Blood myeloid dendritic cells from HIV-1-infected individuals display a proapoptotic profile characterized by decreased Bcl-2 levels and by caspase-3+ frequencies that are associated with levels of plasma viremia and T cell activation in an exploratory study

Abstract

Reduced frequencies of myeloid and plasmacytoid dendritic cell (DC) subsets (mDCs and pDCs, respectively) have been observed in the peripheral blood of HIV-1-infected individuals throughout the course of disease. Accumulation of DCs in lymph nodes (LNs) may partly account for the decreased numbers observed in blood, but increased DC death may also be a contributing factor. We used multiparameter flow cytometry to evaluate pro- and antiapoptotic markers in blood mDCs and pDCs from untreated HIV-1-infected donors, from a subset of infected donors before and after receiving antiretroviral therapy (ART), and from uninfected control donors. Blood mDCs, but not pDCs, from untreated HIV-1-infected donors expressed lower levels of antiapoptotic Bcl-2 than DCs from uninfected donors. A subset of HIV-1-infected donors had elevated frequencies of proapoptotic caspase-3(+) blood mDCs, and positive correlations were observed between caspase-3(+) mDC frequencies and plasma viral load and CD8(+) T-cell activation levels. Caspase-3(+) mDC frequencies, but not mDC Bcl-2 expression, were reduced with viral suppression on ART. Apoptosis markers on DCs in blood and LN samples from a cohort of untreated, HIV-1-infected donors with chronic disease were also evaluated. LN mDCs displayed higher levels of Bcl-2 and lower caspase-3(+) frequencies than did matched blood mDCs. Conversely, LN pDCs expressed lower Bcl-2 levels than their blood counterparts. In summary, blood mDCs from untreated HIV-1-infected subjects displayed a proapoptotic profile that was partially reversed with viral suppression, suggesting that DC death may be a factor contributing to blood DC depletion in the setting of chronic, untreated HIV disease.

FIG. 1.

Gating strategy used to identify DC subsets and expression of markers involved in apoptosis. An initial gate was selected based on forward- and side-scatter properties that eliminated the majority of dead cells and cellular debris. Multiparameter flow cytometry techniques were then used to identify total DCs defined as Lineage/CD34⁻ HLA-DR⁺ and then further subdivided into myeloid DCs (mDCs; CD11c^high CD123^lo) and plasmacytoid DCs (pDCs; CD11c⁻ CD1123^high). Surface expression of Fas and Fas ligand and intracellular expression of Bcl-2 and caspase-3 by mDCs and pDCs was then assessed. Isotype controls were used for all apoptotic markers. Profiles shown are representative of seronegative donors.

FIG. 2.

Expression of Bcl-2 and caspase-3 by blood mDCs. Intracellular expression of Bcl-2 by blood mDCs from seronegative (SN; n = 14) or HIV-1-infected (HIV-1⁺; n = 17 to 18) donors was assessed either directly ex vivo (A) or after overnight in vitro culture (B). Values are shown as the mean fluorescence intensity (MFI) with background isotype staining removed (net MFI). Intracellular caspase-3 expression by blood mDCs was assessed directly ex vivo (C) from seronegative (n = 18) or HIV-1-infected (n = 20 to 23) donors and expressed as the fraction of mDCs positive for caspase-3 within the total mDC population (%). Viral load (D) and CD8⁺ T-cell activation, defined by CD38 expression (E), positively correlate with caspase-3 expression by blood mDCs directly ex vivo. Statistical analysis was performed by using the Mann-Whitney t test to compare Bcl-2 and caspase-3 expression by blood mDCs between seronegative and HIV-1-infected donors, and the Spearman test was performed for plasma viral load and CD8⁺ T-cell correlations.

FIG. 3.

Expression of caspase-3 and Bcl-2 by blood mDCs following ART. To determine the effects of ART on expression of caspase-3 (A) and Bcl-2 (B) by blood mDCs directly ex vivo, a subset of donors (Table 1; Pre-ART) from the untreated, HIV-1-infected cohort were further evaluated for changes in caspase-3 (n = 11) and Bcl-2 (n = 7) expression at least 3 months after starting treatment (Post-ART). Values are expressed as either the fraction of mDCs positive for caspase-3 within the total mDC population (%) (A) or as the mean fluorescence intensity (MFI) with background isotype staining removed (net MFI) for Bcl-2 expression (B). Statistical analysis was performed by using the Wilcoxon signed-rank test.

FIG. 4.

Comparison of expression of CD40, Bcl-2, and caspase-3 by DC subsets from PBMC and matched lymph node samples from HIV-1-infected donors. Expression of CD40 (A and D), Bcl-2 (B and E), and caspase-3 (C and F) were evaluated in mDCs (A to C) or pDCs (D to E) from matched PBMC and lymph node samples from HIV-1-infected donors (n = 6). Values are expressed as the mean fluorescence intensity (MFI) with background isotype staining removed (net MFI) for CD40 and Bcl-2 expression or as the fraction of mDCs or pDCs positive for caspase-3 within the total mDC or pDC population (%). Statistical analysis was performed by using the Wilcoxon signed-rank test.

FIG. 5.

Changes in expression of CD40, Bcl-2, and caspase-3 expression by mDCs and pDCs from seronegative donors in response to TLR ligation. Surface expression of CD40 (A) and intracellular expression of Bcl-2 (B) and caspase-3 expression (C) was assessed on mDCs and pDCs in PBMC from seronegative donors, after overnight (13 to 20 h) in vitro culture with or without TLR ligand (TLRL) stimulation. TLR-specific stimulation of mDCs was assessed using a TLR4L (LPS; 10 μg/ml, n = 7 to 11) or a TLR7/8L (CL097, a derivative of the imidazoquinoline compound R848; 5 μg/ml, n = 11). TLR-specific stimulation of pDCs was assessed using a TLR7L (Imiquimod, [R837], 10 μg/ml, n = 7 to 11). Values are expressed as the mean fluorescence intensity (MFI) minus background isotype staining (net MFI) for CD40 (A) and Bcl-2 (B) or as the fraction of mDCs or pDCs positive for caspase-3 as a percentage of total mDCs or pDCs (C). Statistical analysis was performed comparing the change in expression with or without TLR stimulation using the Wilcoxon signed-rank test.