Variants of CCR5, which are permissive for HIV-1 infection, show distinct functional responses to CCL3, CCL4 and CCL5

Abstract

CCR5 is one of the primary coreceptors for Env-mediated fusion between cells and human immunodeficiency virus type 1 (HIV-1). Analyses of CCR5 variants in cohorts of HIV-1 high-risk individuals led to the identification of multiple single amino-acid substitutions, which may have functional consequences. This study focused on eight naturally occurring allelic variants located between amino-acid residues 60 and 334 of CCR5. All studied allelic variants were highly expressed on the cell surface of HEK-293 cells and permissive for HIV-1 infection. Variant G301V showed 3.5-fold increase in 50% effective concentration (EC(50)) for CCL4 (MIP 1beta) in a competitive binding assay. There was also a significant reduction in CCL5 (RANTES) EC(50) for the R223Q, A335V and Y339F variants. The most unexpected functional abnormality was exhibited by the R60S variant that exhibited a loss of ligand-induced desensitization in chemotaxis assays, but showed normal CCL4 and CCL5 binding avidity. This mutation is located in the first intracellular loop, a domain that has not previously been shown to be involved in receptor desensitization. In conclusion, our results support earlier studies showing that these naturally occurring point mutations do not limit HIV-1 infection, and indicated that single amino-acid changes can have unexpected functional consequences.

Figure 1

Graphic illustration of CCR5—showing predicted amino-acid positions in the cell membrane for human CCR5—the positions of studied variants are marked in black.

Figure 2

Overlay of the FACS tracings for HEK-293 parental cells vs CCR5 variant-transfected HEK cells. Cells were stained with FITC-conjugated 2D7 antibody followed by FACS. The relative fluorescence is shown on the x-axis and the cell number on the y-axis. Each variant is illustrated by a different color line indicated in the inset.

Figure 3

CCL4 competitive binding assays CCR5 WT receptor and G301V variant. Competition binding curves were performed on HEK-293 cells expressing individual CCR5 variants, indicated at the top of each representative curve. Results were analyzed with Graph Pad Prism software using a single site competition model. The data were normalized for specific binding and the percentage shown. The error bars indicate s.e.m. and a representative curve is shown. A mean EC₅₀ was determined for three or more independent experiments and reported in Table 2.

Figure 4

CCL5 competitive binding assays CCR5 WT receptor, R223Q, DelK228, G301V and A335V variants. Competition binding curves were performed on HEK-293 cells expressing individual CCR5 variants, indicated at the top of each representative curve. Results were analyzed with Graph Pad Prism software using a single site competition model. The data were normalized for specific binding and the percentage shown. The error bars indicate s.e.m. and a representative curve is shown. A mean EC₅₀ was determined for three or more independent experiments and reported in Table 2.

Figure 5

CCL5 (RANTES), CCL3 (MIP 1alpha) and CCL4 (MIP 1β-induced chemotaxis of HEK-293 cells expressing CCR5 variants. Chemotaxis assays were preformed in triplicate with four or more independent experiments performed. A representative experiment is shown for select variants. The CCR5 variant being tested is noted at the top of individual graphs. The media control are shown as ‘0’ at the far left side of each graph. CCL5-induced chemotaxis is graphed in black bars. The CCL3-induced chemotaxis is graphed in white bars. The CCL4-induced chemotaxis is graphed in gray bars. The CI is graphed on the y-axis with error bars indicating the s.d., while the chemokine concentration in ng/ml is graphed on the x-axis.

Figure 6

CCL5 induced calcium flux. HEK cells expressing either CCR5 WT or G301V were loaded with Fura-2 and stimulated with 1–10 μg/ml (129–1290 nm) of CCL5 was. The G301V-expressing cells were stimulated at 60 s with varying amounts of CCL5 indicated by an arrow below the x-axis. The green trace shows the 1 μg/ml CCL5 stimulation of G301V-expressing cells. The blue trace shows the 5 μg/ml CCL5 stimulation of G301V cells. The brown trace shows the 10 μg/ml CCL5 stimulation of G301V cells. CCR5 WT-expressing cells were stimulated at 80 s, indicated by an arrow below the x-axis, with 1 μg/ml of CCL5, this trace is shown in pink. The tracings show the relative fluorescence induced.

Figure 7

Kinetics of CCL5-induced receptor internalization. HEK-293 cells stably expressing individual CCR5 receptor variants were exposed at 37°C to 1 μg/ml of CCL5 for various times ranging from 1 to 60 min, indicated graphically in the legend to the right of the bar graph. Data are presented as a percentage decrease from initial surface CCR5 expression for each variant ± s.e.m. for three independent determinations. Matched time point data for each variant were compared with WT receptor data and subjected to one-tailed paired t-test analysis to determine significant differences. *P = 0.04.

Figure 8

HIV-1 infection of CD4 + CXCR4/HEK and CD4 + CCR5 variant/HEK cells. CD4 + CXCR4/HEK or CCR5 variant/HEK cells were exposed to either HIV-1 RF or HIV-1 BaL virus stocks at an MOI of 0.1–1.0. Proviral DNA expression was determined by PCR amplification of the gag proviral DNA at 24 h. The presence of a 200 bp PCR fragment in a lane indicates HIV-1 infection. The samples are correspondingly labeled. Select cultures were pre-treated for 30 min with 1–5 μg/ml of CCL4, CCL5, and 10 μM of NSC 651016. The transfection and PCR were performed at least three times each. The receptor variants are correspondingly labeled. A densitometric analysis of the PCR data is shown with the CCL4 data shown in the same lane for each reported variant. The densitometry units were made relative to each other by setting the CCR5WT RF value to 1.0.