CCL8/MCP-2 is a target for mir-146a in HIV-1-infected human microglial cells

Abstract

MicroRNA-mediated regulation of gene expression appears to be involved in a variety of cellular processes, including development, differentiation, proliferation, and apoptosis. Mir-146a is thought to be involved in the regulation of the innate immune response, and its expression is increased in tissues associated with chronic inflammation. Among the predicted gene targets for mir-146a, the chemokine CCL8/MCP-2 is a ligand for the CCR5 chemokine receptor and a potent inhibitor of CD4/CCR5-mediated HIV-1 entry and replication. In the present study, we have analyzed changes in the expression of mir-146a in primary human fetal microglial cells upon infection with HIV-1 and found increased expression of mir-146a. We further show that CCL8/MCP-2 is a target for mir-146a in HIV-1 infected microglia, as overexpression of mir-146a prevented HIV-induced secretion of MCP-2 chemokine. The clinical relevance of our findings was evaluated in HIV-encephalitis (HIVE) brain samples in which decreased levels of MCP-2 and increased levels of mir-146a were observed, suggesting a role for mir-146a in the maintenance of HIV-mediated chronic inflammation of the brain.

Figure 1.

Increased levels of mir-146a in HIV-1-infected microglial cells. A) ELISA to detect p24 in medium collected at indicated time points from 2 different preparations of HIV-1-infected primary human fetal microglia (brains 7739 and 7490). B) qRT-PCR to detect levels of mature mir-146a in HIV-1-infected and control microglial cells at 0, 4, 8, and 12 d postinfection. qRT-PCR was performed on 25 ng of template, each in quadruplicate. Each sample was normalized for corresponding values of U6 control. Relative amount of mir-146a was calculated by the Roche LightCycler software. Graph represents average values obtained from microglial cell cultures isolated from brain samples 7739 and 7490.

Figure 2.

mir-146a does not affect viral replication. ELISA to determine amounts of p24 protein in medium of HIV-1-infected human fetal microglia. Cells were transfected with mir-146a or nonrelevant miRNA, mir-128a, 72 h prior to infection with HIV-1 or control mock (see Materials and Methods). Supernatant was collected after 4, 7, and 10 d of infection. Each time point was performed in triplicate; whole experiment was repeated twice.

Figure 3.

mir-146a inhibits secretion of CCL8/MCP-2 by HIV-1-infected microglial cells. A) Diagram showing putative mir-146a binding sites in the 3′UTR-MCP-2 mRNA sequence. B) Graph of a representative ELISA showing amount (μg/ml) of secreted CCL8/MCP-2 chemokine in medium of HIV-1-infected cells and control mock-infected cells. Experiment was repeated 3 times, each in duplicate. C) ELISA to monitor amount of secreted MCP-2 on supernatant of primary human fetal microglial cells transfected with mir-146a, a nonrelevant miRNA (mir-128a), or control empty vector (mock), followed by infection with HIV-1. Experiment was repeated 3 times, each in quadruplicate.

Figure 4.

Inhibition of MCP-2 3′UTR by mir-146a. Luciferase assay on 293T cells transfected with empty vector alone (PsiCheck2), or cotransfected with mir-146a and MCP-2 3′UTR, with mir-146a and the 3′UTR-mutated sequence (mut MCP-2), or with the control mir-146a perfect matching sequence (PM). Luciferase and Renilla values were determined 48 h post-transfection. Relative units represent ratio between Renilla values and luciferase internal control. Experiment was repeated twice, each in triplicate. *P < 0.05; **P < 0.001.

Figure 5.

Effects of MCP-2 on HIV-1 infection. ELISA showing amounts of p24 in medium of HIV-1-infected microglial cells in absence or presence of MCP-2 peptide at 4, 7, and 10 d postinfection. Experiment was repeated twice, each in duplicate. *P < 0.05; **P < 0.001.

Figure 6.

Detection of MCP-2 protein in HIVE and control brains. A) Immunohistochemical analysis of MCP-2 in cortical sections of a control patient, an HIV⁺ patient with no brain pathology, and an HIVE patient revealed a gradual reduction in levels of MCP-2 from the control patient, in which MCP-2 in robustly labeled (a), to lesser expression in the HIV⁺ patient (b), to a significant reduction in HIVE (c, d). MCP-2 is also noticeably decreased within the same HIVE sample, from nonaffected areas (c) to areas of severe inflammation, where MCP-2 is almost undetectable (d). Original view: ×200 (top panels); ×1000 (bottom panels). B) As expected, MCP-2 is robustly expressed in leukocytes inside a subarachnoid blood vessel in an HIV⁺ patient with no brain pathology (a); in HIVE, leukocytes show a dramatic reduction in MCP-2 (b). Microglial nodules in HIVE demonstrate robust cytoplasmic expression of MCP-2 (c). Original view: ×400.

Figure 7.

mir-146a is up-regulated in HIVE brain samples. A) qRT-PCR of mir-146a on RNA samples obtained from 4 HIVE patients (HIVE 030022, 010063, and 01057), 4 control brains (CB), and 2 HIV patients without brain pathology (HIV). qRT-PCR of mir-146a was performed in quadruplicate using 25 ng of RNA. B) Detection of hsa-mir-146a in HIVE 010063 and control CB 528B archived frozen samples. In situ hybridization was performed with miRCURY-LNA-DIG mir-146a (and control mir-128a) detection probes. Arrows indicate positive signal in cytoplasm of neurons. Original view: ×1000.