Aminooxyacetic acid attenuates post‐infarct cardiac dysfunction by balancing macrophage polarization through modulating macrophage metabolism in mice

Abstract Excessive activation of pro‐inflammatory M1 macrophages following acute myocardial infarction (MI) aggravates adverse cardiac remodelling and heart dysfunction. There are two break points in the tricarboxylic acid cycle of M1 macrophages, and aspartate‐arginosuccinate shunt compensates them. Aminooxyacetic acid (AOAA) is an inhibitor of aspartate aminotransferase in the aspartate‐arginosuccinate shunt. Previous studies showed that manipulating macrophage metabolism may control macrophage polarization and inflammatory response. In this study, we aimed to clarify the effects of AOAA on macrophage metabolism and polarization and heart function after MI. In vitro, AOAA inhibited lactic acid and glycolysis and enhanced ATP levels in classically activated M1 macrophages. Besides, AOAA restrained pro‐inflammatory M1 macrophages and promoted anti‐inflammatory M2 phenotype. In vivo, MI mice were treated with AOAA or saline for three consecutive days. Remarkably, AOAA administration effectively inhibited the proportion of M1 macrophages and boosted M2‐like phenotype, which subsequently attenuated infarct size as well as improved post‐MI cardiac function. Additionally, AOAA attenuated NLRP3‐Caspase1/IL‐1β activation and decreased the release of IL‐6 and TNF‐α pro‐inflammatory cytokines and reciprocally increased IL‐10 anti‐inflammatory cytokine level in both ischaemic myocardium and M1 macrophages. In conclusion, short‐term AOAA treatment significantly improves cardiac function in mice with MI by balancing macrophage polarization through modulating macrophage metabolism and inhibiting NLRP3‐Caspase1/IL‐1β pathway.


| INTRODUC TI ON
In the past 20 years, owing to implementation of evidence-based treatment, the in-hospital, 30-day and 1-year all-cause mortality of acute myocardial infarction (AMI) were approximately 43%, 42% and 36% lower, respectively, in 2013-2014 than in 1995-1996. 1 Even so, AMI leads to irreversible loss of myocardial cells and poor left ventricular remodelling, resulting in reduced ejection fraction (<50%, 72% in 1997-1998 and 52% in 2013-2014) 1 and subsequent heart failure. More importantly, the mortality seems to have remained constant and not fallen further since around 2008. 1,2 Therefore, new effective therapeutic strategies for MI are still desirable.
The infiltration of immune cells, particularly macrophages, plays important roles in poor left ventricular remodelling and subsequent heart failure after MI. 3,4 Macrophages have peculiar plasticity that allows them to functionally polarize into classically activated M1 macrophages or alternatively activated M2 macrophages in response to different stimuli, and these different types play opposite roles. Studies showed macrophages were differentially activated during different phases after MI. The M1 subtype dominates the early inflammatory phase and secretes pro-inflammatory cytokines, causing damage, which then transits to the M2 macrophages and expresses high level of IL-10 in the infarct scar formation stage after MI, promoting repair. [5][6][7] In AMI, the transformation of pro-inflammatory macrophages to anti-inflammatory macrophages is considered to be the beginning of myocardial repair after MI, and the failure of the transformation may result in sustained M1 macrophages activation and adverse cardiac remodelling. 5 Therefore, modulating the balance of pro-inflammatory macrophages and anti-inflammatory macrophages is speculated as a novel treatment method. 5,8 Interestingly, M1 macrophages preferentially metabolize glucose as an energy substrate, converting glucose into lactate. 9,10 However, M2 macrophages utilize fatty acids as a source of fuel, which fuel an oxidative tricarboxylic acid cycle (TCA) and following oxidative phosphorylation (OXPHOS). 10 Recent studies have shown that enhanced glucose metabolism drives a hyper-inflammatory response in macrophages 11 and inhibiting glycolysis reverses pro-inflammatory cytokines elevations, 12 indicating that manipulating macrophage metabolism may control macrophage polarization and inflammatory response. In short, glucose metabolism is central to the function of M1 macrophages and strategies for modulating its glucose metabolism would be innovative approaches to attenuate inflammatory responses and thus address MI.
Outstandingly, in M1 macrophages, the TCA cycle metabolic flux is discontinued at isocitrate dehydrogenase (IDH) and succinate dehydrogenase, and aspartate-arginosuccinate shunt (AASS) compensates for the breaks. 13 Aminooxyacetic acid (AOAA) is a broad-spectrum inhibitor of pyridoxal phosphate-dependent enzymes, including aspartate aminotransferase (AST), which is a key enzyme of AASS. Previous studies have indicated that manipulating macrophage metabolism may control macrophage polarization and inflammatory response, 11,12 inspiring us to study AOAA potential role in macrophages. In the present study, we set out to determine the effects of AOAA on macrophage metabolism and polarization and cardiac function after MI in a mouse model.

| Laboratory animals
Male wild-type C57BL/6J mice aged 8-10 weeks were purchased from the Experimental Animal Center of the Chinese Academy of Medicine Sciences of Soochow University. All animal procedures were in conformity with the local ethics legislation of animal experimentation in the Institute for Cardiovascular Science, Soochow University, Suzhou, China. Mice were allowed free access to food and water and housed in a room with controlled temperature (22 ± 1℃) and a 12-hour light-dark cycle.

| Generation of primary bone marrow-derived macrophages (BMDMs)
Bone marrow cells were flushed out from femurs and tibias of 8-10 weeks old male C57BL/6 mice. The bone marrow cells were resuspended, cultured and differentiated into macrophages in RPMI-1640 supplemented with 10% FBS, 1% P/S and 20 ng/mL recombinant murine macrophage colony-stimulating factor (PeproTech) in a humidified incubator containing 5% CO 2 at 37℃ for 7 days. On day 7, bone marrow-derived macrophages were harvested using Accutase (Sigma) and replated for further experimentation. BMDMs were characterized by flow cytometry analysis with antibodies specific for F4/80 and CD11b.

| Macrophages experimental protocol
Bone marrow-derived macrophages were treated without or with indicated concentration of AOAA (Sigma) for 1 hour, followed by 10 ng/mL LPS and 10 ng/mL IFN-γ or 20 ng/mL recombinant murine IL-4 (PeproTech) stimulation for 24 hours. Supernatants were harvested for pH and lactic acid measurements, and the cells were collected for ATP detection, RNA or Western blot analysis.

| Measurement of lactic acid
The lactic acid level was determined using a lactic acid assay kit (Beyotime) following the instructions of the manufacturer. Briefly, the supernatant of BMDMs with various treatments was collected.
After incubating with enzyme reagent and developer for 10 minutes at 37°C, the reaction was stopped and lactic acid level of the supernatant was determined by measuring absorbance at 530 nm on a multifunctional microplate reader (BIO-TEK).

| Measurement of extracellular acidification rate (ECAR)
To evaluate glycolysis function, we measured ECAR in BMDMs using a glycolysis assay kit (Abcam) following the manufacturer's instructions. Briefly, BMDMs were seeded in 96-well plate at a density of 3 × 10 5 cells/well in 200 μL culture media for 12 hours to allow cells to adhere in a CO 2 incubator at 37°C The BMDMs were then unstimulated (M0) or stimulated with LPS/INF-γ for 24 hours with or without 1 hour pre-treatment with AOAA (1, 5 mmol/L). Culture media were discarded from all assay wells, and BMDMs were washed with 100 μL respiration buffer two times. About 150 μL respiration buffer was added to all wells containing cells and blank controls wells. BMDMs were incubated in a CO 2 -free incubator at 37°C for 3 hours to purge CO 2 . About 10 μL reconstituted glycolysis assay reagent was added to each sample well, and 10 μL respiration buffer was added to blank control wells. The 96-well plate was inserted into a multifunctional microplate reader (BIO-TEK) pre-set to 37°C.
Glycolysis assay signal was measured at 1.5 minutes intervals for 180 minutes using excitation and emission wavelengths of 380 and 615 nm, respectively.

| Intracellular ATP assays
Intracellular ATP content was determined by using an enhanced ATP assay kit (Beyotime) as previously described 15 with some modifications. Briefly, cells were lysed and boiled for 2 minutes to fully release ATP, and then centrifuged at 12 000 g at 4°C for 5 minutes.
The supernatant was added to the test plate pre-treated with ATP detection reagent. Luminescence was measured by using a microplate reader (BIO-TEK). However, cells lysate for protein content measurement was not boiled and the protein content was measured with the BCA protein assay kit (Takara). Finally, ATP concentration was expressed as μmol/g protein.

| Animal experimental protocol
Myocardial infarction model was established in male C57BL/6 mice as previously described. 16 Briefly, male mice with 8-10 weeks old were anaesthetized by intraperitoneal injection with a mixture of 70 mg/ kg ketamine and 6 mg/kg xylazine and were mechanically ventilated using a rodent ventilator attached to an endotracheal tube during the surgical procedure. Thoracotomy was performed between the 4th and 5th intercostal space to expose the heart. MI was achieved through permanent ligation of the left anterior descending coronary artery with a 6-0 polyester suture. Finally, the thoracic wall was carefully closed. The surgeon was blinded for the mouse grouping. MI mice were allocated to intraperitoneal injection with a daily dose of 10 mg/kg BW of AOAA diluted in saline (1 mg AOAA in 1 mL saline, ie 10 mL/kg solution) or 10 mL/kg saline after operation for three consecutive days. The dose of AOAA was determined according to previously published studies. [17][18][19] Mice were killed at day 3 or day 28.

| Echocardiography
For echocardiographic acquisition, mice were anaesthetized by 1%- All measurements were averaged over three consecutive cardiac cycles.

| Histological preparation
At day 3 or day 28 post MI surgery, the animals were anaesthetized and the hearts were exposed. For pathological examination, the hearts were arrested via the left ventricle injection of 1 mL 1 mol/L KCl followed by 5 mL PBS and 10 mL 4% paraformaldehyde perfusion. Then, the hearts were carefully dissected and fixed in 4% paraformaldehyde overnight.
For biochemical and molecular analysis, the hearts were perfused with PBS, and then snap-frozen in liquid nitrogen and stored at −80°C.

| Haematoxylin and eosin (H&E) staining
After fixed in 4% paraformaldehyde overnight, the hearts of day 3 post MI surgery were processed as standard paraffin embedded.
The heart tissues were then serially sectioned at 5 μm in the left ventricle transverse direction. H&E was performed to evaluate the inflammatory cell infiltration.

| Masson's Trichrome staining
The heart tissues of day 28 after MI were sectioned in the left ventricle transverse direction from the ligation level down to the apex.

| RNA extraction and quantitative real-time PCR (qPCR) analysis
Total RNA was isolated using TRIzol reagent and quantified with ND2000 spectrophotometer (NanoDrop Technologies). DNase I-treated RNA was reverse transcribed into cDNA using the    The signalling was quantified by Image J software and presented as normalized to β-tubulin.

| Statistical analysis
Statistical analysis was performed using the GraphPad Prism 7 software (GraphPad Software). All data were presented as mean ± SEM.
Normal distribution was tested using the Shapiro-Wilk test.
Differences between two groups were compared by unpaired t test. More than two groups were compared by one-way ANOVA, followed with a post hoc Tukey's multiple comparisons test. All P values were two-tailed, and a P value <.05 was considered statistically significant.

| AOAA inhibits glycolysis and enhances ATP levels in classically activated M1 macrophages
Bone marrow-derived macrophages were identified as the cells ex- Pre-treatment with AOAA gradually diminished the glycolysis in a dose-dependent manner. It is well known that a modest amount of ATP is produced in glycolysis and much more ATP is formed through TCA.
Increased glycolysis in classically activated M1 macrophages means reduced ATP production. On the contrary, decreased glycolysis means raised ATP production. Therefore, we confirmed the effects of AOAA on the ATP levels in classically activated M1 macrophages. As expected, cellular ATP levels rose sharply with AOAA pre-treatment in a dose-dependent manner while glycolysis was inhibited ( Figure 1D). Macrophages are typically divided into classically activated M1 subtype and alternatively activated M2 subtype. Classically activated M1 macrophages rely on glycolysis, whereas alternatively activated M2 macrophages acquire energy from OXPHOS. 22 We found that AOAA seems to switch the metabolism of M1 macrophages towards M2-like metabolism, as displayed by reduction in lactic acid production and glycolytic rates in conjunction with increment in ATP production.

| AOAA modulates classically activated M1 macrophages phenotype and alternative activated M2 phenotype
After knowing the effects of AOAA on macrophage metabolism, we  Figure 2D).
Finally, we tested the expressions of M1 associated pro-inflammatory cytokines by qPCR. Intriguingly, the expressions of iNOS and IL-6 were also down-regulated by AOAA in a concentration-dependent manner in BMDMs under the LPS/INF-γ stimulation ( Figure 2E).
However, the TNF-α expression was frustrated only by high dose AOAA ( Figure 2E).  Figure 3A,B). Also, the expression of Arg1 was further assessed by Western blot, which revealed that the Arg1 protein level was enhanced when macrophages were pre-treated with AOAA before IL-4 stimulation ( Figure 3C). At last, we evaluated the expressions of M2 associated anti-inflammatory cytokines by qPCR. We found that the expressions of CD206 and IL-10 were dramatically increased by AOAA in a dose-dependent manner. However, only high dose AOAA amplified the mRNA expression of Arg1 ( Figure 3D).
These data suggest that AOAA adjust macrophage polarization by restraining M1 macrophages phenotype and boosting M2 phenotype, indicating a possible link between metabolic reprogramming and macrophage polarization.  Figure 4A,B, AOAA treatment increased 27% of EF and 32% of FS, respectively, compared with saline treatment in MI mice (all P < .05), suggesting an improvement of cardiac function. The results of echocardiogram were supported by histological analysis. Masson's trichrome staining revealed that AOAA reduced the infarct size relative to saline treatment ( Figure 4C,D).

| AOAA administration attenuates post-MI cardiac dysfunction and infarct size in mice
Collectively, AOAA suppressed myocardial scar expansion and promoted cardiac function recovery from infarction.

| AOAA reduces the proportion of M1 macrophages and boosts M2-like phenotype in the ischaemic border zone
To evaluate the effects of AOAA on macrophage polarization in heart tissues of MI mice, we stained heart sections with iNOS/F4/80 and Arg1/F4/80 to illustrate M1-like and M2-like macrophages by immunofluorescent staining. There was no significant difference in F4/80 + macrophages accumulation between AOAA and salinetreated MI hearts ( Figure 5A,B). However, we found that AOAA treatment reduced iNOS positive macrophages (M1-like) proportion correlated with augmentation of Arg1 positive macrophages (M2like) proportion in the heart as compared with saline-treated MI mice ( Figure 5A,B). We further measured the expression of iNOS and Arg1 in the heart by Western blot and qPCR. We observed that protein and mRNA levels of iNOS were sharply attenuated by AOAA treatment compared with saline treatment in MI heart ( Figure 5C,D).
On the contrary, AOAA treatment led to increments of Arg1 in protein and mRNA levels in MI heart compared with those treated with saline ( Figure 5C,D).
Accordingly, we assessed inflammation cell infiltration in MI heart. H&E staining analysis showed that AOAA treatment lowered the inflammatory cell infiltration in the heart tissues of day 3 after MI compared with saline treatment ( Figure 6A). As indicated in Figure 6B,C, immunofluorescent staining further revealed significant TNF-α positive signal in the peri-infarct zones of the heart tissues, and AOAA treatment notably diminished TNF-α positive signal.
In contrast, AOAA treatment urged CD206 accumulation in MI heart manifested by immunofluorescent staining (Figure 6D,E).
To further verify the modulation of AOAA on inflammatory factor, we measured the mRNA expression of macrophage associated pro-inflammatory and anti-inflammatory cytokines in the heart tissues of day 3 post MI. Notably, AOAA treatment significantly inhibited the expression of TNF-α and IL-6 combined with a remarkable rising of IL-10 in the heart of MI mice compared with saline treatment ( Figure 6F). These data suggested that AOAA modulated macrophage polarization and early inflammation response following MI.   [11][12][13] and the role of macrophages in AMI. [5][6][7] As illustrated in Figure  In this study, we found AOAA reduced lactic acid production, inhibited PPP, and increased ATP production, suggesting a switch from the metabolism of M1 macrophages to M2-like metabolism. Further, we NLRP3 inflammasome has been shown to be activated in MI mice 23,24 and knockdown of its expression can reduce infarct size and improve myocardial function. 23,25 Besides, NLRP3 inflammasome has been proved to be activated predominantly in macrophages. 26,27 Studies showed that ROS derived from NADPH oxidase and mitochondria are involved in NLRP3 inflammasome activation, and inhibition of ROS prevents the activation of NLRP3 inflammasome. 35,36 However, the exact mechanism is not clear. The structure of NLRP3 contains a highly conserved disulphide bond, which is highly sensitive to redox states. 37 It is speculated that NLRP3 inflammasome is activated upon the disulphide bond triggered by ROS. A wide range of stimuli are also supposed to activate NLRP3 inflammasome by producing ROS. 38 In M1 macrophages, as shown in Figure  We selected AOAA in our study for its strong inhibitory effect on AST. 44 However, it is a non-specific inhibitor of AST, since it can inhibit several pyridoxal phosphate-dependent enzymes. 44, 45 We also could not eliminate the possibility of other mechanisms. For example, a recent study by Jespersen and colleagues showed pre-ischaemic administration of AOAA reduced infarct size and protected the heart against ischaemia-reperfusion injury. And, the authors owed its cardioprotection to the inhibiting of malate-aspartate shuttle. 46 Therefore, it is warranted to further investigate the cardioprotective mechanisms of AOAA using specific inhibitors of different enzymes separately.

| D ISCUSS I ON
In conclusion, short-term AOAA treatment during the peak inflammatory phase of the immune response significantly improves cardiac function in mouse with MI, and its effect may be achieved by balancing the macrophage polarization, especially through modulating macrophage metabolism and inhibiting NLRP3-Caspase1/ IL-1β pathway.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The raw data supporting the conclusion of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.