Quantification of CD4, CCR5, and CXCR4 levels on lymphocyte subsets, dendritic cells, and differentially conditioned monocyte-derived macrophages

Abstract

CCR5 and CXCR4 are the major HIV-1 coreceptors for R5 and X4 HIV-1 strains, respectively, and a threshold number of CD4 and chemokine receptor molecules is required to support virus infection. Therefore, we used a quantitative fluorescence-activated cell sorting assay to determine the number of CD4, CCR5, and CXCR4 antibody-binding sites (ABS) on various T cell lines, T cell subsets, peripheral blood dendritic cells (PBDC), and monocyte-derived macrophages by using four-color fluorescence-activated cell sorting analysis on fresh whole blood. Receptor levels varied dramatically among the various subsets examined and typically varied from 2- to 5-fold between individuals. CCR5 was expressed at much higher levels in CD4+/CD45RO+/CD62L-true memory cells compared with CD4+/CD45RO+/CD62L+ cells. Fresh PBDC had the highest number of CCR5 ABS among the leukocyte subsets examined but had few CXCR4 ABS, affording a strategy for sort-purifying PBDC. In vitro maturation of PBDC resulted in median 3- and 41-fold increases in CCR5 and CXCR4 ABS, respectively. We found that macrophage colony-stimulating factor caused the greatest up-regulation of both CCR5 and CXCR4 on macrophage maturation (from approximately 5,000 to approximately 50, 000 ABS) whereas granulocyte-macrophage colony-stimulating factor caused a marked decrease of CXCR4 (from approximately 5,000 ABS to <500) while up-regulating CCR5 expression (from approximately 5,000 to approximately 20,000 ABS). Absolute ABS for CD4 and the major HIV-1 coreceptors serve as a more quantitative measure of cell surface expression, and we propose that this be used for future studies looking at the modulation of CD4 or chemokine receptor expression by cytokines, HIV-1 infection, or receptor polymorphisms.

Figure 1

CD4, CXCR4, and CCR5 ABS on T cell lines. (Upper) The ABS for CD4 and CXCR4. ∗CXCR4 ABS were not determined for the GHOST-R5 cell line. (Lower) The ABS for CCR5. Jurkat-D and Jurkat-M refers to different sources of Jurkat cells from two independent labs. The dashed line (Lower) indicates the threshold of detection. Experiments were performed at least three independent times, and data are presented as absolute number of ABS ±SEM.

Figure 2

CXCR4, CCR5, and CD4 ABS on PBMC. Four-color FACS analysis was used to identify the various subpopulations of PBMC indicated. Fresh whole blood was stained with fluorochrome-conjugated mAb immediately after venupucture, and FACS analysis was performed after ammonium chloride-mediated selective red blood cell lysis. B cells are CD19+, NK cells are CD56+, monocytes are the cluster of cells with high forward scatter and low side-scatter profiles (independently confirmed to be CD14+), and ImmDC are immature PBDC (Fig. 3), which can be CD11c+ or CD11c−. MatDC are mature DC from Ficoll-purified PBMC after 24 hours of culture in RPMI medium 1640 with 10% FCS. MatDC are almost exclusively (>95%) CD11c+. ABS for CXCR4 and CCR5 are indicated on a log scale while that for CD4 is indicated on a linear scale. Median values are indicated by a solid bar across each data group. Data is compiled from simultaneous analysis of 12 donors.

Figure 3

FACS gating strategy for identifying fresh PBDC. Fresh whole blood was processed as in Fig. 2. (A) Tri-color-conjugated CD3, CD14, CD16, CD19, and CD56 was used to exclude lineage-positive cells. Lineage-negative cells that were HLA-DR+ (R1 gate) were identified as candidate PBDC. R1-gated cells were then analyzed for CD11c vs. CCR5, CXCR4, or CD4 expression to obtain values for ABS. CD1a was alternatively used to analyze R1-gated cells and gave similar profiles to that obtained with CD11c expression (data not shown). (B) A proposed strategy for identifying PBDC. An analysis of lineage markers vs. CCR5 expression reveals two distinct populations of lineage-negative cells that are either CCR5-high (R2 gate) or CCR5-low (R3 gate). R2-gated cells are almost exclusively HLA-DR+, i.e., candidate DCs, which can be either CD11c+(51.2 ± 10.5%) or CD11c− (48.8 ± 10.5%) whereas R3 gated cells are exclusively HLA-DR− and CD11c−.

Figure 4

Cytokine effects on CD4, CXCR4, and CCR5 expression on MDM. Absolute numbers of CD4, CC5, and CXCR4 ABS were determined on MDM differentiated in the presence of human serum alone (NoRx) plus either M-CSF (50 units/ml) or GM-CSF (10 ng/ml). Data from six to eight donors in five independent experiments were used for this analysis. Median values for each data group are indicated by the solid bars. Although there was considerable donor variability, M-CSF coordinately increased CD4, CCR5, and CXCR4 ABS significantly compared with GM-CSF (P < 0.0001). GM-CSF significantly reduced CXCR4 expression compared with M-CSF (P < 0.01). The differences in CD4 and coreceptor expression were not as significant when comparing NoRx with either M-CSF (P = 0.05) or GM-CSF (P = 0.06).