HuR inhibition reduces post-ischemic cardiac remodeling by dampening myocyte-dependent inflammatory gene expression and the innate immune response

Abstract

The RNA-binding protein human antigen R (HuR) has been shown to reduce cardiac remodeling following both myocardial infarction and cardiac pressure overload, but the full extent of the HuR-dependent mechanisms within cells of the myocardium has yet to be elucidated. Wild-type mice were subjected to 30 min of cardiac ischemia (via LAD occlusion) and treated with a novel small molecule inhibitor of HuR at the time of reperfusion, followed by direct in vivo assessment of cardiac structure and function. Direct assessment of HuR-dependent mechanisms was done in vitro using neonatal rat ventricular myocytes (NRVMs) and bone marrow-derived macrophages (BMDMs). HuR activity is increased within 2 h after ischemia/reperfusion (I/R) and is necessary for early post-I/R inflammatory gene expression in the myocardium. Despite an early reduction in inflammatory gene expression, HuR inhibition has no effect on initial infarct size at 24 h post-I/R. However, pathological remodeling is reduced with preserved cardiac function at 2 weeks post-I/R upon HuR inhibition. RNA sequencing analysis of gene expression in NRVMs treated with LPS to model damage-associated molecular pattern (DAMP)-mediated activation of toll-like receptors (TLRs) demonstrates a HuR-dependent regulation of pro-inflammatory chemokine and cytokine gene expression in cardiomyocytes. Importantly, we show that conditioned media transfer from NRVMs pre-treated with HuR inhibitor loses the ability to induce inflammatory gene expression and M1-like polarization in bone marrow-derived macrophages (BMDMs) compared to NRVMs treated with LPS alone. Functionally, HuR inhibition reduces macrophage infiltration to the post-ischemic myocardium in vivo. Additionally, we show that LPS-treated NRVMs induce the migration of peripheral blood monocytes in a HuR-dependent endocrine manner. These studies demonstrate that HuR is necessary for early pro-inflammatory gene expression in cardiomyocytes following I/R injury that subsequently mediates monocyte recruitment and macrophage activation in the post-ischemic myocardium.

Keywords: HuR; cardiac remodeling; heart; inflammation; innate immune response; ischemia/reperfusion injury.

FIGURE 1

HuR activation and expression is increased following I/R injury. HuR cytoplasmic translocation, indicative of activation, is increased at 2 h post‐I/R (N and C represent nuclear and cytosolic fractions, respectively) (A). Cytoplasmic translocation of HuR is quantified in panel B. Total HuR protein expression is increased in hearts at 7 days post‐I/R (C, D). Treatment with the HuR inhibitor KH‐3 (50 mg/kg) at the time of reperfusion reduces expression of pro‐inflammatory gene expression of IL‐6 (E) and TNF‐⍺ (F) at 2 h post‐reperfusion. *p ≤ .05. **p ≤ .01.

FIGURE 2

HuR inhibition reduces post‐ischemic cardiac remodeling and preserves cardiac function. HuR inhibition significantly reduced post‐ischemic cardiac remodeling as assessed by LV end‐diastolic volume (A) and end‐systolic volume (B) and preserved LV function (ejection fraction; C) at 2 weeks post‐I/R. Treatment with HuRi also reduced total cardiac fibrosis as determined by picrosirius red staining (D; quantified in E) and myofibroblast activity as assessed by total expression of the myofibroblast marker periostin (F; quantified in G). *p ≤ .05. **p ≤ .01.

FIGURE 3

HuR mediates inflammatory gene expression following TLR stimulation in cardiomyocytes. Principle component analysis (PCA) (A) and heat map analysis (B) show a distinct and consistent differential gene expression profile in NRVMs treated with LPS alone or LPS plus HuRi (KH‐3). Volcano plots of differentially expressed genes show a predominant increase in gene expression in NRVMs following LPS treatment (C) that is largely reversed in the presence of HuR inhibition (D). Gene labels in C and D represent genes present in the Immune System Process Gene Ontology class (GO0002376) that was subsequently identified as the most strongly enriched HuR‐dependent GO Term (H). Donut plot analysis of all LPS‐dependent significant upregulated (E) and downregulated (F) gene expression changes grouped by gene ontology (GO) term. GO grouping of all HuR‐dependent gene expression changes among those upregulated by LPS shows that HuR overwhelmingly regulates immune and inflammation related GO term‐associated genes (G). The top 10 HuR‐dependent GO terms from panel E are graphed out by p‐value and fold enrichment (H).

FIGURE 4

HuR inhibition reduces macrophage infiltration to the post‐ischemic myocardium. Immunofluorescent staining of CD68⁺ cells shows a reduction in macrophage infiltration to the heart at 7 days post‐I/R upon HuR inhibition (A; total area of CD68⁺ stain is quantified in B). *p ≤ .05.

FIGURE 5

TLR signaling in myocytes induces inflammatory gene expression in macrophages in a HuR‐dependent endocrine manner. Schematic representation of NRVM treatment and conditioned media transfer to BMDMs (A). HuR inhibition reduces mRNA expression of IL‐1β (B) and TNF‐α (C) directly in NRVMs. Conditioned media from LPS‐treated NRVMs induces expression of TNF‐α in BMDMs that is blunted when NRVMs are treated with HuR inhibitor (D). siRNA‐mediated knockdown of HuR in NRVMs inhibits the LPS‐mediated increase in TNF‐α (E), IL‐1β (F), and IL‐6 (G) in BMDMs treated with conditioned media from NRVMs. Flow cytometric analysis of BMDMs shows an increase in the percentage of F4/80^hi, CD11c^hi BMDMs following treatment with conditioned media from LPS treated NRVMs that is blunted when NRVMs are treated with a HuR inhibitor (H, I). Conditioned media transfer from NRVMs treated with LPS induces monocyte (PBMC) migration that is lost when NRVMs are pre‐treated with HuR inhibitor (J). All experimental groups in panels D–H received conditioned media. *p ≤ .05. **p ≤ .01. ***p ≤ .001.