IP-10 Promotes Latent HIV Infection in Resting Memory CD4+ T Cells via LIMK-Cofilin Pathway

Abstract

A major barrier to HIV eradication is the persistence of viral reservoirs. Resting CD4⁺ T cells are thought to be one of the major viral reservoirs, However, the underlying mechanism regulating HIV infection and the establishment of viral reservoir in T cells remain poorly understood. We have investigated the role of IP-10 in the establishment of HIV reservoirs in CD4⁺ T cells, and found that in HIV-infected individuals, plasma IP-10 was elevated, and positively correlated with HIV viral load and viral reservoir size. In addition, we found that binding of IP-10 to CXCR3 enhanced HIV latent infection of resting CD4⁺ T cells in vitro. Mechanistically, IP-10 stimulation promoted cofilin activity and actin dynamics, facilitating HIV entry and DNA integration. Moreover, treatment of resting CD4⁺ T cells with a LIM kinase inhibitor R10015 blocked cofilin phosphorylation and abrogated IP-10-mediated enhancement of HIV latent infection. These results suggest that IP-10 is a critical factor involved in HIV latent infection, and that therapeutic targeting of IP-10 may be a potential strategy for inhibiting HIV latent infection.

Keywords: CXCL10; HIV reservoir; cofilin; latent infection; resting memory CD4+ T cells.

Figure 1

IP-10/CXCL10 level is elevated in the chemokine profile of HIV⁺ participants, and positively correlated with viral load and HIV reservoir size. (A–C) Comparison of 11 CXC chemokines and 1 CX3C chemokine (A); 8 CC chemokines (B); and 10 cytokines (C) between HIV+ participants (gray bars, n=28) and NCs (white bars, n=24). Graphs show data as median ± interquartile range. The nonparametric Wilcoxon matched-pairs test was used for intergroup comparisons. (D–F) Spearman correlation analyses of chemokine (Gro-b/CXCL2, GCP-2/CXCL6, MIG/CXCL9, IP-10/CXCL10, I-TAC/CXCL11, BCA-1/CXCL13, TARC/CCL17, and MIP-3β/CCL19) levels and viral load at indicated time points (D); and between total HIV DNA of PBMCs and IP-10 level after ART (E) or without treatment (F). (G) Comparison the total HIV DNA of PBMCs in IP-10 high vs IP-10 low groups (separated according to median IP-10 level in 29 ART-treated HIV⁺ participants). The nonparametric Mann–Whitney U test was used for intergroup comparisons (IP-10 high: N=13; IP-10 low: N=16, P=0.001). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2

IP-10 promotes HIV AD8 and NL4-3 infection of resting memory CD4⁺ T cells. (A) Gating strategy for isolating a pure population of resting memory CD4⁺ T cells. (B) Schematic illustration of the procedure for evaluating the role of IP-10 in HIV infection of resting memory CD4⁺ T cells. (C) Kinetics of the effect of AD8 (n = 6)/NL4-3(n = 11), AD8/NL4-3+IP-10, AD8/NL4-3+IP-10+mAb, and AD8/NL4-3+IP-10+CXCR3 on resting memory CD4⁺ T cell from 2 representative NCs. P24 levels were compared between groups on the 14^th day post infection by one-way analysis of variance. *P < 0.05.

Figure 3

IP-10 promotes HIV AD8 and NL4-3 infection of resting memory CD4⁺ T cells at the resting state. (A) Schematic illustration of the procedure for evaluating the role of IP-10 in HIV infection of resting memory CD4⁺ T cells at the resting state. (B) Kinetics of the effect of AD8(n = 7)/NL4-3(n = 8), AD8/NL4-3+IP-10, AD8/NL4-3+IP-10+mAb, and AD8/NL4-3+IP-10+CXCR3 on resting memory CD4⁺ T cell at resting state from 2 representative NCs. P24 levels were compared between groups on the 14^th day post infection by one-way analysis of variance. *P < 0.05.

Figure 4

IP-10 does not promote resting memory CD4⁺ T cell activation. (A, B) Effect of IP-10 on the expression of the resting memory CD4⁺ T cell activation markers HLA-DR, CD69, and CD25 (A) and on the activation of resting memory CD4⁺ T cells by CD3/CD28 (B) at 6, 12, 24, and 48 h and 5 days post-treatment; representative flow cytometry plots from 3 participants and quantitative analyses are shown. The time point for HLA-DR, CD69, and CD25 of the representative flow cytometry plots are 5 days, 6 h and 24 h, respectively. The nonparametric Wilcoxon matched-pairs test was used for intergroup comparisons.

Figure 5

IP-10 promotes HIV NL4-3 entry into resting memory CD4⁺ T cells, viral DNA synthesis and integration, and actin activation. (A) Gating strategy and representative flow cytometry plots from the experiment evaluating the role of IP-10 in the infection of resting memory CD4+ T cells by HIV NL4-3. Viral entry into resting memory CD4+ T cells was evaluated with the BlaM-Vpr entry assay. For no CCF2, cells were pretreated for 1 h with IP-10 (5 ng/ml) and then infected with NL4-3. For no NL4-3, cells were treated with CCF2 without NL4-3 infection. For NL4-3, cells were treated with CCF2 before infection with NL4-3. For NL4-3+IP-10, cells were treated with CCF2 before pretreatment for 1 h with IP-10 (5 ng/ml), and then infected with NL4-3. The Wilcoxon matched-pairs test was used to compare median values between two groups (N=7, P=0.016). (B) Quantification of HIV+ cells in NL4-3 and NL4-3+IP-10 groups. The Wilcoxon matched-pairs test was used to compare median values between groups. (C) For NL4-3, cells were infected with NL4-3; for NL4-3+IP-10, cells were pretreated for 1 h with IP-10 (5 ng/ml) and then infected with NL4-3. Cells were infected for 2 h, washed, and then collected at different time points for ddPCR quantification of total viral DNA at 4 days. Results are expressed as copies/cell. The Wilcoxon matched-pairs test was used to compare median values between groups (N=6, P=0.031). (D) Quantitative (Alu) PCR analysis of HIV DNA integration in negative control, NL4-3, NL4-3+IP-10, NL4-3+raltegravir, and NL4-3+raltegravir+IP-10 groups. As the variances are different, the Wilcoxon matched-pairs test was used to compare median values between groups (N=8, NL4-3 vs. NC: P=0.008; NL4-3+IP-10 vs. NC: P=0.008; NL4-3+IP-10 vs. NL4-3: P=0.008; NL4-3+Raltegravir vs. NL4-3+IP-10: P=0.008; NL4-3+Raltegravir +IP-10 vs. NL4-3+IP-10: P=0.039; NL4-3+Raltegravir vs. NC: P=0.062; NL4-3+Raltegravir +IP-10 vs. NC: P=0.016; NL4-3+Raltegravir vs. NL4-3: P=0.078; NL4-3+Raltegravir+IP-10 vs. NL4-3+Raltegravir: P=0.469). (E) Representative flow cytometry plots from the experiment evaluating the effect of IP-10 on actin polymerization and depolymerization over time. The red dashed line represents localized baseline unstimulated F-actin level. (F) Temporal profile of MFI in control and IP-10–treated resting memory CD4⁺ T cells (The result for one of the three separated experiments is showed in the figure). *P < 0.05, **P < 0.01.

Figure 6

IP-10 activates LIMK–cofilin signaling in resting memory CD4⁺ T cells. (A) Gating strategy for the experiment evaluating the effect of IP-10 on cofilin activation. (B, C) Representative flow cytometry plots from the experiment evaluating the effect of IP-10 on cofilin phosphorylation over time. Resting memory CD4⁺ T cells were treated with IP-10 (5 ng/ml) and IP-10 (5 ng/ml)+R10015 (100 μM) (B) or NL4-3 and NL4-3+IP-10 (5 ng/ml) (C) for the indicated times, and p-cofilin was marked with Ser3, followed by flow cytometry analysis. The red dashed lines represent localized baselines for cofilin phosphorylation. Cofilin dynamics were evaluated at 1, 5, 10, 15, 30, 45, and 60 min, with the measurement at 0 min serving as the reference value. (D, E) LIMK inhibition abrogates the enhancement of HIV AD8/NL4-3 infection by IP-10. Representative plots of 2 participants are shown. P24 levels of day 14 were compared between groups by one-way analysis of variance (NL4-3: n = 7, AD8: n = 8). *P < 0.05.

Figure 7

Model of IP-10-induced enhancement of latent HIV infection and reservoir establishment in resting memory CD4⁺ T cells. A signaling pathway common to HIV (gray solid line) and IP-10 (black solid line) induces actin activation. IP-10 binds to CXCR3 expressed on the surface of resting memory CD4⁺ T cells; Rac stimulates p21-activated kinase (PAK), which activates LIMK, leading to cofilin phosphorylation and actin polymerization. Another reported signaling pathway for latent HIV infection of resting CD4⁺ T cells (dotted line) involves HIV binding to CXCR4; the signal is transferred to CDC42, which activates Wiskott-Aldrich Syndrome family protein (WASP) and actin-nucleating protein (Arp2/3) to promote actin assembly.