Early Antiretroviral Therapy Prevents Viral Infection of Monocytes and Inflammation in Simian Immunodeficiency Virus-Infected Rhesus Macaques

Abstract

Despite early antiretroviral therapy (ART), treatment interruption is associated with viral rebound, indicating early viral reservoir (VR) seeding and absence of full eradication of human immunodeficiency virus type 1 (HIV-1) that may persist in tissues. Herein, we address the contributing role of monocytes in maintaining VRs under ART, since these cells may represent a source of viral dissemination due to their ability to replenish mucosal tissues in response to injury. To this aim, monocytes with classical (CD14⁺), intermediate (CD14⁺ CD16⁺), and nonclassical (CD16⁺) phenotypes and CD4⁺ T cells were sorted from the blood, spleen, and intestines of untreated and early-ART-treated simian immunodeficiency virus (SIV)-infected rhesus macaques (RMs) before and after ART interruption. Cell-associated SIV DNA and RNA were quantified. We demonstrated that in the absence of ART, monocytes were productively infected with replication-competent SIV, especially in the spleen. Reciprocally, early ART efficiently (i) prevented the establishment of monocyte VRs in the blood, spleen, and intestines and (ii) reduced systemic inflammation, as indicated by changes in interleukin-18 (IL-18) and IL-1 receptor antagonist (IL-1Ra) plasma levels. ART interruption was associated with a rebound in viremia that led to the rapid productive infection of both CD4⁺ T cells and monocytes. Altogether, our results reveal the benefits of early ART initiation in limiting the contribution of monocytes to VRs and SIV-associated inflammation.IMPORTANCE Despite the administration of antiretroviral therapy (ART), HIV persists in treated individuals and ART interruption is associated with viral rebound. Persistent chronic immune activation and inflammation contribute to disease morbidity. Whereas monocytes are infected by HIV/SIV, their role as viral reservoirs (VRs) in visceral tissues has been poorly explored. Our work demonstrates that monocyte cell subsets in the blood, spleen, and intestines do not significantly contribute to the establishment of early VRs in SIV-infected rhesus macaques treated with ART. By preventing the infection of these cells, early ART reduces systemic inflammation. However, following ART interruption, monocytes are rapidly reinfected. Altogether, our findings shed new light on the benefits of early ART initiation in limiting VR and inflammation.

Keywords: CD4; HIV; IL-18; IL-1ra; SIV; antiretroviral therapy; inflammation; intestine; monocytes; spleen; viral reservoir.

FIG 1

Study protocol and viral loads. (A) Sixteen rhesus macaques (RMs) were infected intravenously with SIVmac251 (20 AID₅₀). At day 4 postinfection, 11 were treated with tenofovir (TFV, 20 mg/kg; Gilead) and emtricitabine (FTC, 40 mg/kg; Gilead) subcutaneously and raltegravir (RGV, 20 mg/kg; Merck) or dolutegravir (DTG, 5 mg/kg; ViiV) and ritonavir (RTV, 20 mg/kg; Abbvie) by the oral route. RMs were sacrificed at different time points postinfection during natural infection (No ART) (n = 5), under ART (ART) (n = 5), and after ART interruption (ATI) (n = 6). (B) Levels of viral load in the sera of SIV-infected RMs were quantified by qRT-PCR during natural infection (No ART) and under ART (ART). Each symbol represents the value for an individual. Results are expressed as viral load copies per ml.

FIG 2

Frequencies of cell-associated SIV DNA in ART-naive SIV-infected RMs. (A) Myeloid cell sorting. Representative dot plots depicting the expression of CD14 and CD16 from cells derived from blood, spleen, colon, and small intestine. Gated cells are CD3⁻ CD20⁻ cells, which are then separated into HLA-DR⁺ cells and CD11b⁺ cells, leading to three different monocyte cell subsets. Thus, cells were identified as CD14⁺ CD16⁻, CD14⁺ CD16⁺, and CD14⁻ CD16⁺. (B to D) Frequencies of SIV DNA in sorted CD4⁺ T cells and monocyte cell subsets (CD14⁺, CD14⁺ CD16⁺, and CD16⁺) from blood and spleen. For intestines (small intestine, including jejunum/ileum, closed symbols; colon, open symbols), only two monocyte populations, CD14⁺ and CD16⁺, were sorted from SIV-infected RMs. Each symbol represents the value for one individual. Due to the limited amounts of cells, some RMs were not tested for all conditions. Results are expressed as copies per 10⁶ cells. The Mann-Whitney test was performed. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.

FIG 3

Frequencies of cell-associated SIV R-U5 in ART-naive SIV-infected RMs. (A) Frequencies of SIV R-U5 in CD4⁺ T cells and monocyte cell subsets (CD14⁺, CD14⁺ CD16⁺, and CD16⁺) sorted as described in the legend to Fig. 2. Each symbol represents the value for one individual. Results are expressed as copies per 10⁶ cells. (B) Ratios of SIV R-U5 against viral DNA. Ratios in blood, spleen, and intestines (small intestine, closed symbol; colon, open symbol) and the different monocytic cell subsets are shown. A ratio higher than 3, represented by a dashed line, is considered significant. The Mann-Whitney test was performed. *, P < 0.05; ND, not done due to small amounts of cells.

FIG 4

Frequencies of cell-associated SIV RNA in ART-naive SIV-infected RMs. (A and B) Frequencies of SIV RNA in CD4⁺ T cells and monocyte cell subsets (CD14⁺, CD14⁺ CD16⁺, and CD16⁺) sorted as described in the legend to Fig. 2. Each symbol represents the value for one individual. Results are expressed as copies per 10⁶ cells. The Mann-Whitney test was performed. **, P < 0.01; ND, not done due to small amounts of cells.

FIG 5

Levels of CD3E in sorted monocytic cell subsets. The expression of CD3E mRNA was quantified by qRT-PCR in sorted monocytic cell subsets (including CD14⁺ CD16⁻, CD14⁺ CD16⁺, and CD14⁻ CD16⁺) from blood and spleen. The frequencies were compared with the viral DNA (A) and viral RNA (B) frequencies. Results are expressed as copies per 10⁶ cells. Paired t test was performed. **, P < 0.01; ****, P < 0.0001.

FIG 6

Frequencies of cell-associated SIV DNA and RNA in ART-treated RMs. Frequencies of SIV DNA (A), SIV R-U5 (B), and SIV RNA (C) were quantified by qRT-PCR in CD4⁺ T cells and monocytic cell subsets sorted as described in the legend to Fig. 2. Each symbol represents the value for one individual. Results are expressed as copies per 10⁶ cells.

FIG 7

Replication-competent SIV. SIV p27 levels in cocultures of splenocytes (1.5 × 10⁶ cells) stimulated with either LPS (10 ng/ml) or ConA (2 μg/ml) with CEMx174 cells after 2 weeks. At day 1, cells were cocultured with the cell line CEMx174 (10⁴ cells). Splenocytes are derived from two ART-naive RMs (RM 09051222, day 35 postinfection, and RM PB051, day 14 postinfection) and one ART-treated RM (RM R110562, day 27 postinfection). Viral p27 antigen was quantified by flow cytometry. SSC, side scatter.

FIG 8

Levels of IL-18 and IL-1Ra in SIV-infected RMs. (A and B) IL-18 (A) and IL-1Ra (B) were quantified by ELISA in uninfected RMs (day 0), ART-naive RMs (at days 14 and >46), and ART-treated RMs (at days 14 and >46). Results are expressed as picograms per milliliter of serum. Each symbol represents the value for one individual. In ART-naive RMs, we included additional samples, given that RMs were sacrificed earlier. The Mann-Whitney test was performed. **, P < 0.01; ***, P < 0.001.

FIG 9

Viral loads and frequencies of cell-associated SIV DNA after ART interruption (ATI). (A) Viral loads after ATI were quantified by qRT-PCR. Each color represents the data for one individual. (B to D) Frequencies of SIV DNA were quantified by qRT-PCR in CD4 cells and monocytes sorted as described in the legend to Fig. 2. Each symbol represents the value for one individual. Results are expressed as copies per 10⁶ cells. A Mann-Whitney test was performed. *, P < 0.05.

FIG 10

Frequencies of cell-associated SIV RNA after ART interruption (ATI). Frequencies of cell-associated SIV RNA in individual RMs sacrificed at each time point as in Fig. 9. Each symbol represents the value for one individual. Results are expressed as copies per 10⁶ cells. ND, not done due to small amounts of cells. The Mann-Whitney test was performed, and no statistically significant difference was observed.