Tristetraprolin expression and microRNA-mediated regulation during simian immunodeficiency virus infection of the central nervous system

Abstract

Background: The RNA-binding protein tristetraprolin (TTP) participates in normal post-transcriptional control of cytokine and chemokine gene expression, dysregulation of which contributes to the HIV-associated neurocognitive disorders. Transcriptional and post-transcriptional regulation of TTP has been described, including regulation by microRNA-29a. In the simian immunodeficiency virus (SIV) model of HIV CNS disease, control of cytokine/chemokine expression coincides with the end of acute phase infection. This control is lost during progression to disease. In this study, we assessed TTP regulation and association with cytokine regulation in the brain during SIV infection.

Results: Quantitation of TTP expression over the course of SIV infection revealed downregulation of TTP during acute infection, maintenance of relatively low levels during asymptomatic phase, and increased expression only during late-stage CNS disease, particularly in association with severe disease. The ability of miR-29a to regulate TTP was confirmed, and evidence for additional miRNA targeters of TTP was found. However, increased miR-29a expression in brain was not found to be significantly negatively correlated with TTP. Similarly, increased TTP during late-stage disease was not associated with lower cytokine expression.

Conclusions: TTP expression is regulated during SIV infection of the CNS. The lack of significant negative correlation of miR-29a and TTP expression levels suggests that while miR-29a may contribute to TTP regulation, additional factors are involved. Reduced TTP expression during acute infection is consistent with increased cytokine production during this phase of infection, but the increases in TTP observed during late-stage infection were insufficient to halt runaway cytokine levels. While antisense inhibitors of the post-transcriptional targeters of TTP identified here could conceivably be used further to augment TTP regulation of cytokines, it is possible that high levels of TTP are undesirable. Additional research is needed to characterize members of the miRNA/TTP/cytokine regulatory network and identify nodes that may be best targeted therapeutically to ameliorate the effects of chronic inflammation in retrovirus-associated CNS disease.

Figure 1

Tristetraprolin mRNA expression during SIV infection. TTP was quantitated by TaqMan assay using total RNA isolated from thalamus of control, uninfected (‘0’) or SIV-infected macaques at different stages of infection and disease (A) and at the indicated number of days post-inoculation (B). Data were normalized by delta-deltaCt method, with 18S rRNA as the control, and graphed relative to the average expression of the control group. Bars represent mean (A) or median (B) fold change in each stage or at each time point. Shown for late-stage infection (B), CNS disease severity as determined by pathology examination is denoted by color: green = none, turquoise = mild, blue = moderate, red = severe. Inset: data sorted by stage of infection. Acute phase comprises days 4-10; persistent includes time points with lower viral load and cytokine expression (14 and 21); late-stage disease begins at day 42. ** denotes p < 0.01 by Kruskal-Wallis test with Dunn’s multiple comparison test.

Figure 2

miR-29a-mediated regulation through the TTP 3′ UTR. HEK-293 T cells (A) or monocyte-derived macrophages (B) were transfected with reporter plasmids (GFP-TTP-3′ UTR or GFP empty vector), red fluorescent protein (RFP) transfection control, and small RNA molecules as indicated (scrambled mimic control, miR-29a mimic, antagonist control, or miR-29a antagonist). 24 hours post-transfection, fluorescence of cells transfected with reporter and empty vector were determined by flow cytometry. A) Exogenous miR-29a caused a significant reduction (***, p < 0.0001 by t-test, p < 0.01 by ANOVA) in the normalized fluorescence ratio of HEK-293 T cells transfected with the GFP-TTP-3′ UTR construct vs. GFP vector alone. B) Because of sample-to-sample variability, the consistent decrease in normalized GFP-TTP 3′UTR fluorescence in the presence of exogenous miR-29a in macrophages approached but did not reach statistical significance as determined by ANOVA (p > 0.05). Error bars are standard error of the mean.

Figure 3

Longitudinal expression of miR-29a in SIV-infected brain. Expression of miR-29a was determined by miRNA stem-loop RT-qPCR using the delta-deltaCt method with normalization to snRNA U6 (similar results were obtained with normalization to the geometric mean of multiple invariant miRNAs—data not shown). Fold changes relative to the average of the control animals (‘0’) are shown. Bars represent the mean at each time point. Color coding of severity during late-stage infection is as described for Figure 1. Statistical significance was assessed for each time point versus control by Kruskal Wallis test with Dunn’s test for multiple comparisons; * represents p < 0.05.

Figure 4

Novel interactions of miRNAs with TTP 3′ UTR elements. A) Alignment representing the previously-described miR-29a/TTP 3′ UTR interaction, along with alignments for additional miR-29 family members and a novel predicted interaction with miR-145. In each alignment, TTP (top) and miRNA (bottom) canonical base pairings are shown with ‘|’, while non-Watson-Crick “wobble” pairing of G and U is shown as ‘:’. B. Reduction in mean fluorescence intensity (MFI) of the TTP (ZFP36) 3′ UTR in the presence of the transfected synthetic miRNAs (HEK-293 T cells). Error bars are standard error of the mean.

Figure 5

Longitudinal miRNA expression in SIV-infected brain. miRs-145, 361-3p, and -155 were measured in thalamus using stem-loop RT-qPCR. Normalization was to U6 snRNA expression, and results are shown in comparison with the average of the control, uninfected (‘0’) samples. Error bars are SEM. Significance of expression differences at each time point versus control was assesssed by Kruskal-Wallis test with Dunn’s multiple comparison test: * (p < 0.05), ** (p < 0.01), *** (p < 0.001).