C1q/TNF‐related protein‐9 promotes macrophage polarization and improves cardiac dysfunction after myocardial infarction

Abstract The timely regulation of inflammatory M1 macrophage polarization toward regenerative M2 macrophages suggests the possibility of immunotherapy after myocardial infarction (MI). C1q/TNF‐related protein‐9 (CTRP9) has anti‐inflammatory effects and can ameliorate heart function in mice after long‐term myocardial infarction. The role of CTRP9 in macrophage polarization remains completely unclear. This study determined whether CTRP9 can preserve post‐MI early cardiac function through the regulation of macrophage polarization. In the present study, an adenovirus‐delivered CTRP9 supplement promoted macrophage polarization at Day 3 post MI and improved cardiac function at Day 7 post MI. Pretreatment with gCTRP9 promoted the M1 to M2 polarization transition and attenuated inflammation after lipopolysaccharide + interferon‐γ stimulation; the effects were partly abrogated by the adenosine monophosphate kinase (AMPK) inhibitor compound C and were obviously reinforced by pyrrolidine dithiocarbamate, a nuclear factor‐κB (NF‐κB) inhibitor. Meanwhile, CTPR9 markedly reduced the expression of toll‐like receptor 4 (TLR4), myeloid differentiation factor 88 (MyD88), and NF‐κB p65 phosphorylation by promoting AMPK phosphorylation in vivo and in vitro. Moreover, the competitive binding of gCTRP9 and LPS to the myeloid differentiation protein 2 (MD2)/TLR4 complex was associated with direct binding to MD2, thereby inhibiting the downstream signaling molecule MyD88. Taken together, we demonstrated that CTRP9 improved post‐MI early cardiac function, at least in part, by modulating M1/M2 macrophage polarization, largely via the TLR4/MD2/MyD88 and AMPK‐NF‐κB pathways.


| INTRODUCTION
Inflammation subsidence is important for improving tissue injury and cardiac function after myocardial infarction (MI) and requires the concerted action of macrophages. Peripheral monocytes/macrophages and cardiac-resident macrophages migrate to the infarct and border region, differentiate into M1 macrophages, and exacerbate cardiac function. Thereafter, entering the inflammation resolution stage, the dominant macrophage populations of myocardial tissues become reparative M2 macrophages that accelerate cardiac repair (Gombozhapova et al., 2017;Tokutome et al., 2018).
Inflammation from M1 macrophages is an essential component of early ventricle remodeling, and prolonged inflammation may damage the physiology of the left ventricle (LV) by raising LV expansion and redundant scar transformation (Ben-Mordechai et al., 2015;Gombozhapova et al., 2017). Many studies have emphasized the therapeutic potential of pharmacological regulators in the treatment of myocardial injury through the modulation of macrophage polarization (Cheng & Rong, 2018;Shirakawa et al., 2018). Regarding the features of phenotypic dynamics after MI, M1 macrophages predominate early at 1-3 days and peak on Day 3, whereas M2 macrophages gradually predominate during inflammation regression after 5 days (Christia et al., 2013;Mouton et al., 2018;Yan et al., 2013). Earlier phenotype transformation of M1 to M2 macrophages has resulted in prominent improvements in MI wound healing and cardiac function (Choo et al., 2017). Hence, positively transforming to the M2 phenotype at 3 days is a therapeutic key to reduce adverse LV remodeling after MI.
The C1qTNF-related proteins (CTRPs) are a recently discovered and highly conserved family of adiponectin (APN) paralogs. Among CTRPs, CTRP9 is highly expressed in heart and adipose tissue (Appari et al., 2017). The globular domain isoform of CTRP9 (gCTRP9) is generated via proteolytic cleavage of full-length CTRP9 (fCTRP9) and is the main active isoform (Y. Yuan et al., 2015). Studies have shown that CTRP9 has anti-inflammatory effect via AMPK activation (Kambara et al., 2015;Zhang et al., 2016) and can attenuate adverse cardiac remodeling in long-term MI, largely via a protein kinase A (PKA)-dependent pathway (Sun et al., 2013). Simultaneously, we consider that cardiac CTRP9 expression might be regulated by PPAR-γ (Su et al., 2013), and CTRP9 prevents activation of the TLR4-MyD88-p65 pathway by activating AMPK to inhibit the cholesterol-induced vascular smooth muscle cell (VSMC) phenotype switch (Q. Liu et al., 2017). Conversely, we have known that the signaling molecules TLR4, AMPK, NF-κB, and PPAR-γ, which are associated with CTRP9, are involved in macrophage polarization.
Given this evidence, whether CTRP9 can regulate cardiac function in the early stage post MI by affecting macrophage polarization and the specific mechanism remain largely unknown.
In the present study, we found that overexpression of CTRP9 in an MI model can enhance M2 macrophage polarization and improve cardiac function in the early stage post MI. gCTRP9 can reduce the inflammatory response by promoting the M1 to M2 macrophage transition in vitro, which was mechanistically associated with the TLR4/MD2/MyD88 and AMPK-NF-κB signaling pathways. In general, we provided evidence that CTRP9 acts as a critical regulator of cardiac function and macrophage polarization through suppression of the NF-κB signaling pathway.

| Adenoviral vector construction
A recombinant adenovirus (Ad) encoding full-length rat CTRP9 (Ad-CTRP9; PubMed No. NM_001191891) or green fluorescent protein (Ad-GFP) for the control virus was constructed and amplified.
After confirming the sequence, we purified the recombinant virus and quantified the final plaque-forming units (pfu; Genechem Co., Ltd., Shanghai, China).
Cells were transfected with these shRNAs, and messenger RNA (mRNA) expression of CTRP9 was detected by reverse-transcription quantitative polymerase chain reaction (RT-qPCR). CTRP9-shRNA3 was the most efficient construct for knocking down CTRP9 expression.  Yuan et al., 2016). For all procedures, samples were randomized and analyzed in a blinded manner. Animals were randomly divided into four groups: Sham, MI, Ad-GFP + MI, and Ad-CTRP9 + MI. Then, 3 × 10 9 pfu of Ad-CTRP9 or Ad-GFP was injected into the jugular vein of each rat 5 days before MI. The Ad-CTRP9 dose was based on previous research (Kambara et al., 2015), according to the conversion of the dose/body weight from mice.

| Experimental protocols in vivo and cardiac function
Nonsurviving rats underwent an autopsy and cardiac rupture was confirmed by the presence of coagulated blood in the thoracic cavity or observation of the crevasse on the ventricle. Echocardiography was performed as previously described at 7 days (per group, successively, n = 7, 6, 6, and 7; M. Yang, Xiong, Zou, Wang & Huang, 2018). After 3 or 7 days, rats were killed by intracardiac injection of KCl to induce diastolic arrest of cardiac activity. At the end of the study, all hearts were used to perform PCR, western blot analysis and enzyme-linked immunosorbent assay (ELISA) at the peri-infarct area (>2 mm outside the infarct).

| Quantitative PCR analysis
Total RNA was extracted from cells by TRIzol reagent (15596-026; Invitrogen, Carlsbad, CA), and the conversion of mRNA into cDNA was performed using an RT kit with genomic DNA Eraser (RR047A; TaKaRa, Kusatsu, Shiga, Japan). RT-qPCR was performed using SYBR

| Cytokine quantification
Following treatment, IL-1β, IL-6, and IL-10 levels in cell culture supernatants and heart tissue were measured using rat ELISA kits according to the manufacturer's instructions (Elabscience, Wuhan, China).
Protein levels were calculated from a cytokine standard curve. Tissue ELISA measurements were normalized to the protein content of the homogenates (μg proteins/ml). All samples were analyzed in triplicate. FlowJo software (Tree Star, Ashland, OR). iNOS + /F4/80 + macro-phages were regarded as M1 macrophages, and CD163 + /F4/80 + macrophages were regarded as M2 macrophages.

| Masson's trichrome staining
Masson's trichrome staining was performed to evaluate infarct size in heart tissue sections. Fibrosis was measured via Image-Pro Plus (Media Cybernetics, Maryland, GA). With respect to clinical significance, only rats with large infarctions (>30%) were selected for analysis. Digitized pictures were analyzed by planimetry, and the infarcted area was measured as the percentage of stained fibrosis area/total LV area.
2.10 | Effects of CTRP-9 on binding to a TLR4/MD2 fusion protein We validated whether CTPR9 could compete with LPS for binding to TLR4/MD2-MyD88. We simulated the process using the following steps, which were derived from an ELISA-based assay at a wavelength of 450 nm (Brandl, Gluck, Hartmann, Salzberger & Falk, 2005;Kopp et al., 2010).
First, we designed the TLR4/MD2 molecule (Bioyeartech, Wuhan, China), which was prepared by fusing MD2 with a C-terminal His tag to the C terminus of soluble TRL4 via a flexible linker during plasmid construction (Flag-TLR4-Linker-MD2-6 × His-pcDNA3.1 + vector; LPS trap) and transfected into macrophages. The primer sequences used are listed in Table 2. This so-called LPS trap was successfully proven to bind LPS in RAW 264.7 cells (Brandl et al., 2005). An anti-Flag antibody (M2; 66008-2-Ig; Wuhan Sanying) was coated on a plate at an appropriate dilution (2 μg/ml) in carbonate buffer and then added to the cell lysate (containing the LPS trap) and specifically bound to M2 on the ELISA plate. Second, we constructed bio-LPS (B2643; Sigma-Aldrich). Third, a 100-fold dose of nonbiotinylated LPS and three different concentrations of gCTRP-9 were added in groups to simultaneously compete with bio-LPS for binding to the LPS trap. Then, to rule out the possibility that the effects of gCTRP9 on the LPS trap were attributable to the fact that gCTRP9 binds directly to bio-LPS, we coated gCTRP-9 (0 and 10 μg/ml) and then added different concentrations of bio-LPS. Streptavidin-HRP (SA00001-0; Wuhan Sanying) was diluted to 0.5 μg/ml, and the reaction was terminated after adding the substrate solution to each well. Finally, to confirm that gCTRP9 effectively coated the enzyme labeling board, we coated a plate with a different concentration of gCTRP-9 per well and recorded the fluorescence values with an ELISA-based system after the application of an anti-CTRP-9 (1 μg/ml) antibody and a corresponding secondary antibody (peroxidase-labeled antibody to rabbit immunoglobulin G (H + L; 0.2 μg/ml; 5220-0336; seracare, Gaithersburg, MD).

| Coimmunoprecipitation
Macrophages were treated with LPS (100 ng/ml) and 3 µg/ml gCTRP9, protein, and membranes were then probed with specific primary and secondary antibodies. The immune complexes were exposed in a dark box instrument after adding enhanced ECL reagent, and the AlphaEaseFC software (Kreuzbergstr, Berlin, Germany) processing system analyzed the optical density of the target band.

| Statistical analysis
Unpaired Student's t tests were performed between any two groups, and multiple comparisons were evaluated by one-way analysis of variance and Tukey's test using GraphPad Prism 5.01 (San Diego, CA).
Kaplan-Meier survival analysis was performed by using the log-rank test. Measurement results are presented as the mean ± standard error of the mean (SEM). Each experiment was repeated at least three times, and data were considered to be significantly different at p < 0.05.

| fCTPR9 suppresses the inflammatory response on Day 3 and stimulates M2 macrophage polarization in the infarct border zone
Earlier conversion from an inflammatory to a reparative macrophage phenotype is potentially cardioprotective. The turning point for M1 to M2 macrophages is 3 days after MI localization at the border zone (Shirakawa et al., 2018). Therefore, to determine the effect of Ad-CTRP9 on macrophage markers and inflammatory substances, tissues were detected from the infarct border zone at Day 3 post MI. As shown in

| CTRP9 enhances an anti-inflammatory phenotype transition and decreases inflammatory factors produced by LPS + INF-γ-triggered macrophages
To investigate the phenotypic character of macrophages in the presence of gCTRP9 in vitro, we collected peritoneal macrophages from SD rats. concentration-dependent manner. The results showed that the effect of gCTRP9 comprehensively began at a concentration of 3 μg/ml, and the expression of M2 instead of M1 marker was improved at 1 μg/ml, which differed from the 1 μg/ml inhibitory concentration of CTRP9 in the oxidized low-density lipoprotein-induced inflammatory reaction (Zhang et al., 2016); therefore, we adopted the concentration of 3 μg/ml in the following experiments. gCTRP9 had a similar effect on macrophage polarization in the absence of LPS IFN-γ intervention (Figure 3a-i).
F I G U R E 3 CTRP9 promotes M2 macrophage polarization and decreases inflammatory cytokine production in vitro. Peritoneal macrophages (1 × 10 6 cells/ml per well) were preincubated with different concentrations of gCTRP9 (1, 3, and 5 μg/ml) for 30 min. After stimulation with 100 ng/ml LPS + 20 ng/ml IFN-γ or treatment with PBS for 24 hr, the cells were collected. We also observed the elevation of this phenomenon, in which the percentage of M1-like macrophages was lower while M2-like macrophages were increased, upon cotreatment of lipopolysaccharide + interferon-γ (LPS + IFN-γ) with gCTRP9 by flow cytometry analysis (Figure 3j).
To evaluate the effects of gCTRP9 on the release of proinflammatory cytokines, the supernatant protein levels of IL-1β, IL-6, and IL-10 were determined by ELISA kits. As shown in

| CTRP9 deficiency increases the M1 macrophage phenotype at baseline and among LPS + IFN-γ-induced macrophages
Next, we investigated whether reduced CTRP9 levels affect macrophage phenotype. First, we detected the mRNA and protein expression of CTRP9 in four types of cells (Figure 4a,b). Subsequently, we knocked down CTRP9 in macrophage cells to verify the and IL-6 and increased the release of IL-10 (Figure 5e-g). Therefore, we deem that CTRP9 has a similar effect as IL-4 in terms of motivating M2 polarization and anti-inflammatory ability, rather than decreasing function.
3.6 | CTRP9 promotes macrophage polarization partly through AMPK-mediated activation and largely via an NF-κB-dependent pathway As summarized in Figure 6, Ad-CTRP9 replenishment significantly activated AMPK and inhibited NF-κB p65 phosphorylation in a post-MI F I G U R E 5 CTRP9 has a similar function to that of IL-4 and promotes an anti-inflammatory macrophage phenotype. (a-d) mRNA levels of specific M1 and M2 markers under treatment with gCTRP9 (3 μg/ml), IL-4(20 ng/ml), or both by RT-PCR. (e-g) The supernatant protein levels of IL-1β, IL-6, and IL-10 were measured by ELISA. Data are shown as the mean ± SEM from three independent experiments. *p < 0.05 and **p < 0.01, the connection represents a comparison between two groups. Arg1: arginase1; CTRP9: C1q/TNF-related protein-9; ELISA: enzymelinked immunosorbent assay; IL: interleukin; mRNA: messenger RNA; RT-PCR: reverse-transcription polymerase chain reaction; SEM: standard error of the mean F I G U R E 6 CTRP9 promotes macrophage polarization partly through AMPK-mediated activation and largely via an NF-κB-dependent pathway.
(a-c) The effects of Ad-GFP and Ad-CTRP9 transfection on the phosphorylation of AMPK and NF-κB p65 in the infarct border zone at Day 3 post MI by western blot. (d-f) Effects of treatment of macrophages with gCTRP9 (3 μg/ml) at different time points on AMPK (Thr172) and NF-κB p65 phosphorylation by western blot analysis. (g-i) Western blot analysis assessed the effects of the AMPK inhibitor compound C on pAMPKα and pNF-κB p65 protein expression in macrophages responding to gCTRP9 by LPS + IFN-γ stimulation. Cells were treated with compound C (10 mol/l) for 2 hr and then incubated with CTRP9 (3 µg/ml) for 30 min, followed by LPS + IFN-γ challenge for 24 hr. (j-n) Effects of two inhibitors (compound C or PDTC; 100 μM) on M1 and M2 marker protein expression in macrophages in response to gCTRP9 via LPS + IFN-γ stimulation using western blot analysis. Data are shown as the mean ± SEM from three independent experiments; *p < 0.05, **p < 0.01 (vs. control group), &,# p < 0.05, and NS = p > 0.05, the connection represents a comparison between two groups. AMPK: adenosine monophosphate kinase; Arg1: arginase1; CTRP9: C1q/TNF-related protein-9; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; GFP: green fluorescent protein; IFN-γ: interferon-γ; LPS: lipopolysaccharide; MI: myocardial infarction; NF-κB: nuclear factor-κB; NS: not significant; PDTC: pyrrolidine dithiocarbamate; SEM: standard error of the mean model (Figure 6a-c). To obtain more reliable evidence, we utilized peritoneal macrophages for an in vitro experiment in which gCTRP9induced AMPK phosphorylation (at Thr172 of its α subunit; Figure   6d,e) and NF-κB p65 dephosphorylation (Figure 6d,f) at 3 μg/ml was observed within 30 min of gCTRP9 administration in a time-dependent manner, followed by reduction to approximately normal levels with time; the chronergy was analogous to the effect of CTRP9 in cardiomyocytes (Zhang et al., 2016). We further assessed the role of gCTRP9 in LPS + INF-γ-induced AMPK and NF-κB p65 phosphorylation expression in macrophages by western blot, and we found that gCTRP9 promoted AMPK phosphorylation (Figure 6g,h) and attenuated NF-κB p65 phosphorylation (Figure 6g,i). We also observed that silencing CTRP9 led to increased p-NF-κB p65 in macrophages (Figure 7j).
Simultaneously, we treated macrophages with the specific AMPK

| DISCUSSION
In the present study, we focused on the role of the presence of CTRP9 improving cardiac function through macrophage polarization in the early stage of MI. Our results affirmed that CTRP9 in vivo attenuated post-MI cardiac dysfunction and was dispensable for M1 to M2 macrophage polarization transition, which was further proven in the in vitro experiments. Then, we identified that these effects of CTRP9 on macrophages were regulated mainly by depressing NF-κB p65 phosphorylation, which is involved in inhibiting the TLR4/MD2/MyD88 pathway and elevating AMPK phosphorylation.
We know that inappropriate adjustment of macrophage phenotypes and extended inflammation can impair proper tissue remodeling after MI. Hence, the early resolution of the inflammatory reactions is a pivotal juncture of cardiac remodeling. Currently, the modulation of macrophage polarization is a potential therapeutic object for the treatment of MI. CTRP9 is a well-known adipokine that can suppress production of proinflammatory cytokines and has protective functions in the heart (Kambara et al., 2015;Sun et al., 2013;Zhang et al., 2016). Whether CTRP9 mediated macrophage polarization and specific regulatory mechanisms that directed macrophages into various functional subsets need further study to explain the protective effects of CTRP9 against anti-inflammatory activities in the MI model of the disease.
To elucidate the effects of CTRP9 in the post-MI heart in vivo, overexpressing CTRP9 post MI alleviated adverse cardiac dysfunction without improving the 7-day post-MI infarct area, rupture rates or survival rates. We can see that CTRP9 does not ameliorate early prognosis; existing research showed that exogenous CTRP9 improved the survival rate at 42 days post MI but did not decrease mortality at 7 days post MI (Sun et al., 2013). The effect of CTRP on prognosis in the early and late stages is inconsistent, which may be related to the profibrosis of M2 macrophages in the early stage of MI . MMP-2, which is notably secreted by M1 macrophages and alters the extracellular matrix, participates in cardiac remodeling and boosts cardiac rupture (Sica & Mantovani, 2012). Lower MMP-2 expression in vivo and in vitro was observed with CTPR9 pretreatment in the present study, which may have contributed to the better post-MI contractile function but did not reduce post-MI heart rupture. An in vivo study showed that CTRP9 obviously promoted the M1 to M2 macrophage transition, suggesting that the partial recovery of the LV function may be related to not only the regulation of inflammation via M1 macrophage inhibition but also the promotion of the M2 macrophage population.
We used peritoneal macrophages and induced them to differentiate into M1 macrophages. Conventionally, M1 macrophages were induced when macrophages were exposed to LPS + IFN-γ, and then we conducted a series of cell studies. M1 macrophages expressed high levels of M1 markers and diminished the expression of M2 markers, which was consistent with previous research Mo et al., 2014), whereas pretreatment with gCTRP9 significantly reversed these effects and was further improved by CTRP9 deficiency and the Th2 cytokine IL-4. The antiinflammatory effects of CTRP 9 on inflammatory factors have also been confirmed. CTRP9 alone, in the absence of stimulators, had relatively homologous effects. Similarly, APN, a cytokine in the same family as CTRP9, also played a direct role in lung macrophage polarization (Ohashi et al., 2010). We examined the expression of CTRP9 in macrophages and found that expression in macrophages was lower than the abundant expression on cardiomyocytes, adipocytes, and endotheliocytes (Appari et al., 2017;Su et al., 2013). Taken together, our results suggested the positive regulatory function of CTRP9 in macrophage polarization and anti-inflammatory reactions.
AMPK is not only a cellular stress sensor but also plays an important role in the suppression of inflammatory responses, and phosphorylation of the threonine-172 residue of the α subunit is critical for AMPK activity (Steinberg & Schertzer, 2014). AMPK can attenuate inflammatory responses by suppressing the activation of NF-κB (Yi et al., 2011). AMPK and NF-κB p65 are involved in macrophage polarization and were both identified as effector molecules of CTRP9, as activation of AMPK exerts some potential protective functions in an MI disease model (Gu et al., 2018;J. Wang et al., 2013). In the present study, fCTPR9 increased AMPK activation and suppressed NF-kB activation at Day 3 post-MI in vivo. We further found that administration of CTRP9 to macrophages led to AMPK phosphorylation and NF-κB dephosphorylation in the absence or presence of LPS + IFN-γ in an in vitro experiment, and the timedependent effects were similar to those in previous studies exploring different stimulators and cells (Kambara et al., 2015;Sun et al., 2013;Zhang et al., 2016), which was further confirmed by the application of compound C and PDTC. As expected, we obtained similar results.
Previous research deemed that CTRP9 can activate AMPK after MI, but CTRP9 attenuates adverse post-MI cardiac remodeling, largely via a PKA-dependent pathway (Sun et al., 2013); this previous study utilized H9C2 and AMPK-DN cardiomyocytes and mainly studied apoptosis and fibrosis mechanisms after long-term MI, whereas in this study, we mainly studied the macrophage polarization process mediated by CTRP9 in the context of early inflammatory transformation post MI. Indeed, CTRP9 protected against acute heart injury after ischemia-reperfusion through an AMPK-dependent mechanism (Kambara et al., 2012). Furthermore, inflammatory lesions in the early stage of myocardial infarction had a greater effect on cardiac remodeling than fibrotic processes (Prabhu & Frangogiannis, 2016).
Thus, we presumed that the macrophage polarization-promoting effect of CTRP9 may contribute to early cardioprotective function post MI by partially regulating AMPK signaling pathways. Interestingly, the extent of LPS-induced NF-κB p65 phosphorylation and the changes in M1 and M2 markers under CTRP9 intervention were not fully restored by preconditioning with compound C; these data suggested that other signaling pathways besides AMPK activation participate in the NF-κB phosphorylation suppression process mediated by CTRP9, thereby affecting macrophage phenotype transformations.
The MD2/TLR4 complex acts as an upstream signaling molecule of NF-κB, and thus, the LPS receptor is brought to our attention.
MD2 is an essential extracellular molecule for LPS recognition. The binding of LPS with MD2 leads to the recruitment of the adapter protein MyD88 and the production of pNF-κb (Y. Yang et al., 2016).
CTRP9 can restrain the cholesterol-induced activation of the TLR4 signaling pathway in VSMCs (Q. Liu et al., 2017). However, the precise molecular mechanisms remain unclear. Combined with the important role of the TLR/MyD88 downstream molecules IRF5, AP1, and NF-κB in M1 macrophage polarization, we deduced that CTRP9 also affected macrophage polarization to participate in post-MI protective effects via the TLR4/NF-κB pathway.
The two key results were as follows: (a) Negative regulation of TLR4 and MyD88 protein expression. Activation of TLR4 is the pivotal process in LV remodeling following MI, and TLR4 defects and reduction of TLR4 ameliorate LV remodeling (Sheng et al., 2009;Timmers et al., 2008).
Thus, in this experiment, CTRP9 reduced the expression of TLR4 and MyD88 after MI and also contributed to anti-ventricular remodeling, which was one of the reasons for the improvement of post-MI cardiac function. Previous research studies showed that reducing the expression of TLR4 in macrophages can decrease p-NF-κB, attenuate inflammatory responses and promote M1 toward M2 macrophage polarization (Sheng et al., 2009;C. Zhang et al., 2018). It has been reported that TLRmediated proinflammatory pathways, particularly NF-κB, are correlated with the anti-inflammation role of AMPK and activation of AMPK inhibits TLR4 expression (Tao et al., 2017;Y. Yang et al., 2016;C. Zhang et al., 2018). LPS + INF-γ stimulation elevated the expression of TLR4 and MyD88, and similar results were obtained in this study, but AMPK phosphorylation increased, resulting in similar or inconsistent reports (Tao et al., 2017;C. Zhang et al., 2018). Our results indicated that AMPK was activated by CTRP9 and subsequently suppressed the TLR4 and MyD88 activity of macrophages in the presence or absence of an inflammatory stimulus and in the adaptive response to hypoxic and ischemic stress of the post-MI heart. (b) CTRP9 directly bound to the MD2/TLR4 complex. We discuss the possibility of other mechanisms through which CTRP9 may affect p-NF-κB via TLR4. More studies have confirmed that certain substances and components can inhibit the downstream pathways by directly binding to the extracellular structure of TLR receptors (Kopp et al., 2010;M. Wang et al., 2011). That being the case, can CTRP9 also play a corresponding role by combining with the extracellular structure of TLR4? Our evidence supported a mechanism by which CTRP9 directly binds to MD2 to activate TLR4 signaling (MyD88-dependent pathway), based on our construction of a TLR4/MD2 fusion protein for ELISA and immunoprecipitation analyses.
Therefore, we confirmed that CTRP9 acts as a novel and endogenous LPS antagonist that directly binds to MD2 and impedes MD2/TLR4 complex formation and MyD88 recruitment. We suggest that this mechanism supplements the inhibition of NF-κB phosphorylation by CTRP9. Competing with MD2 for TLR4 binding can block inflammatory responses (O'Reilly & van Laar, 2018), which represents a promising drug target for the treatment of inflammation-associated diseases.
In summary, the present study established a better understanding of the protective role of CTRP9 in the early-stage post-MI rat. Our findings indicated that the effects associated with macrophage polarization were likely regulated by AMPK and direct binding with the MD2/TLR4 complex to influence the phosphorylation of the downstream molecule NF-κB. The findings represent new mechanistic insights into MI and the immunity of cardiovascular diseases.

ACKNOWLEDGMENTS
We would like to thank Lin Yin, Wei Li, Juan Hu, and Bingjie Ye in their assistance during experiments. Huibo Wang and Yanhong Tang for their advice on study design and data analysis and interpretation. This study was financially supported by the Special