HSV-2 infection of dendritic cells amplifies a highly susceptible HIV-1 cell target

Abstract

Herpes simplex virus type 2 (HSV-2) increases the risk of HIV-1 infection and, although several reports describe the interaction between these two viruses, the exact mechanism for this increased susceptibility remains unclear. Dendritic cells (DCs) at the site of entry of HSV-2 and HIV-1 contribute to viral spread in the mucosa. Specialized DCs present in the gut-associated lymphoid tissues produce retinoic acid (RA), an important immunomodulator, able to influence HIV-1 replication and a key mediator of integrin α₄β₇ on lymphocytes. α₄β₇ can be engaged by HIV-1 on the cell-surface and CD4⁺ T cells expressing high levels of this integrin (α₄β₇ (high)) are particularly susceptible to HIV-1 infection. Herein we provide in-vivo data in macaques showing an increased percentage of α₄β₇ (high) CD4⁺ T cells in rectal mucosa, iliac lymph nodes and blood within 6 days of rectal exposure to live (n = 11), but not UV-treated (n = 8), HSV-2. We found that CD11c⁺ DCs are a major target of HSV-2 infection in in-vitro exposed PBMCs. We determined that immature monocyte-derived DCs (moDCs) express aldehyde dehydrogenase ALDH1A1, an enzyme essential for RA production, which increases upon HSV-2 infection. Moreover, HSV-2-infected moDCs significantly increase α₄β₇ expression on CD4⁺ T lymphocytes and HIV-1 infection in DC-T cell mixtures in a RA-dependent manner. Thus, we propose that HSV-2 modulates its microenviroment, influencing DC function, increasing RA production capability and amplifying a α₄β₇ (high)CD4⁺ T cells. These factors may play a role in increasing the susceptibility to HIV-1.

Figure 1. Rectal HSV-2 challenge increases the percentage of α₄β₇ ^highCD3⁺CD4⁺ T cells in in-vivo.

A) The gating strategy for α₄β₇ ^highCD3⁺CD4⁺CD95⁺ T cells in blood is shown from one representative animal. The majority of the α₄β₇ ^high T cells are CD95⁺CD28⁺CCR7⁺ (last plot on the right). B) The percentages (mean ± SEM) of CD3⁺CD4⁺CD95⁺ T cells that are α₄β₇ ^high in blood are shown at 4 days before infection (BL) and at 6 days p.i. for HSV-2-infected (LIVE, n = 9) and UV-HSV-2-treated (UV, n = 8) animals. C) The percentages (mean ± SEM) of CD3⁺CD4⁺ T cells that are α₄β₇ ^high in rectal mucosa are shown for HSV-2-infected (LIVE, n = 7) and UV-HSV-2-treated (UV, n = 5) animals 6 days p.i.. D) The percentages (mean ± SEM) of CD3⁺CD4⁺CD95⁺ T cells that are α₄β₇ ^high in axillary (AX), mesenteric (M), inguinal (ING), and iliac LNs are shown for HSV-2-infected (LIVE, n = 5) and UV-HSV-2-treated (UV, n = 7) animals. B–C) Each symbol represents an animal. (*p<0.05, **<0.01, ***p≤0.001).

Figure 2. Blood CD11c⁺ DCs are susceptible to HSV-2 infection.

A) Human (n = 3) and macaque (n = 3) PBMCs were HSV-2-infected (5 MOI, LIVE) or mock-treated and the ICP-8 expression measured 24 h later. ICP-8 versus CD11c expression is shown for Lin⁻HLA-DR⁺ cells on representative examples. The percentages of the positively stained cells are provided. B) The percentages (mean ± SEM of 3 donors each) of total live PBMCs that are ICP8⁺, of ICP8⁺ cells that are Lin⁻, and of the ICP8⁺Lin⁻ cells that are HLA-DR⁺CD11c⁺ are shown for human (HU) and macaque (RM) cultures.

Figure 3. Low level HSV-2 infection modulates DC function.

A) Immature moDCs were exposed to 0.04, 0.2, or 1 MOI of replication competent HSV-2 (or medium, MOCK) and infection was monitored by measuring the expression of ICP-8 at 4 h and 24 h. The percentage of ICP-8⁺ (infected) cells is indicated in each panel. One representative experiment of 5 is shown. B) DCs were treated with 0.2 MOI of replication competent (LIVE) or UV-HSV-2 (UV) versus mock (MOCK) supernatant and the surface phenotype assessed after 24 h. The mean fluorescent intensities (MFI) (means ± SEM, 5 independent experiments) of the indicated markers are shown for the total mock (MOCK) and total UV-HSV-2-treated (UV) DCs versus the ICP-8⁻ (−) and ICP-8⁺ (+) fractions of the LIVE condition. C) CK/CC concentrations (means ± SEM, 11 independent experiments) in the cell-free supernatants from LIVE-, UV-, and MOCK-treated moDCs are shown. (*p<0.05, **<0.01, ***p≤0.001).

Figure 4. HSV-2 infection of moDCs induces α₄β₇ on CD4⁺ T cells.

Mock-, UV-HSV-2-, or Live HSV-2-treated DCs (Figure 3) were mixed with autologous CD4⁺ T cells and cultured for 5 days. A) The gating strategy for the definition of the α₄β₇ ^high (HIGH) (memory cells), α₄β₇ ^int (INT) (memory cells) and α₄β₇ ^low (LOW) (naïve cells) and CD69⁺ subsets is shown for 1 representative of 11 independent experiments. B–C) The fold changes (mean ± SEM, 11 independent experiments) in the percentage of α₄β₇ ^highCD3⁺CD4⁺ T cells (B) and CD69⁺CD3⁺CD4⁺ T cells (C) in the absence (−) or presence (+) of an RARα antagonist are shown relative to the mock-treated controls (set as 1). A T cell only control (no DCs, T ONLY), not exposed to HSV-2, was included alongside. (*p<0.05, **<0.01, ***p≤0.001).

Figure 5. HSV-2 infection increases ALDH1A1 expression in moDCs.

A) Expression of ALDH1A1, ALDH1A2, and ALDH1A3 RNA in immature moDCs and PBMCs by RT-PCR. Results shown are representative of data from 2 different donors. A negative control (−) without DNA was included to control for contamination. GAPDH was amplified in each sample as an internal PCR and loading control. B) The fold changes (mean ± SEM, 13 independent experiments) in ALDH1A1 mRNA of LIVE- and UV-HSV-2-treated moDCs (24 h post treatment) are shown relative to mock-treated moDCs (set as 1). C) The fold changes (mean ± SEM, 5 independent experiments) in the percentage of ALDEFLUOR positive (ALDH enzymatic activity) moDCs treated with LIVE or UV-HSV-2 (24 h post treatment) are shown relative to mock-treated moDCs (set as 1). D) Gating strategy for ALDH enzymatic activity in mock-, UV-HSV-2-, or Live HSV-2-treated moDCs (summarized in panel C). One representative of 5 independent experiments is shown. The numbers indicate the percentage of ALDH⁺ cells. (*p<0.05, **<0.01, ***p≤0.001).

Figure 6. HSV-2 infection enhances HIV-1 replication in a RA-dependent manner.

Mock-, UV-HSV-2-, or Live HSV-2-treated DCs were loaded with HIV-1 and co-cultured with autologous CD4⁺ T cells for 3 to 7 days (A) or 5 days (B–E). A) The kinetics of infection is shown for 1 of 3 similar experiments. B) The fold changes (mean ± SEM of 15 independent experiments) in HIV-1 DNA copies/cell for the 3 conditions after 5 days of co-culture are shown relative to the MOCK controls (set as 1). C) The percentage of p24⁺CD4⁺ T cells within each α₄β₇ subset was measured 5 days after co-culture with the differently treated DCs. 1 of 3 independent experiments is shown. D) The fold changes (mean ± SEM, 4 independent experiments) in the HIV-1 copies/cell in presence (+) or absence (−) of the RARα antagonist are shown relative to the mock-treated controls (set as 1). One experiment, in which the HIV-1 infection was about 100 fold higher in the live HSV-2 condition compared to the mock treated, was excluded. In the excluded experiment the RARα antagonist did not reverse the HSV-2-infected DC enhancement effect (reduced by 20%). (*p<0.05, **<0.01, ***p≤0.001).

Figure 7. Potential mechanism of HSV-2-infected DCs enhancement of HIV-1 infection.

Rapid replication of HSV-2 in epithelial cells could drive a low level HSV-2 infection in neighboring DCs, imprinting on them a “mucosal-like” phenotype and inducing the ability to release RA. RA can profoundly affect resident and/or recruited lymphocytes and expand the pool of α₄β₇ ^highCD4⁺ T cells at the mucosal site of infection. These highly susceptible cells could contribute to increase HIV-1 replication rate and HIV-1 rapid access gut inductive sites, PPs and MLNs. Therefore, the RA-driven immunomodulatory effect and other HSV-2 driven changes – that need to be investigated - could partner to create an environment particularly susceptible to HIV-1 infection.