The design and validation of a novel phenotypic assay to determine HIV-1 coreceptor usage of clinical isolates

Abstract

A phenotypic assay to determine coreceptor usage of HIV-1 has been developed for rapid testing of clinical samples. The assay is based on the synthesis of viral stock from full-length env amplicons isolated from patient's plasma. Pseudoviral stock is generated rapidly by using an overlapping PCR method to assemble a CMV promoter to env, followed by co-transfection into producer cells with a HIV plasmid (pNL4-3.Luc.R(-)E(-)) containing a non-functional env. The coreceptor used by the viral quasispecies is tested by infection into U87.CD4.CCR5 and U87.CD4.CXCR4 cells. Viral entry is indicated by the expression of the luciferase gene in relative light units (RLU). The use of CXCR4 coreceptor by minor variants is confirmed with sufficient suppression of RLU by a CXCR4 inhibitor. Two statistical tests are employed to confirm viral entry. This assay accurately assigned coreceptor usage of isolates of various subtypes and in the majority of samples of various viral loads. The sensitivity to detect minor species of CXCR4-using env is 1% at higher viral loads and 5% at less than 1,000 copies/ml. This assay provides a sensitive, efficient and relatively low-cost approach suitable for use by research laboratories for assessing HIV-1 coreceptor usage of plasma samples.

Figure 1. Schematic diagram of the methodology to determine HIV-1 coreceptor usage from clinical samples

(A) Generation of promoter env PCR amplicons using overlapping PCR (Kirchherr et al, 2007). The env amplicon is generated from patient-derived viral RNA using RT-PCR followed by nested PCR to generate a 3.5 kb amplicon. The CMV promoter is amplified in a separate PCR and attached to the env amplicons by an overlapping PCR step. (B) Viral stock generation and viral infection. Env amplicons with attached CMV promoter are cotransfected into 293T cells with a plasmid containing a full-length HIV genome deleted in env and carrying a luciferase reporter gene in the nef region (pNL4-3.Luc.R-E-; (AIDS Research and Reference Reagent Program). The pseudotyped virons are harvested and used to infect two separate indicator cell lines, CCR5-CD4-U87 and CXCR4-CD4-U87 cells, in the absence and presence of inhibitors, TAK779 and AMD3100, respectively. Viral entry using the coreceptors is assessed by measuring the luciferase activity in RLU in each cell lines and is confirmed by greater than 50% inhibition of luciferase activity in the presence of the specific inhibitor. Two statistical tests are used to evaluate the raw RLU data to make an assignment of coreceptor usage.

Figure 1. Schematic diagram of the methodology to determine HIV-1 coreceptor usage from clinical samples

(A) Generation of promoter env PCR amplicons using overlapping PCR (Kirchherr et al, 2007). The env amplicon is generated from patient-derived viral RNA using RT-PCR followed by nested PCR to generate a 3.5 kb amplicon. The CMV promoter is amplified in a separate PCR and attached to the env amplicons by an overlapping PCR step. (B) Viral stock generation and viral infection. Env amplicons with attached CMV promoter are cotransfected into 293T cells with a plasmid containing a full-length HIV genome deleted in env and carrying a luciferase reporter gene in the nef region (pNL4-3.Luc.R-E-; (AIDS Research and Reference Reagent Program). The pseudotyped virons are harvested and used to infect two separate indicator cell lines, CCR5-CD4-U87 and CXCR4-CD4-U87 cells, in the absence and presence of inhibitors, TAK779 and AMD3100, respectively. Viral entry using the coreceptors is assessed by measuring the luciferase activity in RLU in each cell lines and is confirmed by greater than 50% inhibition of luciferase activity in the presence of the specific inhibitor. Two statistical tests are used to evaluate the raw RLU data to make an assignment of coreceptor usage.

Figure 2

pPCR tropism assay determination of previously characterized reference strains. U87-CXCR4 and -CCR5 cells were infected with pseudoviral stock from pPCR of env gene from R5-tropic MJ4 (A) and YU2 (B); X4-tropic NL4-3 (C) and HXB2 (D), and dual-tropic SF2 (E) and 89.6 (F). Infection was indicated by a mean RLU which was statistically above background (infection of pseudoviral stock generated by HIV-1 backbone; NL4-3delEnv-LUC plasmid with a non-functional env gene) and suppression of 50% or greater by an X4 or R5 inhibitor, AMD3100 or TAK 775, respectively.

Figure 3

Sensitivity to detect CXCR4-using variants within mixed viral populations. The results are the geometric mean RLU from multiple replicates with SE error bars. Minor species of X4-tropic env from HXB2 was detected consistently at 1% and approximately 50% of the times at 0.5%. Use of CXCR4 for entry was confirmed by ablation of the luciferase signal with the addition of AMD3100, which decreased the RLU to minimal levels. Luciferase activity in U87-CXCR4 cells decreased proportionally to the proportion of input DNA from X4 virus.