MiR-181c-5p Promotes Inflammatory Response during Hypoxia/Reoxygenation Injury by Downregulating Protein Tyrosine Phosphatase Nonreceptor Type 4 in H9C2 Cardiomyocytes

Background Constitutive nuclear factor kappa B (NFκB) activation has been shown to exacerbate during myocardial ischemia/reperfusion (I/R) injury. We recently showed that miR-181c-5p exacerbated cardiomyocytes injury and apoptosis by directly targeting the 3′-untranslated region of protein tyrosine phosphatase nonreceptor type 4 (PTPN4). However, whether miR-181c-5p mediates cardiac I/R injury through NFκB-mediated inflammation is unknown. Thus, the present study aimed to investigate the role of miR-181c-5p during myocardial I/R injury and explore its mechanism in relation to inflammation in H9C2 cardiomyocytes. Methods and Results In hypoxia/reoxygenation (H/R, 6 h hypoxia followed by 6 h reoxygenation)-stimulated H9C2 cardiomyocytes or postischemic myocardium of rat, the expression of miR-181c-5p was significantly upregulated, which was concomitant increased NFκB activity when compared to the nonhypoxic or nonischemic control groups. This is indicative that miR-181c-5p may be involved in NFκB-mediated inflammation during myocardial I/R injury. To investigate the potential role of miR-181c-5p in H/R-induced cell inflammation and injury, H9C2 cardiomyocytes were transfected with the miR-181c-5p agomir. Overexpression of miR-181c-5p significantly aggravated H/R-induced cell injury (increased lactate dehydrogenase (LDH) level) and exacerbated NFκB-mediated inflammation (greater phosphorylation and degradation of IκBα, phosphorylation of p65, and increased levels of proinflammatory cytokines tumor necrosis factor α (TNFα), interleukin (IL)-6, and IL-1β). In contrast, inhibition of miR-181c-5p by its antagomir transfection in vitro had the opposite effect. Furthermore, overexpression of miR-181c-5p significantly enhanced lipopolysaccharide-induced NFκB signalling. Additionally, knockdown of PTPN4, the direct target of miR-181c-5p, significantly aggravated H/R-induced phosphorylation and degradation of IκBα, phosphorylation of p65, and the levels of proinflammatory cytokines. PTPN4 knockdown also cancelled miR-181c-5p antagomir mediated anti-inflammatory effects in H9C2 cardiomyocytes during H/R injury. Conclusions It is concluded that miR-181c-5p may exacerbate myocardial I/R injury and NFκB-mediated inflammation via PTPN4, and that targeting miR-181c-5p/PTPN4/NFκB signalling may represent a novel strategy to combat myocardial I/R injury.


Introduction
Ischemic heart disease is one of the major causes of death worldwide [1,2]. During cardiac ischemia/reperfusion (I/R) injury, cellular damage such as excessive apoptosis can result in protease and danger-associated molecular patterns that favours a proinflammatory environment through the activation of nuclear factor kappa B (NFκB) [3]. In the ischemic heart, prolonged activation of NFκB is generally considered to be detrimental by eliciting signals that trigger chronic inflammation through enhanced elaboration of proinflammatory cytokines, including interleukin (IL)-6, IL-1β, and tumor necrosis factor α (TNFα), leading to cardiac injury [4,5]. Constitutive NFκB activation has been demonstrated in various models of experimental myocardial ischemia and reperfusion [3,6]. In vivo transfer of NFκB decoy oligodeoxynucleotides to bind transcriptional factor, blocking inflammatory gene activation, reduced the extent of myocardial infarction following reperfusion [7]. Thus, a strategy that can inhibit excessive NFκB-mediated inflammation should be an effective therapy to combat ischemic heart disease.
The miR-181 family (including miR-181a, miR-181b, miR-181c, and miR-181d) plays diverse roles in regulating various cellular and biological processes through posttranscriptional regulation of target genes [8][9][10][11][12]. Accumulating evidence suggests a central role for the miR-181 family in inflammation by regulating target proteins invovled in critical inflammatory signalling pathways, such as NFκB signalling [13,14]. For example, miR-181b can reduce inflammation through targeting the 3 ′ -untranslated region of the importin-α3 (a protein critical for the translocation of NFκB from the cytoplasm to the nucleus), further inhibiting the activation of NFκB signalling pathway [14]. In addition, previous study has demonstrated that miR-181c-5p (named miR-181c in other studies) can directly target the 3 ′ -untranslated region of TNFα mRNA, suppressing its mRNA and protein expression in rat microglial cells after ischemia injury [15]. However, the anti-inflammatory effect of miR-181c-5p in neuroinflammation was challenged as miR-181c-5p exacerbated brain injury in acute ischemic stroke [16]. Thus, it is still unclear whether miR-181c-5p exerts pro-or antiinflammatory effect in the context of I/R in general, and in myocardial I/R injury in specific.
We have recently shown that miR-181c-5p exacerbated hypoxia/reoxygenation (H/R)-induced cardiomyocyte injury and apoptosis via targeting protein tyrosine phosphatase nonreceptor type 4 (PTPN4) [17]. Of note, PTPN4 suppresses Toll-like receptor 4/NFκB signalling in mouse peritoneal macrophages [18]. However, it is unknown whether or not PTPN4 may inhibit NFκB-mediated proinflammatory responses in cardiomyocytes. During myocardial I/R injury, the cardiac resident cells, such as cardiomyocytes may elaborate proinflammatory cytokines in response to various stimuli and thus favour a proinflammatory environment. Therefore, the present study aimed to determine whether or not miR-181c-5p enhances NFκB-mediated inflammation via inhibiting PTPN4 during cardiac I/R injury by using rat origin cardiomyocytes (H9C2) subjected to H/R stimulation.

Cell
Treatment. The H9C2 cardiomyocytes were seeded into six-well plate (2 × 10 5 cell/well) overnight and transfected with micrON rno-miR-181c-5p agomir (50 nM, RIBOBIO, Guangzhou, China), micrOFF rno-miR-181c-5p antagomir (50 nM, RIBOBIO), small RNA (siRNA) against PTPN4 (50 nM, RIBOBIO), or their negative controls using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) for 24 h according to the manufacturer's instructions. The subgroups of these cardiomyocytes were subsequently subjected to H/R or lipopolysaccharide (LPS, 3 μg/ml, 21 h, Sigma) stimulation before harvesting. H/R stimulation was achieved as previously described [19]. Briefly, the H9C2 cardiomyocytes were cultured with DMEM medium (no glucose or FBS) for 6 h in a humidified Plexiglas chamber containing 95% N 2 and 5% CO 2 . The cells were then exposed to fresh culture medium and room air atmosphere containing 5% CO 2 and 95% O 2 for an additional 6 h to achieve reoxygenation. The transfection effects were verified by detecting the expression of miR-181c-5p, mRNA, or protein expression of PTPN4 through real-time polymerase chain reaction (PCR) or Western blotting.

Measurement of Lactate Dehydrogenase (LDH) Activity.
The content of LDH, which was released in the culture medium, was measured by LDH cytotoxicity assay kit (Roche, Germany) according to the manufacturer's instructions.

In Vivo Left Anterior Descending Artery Ligation Model.
All experimental procedures were approved by The University of Hong Kong Committee on the Use of Live Animals for Teaching and Research. Male adult Sprague-Dawley rats (8 weeks of age) were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg). The in vivo myocardial I/R injury model was induced by occluding the left anterior descending (LAD) artery with a 7-0 silk suture for 30 min followed by 2 h of reperfusion [20]. A sham operation was performed by passing a silk thread under the LAD without occlusion. Myocardial infarct size (IS) was measured by using Evans blue/TTC (1% 2, 3, 5-triphenyltetrazolium chloride) staining and expressed as a percentage of the area at risk (AAR). At the harvest time, the heart was quickly collected for further measurement of miRNA, mRNA, and protein expression.
2.7. Statistical Analysis. All data are presented as means ± standard error of means (S.E.M.). Comparison between groups was carried out by two-tailed unpaired Student's t -test, one-way ANOVA, or two-way ANOVA followed by Bonferroni post hoc test, where appropriate, using the GraphPad Prism 8.0 software (San Diego, CA, USA). In all comparisons, P value less than 0.05 was considered as statistically significant difference.

Results
3.1. miR-181c-5p Was Upregulated Concomitantly with Enhanced NFκB Activity in Posthypoxic H9C2 Cardiomyocytes and Postischemic Myocardium of Rat. We have previously reported that the expression of miR-181c-5p in H/R-stimulated H9C2 cardiomyocytes or postischemic myocardium of rat was significantly increased when compared to control groups [17]. Of note, the current study further demonstrated that upregulation of miR-181c-5p (Figure 1(a)) was paralleled by enhanced NFκB activity, as evidenced by enhanced degradation of IκBα and phosphorylation of IκBα (Ser 32/36 ) and p65 (Ser 536 ) (Figure 1(b)) in H/R-stimulated H9C2 cardiomyocytes. Furthermore, in the in vivo myocardial I/R model, increased myocardial infarction size (Figure 1(c)) was accompanied by upregulation of miR-181c-5p (Figure 1(d)) and enhanced NFκB activity (Figure 1(e)), suggesting that miR-181c-5p may be involved in the NFκB-mediated proinflammatory responses of cardiomyocytes during the pathology of myocardial I/R injury.

Overexpression of miR-181c-5p Exacerbated NFκB-Mediated Inflammation in H9C2 Cardiomyocytes in
Response to H/R Stimulation. To determine the role of miR-181c-5p in NFκB-mediated inflammation in cardiomyocytes under hypoxic condition, the effect of miR-181c-5p on the key enzymes in the NFκB signalling pathway and the expression of NFκB-dependent genes were examined in H9C2 cardiomyocytes by using gain-of-function experiments.
Overexpression of miR-181c-5p was achieved by transfection of miR-181c-5p agomir into H9C2 cardiomyocytes, which resulted in significant increases in the expression of miR-181c-5p ( Figure 2(a)) and in the levels of H/R-stimulated release of LDH ( Figure 2(b)), which were consistent with our previous report [17]. Overexpression of miR-181c-5p did not alter the presence of total p65 but significantly enhanced the degradation of IκBα and increased the level of phosphorylated IκBα (Ser 32/36 ) and phosphorylated p65 (Ser 536 ) in response to H/R stimulation (Figure 2(c)). Furthermore, the H/R-stimulated induction of NFκBmediated proinflammatory cytokines (including IL-1β, IL-6, and TNFα) was further increased in the presence of overexpression of miR-181c-5p by 58%, 72%, and 435%, respectively ( Figure 2(d)). Taken in conjunction, these observations suggested that miR-181c-5p can exacerbate H/R-induced NFκB signalling by facilitating the phosphorylation of IκBα and p65, and thus lead to augmentation of NFκB transcription activity.
3.4. miR-181c-5p Exacerbated LPS-Induced NFκB Signalling in H9C2 Cardiomyocytes. To further explore whether or not miR-181c-5p could exacerbate NFκB-mediated proinflammatory responses in cardiomyocytes, experiments were performed using another stimulus-LPS, which has been proved to induce NFκB-dependent proinflammatory cytokines in cardiomyocytes [23,24]. In response to LPS stimulation, the expression of miR-181c-5p was significantly increased to a level over 2-fold of that of unstimulated cells (Figure 4(a)), suggesting that miR-181c-5p may be involved   in (b, c)), n = 5.

Reduction of PTPN4 Mediated the Proinflammatory
Effect of miR-181c-5p in H9C2 Cardiomyocytes. We have previously reported that miR-181c-5p can directly bind to the 3′ -untranslated region of PTPN4 [17]. Consistently, overexpression of miR-181c-5p leads to significant reduction of protein expression of PTPN4 in cardiomyocytes ( Figure 5(a)). In addition, the expressions of PTPN4 mRNA and protein were both significantly reduced in the H/R or LPS-treated H9C2 cells when compared with unstimulated cells (Figures 5(b)-5(e)). Furthermore, the levels of PTPN4 mRNA and protein were also significantly suppressed in the postischemic myocardium of rat (Figures 5(f) and 5(g)), indicating that PTPN4 may be the potential target of miR-181c-5p. The protein level of PTPN4 was also measured in the H9C2 cells transfected with miR-181c-5p agomir or antagomir with or without H/R stimulation. Overexpression of miR-181c-5p led to significant reduction of PTPN4 at basal condition, and the PTPN4 expression was further reduced in miR-181c-5p agomirtransfected cells after H/R stimulation ( Figure 5(h)). In contrast, inhibition of miR-181c-5p led to significant increment of PTPN4 in both basal and H/R-stimulated condition ( Figure 5(i)). Taken together, these observations lend further support to the interpretation that PTPN4 is the downstream target of miR-181c-5p.
Although emerging evidence has demonstrated that PTPN4 inhibits Toll-like receptor 4/NFκB signalling in mouse peritoneal macrophages [18], it is still unclear whether or not PTPN4 suppresses NFκB-mediated proinflammatory responses in cardiomyocytes, especially in the context of H/R-stimulation. To further explore whether PTPN4 knockdown can reproduce the proinflammatory effect of miR-181c-5p, PTPN4 knockdown model was established in H9C2 cardiomyocytes by using siRNA technology, and these cells were subsequently subjected to normoxia or H/R stimulation. Transfection of PTPN4 siRNA significantly reduced the endogenous protein levels (Figure 6(a)) of PTPN4 in H9C2 cells when compared with mock-transfected cells. As anticipated, PTPN4 knockdown significantly enhanced H/R-induced degradation of IκBα, phosphorylation of IκBα (Ser 32/36 ) and p65 (Ser 536 ) (Figure 6(b)), and mRNA expression of NFκB-mediated proinflammatory cytokines (including IL-1β, IL-6, and TNFα) (Figure 6(c)). To strengthen the notion that reduction of PTPN4 mediates the proinflammatory effect of miR-181c-5p during H/R-induced cell injury, the H9C2 cells were cotransfected with miR-181c-5p antagomir and PTPN4 siRNA and subjected to H/R stimulation. Transfection of miR-181c-5p antagomir alone significantly attenuated phosphorylation of IκBα (Ser 32/36 ), degradation of IκBα, and phosphorylation of p65 (Ser 536 ) upon H/R stimulation, while cotransfection of miR-181c-5p antagomir and PTPN4 siRNA cancelled the anti-inflammatory effect of miR-181c-5p antagomir, as evidenced by enhanced phosphorylation of IκBα (Ser 32/36 ), degradation of IκBα, and phosphorylation of p65 (Ser 536 ) (Figure 6(d)). Collec-tively, these results demonstrated that miR-181c-5p may exacerbate NFκB signalling pathway and thus aggravate cardiomyocyte inflammation and cell injury by directly targeting PTPN4 expression in H9C2 cardiomyocytes.

Discussion
Given the detrimental effect of the sustained NFκB activation in the ischemic heart disease [5,6], examining ways to attenuate excessive NFκB-mediated inflammation is of clinical interest to combat cardiac I/R injury. The present study demonstrated that the NFκB activity was significantly increased, with concomitantly upregulated miR-181c-5p level in the postischemic myocardium and H/R-stimulated H9C2 cardiomyocytes when compared to the control groups, suggesting that increased level of miR-181c-5p may be involved in the NFκB-mediated inflammation during myocardial I/R injury. Indeed, overexpression of miR-181c-5p exacerbated H/Rinduced cell injury (greater LDH leakage), and its proinflammatory effect in cardiomyocytes involves activation of NFκB signalling pathway, as evidenced by enhanced degradation of IκBα, increased level of phosphorylated IκBα (Ser 32/36 ) and phosphorylated p65 (Ser 536 ), and augmented expression of proinflammatory cytokines in response to H/R stimulation. In contrast, inhibition of miR-181c-5p in vitro had the opposite effect in NFκB-mediated inflammation. Of note, neither overexpression nor inhibition miR-181c-5p altered the phosphorylated IκBα (Ser 32/36 ) or total IκBα at basal condition. The proinflammatory effect of miR-181c-5p may require the suppression or elevation of some other molecules during H/R. Indeed, in response to H/R, multiple signalling pathways were altered, such as hypoxia-inducible factor 1-α (HIF-1α) [20] and cyclooxygenase-2 (COX-2) [19], both of which are involved in the H/R-induced inflammation. However, whether HIF-1α, COX-2, or other molecules work as cofactors and participate in the proinflammatory effect of miR-181c-5p are still unclear and warrants further investigation. Taken together, these results indicated that miR-181c-5p enhances NFκB-mediated inflammation in cardiomyocytes in response to H/R stimulation.
In addition to explore the role of miR-181c-5p in H/R stimulation induced inflammation, LPS was used in the present study as another stimulus to activate NFκB signalling pathway in H9C2 cardiomyocytes. In response to LPS stimulation, IκBα is phosphorylated at serine 32 and 36, followed by ubiquitination and proteasome-mediated degradation [25,26], leading to the dissociation of IκBα from NFκB. The activated NFκB moves into the nucleus and binds to specific sequences of DNA κB sites resulting in the transcription of NFκB-mediated genes [25,26]. As a component of the bacterial cell wall, LPS has been widely used to establish sepsis model in vivo and in vitro, because LPS induces profound inflammation and pathological consequences similar to those found during sepsis [27,28]. Sepsis, an acute inflammatory disease, is a life-threatening condition that follows bacterial infection [29,30]. Cardiac dysfunction could be an important consequence of sepsis/septic shock and contributes to the high mortality because of the elevated inflammation [29,30]. In the present study, overexpression of miR-181c-  proinflammatory effect of miR-181c-5p in sepsis, especially sepsis-induced myocardial dysfunction has not been explored yet but is worth further investigation.
In response to different stimuli, besides to the classical phosphorylation sites of IκBα at serine 32 and 36, phosphorylation of IκBα at tyrosine residue 42 also mediates the degradation of IκBα and the subsequent NFκB activation [31,32]. In a T lymphocytic cell line (EL4), mutation of Ser 32/36 in IκBα had no effect on H 2 O 2 -induced NFκB activation, but mutation of tyrosine 42 abolished NFκB activation by H 2 O 2 [31]. Consistently, the cardiac NFκB activation was completely blocked in a murine model which expressed the mutant IκBα (S32A, S36A, Y42F) in a cardiac-specific manner, while in the mice expressed two mutant IκBα (S32A, S36A), the NFκB activation was only partially blocked (70-80%), indicating that phosphorylation of IκBα at tyrosine residue 42 mediates NFκB activation independent of Ser 32/36 phosphorylation in IκBα [32]. In addition, in HepG2 liver cells, in response to TNFα, activated cytosolic calpains has been shown to degrade IκBα and activate NFκB signalling independently of the ubiquitin-proteasome pathway [33]. In the present study, overexpression of miR-181c-5p leads to the phosphorylation of IκBα at Serine 32 and 36, IκBα degradation and NFκB activation in H9C2 cells in response to H/R. However, whether or not miR-181c-5p may affect the phosphorylation of IκBα at tyrosine residue 42 or cytosolic calpains activity is still unclear and merits further studies.
During the course of our study, we noticed that there are some studies reported that miR-181c-5p can attenuate excessive neuroinflammation through directly targeting the 3′-untranslated region of TNFα mRNA, suppressing its mRNA and protein expression in rat microglial cells after ischemia injury [15], which is opposite to our findings that miR-181c-5p exerts proinflammatory effect through enhanced NFκB signalling in cardiomyocytes. The different cell types (BV-2 microglial cells vs. H9C2 cardiomyocytes) could be the main explanation regarding the discrepancy between our and others' study. The proinflammatory effect of miR-181c-5p may be cell specific. Similarly, the inhibitory effect of miR-181b (another member in miR-181 family) on NFκB activation was specific to endothelial cells but not observed in other cell types, such as peripheral blood mononuclear cells [14]. Additionally, the expression pattern of miR-181c-5p varies in different cell types in response to stimulation. In oxygen-glucose deprivation activated BV-2 cells, there was significantly reduced level of miR-181c-5p in a time-dependent manner when compared with nonstimulated cells [15]. However, in the present study, in H9C2 cardiomyocytes, the expression of miR-181c-5p was significantly upregulated in response to H/R or LPS stimulation. Furthermore, in our study, transfection with miR-181c-5p agomir leads to an approximately 6000-fold increase of miR-181c-5p level, whereas there is only 20-fold increase of miR-181c-5p in others' work [15], which may also be viewed as a discrepancy between our and others' study. Moreover, the unaltered mRNA expression of TNFα in the H9C2 cardiomyocytes with miR-181c-5p overexpression also rules out the possibility that miR-181c-5p can directly target 3 ′ -untranslated region of TNFα mRNA in the current experimental setting. In addition, loss of miR-181c-5p in the mitochondrial compartment shows cardioprotective effects during myocardial I/R injury [34]. In another study, miR-181c-5p aggravates brain injury in acute ischemic stroke through enhancement of apoptosis of microglia and neurons [16]. Taken together, these studies provide additional evidence that miR-181c-5p may play detrimental (proinflammatory) roles during ischemic attack.
We recently reported that miR-181c-5p exacerbates cardiomyocytes injury and apoptosis by directly targeting the 3′ -untranslated region of PTPN4 [17]. This target was substantiated through several lines of evidence: (1) overexpression of miR-181c-5p results in the significantly reduction of protein level of PTPN4 in H9C2 cardiomyocytes, which has also been repeated in the present study; (2) significant reduction of mRNA and protein level of PTPN4 in H9C2 in H/R or LPStreated H9C2 cardiomyocytes and postischemic myocardium of rat; (3) complementary sequence of miRNA-181c-5p was located on the positions from 4915 to 4921 or from 6333 to 6339 (or both) on the 3′UTR of rat PTPN4 mRNA, and there were 7 pairs of Watson-Crisk match; (4) mutation of miR-181c-5p binding sites blocked miR-181c-5p-mediated repression of PTPN4 in 293T cells [17]; (5) PTPN4 knockdown recapitulated the proapoptotic effect of miR-181c-5p in H9C2 cardiomyocytes. In addition to its reported effects in protecting against cell apoptosis, PTPN4 has been also reported to suppress Toll-like receptor 4 and may inhibit its downstream NFκB signalling in mouse peritoneal macrophages [18]. These findings prompted us to hypothesize that miR-181c-5p may enhance NFκB-mediated inflammation through targeting PTPN4 in H9C2 cardiomyocytes. In line with this speculation, siRNA-mediated knockdown of PTPN4 expression reproduced the proinflammatory effect of miR-181c-5p on NFκB signalling in cardiomyocytes, as evidenced by the increased H/R-induced degradation of IκBα, phosphorylation of IκBα (Ser 32/36 ) and p65 (Ser 536 ), and mRNA expression of NFκB-mediated proinflammatory cytokines (including IL-6, IL-1β, and TNFα]. Furthermore, cotransfection with miR-181c-5p antagomir and PTPN4 siRNA cancelled the anti-inflammatory effect of miR-181c-5p antagomir. Taken in conjunction, these findings implicate that miR-181c-5p may serve as a regulator of NFκB-mediated inflammation through targeting PTPN4 in H9C2 cardiomyocytes. It is still unclear whether or not miR-181c-5p targets PTPN4 and exerts proinflammatory effect in vivo during I/R injury and warrants investigation in the future study by using miR-181c-5p knockout/overexpression mice.

Conclusion
In conclusion, the present study demonstrates that miR-181c-5p is involved in the enhanced NFκB-mediated inflammation through targeting PTPN4 during myocardial I/R injury or H/R-stimulated cardiomyocyte injury. These observations suggest that increased miR-181c-5p level may serve as a potential risk factor, and future studies will focus on the potential clinical use of miR-181c-5p to combat inflammatory diseases, including ischemic heart disease.

11
Oxidative Medicine and Cellular Longevity