MicroRNA210 Suppresses Mitochondrial Metabolism and Promotes Microglial Activation in Neonatal Hypoxic-Ischemic Brain Injury

Abstract

Neuroinflammation is the major contributor to the pathology of neonatal hypoxic-ischemic (HI) brain injury. Our previous studies have demonstrated that microRNA210 (miR210) inhibition with antisense locked nucleic acid (LNA) inhibitor mitigates neuroinflammation and provides neuroprotection after neonatal HI insult. However, the underlying mechanisms remain elusive. In the present study, using miR210 knockout (KO) mice and microglial cultures, we tested the hypothesis that miR210 promotes microglial activation and neuroinflammation through suppressing mitochondrial function in microglia after HI. Neonatal HI brain injury was conducted on postnatal day 9 (P9) wild-type (WT) and miR210 knockout (KO) mouse pups. We found that miR210 KO significantly reduced brain infarct size at 48 h and improved long-term locomotor functions assessed by an open field test three weeks after HI. Moreover, miR210 KO mice exhibited reduced IL1β levels, microglia activation and immune cell infiltration after HI. In addition, in vitro studies of microglia exposed to oxygen-glucose deprivation (OGD) revealed that miR210 inhibition with LNA reduced OGD-induced expression of Il1b and rescued OGD-mediated downregulation of mitochondrial iron-sulfur cluster assembly enzyme (ISCU) and mitochondrial oxidative phosphorylation activity. To validate the link between miR210 and microglia activation, isolated primary murine microglia were transfected with miR210 mimic or negative control. The results showed that miR210 mimic downregulated the expression of mitochondrial ISCU protein abundance and induced the expression of proinflammatory cytokines similar to the effect observed with ISCU silencing RNA. In summary, our results suggest that miR210 is a key regulator of microglial proinflammatory activation through reprogramming mitochondrial function in neonatal HI brain injury.

Keywords: hypoxia–ischemia; microglia; mitochondrial dysfunction; neuroinflammation.

Figure 1

miR210 knockout (miR210 KO) provided neuroprotection in neonatal hypoxic–ischemic (HI) brain injury in mice. HI brain injury was conducted on postnatal day 9 (P9) miR210 KO or wild-type (WT) mouse pups. (a) RT-qPCR result of miR210 levels in the ipsilateral hemisphere of WT and miR210 KO pups at 24 h after HI. Data are presented as mean ± SEM. n = 4–6 pups per group. One-way ANOVA followed by Bonferroni post hoc test. (b) Representative images and quantitative result of brain infarct at 48 h after HI. Data are presented as mean ± SEM. n = 10 pups per group. Student’s t test (unpaired, two-tailed). (c) Open field test for locomotor function evaluation performed three weeks after HI. Data are presented as mean ± SEM. n = 7–10 pups per group. One-way ANOVA followed by Bonferroni post hoc test. ns, not significant.

Figure 2

miR210 knockout reduced neuroinflammation in neonatal hypoxic–ischemic (HI) brain injury. (a) ELISA of IL1β levels in the ipsilateral hemisphere of WT and miR210 KO pups 24 h after HI insult. Data are presented as mean ± SEM. n = 4–6 pups/group. One-way ANOVA followed by Bonferroni post hoc test. (b–d) Representative images and quantitative result of flow cytometry of the proportion of activated microglia–macrophages (CD45 high/CD11b+) (b,c) and infiltrated lymphocytes (CD45+/CD11b-) (b,d) in the ipsilateral hemisphere of WT and miR210 KO pups 24 h after sham or HI insult. Colors show different particle density (high, red/orange; low, blue/green). Data are presented as mean ± SEM. n = 5–6 pups/group. One-way ANOVA followed by Bonferroni post hoc test. ns, not significant.

Figure 3

miR210 inhibition reduced the expression of proinflammatory cytokine and preserved the expression of mitochondrial OXPHOS-related genes in BV2 microglia after oxygen–glucose deprivation (OGD). BV2 cells were transfected with either miR210-LNA or LNA scramble (SCR) overnight followed by OGD treatment for 3 h. The RT-qPCR was performed for the transcript levels of Il1b (a), ISCU (b), Ndufa4 (c), and Sdhd (d). Data are presented as mean ± SEM. n = 3 independent cell culture. One-way ANOVA followed by Bonferroni post hoc test. ns, not significant.

Figure 4

miR210 inhibition preserved mitochondrial OXPHOS and weakened glycolysis in BV2 microglia after OGD. BV2 cells were transfected with either miR210-LNA or LNA scramble (SCR) overnight followed by OGD treatment for 3 h. (a,e) Extracellular flux assay profiles including oxygen consumption rate (OCR) (a) and extracellular acidification rate (ECAR) (e), measured using the Agilent Seahorse analyzer. (b–d) OCR changes in basal respiratory (b), ATP production (c), and maximal respiration (d). (f,g) ECAR changes in basal glycolysis (f) and glycolysis capacity (g). Data are presented as mean ± SEM. n = 3–5 independent cell culture. One-way ANOVA followed by Bonferroni post hoc test. ns, not significant.

Figure 5

miR210 mimic upregulated proinflammatory cytokines and inhibited ISCU expression in microglia. (a) Representative image of isolated primary murine microglia stained with microglia marker Iba-1 (green). Scale bar: 50 µm. Primary microglia were transfected with miR210 mimic or control miRNA (Neg. Ctrl) for 48 h. Student’s t test (unpaired, two-tailed). (b,c) RT-qPCR results of the transcript levels of Il1b, Il6, Tnf-α (b), and ISCU (c). Data are presented as mean ± SEM. n = 3 independent cell culture. Student’s t test (unpaired, two-tailed). (d) Western blot result of the expression of ISCU. Data are presented as mean ± SEM. n = 3 independent cell culture. Student’s t test (unpaired, two-tailed). The blots were transferred to the PVDF from the same gel and cropped between 25 and 37 kDa for different primary antibody probing. Primary microglia were transfected with ISCU silencing RNA or control siRNA (Neg. Ctrl) for 48 h. (e,f) RT-qPCR results of the transcript levels of ISCU (e) and Il1b (f). Data are presented as mean ± SEM. n = 3 independent cell culture. Student’s t test (unpaired, two-tailed). ns, not significant.