|Home | About | Journals | Submit | Contact Us | Français|
Circulating HIV-1-infected monocytes have been identified in patients on highly active antiretroviral therapy and may represent an important barrier to viral eradication. The nature of these cells in HIV-1-infected patients who maintain undetectable viral loads and preserved CD4+ T cell counts without antiretroviral therapy (known as elite controllers or elite suppressors [ES]) is unknown. We describe here infrequent recovery of proviral HIV-1 DNA from circulating monocytes relative to CD4+ T cells in ES, despite permissiveness of these cells to HIV-1 viral entry ex vivo. Thus, monocytes do not appear to be a major reservoir of HIV-1 in ES.
Cells of the monocyte/macrophage lineage have long been recognized as targets of HIV-1 infection and contributors to disease pathogenesis (20–22, 25, 42, 47, 50). CD4+ T cells and monocyte/macrophages both serve as HIV-1 reservoirs that preclude viral eradication in HIV-1-infected patients despite the ability of highly active antiretroviral therapy (HAART) to block replication and durably suppress plasma viral load (1, 6, 11, 12, 17, 18, 45, 61). These reservoirs can lead to a rebound in viremia if HAART is stopped (12), and in the case of HIV-1-infected macrophages may contribute to tissue inflammation and damage despite ongoing suppressive HAART (15, 48, 56). Circulating monocytes in the peripheral blood are relatively resistant to productive HIV-1 infection ex vivo prior to differentiation into tissue macrophages (38, 39, 52, 57), which are more permissive to HIV-1 infection and viral replication (2, 37, 58). Despite this finding, HIV-1-infected monocytes have been identified in the peripheral blood of viremic and HAART-treated patients (3, 9, 14, 19, 26–28, 31, 33, 34, 36, 43, 49, 51, 53, 54, 62) and are thought to contribute to viral persistence (13, 32).
The mechanisms of virologic and immunologic control that allow a minority of HIV-1-infected individuals to maintain undetectable viral loads without antiretroviral therapy are not known. While low-level viral replication appears to occur in these patients, known as elite controllers or elite suppressors (ES) (40), the quantity of total (24, 30, 46) and integrated (23) proviral DNA and replication-competent HIV-1 (8) in circulating CD4+ T cells is small compared to HIV-1-infected patients on HAART. The frequency of HIV-1-infected monocytes in the peripheral blood of ES has not been determined, and the role HIV-1-infected macrophages play in viral persistence in ES is unknown. We demonstrate here that the frequency of circulating monocytes containing HIV-1 proviral DNA is much lower than the frequency of HIV-1-infected CD4+ T cells in a cohort of ES, despite evidence of permissiveness to HIV-1 entry in monocytes from ES ex vivo.
In order to obtain pure populations of CD4+ T cells and monocytes, peripheral blood was obtained from HIV-1-infected patients on HAART and ES according to an approved Institutional Review Board protocol, and Ficoll gradient centrifugation was performed. All participants had undetectable viral loads that were confirmed by commercial assays. The median CD4+ T cell count for patients on HAART was 593 cells per μl (interquartile range, 499 to 823), the median CD8+ T cell percentage was 44.3% (range, 35.2 to 57.3), the median time with an undetectable viral load was 3 years (interquartile range, 2 to 5), and the median number of years from diagnosis was 11 years (interquartile range, 5 to 14). For ES, the median CD4+ T cell count was 989 cells per μl (interquartile range, 735 to 1282), the median CD8+ T cell percentage was 31.5% (range, 15.4 to 54.3), and the median number of years from diagnosis was 11 years (interquartile range, 3 to 19). Flow cytometry was performed on peripheral blood mononuclear cells (PBMCs) from each patient group along with PBMCs from healthy non-HIV-infected donors. CD14+ monocytes and CD4+ T cells were isolated from PBMCs by using magnetic microbeads according to the manufacturer's instructions (Miltenyi Biotec). CD14+ cells then underwent magnetic bead depletion according to the manufacturer's instructions (Invitrogen) to remove remaining CD3+ cells. The CD4+ T cell suspension was incubated with mouse anti-human antibodies to CD8, CD14, CD16, and CD19, followed by addition of sheep anti-mouse magnetic beads and magnetic bead depletion. Flow cytometry was used to determine the purity of each cell population (in a representative monocyte population, 1.5% of cells are CD3+/CD14−; in a representative CD4+ T cell population, 0.2% are CD14+ [data not shown]). These cells were lysed to obtain genomic DNA according to the manufacturer's instructions (Qiagen). Quantitative PCR (qPCR) was performed using long terminal repeat (LTR) primers to determine the frequency of proviral HIV-1 DNA in circulating CD4+ T cells and monocytes. Single-step qPCR was performed on each sample under previously published conditions (29). The proviral HIV-1 DNA copy number was calculated relative to the RNase P gene copy number (Applied Biosystems), which had been previously quantified with a standard curve. Nested PCR using previously described nef primers (4) was performed in replicates of four on sample aliquots containing genomic DNA from 105 cells and from 104 cells as a complementary, semiquantitative means of determining the frequency of HIV-1 proviral DNA. A separate in vitro viral entry assay was used to compare the relative permissiveness of each cell type to viral entry in each cell population, using a previously described protocol (10).
Flow cytometry performed on PBMCs of uninfected controls, patients on HAART, and ES demonstrated similar frequencies of CD14+ monocytes between uninfected donors and ES (median values, 5.38% versus 4.28%, respectively; P = 0.91) (Fig. 1B). There was a significantly higher percentage of circulating monocytes expressing CCR5 among ES than in uninfected donors or patients on HAART (7.43% versus 2.44% [uninfected] or 0.73 [HAART]; P = 0.001 for ES versus uninfected and P = 0.0001 for ES versus HAART) (Fig. 1C). The mean fluorescence intensity (MFI) of CCR5+ monocytes was similar between ES and uninfected donors (Fig. 1D). CD16+ monocytes, a subset of CD14+ cells shown to be more permissive to HIV-1 infection (16), were overrepresented in ES compared to uninfected controls (7.87% versus 3.44%; P = 0.06) and patients on HAART (7.87% versus 1.86%; P = 0.02) (Fig. 1E). There was no correlation between CD4+ T cell count or percentage or CD8+ T cell percentage and the percentage of CD14+ monocytes or CD14+ CD16+ monocytes in ES versus patients on HAART (data not shown).
Quantitative PCR using LTR primers demonstrated that the level of HIV-1 proviral DNA in circulating monocytes was below the limit of detection of this assay in all 11 ES patients and 11 of 13 patients on HAART. Proviral DNA was also not detected in CD4+ T cells of 10 of 11 ES. In contrast, proviral DNA was easily amplified from CD4+ T cells in 8 of 13 patients on suppressive HAART regimens (Fig. 2).
To further quantify the difference in the frequency of HIV-1-infected monocytes compared to CD4+ T cells in ES and patients on HAART, we utilized a second PCR methodology. We performed four replicates of a nested PCR with nef primers on DNA isolated from these two cell types obtained from patients in both groups (Fig. 3). While virus was amplified from at least one out of four aliquots of 105 CD4+ T cells in seven of the eight ES sampled, only one aliquot of 105 monocytes from a single ES was positive for virus (Fig. 3A).
In patients on HAART, virus was obtained from at least 1 of 4 aliquots of 105 cells from all 10 patients when CD4+ T cells were analyzed and from 4 of 10 patients when monocytes were examined (Fig. 3A). In ES, virus could still be amplified from CD4+ T cell aliquots containing 1 log fewer cells (104 CD4+ T cells) in 5 out of 8 ES, whereas no virus was amplified in 104 monocyte aliquots in any ES (Fig. 3B). Virus was amplified with DNA from 104 CD4+ T cells in all 10 patients on HAART and from 1 out of 7 of these patients when using DNA from 104 monocytes (Fig. 3B). Thus, the frequency of infected circulating CD4+ T cells in both ES and patients on HAART appears to be at least 1 log higher than the frequency of HIV-1-infected monocytes.
Given the low recovery of HIV-1 proviral DNA from monocytes of ES patients, we investigated whether these cells are resistant to HIV-1 entry. Flow cytometry data (Fig. 1) demonstrated that a significant percentage of circulating monocytes in ES patients expressed the cell surface markers CCR5 and CD16, suggesting a phenotype conducive to viral infection. We made use of a previously described VPR-BlaM entry assay (10) and found that between 3 and 34% of monocytes were susceptible to HIV-1 entry among individuals; however, no difference with regard to viral entry susceptibility was observed between the three patient groups tested (ES patients, viremic patients, and uninfected controls) (Fig. 4). Thus, consistent with our recent findings in CD4+ T cells (44), resistance to viral entry does not explain the low frequency of virus present in ES monocytes.
This is the first study to examine the frequency of infected monocytes in ES. We showed that among circulating target cells, monocytes contain a minimal amount of detectable proviral HIV-1 DNA compared to CD4+ T cells in ES, despite the permissiveness of monocytes to HIV-1 entry demonstrated ex vivo. CCR5 and CD16 expression levels on a fraction of monocytes from ES suggest that a subset of these cells may be permissive to infection, either as precursor cells in the bone marrow or in the circulation (16, 37). Based on the PCR results shown here, however, it appears that circulating monocytes are not a significant source of residual HIV-1 viremia in ES.
Previous studies have generally shown a lower frequency of infected circulating monocytes than CD4+ T cells in patients receiving HAART (3, 9, 14, 19, 26–28, 31, 33, 34, 36, 43, 49, 51, 54, 62), although significant variability exists in these studies with regard to methodology, cell purity, and frequency of proviral DNA recovered. Our data suggest that a similar trend is present in ES, and we cannot rule out the possibility that the low level of infection we detected in monocytes was in fact due to contaminating CD4+ T cells.
One explanation for this phenomenon lies in the different half-lives of these cell types in the circulation: monocytes typically migrate into tissue after 72 h (35, 55, 59, 60), while CD4+ T cells recirculate in the peripheral blood. We have previously documented a marked discordance in HIV-1 sequence present in the plasma compared to that present in resting (4, 5, 7) and activated (41) CD4+ T cells in ES. One possible explanation for this finding is that monocytes are the source of plasma virus in these patients. Due to the very limited amount of HIV-1 proviral DNA recovered from the monocytes of ES, we were unable to perform the phylogenetic analysis required to test this hypothesis. However, our results support the conclusion that monocytes are infrequently HIV-1 infected compared to CD4+ T cells and are unlikely to significantly contribute to persistent, low-level viremia in ES.
This work was supported by NIH grant R01 AI080328 (J.N.B.).
Published ahead of print on 27 July 2011.