Mode of coreceptor use by R5 HIV type 1 correlates with disease stage: a study of paired plasma and cerebrospinal fluid isolates

Abstract

Through the use of chimeric CXCR4/CCR5 receptors we have previously shown that CCR5-tropic (R5) HIV-1 isolates acquire a more flexible receptor use over time, and that this links to a reduced viral susceptibility to inhibition by the CCR5 ligand RANTES. These findings may have relevance with regards to the efficacy of antiretroviral compounds that target CCR5/virus interactions. Compartmentalized discrepancies in coreceptor use may occur, which could also affect the efficacy of these compounds at specific anatomical sites, such as within the CNS. In this cross-sectional study we have used wild-type CCR5 and CXCR4 as well as chimeric CXCR4/CCR5 receptors to characterize coreceptor use by paired plasma and cerebrospinal fluid (CSF) isolates from 28 HIV-1-infected individuals. Furthermore, selected R5 isolates, with varying chimeric receptor use, were tested for sensitivity to inhibition by the CCR5 antagonist TAK-779. Discordant CSF/plasma virus coreceptor use was found in 10/28 patients. Low CD4+ T cell counts correlated strongly with a more flexible mode of R5 virus CCR5 usage, as disclosed by an increased ability to utilize chimeric CXCR4/CCR5 receptors, specifically receptor FC-2. Importantly, an elevated ability to utilize chimeric receptors correlated with a reduced susceptibility to inhibition by TAK-779. Our findings show that a discordant CSF and plasma virus coreceptor use is not uncommon. Furthermore, we provide support for an emerging paradigm, where the acquisition of a more flexible mode of CCR5 usage is a key event in R5 virus pathogenesis. This may, in turn, negatively impact the efficacy of CCR5 antagonist treatment in late stage HIV-1 disease.

FIG. 1.

Schematic representation of the wild-type coreceptors CXCR4 (black) and CCR5 (gray) and the CXCR4/CCR5 chimeric receptors FC-1, FC-2, and FC-4b. The chimeric receptors were constructed by successively exchanging regions of CCR5 with corresponding parts of CXCR4.

FIG. 2.

Receptor use of paired plasma and CSF isolates. Diagrams showing results after infection of U87.CD4 cells expressing CCR5, CXCR4, or chimeric CXCR4/CCR5 receptors, with paired plasma (gray bars) and CSF (open bars) virus isolates. (A–F) The six paired isolates with discordant use of chimeric receptors. (G) Paired isolates of concordant R5^narrow phenotype. (H) Paired isolates of concordant R5^broad3 phenotype. Infections are measured as p24 protein content in the cell culture supernatant. # indicates a p24 value <1. The criterium for discordant use is a >10-fold difference in p24 antigen production between the infections.

FIG. 3.

Correlations between (A) plasma virus FC-2 usage and CD4⁺ T cell counts and (B) CSF virus FC-2 usage and CSF HIV-RNA load. (A) Individuals harboring plasma FC-2⁺ R5 isolates or X4 isolates had significantly lower CD4⁺ T cell counts as compared to individuals with FC-2- R5 isolates (p = 0.004 and p = 0.005, respectively). (B) Individuals harboring CSF FC-2⁺ R5 isolates had significantly higher CSF HIV-RNA levels as compared to individuals with FC-2- R5 isolates (p = 0.02). CSF HIV-RNA levels for individuals harboring CSF X4 isolates did not significantly differ from other groups. Coreceptor use was determined as p24 antigen production >100 pg/ml in the cell culture supernatant. Horizontal lines represent mean values of CD4⁺ T cell counts and CSF HIV-RNA levels.