MicroRNA 146a (miR-146a) is over-expressed during prion disease and modulates the innate immune response and the microglial activation state

Abstract

Increasing evidence supports the involvement of microRNAs (miRNAs) in inflammatory and immune processes in prion neuropathogenesis. MiRNAs are small, non-coding RNA molecules which are emerging as key regulators of numerous cellular processes. We established miR-146a over-expression in prion-infected mouse brain tissues concurrent with the onset of prion deposition and appearance of activated microglia. Expression profiling of a variety of central nervous system derived cell-lines revealed that miR-146a is preferentially expressed in cells of microglial lineage. Prominent up-regulation of miR-146a was evident in the microglial cell lines BV-2 following TLR2 or TLR4 activation and also EOC 13.31 via TLR2 that reached a maximum 24-48 hours post-stimulation, concomitant with the return to basal levels of transcription of induced cytokines. Gain- and loss-of-function studies with miR-146a revealed a substantial deregulation of inflammatory response pathways in response to TLR2 stimulation. Significant transcriptional alterations in response to miR-146a perturbation included downstream mediators of the pro-inflammatory transcription factor, nuclear factor-kappa B (NF-κB) and the JAK-STAT signaling pathway. Microarray analysis also predicts a role for miR-146a regulation of morphological changes in microglial activation states as well as phagocytic mediators of the oxidative burst such as CYBA and NOS3. Based on our results, we propose a role for miR-146a as a potent modulator of microglial function by regulating the activation state during prion induced neurodegeneration.

Figure 1. Hierarchical cluster plot generated from miRNA expression profiling of a repertoire of a variety of CNS cell lineages (neuronal, microglia, astrocytes).

A microglia specific cluster of miRNAs is indicated. Red indicates high levels of miRNA expression, green low levels and gray indicates expression that was undetected by the microarray used.

Figure 2. MiR-146a induction in microglia.

A. LPS at concentrations ranging from 0.1–100 ng/ml was used to stimulate BV-2 cells. After 8 hours, RNA was collected and TaqMan® qRT-PCR was used to determine miR-146a expression relative to PBS treated (unstimulated) control cells. The experiment was performed in triplicate and the average fold induction is shown. B. EOC 13.31 cells were stimulated with 100 ng/ml semi-pure LPS, 100 ng ultra-pure LPS or PBS alone (unstimulated). RNA was collected after 8 hours and miR-146a expression was measured by TaqMan® qRT-PCR. Fold induction relative to untreated cells is shown. The experiment was performed in triplicate and the average fold change is shown. C. EOC 13.31 cells were incubated with increasing concentrations of an anti-TLR2 antibody for 30 minutes prior to stimulation with 100 ng/ml LPS. MiR-146a expression relative to mock-treated control cells was measured by TaqMan® qRT-PCR. Inhibition of miR-146a expression following anti-TLR2 antibody treatment was significant at all concentrations; 10 and 50 ng/ml * p<0.01, 100 ng/ml ** p<0.005. Treatment of BV-2 cells with 100 ng/ml anti-TLR-2 antibody prior to LPS stimulation failed to inhibit miR-146a induction. The experiment was performed in triplicate and the average fold change is shown. D. EOC 13.31 and BV-2 cells were stimulated with with 10⁸ heat-killed Listeria monocytogenes (HKLM) cells/ml. RNA was collected at various time-points over 72 hours and miR-146a expression measured by TaqMan® qRT-PCR. Fold induction relative to unstimulated cells (US) is shown. The experiment was performed in triplicate and the average fold change is shown.

Figure 3. EOC 13.31 cells were stimulated with 100 ng/ml semi-pure LPS and temporal changes in cytokine gene expression relative to mock-treated control were measured by microarray analysis.

MiR-146a expression relative to mock-treated control was measured by TaqMan® qRT-PCR and expression is shown as grey bars.

Figure 4. Altering the expression of miR-146a in microglia.

A. MiR-146a over-expression as measured by TaqMan® qRT-PCR in microglia cells after 8, 24, and 48 hrs of treatment with 30 nM (previously optimized, data not shown) of transfected pre-miR-146a in comparison to 30 nM of transfected scrambled negative control miRNA. An average fold change and standard deviation were measured from triplicate experiments. B. MiR-146a knock-down as measured by TaqMan® qRT-PCR in microglia cells after 8, 24, and 48 hrs of treatment with 50 nM (previously optimized, data not shown) of transfected anti-miR-146a in comparison to 50 nM of transfected scrambled negative control miRNA. An average fold change and standard deviation were measured from triplicate experiments.

Figure 5. Analysis of genes dysregulated by over-expression, or knock-down of miR-146a in resting EOC 13.31 cells using either miRNA mimics or anti-miRs.

A. Venn diagram to show the intersection between genes down-regulated by miR-146a over-expression, up-regulated on miR-146a knock-down and those targets bioinformatically predicted using the TargetScan 5.1 program and IPA software. B. Networks showing the interactions between several predicted miR-146a target genes. Shaded grey are those genes also bioinformatically predicted using the TargetScan 5.1 program and IPA software.

Figure 6. Analysis of genes dysregulated upon over-expression, or knock-down, of miR-146a in stimulated EOC 13.31 cells using either miRNA mimics or anti-miRs.

A. Venn diagram to show the intersection between genes down-regulated in LPS stimulated EOC 13.31 cells by miR-146a over-expression, and those up-regulated upon miR-146a knock-down. Also shown are those targets bioinformatically predicted using the TargetScan 5.1 program and IPA software. B. Schematic showing alterations in expression in key inflammatory response-related genes in LPS stimulated EOC 13.31 cells following miR-146a over-expression. Colored green are those down-regulated genes, colored red are up-regulated genes, while those highlighted in blue are bioinformatically predicted targets of miR-146a using the TargetScan 5.1 program and IPA software.

Figure 7. EOC 13.31 cells were transfected with 30 nM of miR-146a mimic or anti-miR and 16 hrs later stimulated with 10 ng/ml LPS.

At 24 hours post LPS stimulation, culture supernatant was collected and the levels of IL-6 and GM-CSF (coded for by the gene CSF2) were determined by ELISA; significance: ** p<0.001, * p<0.01.

Figure 8. A summary showing key inflammatory response-related genes whose expression is modulated upon the over-expression of miR-146a.