for during progressive C-HIV infection, and which may contribute to C-HIV pathogenicity.

Figure 1

Alternative coreceptor usage by R5 C-HIV Envs. Luciferase reporter viruses pseudotyped with each of the C-HIV R5 Envs (n = 294) were used to infect NP2-CD4 cells expressing CCR5, FPRL1, CCR3 or CCR8, and the levels of HIV-1 entry were determined as described in the Methods. The background level of luciferase activity, as determined by infections with luciferase reporter virus pseudotyped with the non-functional ΔKS Env [44], is shown by the horizontal dashed line. The data shown are the means and standard deviations of entry levels of all the R5 C-HIV Envs in the different cell types, and are representative of 3 independent experiments, each performed in triplicate. The individual dots represent mean entry levels of triplicates from a representative experiment. Statistical comparisons were made using a Mann–Whitney U test. P values < 0.05 were considered statistically significant.

Figure 2

Associations between CCR5 usage and usage of CCR3, FPRL1 and CCR8 by R5 C-HIV Envs. Luciferase reporter viruses pseudotyped with each of the C-HIV R5 Envs (n = 294) were used to infect NP2-CD4 cells expressing CCR5, FPRL1, CCR3 or CCR8, and the levels of HIV-1 entry were determined as described in the Methods. Comparative entry levels for each of the individual Envs are shown in Additional file 1: Table S1. (A), scatterplots of all 294 R5 Envs comparing the levels of HIV-1 entry via CCR5 to that of CCR3 (left), FPRL1 (middle), and CCR8 (right). (B), these data were then stratified according to whether the Envs were cloned from the “enrolment” (top row), “intermediate” (middle row) of “final” (bottom row) plasma samples from the longitudinal cohort [8]. Each dot represents the mean value of triplicate experiments. Statistical analysis of these data is shown in Table 1.

Figure 3

Changes in alternative coreceptor usage during the transition from R5 to X4 phenotype in subject 1109. Luciferase reporter viruses pseudotyped with “enrolment” (R5, n = 5), “intermediate” (R5, n = 4) or “final” (X4, n = 6) Envs from subject 1109 [8] were used to infect NP2-CD4 cells expressing CXCR4, CCR5, FPRL1, CCR3 or CCR8, and the levels of HIV-1 entry were determined as described in the Methods. The dotted lines indicate the limit of detection of coreceptor activity, as determined by infections with luciferase reporter virus pseudotyped with the non-functional ΔKS Env [44]. Comparative entry levels for each of the individual Envs are shown in Additional file 1: Table S1. The data shown are means and standard deviations of entry levels in the different cell types, and are representative of 3 independent experiments, each performed in triplicate. The individual circles represent mean entry levels of each of the individual Env clones that were performed in triplicate, from a representative experiment.

Figure 4

C-HIV Env mutants. Amino acid sequences of the Env mutants, aligned against the gp120 sequence of the X4 1109-F-30 sequence. The sequences and descriptions of these Env mutants have been reported previously [8], and are included again here to assist in the interpretation of the functional data. Dots indicate residues identical to 1109-F-30, dashes indicate gaps. Numbers refer to amino acid positions in the V1 and V3 loop regions. Descriptions of the Env mutants are provided in greater detail in the Methods.

Figure 5

V3 loop alterations important for the association between CCR5 and FPRL usage. (A), luciferase reporter viruses pseudotyped with Env mutants (M1 through M15), or unmodified 1109-E-10 and 1109-F-30 Envs were used to infect NP2-CD4/FPRL1 cells, and the levels of HIV-1 entry were determined as described in the Methods. Open bars indicate Envs where no detectable FPRL1 usage was observed, and shaded bars indicate Envs with detectable FPRL1 usage. The dotted line indicates the limit of detection of coreceptor activity, as determined by infections with luciferase reporter virus pseudotyped with the non-functional ∆KS Env [44]. The results shown are a compilation of 3 independent experiments, each performed in triplicate. The data shown are means, and the error bars represent standard errors of the means. (B), summary of descriptions of the Env mutants. Shaded rows illustrate the Env mutants which had detectable FPRL1 usage. The R318P mutation, which was strongly associated with both CCR5 and FPRL1 usage, is highlighted in bold. The primary coreceptor specificity of the Env mutants (ie., whether they exhibit an R5, R5X4 or X4 phenotype) was determined previously [8]. (C), comparisons of the V3 loop region between the R5 1109-E-10 and X4 1109-F-30 Envs. Shown in bold are the amino acids tested by mutagenesis for modulation of FPRL1 usage, and boxed is the V3 crown motif shown to principally modulate FPRL1 usage.