An immunometabolic patch facilitates mesenchymal stromal/stem cell therapy for myocardial infarction through a macrophage‐dependent mechanism

Abstract Mesenchymal stromal/stem cells (MSCs) have emerged as a promising approach against myocardial infarction. Due to hostile hyperinflammation, however, poor retention of transplanted cells seriously impedes their clinical applications. Proinflammatory M1 macrophages, which rely on glycolysis as their main energy source, aggravate hyperinflammatory response and cardiac injury in ischemic region. Here, we showed that the administration of an inhibitor of glycolysis, 2‐deoxy‐d‐glucose (2‐DG), blocked the hyperinflammatory response within the ischemic myocardium and subsequently extended effective retention of transplanted MSCs. Mechanistically, 2‐DG blocked the proinflammatory polarization of macrophages and suppressed the production of inflammatory cytokines. Selective macrophage depletion abrogated this curative effect. Finally, to avoid potential organ toxicity caused by systemic inhibition of glycolysis, we developed a novel chitosan/gelatin‐based 2‐DG patch that directly adhered to the infarcted region and facilitated MSC‐mediated cardiac healing with undetectable side effects. This study pioneered the application of an immunometabolic patch in MSC‐based therapy and provided insights into the therapeutic mechanism and advantages of this innovative biomaterial.


| INTRODUCTION
Myocardial infarction (MI), which is principally caused by the stenosis and occlusion of coronary arteries, remains the leading cause of most acute cardiac events and deaths worldwide. 1 Many studies and clinical trials have supported the efficacy of mesenchymal stromal/stem cells (MSCs) in patients with MI. 2,3 Our previous studies have explored whether MSC therapy reduces scarring, minimizes perfusion defects, and ameliorates cardiac dysfunction in animal MI models. [4][5][6] Nonetheless, MSC application faces serious obstacles in terms of the low retention and survival of implanted cells, which are attributed to hostile hyperinflammation in the ischemic region. After MI attack, glycolysis occurs in the infarct area due to hypoxia and exhausted nutrients, accompanied by the accumulation of acid metabolites, free radicals, Weizhang Xiao and Ming Chen contributed equally to this study. necrotic cells, and degraded extracellular matrix, which provoke sterile inflammation and the inflammatory activation of immune cells. 7,8 Previous studies have determined the crucial role of macrophages in cardiac injury and established the heterogeneity of macrophages during the process of cardiac injury and repair. 9 Several hours after MI, monocytes are recruited to the infarct zone and differentiate into proinflammatory M1 macrophages. These macrophage subpopulations initially predominate and culminate on Day 3, which generates a cascade of cytokines to trigger intense inflammatory responses and injury. 10 Then, the number of M1 macrophages decreases and these cells are replaced by reparative M2 macrophages, which represent the predominant subset after 5 days. 11 Although M1 macrophages function positively to clear cellular debris, their excessive activities inevitably extend the duration of inflammation and undermine cardiac recovery. 12 To date, treatments targeting M1 macrophages or promoting the M1/M2 switch have been considered novel strategies for MI. 13,14 However, few studies have scrutinized the immunomodulatory effect of metabolism on macrophages.
The main energy source of M1 macrophages is glycolysis, whereas M2 macrophages rely largely on oxidative phosphorylation. 15 Reprogrammed immunometabolism in macrophages contributes to the modulation of the inflammatory response. 16 As shown in our previous report, curbing the aspartatearginosuccinate shunt, which compensates for the fragmented tricarboxylic acid cycle during glycolysis, ameliorates cardiac function in a murine MI model through immunometabolic reprogramming of macrophages. 17 As a classical glycolytic inhibitor, 2-deoxy-D-glucose (2-DG) has long been proved to effectively suppress the growth of tumor cells, the energy production of which depends mainly on aerobic glycolysis. 18,19 However, the steeply increased glycolysis after MI occurs in a confined ischemic area, and researchers have not determined whether 2-DG exerts an immunometabolic modulation on the local hyperinflammation and contributes to cardiac protection against MI.
In this study, we proposed an immunomodulatory effect of 2-DG on macrophages through the repression of glycolysis. By detecting the functional switch of macrophages and the generation of proinflammatory factors, we explored the alteration of the inflammatory response in vivo after 2-DG supplementation. Considering the pitfalls of systemic 2-DG administration, including a limited half-life and potential organ toxicity, 20 local application of 2-DG through a cardiac patch has entered our view. Chitosan and gelatin are two valuable biomaterials with their characteristics of biocompatibility, biodegradation, nontoxicity, and plasticity. 21 Under acidic conditions, cation-rich chitosan reacts with gelatin that contains anions to form polyelectrolyte complexes, which have been widely used in biomedical and tissue engineering applications including wound healing, 22 drug delivery, 23 and cardiac repair. 24 Therefore, we developed an innovative chitosan/gelatin-

| Effect of glycolytic inhibition on immunometabolism of macrophages
2-DG is considered to efficiently blunt glycolysis by competitively restraining hexokinase 2 (HK2), which catalyzes the initial step of glucose metabolism (Figure 1a). 20 Here, lipopolysaccharide (LPS)stimulated macrophages were pretreated with or without 2-DG to elucidate the effect of glycolytic inhibition on immunophenotypic switch of macrophages. We initially analyzed the extracellular acidification rate (ECAR) in macrophages and detected an expectedly nota-  Figure S1A,B), which was consistent with a previous study exploring the M1-M2 repolarization of inflammatory macrophages. 25 Taken together, these results provide important insights into the role of 2-DG in the immunomodulation of macrophages by inactivating glycolytic metabolism.

| Therapeutic effects of systemic 2-DG against hyperinflammation in ischemic myocardium
In order to determine whether 2-DG treatment tempers detrimental inflammation in vivo, we intraperitoneally injected 2-DG into MI mice and analyzed the inflammatory response within the infarcted area at 3 days after MI when the number of proinflammatory macrophages  Transthoracic echocardiography was conducted at consecutive time points after MI to evaluate whether 2-DG amplifies MSCmediated cardiac repair. As illustrated in Figure 3f,g, the MI-induced reduction of left ventricular ejection fraction (LVEF) and fractional shortening (LVFS) was partially reversed by MSCs implantation and was further recovered with greater significance by the addition of 2-DG (LVEF: 55.88% ± 7.12% vs. 29.68% ± 7.37%, p < 0.001; LVFS: 28.51% ± 4.80% vs. 13.80% ± 3.62%, p < 0.001 on Day 28 post-MI).
Consistent with these findings, a decreased fibrotic area was observed in mice treated with the combination of MSCs + 2-DG on Day 28 after surgery (Figure 3h-j). Furthermore, the morphological analysis revealed a diminished relative heart weight in MI-challenged mice treated with the combination of MSCs + 2-DG, representing tempered ventricular hypertrophy (Figure 3k).
Collectively, 2-DG exerts a cardioprotective effect on myocardial injury by enhancing the retention of MSCs, which resist MI-induced cardiac dysfunction.

| 2-DG facilitates MSCs therapy in a macrophage-dependent manner
We selectively depleted macrophages by administering Cl 2 MDP to exclude the possibility that 2-DG exerts its cardioprotective effect by modulating the immunometabolism of cells other than macrophages.
Consistent with a previous report, 26 Cl 2 MDP successfully removed macrophages from the heart, spleen, and blood (Figure 4a

| DISCUSSION
Current treatments for MI focus on rapid revascularization and reperfusion, including vasodilators, anticoagulants, implantation of stents or bridging vessels, and relief of cardiac burden such as betablockers. 31 However, these impressive therapies have not yielded satisfactory effect with the mortality and morbidity of complications associated with MI remain high. 1 In the circumstances, MSCs transplantation has emerged as an innovative and promising tool for their robust potential of paracrine and immunomodulation. 32 The primary obstacle to MSC therapy is extremely poor retention and low survival, which hampers further clinical application. For decades, a variety of strategies have been devoted to increasing the efficiency of MSC delivery. These strategies include modifying the properties of MSCs, such as increasing the expression of adhesion factors or inflammatory cytokine receptors, 33,34 preconditioning MSCs with toxic elements, exposure to a hypoxia-or nutrient-free environment, 35,36 and magnetic targeting techniques. 37 Meanwhile, emerging studies have focused on the encapsulation of MSCs within biomaterials, including injectable hydrogels, 38,39 scaffolds, 40 gelatin coatings, 41 and microneedles. 42 Nevertheless, current strategies are limited by the inefficient interactions between implanted cells and excessively inflammatory tissue.
Ideal approaches are required not only to deliver more "seeds" (beneficial cells) but also to improve the "soil," namely, to reduce the hyperinflammatory response in the ischemic region. In this study, 2-DG application exhibited impressive potential to ameliorate harsh inflammation in the ischemic region (Figure 2), followed by MSC retention extension and cardiac outcome improvement, as expected ( Figure 3).
The reprogrammed inflammatory response in situ may trigger an escalated effect on the biological behavior of MSCs than previously anticipated. 43 Therefore, we assumed that 2-DG facilitates the efficacy of MSC therapy against MI by modulating the immunometabolism of macrophages.
As commonly used biomaterials, the biocompatibility of chitosan and gelatin has been confirmed in previous studies. 44,45 In the present study, we fabricated a chitosan/gelatin composite patch loaded with 2-DG. Unlike other biomaterials that encapsulate MSCs to extend their preservation or reinforce their viability, the 2-DG composite patch directly targets the infarcted area to modulate the disturbed inflammatory response, particularly the accumulation of proinflammatory M1 macrophages, which dramatically hinders the retention and viability of engrafted cells. Our study showed superior effects of 2-DG pat on MSC preservation and myocardial protection (Figures 6   and 7). Moreover, this immunometabolic patch slowly released 2-DG for more than 3 days (Figure 5), the period of peak M1 macrophage accumulation after MI. According to these findings, we infer that 2-DG pat continuously modulates the immunometabolism of local macrophages and subsequently effectively attenuates hostile hyperinflammation in the infarcted myocardium.
As a typical glycolytic suppressor, 2-DG has been utilized for decades in many pathological situations, such as tumors 46 and epilepsy. 47  (g) Quantification of the heart weight/body weight ratio (n = 11-12). All data are presented as the means ± SD. Statistically significant differences were determined using one-way ANOVA followed by Tukey's test. LVEF and LVFS were analyzed using two-way ANOVA followed by Tukey's test (*p < 0.05, **p < 0.01, ***p < 0.001, and ns for not significant).
One unanticipated finding was hepatorenal toxicity, as evidenced by increased ALT levels and hepatorenal vacuolation (Figure 8), suggesting that although as it is a small molecule, 2-DG still exerted a toxic effect on those major metabolic organs. The topical application of 2-DG pat perfectly evades these side effects and exhibits similar benefits as systemic injection (Figures 6 and 7). These results suggest that in the treatment of local lesions, topical application may be a better choice for the clinical utilization of drugs, especially for those metabolic agents.
One important finding of this study is that macrophagedependent attenuation of hyperinflammation is involved in the 2-DGmediated cardioprotective benefits. In the presence of macrophages, MSCs + 2-DG-treated MI mice displayed a greater improvement in the retention of MSCs and subsequent cardiac repair. In contrast, after the depletion of macrophages, the protective effect of 2-DG disappeared completely (Figure 4c-h). When macrophages are absent, 2-DG loses its target for regulating immune stress, and thus Cl 2 MDP-lipo-treated MI mice exhibit aggravated cardiac dysfunction and an increased fibrotic area. Therefore, our purpose is not to obliterate all macrophages or M1 macrophage subsets but to decrease their proinflammatory activation.
In addition, the impact of glycolytic inhibition on MSCs was also explored. Intriguingly, we found that 2-DG enhanced the gene expres-   4). Error bars represent SD, and significance was determined using oneway ANOVA followed by Tukey's test (*p < 0.05, **p < 0.01, ***p < 0.001, and ns for not significant).
One limitation of the study is the indeterminate energy source of cells or tissues when glycolysis is subdued. Due to limited oxygen and nutrients, glycolysis is important for supplying energy to the myocardium after MI. In a review of the literature, restrained glycolysis induces biosynthesis and alternate fuel consumption, 48 providing substrates for subsequent biogenesis and energy demand. 49

| Preparation and functional assay of the chitosan/gelatin-based 2-DG patch
The filter-sterilized 4% chitosan (w/v, in 1% acetic acid, Senopharm) and 2% gelatin (w/v, in 1% acetic acid, Senopharm) solutions were mixed at a weight ratio of 1:1, and then 1% 2-DG was added. After complete dissolution by stirring, the impurities and bubbles were removed by centrifugation at 3000-4000 rpm for 15 min. The mixture was poured on a culture plate and placed in a drying oven overnight at 50-66 C, followed by deacidification with 2% NaOH-80% ethanol.
Finally, deionized water was added to remove the excess acetic acid.
To explore the sustained release of 2-DG pat , ddH2O was added to the patch, and the supernatant after soaking were collected contin- ns equals to no significant difference between the MSCs + 2-DG pat group and MSCs + 2-DG inj group. All data are presented as the means ± SD, and significant differences were determined using ANOVA followed by Tukey's test (*p < 0.05, **p < 0.01, ***p < 0.001, #p < 0.05, and ns for not significant). site. Next, 5 Â 10 5 MSCs in 20 μl of saline were injected intramyocardially at three different sites surrounding the infarct zones. Mice in the sham group underwent only thoracotomy without LAD ligation.
In the 2-DG injection group (2-DG inj ), 500 mg/kg 2-DG was administered intraperitoneally 6 h before MI and 1 and 2 days after MI. For 2-DG pat group, a prepared patch was attached to infarct area with fibrin glue (Sigma-Aldrich) immediately after LAD ligation. The patch size was approximately 3.5 mm in diameter.

| Echocardiography
Cardiac function was continuously measured by echocardiography before and after MI induction as we previously described. 55 Briefly, after inhaled anesthesia, satisfactory two-dimensional long-and short-axis images of the left ventricle were obtained. The M-mode view of the parasternal short axis was recorded for analysis. All measurements and analyses were conducted by an experienced researcher who was blinded to the study groups. All results were averaged over three separate cardiac cycles.

| Histological analysis
Ischemic hearts were fixed with 4% paraformaldehyde and sectioned.
Sections perpendicular to the axis of the LAD were stained with Masson's trichrome (Solarbio) and H&E (Solarbio). Histological images were visualized using a stereoscopic microscope and analyzed with ImageJ software.  (e) Quantification of the heart weight/body weight ratio (n = 11-13). All data are presented as the means ± SD. Statistically significant differences were determined using one-way ANOVA followed by Tukey's test. Cardiac function was analyzed using two-way ANOVA followed by Tukey's test (*p < 0.05, **p < 0.01, #p < 0.05, and ns for not significant). ). (f) Random blood glucose levels measured in mice after MI (n = 10-11). Error bars represent SD, and significance was determined using one-way ANOVA followed by Tukey's test (*p < 0.05, **p < 0.01, ***p < 0.001, and ns for not significant). cardiac function in multiple groups over time, two-way repeated measures ANOVA was performed followed by Tukey's test. Differences were considered statistically significant at p < 0.05.

| CONCLUSIONS
Distinct from most strategies attempting to optimize MSCs potency or their resistance to hostile environments, in this study, we developed a 2-DG-loaded, chitosan/gelatin-based immunometabolic patch, aiming to tender aggravated glycolysis in ischemic region. On one hand, sustained release of 2-DG by the patch calmed hyperinflammatory response in ischemic myocardium, thereby facilitating retention of implanted MSCs and enhancing cardiac healing ultimately. On the other hand, topical release of 2-DG by the patch also avoided potential side effects associated with whole-body inhibition of glycolysis. In summary, our data support the promise of immunometabolic patch as a novel strategy to optimize cell therapy for treating MI.

CONFLICT OF INTEREST
The authors have no conflicts of interest to declare.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.