Microglial Mincle receptor in the PVN contributes to sympathetic hyperactivity in acute myocardial infarction rat

Abstract Malignant ventricular arrhythmias (VAs) following myocardial infarction (MI) is a lethal complication resulting from sympathetic nerve hyperactivity. Numerous evidence have shown that inflammation within the paraventricular nucleus (PVN) participates in sympathetic hyperactivity. Our aim was to explore the role of Macrophage‐inducible C‐type lectin (Mincle) within the PVN in augmenting sympathetic activity following MI,and whether NOD‐like receptor family pyrin domain‐containing 3 (NLRP3) inflammasome/IL‐1β axis is involved in this activity. MI was induced by coronary artery ligation. Mincle expression localized in microglia within the PVN was markedly increased at 24 hours post‐MI together with sympathetic hyperactivity, as indicated by measurement of the renal sympathetic nerve activity (RSNA) and norepinephrine (NE) concentration. Mincle‐specific siRNA was administrated locally to the PVN, which consequently decreased microglial activation and sympathetic nerve activity. The MI rats exhibited a higher arrhythmia score after programmed electric stimulation than that treated with Mincle siRNA, suggesting that the inhibition of Mincle attenuated foetal ventricular arrhythmias post‐MI. The underlying mechanism of Mincle in sympathetic hyperactivity was investigated in lipopolysaccharide (LPS)‐primed naïve rats. Recombinant Sin3A‐associated protein 130kD (rSAP130), an endogenous ligand for Mincle, induced high levels of NLRP3 and mature IL‐1β protein. PVN‐targeted injection of NLRP3 siRNA or IL‐1β antagonist gevokizumab attenuated sympathetic hyperactivity. Together, the data indicated that the knockdown of Mincle in microglia within the PVN prevents VAs by attenuating sympathetic hyperactivity and ventricular susceptibility, in part by inhibiting its downstream NLRP3/IL‐1β axis following MI. Therapeutic interventions targeting Mincle signalling pathway could constitute a novel approach for preventing infarction injury.


| INTRODUCTION
Ventricular arrhythmias (VAs) following acute myocardial infarction (AMI) are common complications which remain a major cause of mortality, and primarily result from sympathetic nerve hyperactivity. 1 The paraventricular nucleus (PVN) is a nucleus controlling cardiovascular regulation and is located in the hypothalamus of the central nervous system(CNS). 2 Recent studies have demonstrated that inflammation within the PVN is related to regulation of the cardiovascular sympathetic tone, [3][4][5] however, the precise mechanism remains largely unknown.
Innate immunity makes a great sense to protect host against pathogens' invasion, 6,7 as well as in the pathological process of several diseases including MI. Various pattern recognition receptors (PRRs) are activated, indicating the initiation of innate immunity. 6 Macrophage-inducible C-type lectin (Mincle), also identified as Ctype lectin domain family 4 (CLEC4E), is a kind of transmembrane germline-encoded PRR 8 and is expressed on surface of various immune cells comprising monocytes/macrophages, neutrophils and others. 9 It could be activated via recognizing a kind of endogenous ligand named Sin3A-associated protein 130 kDa (SAP130) which belongs to subunit of the histone deacetylase. 10 The downstream protein spleen tyrosine kinase (Syk) is phosphorylated after activation of Mincle receptor, followed by the various routes to synthesize pro-inflammatory cytokines like IL-1β and chemokines, 10 resulting in the infiltration of macrophages and phagocytizing dying cells. 11 Previous studies have demonstrated that Mincle is recognized as a launching point in various noninfectious inflammatory diseases such as traumatic brain injury and subarachnoid haemorrhage (SAH). 10,12 However, the role of Mincle in MI has not yet been explored.
Several studies have reported that the activation of microglia and an increase in pro-inflammatory cytokines are observed in the PVN post-MI. 13,14 Among all of the pro-inflammatory cytokines, IL-1β is recognized as a key trigger of inflammation. It is released following the activation of microglia in CNS and is involved in sympathetic hyperactivity. 15 The nucleotide-binding and oligomerization domain-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome containing of apoptosis speck-like protein containing a caspase-recruitment domain (ASC), NLRP3 and pro-caspase-1 is reported to participate in cleaving the IL-1β precursor to form the mature and secreted structure. 16,17 Simultaneously one previous study reported that Mincle receptor plays an important role in NLRP3 inflammasome formation and IL-1β synthesis by activated myeloid cells. 18 Mincle receptor is primarily expressed in microglia known as the monocyte-macrophage cell lineage of the CNS. 19

| Experimental design
The entire experiments were divided into three separate protocols.

| Protocol 2
A total of 187 rats were randomly placed into four groups: (1) sham  in naïve group did not receive any procedure except anesthesia until RSNA measurement. The rats were stimulated with LPS via PVN microinjection, and 3 hours later recombinant SAP130 (rSAP130) was administrated. To promote the gene silencing efficiency, two different NLRP3 siRNA sequences were selected: (a) 5′GAUCCUAUUU- siRNA was injected into the PVN 24 hours prior to LPS stimulation, and pharmacological IL-1β inhibitor gevokizumab was used immediately after rSAP130 administration. On the time duration, animals which did not receive targeted reagents' injection were microinjected with same dosage of saline as control to ensure that each rat was microinjected to PVN for three times. Six hours later, the RSNA was recorded, and tissues were collected for ELISA and Western blot analysis.

| PVN microinjection
In protocol 2 and 3, the rats were mounted on a stereotaxic appara-

| MI surgery
In protocol 1 and 2, MI models were developed as described previously. 25 Procedures were performed in a sterile environment and rats underwent a 12 hours of fasting period before and after their surgical procedures. Buprenorphine (0.05 mg/kg) was injected subcutaneously to control the pain after anesthetization, then rats were intubated via tracheotomy and ventilated with a small-animal ventilator. The heart was exposed via a left thoracotomy at the third and fourth intercostal space and the left coronary artery was ligated 2-3 mm from its origin between the pulmonary artery conus and the left atrium with a 6-0 polypropylene ligature. In the sham animals, we exposed the heart but did not ligate the artery. Postsurgery, the rats were not returned to their own cages but were placed on a heating pad at 37°C to maintain their body temperature until they woke up. Bupivacaine (1 mg/kg) was administered at the intercostal incision during closure, and penicillin (0.8 million units) was injected intramuscularly to prevent infection. Animals were monitored daily after the surgery. Tissue samples were harvested 24 hours following MI surgery in protocol 2.

| HRV measurements
Telemetry was conducted on the rats in protocol 2 after the MI surgery. We placed a radiotelemetry transmitter (TR50B; AD Instruments, Sydney, Australia) body in the abdominal cavity of the rats; the positive lead wire was placed on the xiphoid process, while the negative lead wire was placed between the sternocleidomastoid muscles. The frequency domains were very-low-frequency (VLF; 0.05 Hz), low-frequency (LF; 0.05-0.75 Hz), and high-frequency (HF; 0.75-2.5 Hz) bands. Data from dynamic electrocardiography (ECG) was recorded via the LabChart Pro software (AD Instruments). The ratio of LF to HF was calculated. The ECG was monitored for 24 hours until the RSNA recording.

| RSNA recording
The RSNA recording was conducted to monitor the activity of the renal sympathetic nerve after the rats were anaesthetized with 2% phenobarbital sodium (40 mg/kg). The rats were placed in the left lateral position before receiving a left flank incision. The renal sympathetic nerve was separated from the surrounding tissue in the abdominal cavity under an anatomical microscope and hooked to a pair of bipolar silver wire recording electrodes. The electrodes and the nerve were immersed in liquid paraffin to insulate them from the surrounding tissue and to prevent desiccation. An oscilloscope (AD instruments) was used to amplify, filter and monitor the electrical data with a frequency cut-off from 100 to 3000 Hz. The bipolar platinum electrode was connected to a biological polygraph (AD instruments) to record the RSNA simultaneously. The changes in RSNA were calculated as percentage change from the baseline activity recorded before the experiments. Both the integrated and raw RSNA values are represented in the results; however, only the integrated data were statistically analysed. 26

| In vivo electrophysiological experiments
Programmed electrical stimulation was conducted as previously described. Briefly, the rats were placed in the supine position to receive thoracotomy, and electrodes were implanted on the epicardial surface of the LV. 27 Eight paced beats were applied at a basic cycle length of 150 milliseconds (S0) to induce VAs, followed by single (S1), double (S2), and triple (S3) additional stimuli at shorter coupling intervals. The stimulation intensity was twice the threshold, and the stimulus length was set to 5 milliseconds. The stimulating protocols were completed within 10 minutes. An arrhythmia scoring system was applied to measure the induced arrhythmic degree. All the protocols were conducted via an animal biological function experiment system (LEAD-7000; JJET, Chengdu, China).

| Tissue collection
An intravenous injection of KCl was administrated to kill the animals following the electrophysiological study, after which they were decapitated, and the whole brain was removed. The brain tissue in each group was collected using different methods. (a) The location of the PVN was determined, and a microinjection of methylene blue was administered in line with the Paxinos and Watson rat atlas 28 (supplemental data). The PVN was separated from the whole brain tissue and stored at −80°C immediately for further biochemical analysis. (b) The brain tissues were collected and placed in freshly prepared 4% paraformaldehyde for 12 hours at 4°C and transferred to 10% neutral buffered formalin (NBF) until sectioning. (c) The brain tissue was immersed in PBS solution containing 30% sucrose overnight, embedded in Tissue-Tek ® OCT compound (Sakura Finetek, Tokyo, Japan) and frozen at −80°C before further analysis. The sample size for immunohistochemical studies and immunofluorescence was 2.0 × 1.0 cm 2 . The hearts and blood were additionally harvested. After stored at −80°C for 1 hour, the heart was cut into 2 mm thick slices and stained with 1% triphenyltetrazolium chloride (TTC) for 30 minutes at 37°C. We recognized the infarcted area based on visualization of the pale ventricle wall 29 and measured it via TTC staining using Image ProPlus 5.0 software (supplemental data). An infarct size between 30% and 50% for each rat was chosen in the study.

| Western blot
The PVN tissues were homogenized in RIPA lysis buffer (Beyotime 2.12 | Enzyme-linked immunosorbent assay (ELISA) The peripheral blood and myocardium tissues were treated using a double-antibody sandwich ELISA kit (Cusabio Biotech Co, Wuhan, China) according to the manufacturer's instructions to detect norepinephrine (NE), a marker of sympathetic nerve activity. The NE tissue concentration was reported in pg/mg myocardium tissue. Each treatment was repeated at least three times. The mean values of the concentration of NE were statistically compared among the groups.

| Statistics
The data are presented as the means ± standard deviations (SDs).
Values for more than two groups were compared using ANOVA followed by Tukey's test using SPSS software, version 19.0 (SPSS Inc., Chicago, IL, USA). Statistical significance was considered as P < 0.05.

| Mincle was distributed in microglia within the PVN after MI induction
Double-staining immunofluoresence was conducted in our study.
Mincle was seen as green fluorescence, and Iba-1 was seen as red fluorescence. We observed a similar tendency in Mincle and Iba-1 expression via immunofluorescence. The presence of Mincle receptor was almost co-localized with microglia within the PVN (Figure 3).

| Mincle was involved in sympathetic hyperactivity in MI rats
The data depicted above supported the hypothesis that inhibition of Mincle receptor could be sufficient to blunt microglial expression as well as NLRP3 and IL-1β activation post-MI. Then, we investigated the sympathetic activity and biological function via analysing the levels of NE, a neural transmitter secreted by the sympathetic nerve, in the peripheral blood and myocardium as well as by measuring the RSNA, which indicates the activity of the efferent sympathetic nerve, heart rate variability (HRV) and programming electrical stimulation. We observed that MI could induce an increase in NE in both the plasma and the myocardium. NE secretion in the Mincle-knockdown treated rats was markedly lower compared with that in the MI models ( Figure 4C,D). The NE levels mirrored the RSNA content ( Figure 4B). A higher VA score was also observed via electrical stimulation, suggesting sympathetic nerve hyperactivity ( Figure 5). The HR, which clinically indicates sympathetic nerve excitation, 31 was notably increased in the MI rats vs the baseline in the sham rats.
PVN microinjection of Mincle siRNA in the AMI-treated rats decreased the HR (Table 1). In protocol 3, we used several rats without any experiments as a real control group to compare with scramble siRNA-microinjection-treated groups. There were no significant differences observed between naïve group and scramble siRNA group, therefore we could exclude the potential effects of microinjection itself. It is reported that recombinant SAP130 can cooperate with LPS as a synergistic effect to induce robuster inflammatory gene expression and inflammasome activation in macrophages. 32 Therefore, recombinant SAP130 was applied locally with LPS to stimulate Mincle in the PVN. We observed that Mincle receptor and IL-1β protein expression were significantly increased by LPS+rSAP130 stimulation compared with LPS group, while there was no statistically difference in Iba-1 and NLRP3 levels between LPS+rSAP130 and LPS stimulated rats. We hypothesized that LPS has already been a potent stimulator for microglial activation but not for Mincle receptor. In contrast, the inhibition of NLRP3 or IL-1β did not affect the expression of Mincle and microglia in the LPS+rSAP130-treated rats ( Figure 6).
Next, we explored the mechanism of peripheral sympathetic hyperactivity via Mincle activation in the PVN. Our data showed that LPS+rSAP130 significantly increased the NE levels in the plasma and myocardial tissue ( Figure 7C,D) and the RSNA (P < 0.05, Figure 7B); this tendency was partly attenuated by the administration of NLRP3 siRNA or IL-1β antagonist in serum NE concentration and the RSNA level. We could speculate that Mincle in the PVN participated in sympathetic nerve activity at least partly via the NLRP3/IL-1β pathway.

| DISCUSSION
This study reported the novel findings that MI induces the activation of Mincle expressed in microglia within the PVN, further causing sympathetic hyperactivity, and this process is partly associated with the NLRP3/IL-1β pathway.
Rapid life-threatening VAs including ventricular tachyarrhythmia (VT) and/or ventricular fibrillation(VF) lead to increased mortality and lethality following AMI. 33 In the recent study, arrhythmic mechanism is different compared with previous studies which were investigated Mincle is a key C-type lectin receptor and was originally found based on its strong induction in macrophages by inflammatory stimuli. 35 It is rarely expressed in the normal circumstances, however, it can be activated powerfully following recognizing various stimuli, including microbial products or endogenous signals of tissue injury, such as debris from apoptotic and necrotic cells, heat shock proteins and nucleic acid fragments, then lead to macrophage excitation and induce a local inflammatory response. 36 Previous studies demonstrated that Mincle plays an essential role in the pathologic process of acute renal inflammation, alcohol-induced liver injury, traumatic Each column with a bar represents the mean ± SD. *P < 0.01 and **P < 0.05 vs sham groups; # P < 0.05 versus MI+ scramble siRNA group brain injury and SAH, 9,10,12,37 however, little is known underlying the role of Mincle in sympathetic regulation in MI rats. To determine the effects of activated Mincle on sympathetic nerve activity post-MI, Mincle siRNA was administered. It is reported that various "RNA interference (RNAi) therapeutics" have been applied in clinic trials. 38 The superiority in a Mincle-specific siRNA is to ensure the protective effects for experimental animals and eliminative results for a longer term inhibition in vivo. We observed an increase in RSNA, which is representative of enhanced sympathetic activity, and increased HRV, which may indicate dysfunction of the cardiac autonomic nervous system, following MI. MI-induced VAs are hypothesized to be related to cardiac sympathetic overdrive; we analysed the NE content in the myocardium to test this hypothesis. Knockdown Mincle could induce a reduction in NE in the myocardium, and the ventricular arrhythmia score was reduced, supporting the concept that the VF threshold could be reduced when confronted with sympathetic blockade or vagal up-regulation in animal studies. 39 We further explored the mechanism of the PVN Mincle in sympathetic activation.
SAP130, an endogenous ligand for Mincle, is normally related with assembling the spliceosome and consisting of the U2 small nuclear ribonucleoprotein-associated protein complex. 40 Under steady condition, neither SAP130 nor Mincle is expressed at a high level. 41 We noted that there was an increase in SAP130 release within the PVN post-MI, however, the source of that need more exploration. SAP130 released from injured cells post-MI may be one of the mechanism. Another possibility is that SAP130 is secreted through vesicles from macrophages. 42 One previous study reported that rSAP130 administration markedly increased the expression of IL-1β, however, there was no changes in Mincle expression after SAH. 12 It suggested that Mincle might have various self-and non- | 119 self-ligands other than SAP130. SAP130 has been identified to activate Mincle signalling in macrophages in vitro, 12 however, no abundant evidence was found that exogenous SAP130 alone may active Mincle in vivo of naïve rats, therefore we used LPS and recombinant SAP130 to investigate the mechanism of Mincle function in vivo.
We observed that Mincle was activated in parallel with NLRP3 expression increase as well as IL-1β release after LPS+rSAP130 administration, and the RSNA was high as well as NE increased while they were both inhibited by NLRP3 siRNA or IL-1β inhibitor. IL-1β plays an essential role in regulating various physiological, behavioural, and endocrine actions in the CNS. 43 It is reported that central microinjection of IL-1β could promote the release of serum NE and increase blood pressure. 44 Different from most other pro-inflammatory cytokines, however, IL-1β is not synthesized via gene expression solely. The reason is that IL-1β lacks a signal peptide and is not released through the default ER-Golgi dependent mechanisms employed by other cytokines. It is synthesized and maintained as an inactive pro-form in the cytoplasm. Intracellular signalling complexes called inflammasomes containing caspase-1 are activated following sensing a distinct stimulus or set of stimuli. 45,46 Next, caspase-1 cleaves pro-IL-1β into active and mature IL-1β. Simultaneously caspase-1 plays an important role in the secretion of IL-1β, however, the mechanism remains poorly understood. Previous studies have reported several types of inflammasomes and the best-characterized one is the NLRP3 inflammasome, which contains the sensor molecule NLRP3, the adaptor protein ASC, and pro-caspase-1. 46,47 Reports from us and others demonstrated that the NLRP3 inflammasome is related with several sterile inflammatory diseases, including obesity, atherosclerosis and MI. 25,48 On the other hand, Syk, the downstream protein of Mincle, regulates NLRP3 activity via the production of reactive oxygen species. 18 Qi et al 23    LPS+rSAP130 + NLRP3 siRNA (e) and LPS+SAP130 + gevokizumab group (f). The cytokine levels of NE from serum (C) and myocardium tissue (D) as measured by ELISA. The data are expressed as the mean ± SD. of two experiments. n = 14-16 in each group. *P < 0.01 and **P < 0.05 compared with the naïve group; #P < 0.05 compared with the LPS group; & P < 0.05 vs LPS+rSAP130 group; @ P < 0.05 vs LPS+rSAP130 + NLRP3 siRNA group inhibitory activities from sympathetic neurons. 52,53 It was reported in heart failure and hypertensive rat models which may be also applied in MI rats.

| Outlook
The present results indicate that Mincle expressed in microglia within the PVN is markedly activated in MI rats following sympatho-excitation. The mechanism on the role of inflammation within the PVN in sympathetic regulation was previously conducted, 53 however, our study constitutes a novel perspective exploring Min-

CONFLI CT OF INTEREST
None.