Tanshinone (cid:0)A Enhances the Therapeutic Ecacy of Mesenchymal Stem Cells Derived Exosomes in Myocardial Ischemia/reperfusion Injury via Up-regulating miR-223-5p

Background: Myocardial ischemia-reperfusion (I/R) injury is a serious obstacle for patients with coronary heart disease to benet from post-ischemic reow. After myocardial I/R injury, CCR2 + -resident macrophages are rapidly activated and participate in the subsequent inammatory response, whereas CCR2 - -resident macrophages play a major role in attenuating cardiac inammation and promoting tissue repair. Mesenchymal stem cells (MSCs) have gradually become attractive candidates that aid in understanding the pathogenesis and progression of cardiovascular diseases. The low immunogenicity and low carcinogenicity of stem cell-derived exosomes offer advantage in treating myocardial injuries. In this study, we investigated whether MSC-derived exosomes pretreated with tanshinone IIA (TSA) could exhibit stronger cardioprotective function in an I/R rat model and explored its underlying mechanism. Methods: We investigated the effect of TSA-MSC exo on myocardial I/R injury in vivo. The overexpression of CCR2 in the rat heart was used to determine the regulatory role of CCR2 in I/R injury. High-throughput sequencing of MSC exo and TSA-MSC exo to screen differential genes to explore the mechanism of TSA-MSC exo 's cardioprotective effect. Results: Compared with MSC exo , an intramyocardial injection of TSA-MSC exo was found to be more effective in rats in improving cardiac function, limiting the infarct size, inhibiting CCR2 activation, reducing monocyte inltration and promoting angiogenesis in the heart after myocardial I/R. Moreover, CCR2 had a regulatory effect on monocyte inltration and angiogenesis after I/R. Bioinformatics analysis and miRNA sequencing of MSC exo and TSA-MSC exo revealed miR-223-5p an effective candidate mediator for TSA-MSC exo to exert its cardioprotective function and CCR2 as the downstream target. Conclusion: In summary, our ndings indicated that miR-223-5p packaged in TSA-MSC exo inhibited CCR2 activation

myocardium is very important for tissue repair after myocardial injury [5][6][7]. Hence, studies on inhibiting monocyte in ltration and promotion of angiogenesis after I/R have been receiving increased attention.
In mammalian hearts, there are a large number of resident macrophages in the heart tissue that are divided into CCR2 − and CCR2 + subsets of embryonic and adult blood lineages [8]. Several studies have con rmed that when myocardial I/R injury occurs, CCR2 + tissue-resident macrophages are rapidly activated and monocytes are recruited to in ltrate the damaged myocardium to trigger an in ammatory response [9][10][11][12], whereas CCR2 − tissue-resident macrophages inhibit the recruitment of monocytes [8, 10, [13][14][15][16] and are selectively recruited into the perfused coronary vasculature to stimulate the growth and remodeling of blood vessels to repair the myocardial tissue, thereby improving I/R injury [15]. Therefore, inhibiting CCR2 activation after myocardial damage may effectively reduce the migration and accumulation of in ammatory monocytes to the injured myocardium and may be bene cial in promoting angiogenesis, thus resulting in repair of the myocardial tissue and protecting the injured heart from adverse remodeling. Thus, inhibiting CCR2 activation is of great signi cance in the treatment of cardiac I/R injury.
In recent years, the development of stem cell regenerative medicine has brought new hope for the treatment of cardiovascular diseases. Several studies have indicated that stem cells can promote the repair of damaged myocardial tissue through paracrine exosomes and cytokines. Exosomes are extracellular vesicles that are 30-150 nm in diameter. They are responsible for intercellular communication and play a role in the exchange of cellular substances and information. MicroRNA (miRNA) transported by exosomes facilitates cell-to-cell communication through epigenetic regulation of the receiving cell. Additionally, when maternal cells are stimulated by the environment, the secreted exosomal contents (miRNA, mRNA, protein) change and their biological effects are affected due to remodeling of their epigenetic chromatin, thereby enhancing the therapeutic e cacy of exosomes [17][18][19].
Previous studies have shown that TSA can limit infarct size and ameliorate several consequences of myocardial I/R injury, including myocardial enzyme spectrum, oxidation status, in ammation, cardiac dysfunction, and microstructural disorders [20][21][22][23][24]. In this study, we evaluated the cardioprotective effect of TSA-pretreated MSC-derived exosomes (TSA-MSC exo ) in rats with myocardial I/R injury and explored potential molecular targets. Thus, the development of therapies for exosomes derived from the combination TSA and MSC, and the exploration of therapeutic targets and mechanisms may provide novel strategies for the treatment of patients with heart failure.

Cell transfection
For cell transfection, MSCs were stably transfected with miR-223-5p inhibitory sequence (i223) containing lentivirus (Genechem, Shanghai, China) or lentivirus containing scrambled control sequences. MSCM was used to prepare a cell suspension with a density of 3-5×10 4 cells/mL, which was inoculated into a 6-well plate and cultured at 37℃ for 16-24 h until cells reached a con uence of 20%-30%. The infection reagent and lentivirus were added to MSCs according to the manufacturer's instructions. After 72 h of infection, the medium was changed to MSCM without exosomes containing 10 μmol/L TSA.

Isolation and characterization of exosomes
Exosomes were isolated by ultracentrifugation. Brie y, the conditioned supernatants of MSCs pretreated with or without TSA were collected and centrifuged at 300 ×g for 10 min. The supernatant was centrifuged at 10000 ×g for 30 min and ltered through a sterile 0.22-μm pore lter. The ltered solution was ultracentrifuged at 120,000 ×g for 2 h. Lastly, the supernatant was discarded to obtain the exosome precipitate [25,26]. The structure of exosomes was observed by transmission electron microscopy (TEM) analysis using a Hitachi HT7700 (Hitachi, Japan) and their size distribution was determined by dynamic light scattering (DLS) using a DynaPro NanoStar (WYATT, USA). The exosomes were identi ed using western blotting with the marker proteins CD9 (20597-1-AP; Proteintech, Chicago, USA), CD63 (25682-1-AP; Proteintech), and Alix (12422-1-AP; Proteintech). The protein concentration in exosomes was determined using the bicinchoninic acid kit (Thermo Fisher Scienti c, Waltham, MA, USA).

Internalization of MSC-derived exosomes in vitro and in vivo
To evaluate the internalization of MSC-derived exosomes by H9C2 in vitro, the exosomes were labeled with PKH26 (PKH26PCL; Sigma, USA) according to the manufacturer's protocol. H9C2 (1×10 5 cells/mL) were cultured in a 5% CO 2 incubator at 37°C. When the cells attained 80% con uence, they were supplemented with a medium containing MSC-derived exosomes (75 μg) labeled with PKH26. The cells were washed 3 times with phosphate-buffered saline (PBS) after 6 h. The nuclei were stained with DAPI and the cells were observed and photographed using uorescence microscopy (Leica DM6000M, HE, Germany). After a three-point intramyocardial injection of MSC-derived exosomes (75 µg in 100 µL PBS), the in vivo internalization of PKH26-labeled MSCs-derived exosomes was evaluated. After 0, 6, 12, 24, 48, and 72 h, epi uorescence was detected using an IVIS kinetic imaging system (PerkinElmer, Waltham, MA, USA).

In vitro functional assays
After exosomes were added to HUVECs, a cell scratch test was performed to evaluate the proliferation and migration of endothelial cells in vitro. Next, CD31 (11265-1-AP; roteintech) and vascular endothelial growth factor (VEGF) (19003-1-AP; Proteintech) immuno uorescence were used to evaluate the angiogenesis of HUVECs. ECM was used to prepare a cell suspension with a density of 1×10 5 cells/mL, which was inoculated into a 6-well plate and cultured in 5% CO 2 at 37℃ until 90% con uence was attained. The cells were scratched using a pipette tip and a serum-free medium was added. Then, the cells were taken out at 0 and 24 h of incubation and photographed. The supernatant was discarded and the cells were xed with 4% paraformaldehyde. Goat serum was used to block the activity for 30 min at 25℃. Next, the cells were incubated overnight with CD31 and VEGF at 4°C and subsequently incubated with anti-rabbit horseradish peroxidase (HRP) conjugated to Alexa Fluor 647 (ab190565; Abcam, UK) dye and anti-mouse HRP conjugated to Alexa Fluor 488 (ab171449; Abcam) at 37°C in the dark for 1 h. DAPI was used to stain the nucleus and the cells were observed and photographed using uorescence were anesthetized using an intraperitoneal injection of 5% tribromoethanol (30 mg/kg). Thoracotomy was performed in the space between the third and fourth ribs to expose the heart. A 6-0 suture was used to ligate the left anterior descending coronary artery. One hour after ischemia, the suture was removed for reperfusion [27]. Exosomes (75 μg in 100 μL PBS) or PBS were injected into the boundary region of the infarcted heart at three different locations. The rats in the sham operation group were not ligated. Speci c CCR2 overexpression was achieved in the left ventricles of rats by using adeno-associated virus (AAV)9based delivery vectors (Genechem, Shanghai, China).
Following tracheal intubation and thoracotomy, AAV9 expressing the CCR2 open reading frame (titer: 1.32×10 12 ) or a scrambled control sequence bearing no homology to known gene transcripts (titer: 1.14×10 12 ) were injected into the left ventricle at multiple sites. The myocardial I/R model was established successfully after 8 weeks of transfection.
Echocardiography to assess left ventricular (LV) function Two-dimensional M-mode echocardiography was performed on day 3 post-surgery using Vevo 2100 (VisualSonic, Canada) ultra-high-resolution animal ultrasound imaging system to evaluate LV function after rats were anesthetized by iso urane. Three representative cycles were captured for each animal and the LV ejection fraction (LVEF) and LV fraction shortening (LVFS) were calculated as described previously [28]. A Millar pressure-volume catheter (SPE-869, #840-813G) connected to an MPVS Ultra PV circuit system (Millar, Houston, TX, USA) was used to measure the pressure volume of the left ventricle of rats. The catheter entered the left ventricle through the right common carotid artery after calibration and a data acquisition system (LabChart Pro, ADInstruments, Colorado Springs, CO, USA) was used to calculate the measured values of dP/dt max and dP/dt min.

2,3,5-Triphenyltetrazolium chloride (TTC) staining
Rats were euthanized by cervical dislocation after 3 days of reperfusion. The heart was isolated and 5 horizontal slices were obtained on average. The tissue slices were incubated in TTC (G3005; Solarbio, Beijing, China) staining solution at 37°C in the dark for 30 min, xed with 4% paraformaldehyde, and then photographed. As reported previously, the necrotic myocardium was stained white, whereas the viable myocardium was stained red. The images were processed and analyzed using ImageJ (NIH, Bethesda, MD, USA) [27]. The percentage of infarct area was calculated using the following equation:

Flow cytometry
According to the manufacturer's recommendations, the enzyme mixture in the Multi-Tissue Dissociation kit 2 (130-110-203; Miltenyi, Germany) was used to digest the cells in the myocardial tissue. Brie y, the heart tissue of the infarct marginal zone was cut into small pieces (1-2 mm³) and suspended in 2.5 mL of the enzyme mix in gentleMACS C Tubes (130-093-237; Miltenyi). After incubation at 37°C for 15 min, the C Tubes were attached onto the sleeve of a gentleMACS Dissociator (Miltenyi), and the program "37C_Multi_G" was run until the myocardial tissue was in a single-cell suspension state. The digestion was stopped with DMEM containing 10% FBS and the single-cell suspension was centrifuged at 600 ×g at 4°C for 5 min. After washing with PBS, the cells were incubated with CCR2 (PA5-23040; Thermo Fisher Scienti c), major histocompatibility complex class II (MHCII) (17-0920-82; Thermo Fisher Scienti c) antibodies at 37°C for 1 h in the dark. Flow cytometry was carried out using a BD LSRFortessa X-20 Cell Analyzer (BD Biosciences, NJ, USA), and FlowJo software (Treestar, OR, USA) was used for data analysis.

Western blotting
Protein lysate (PMSF: RIPA=1:100) (Solarbio) was used to lyse the heart tissue from the infarct marginal zone. The extracted proteins were subjected to polyacrylamide gel electrophoresis (GenScript, Nanjing, China) and transferred onto polyvinylidene di uoride membranes (Thermo Fisher Scienti c). The membranes were blocked with QuickBlock Western reagent (Beyotime, Beijing, China) and incubated sequentially with CCR2 overnight at 4°C. Subsequently, the washed membranes were incubated with the corresponding HRP-conjugated secondary antibodies (SA00001-2; Proteintech) for 2 h at 25℃. Lastly, a gel imaging system (GE Healthcare, CHI, USA) was used to quantify the protein bands. The western blots were quanti ed using ImageJ software (NIH, MD, USA).
Quantitative polymerase chain reaction (qPCR) mRNA and miRNA reverse-transcription kits were used to obtain cDNA following the manufacturer's instructions (Roche, Basel, Switzerland) and SYBR Green PCR master mix (Roche) was used for qPCR.

Target gene prediction
Bioinformatics was used to predict the target genes of miR-223-5p and to explore the relationship between miR-223-5p and CCR2. The principle is that the smaller the free energy of miRNA sequence and mRNA sequence binding, the easier is the binding, thereby playing a regulatory role on target genes. The gene sequences of miR-223-5p and CCR2 mRNA were searched in the NCBI database. RIsearch 2.0 software was used for target gene prediction to explore the binding between miR-223-5p and CCR2.

Statistical analyses
Experimental data were statistically processed using SPSS 22.0 software and GraphPad Prism 6 was used for graphical constructions. Data are expressed as mean ± standard deviation (SD). Multiple groups of samples were compared using one-way ANOVA, if they obeyed normal distribution and homogeneity of variance. A t-test was used to compare differences between groups. P < 0.05 was considered statistically signi cant.

Characterization and internalization of exosomes
Exosomes were obtained based on the above methods of exosome extraction. TEM ndings revealed the TSA-MSC exo to be round vesicles with a lipid bilayer (Fig. 1A). DLS analysis revealed that particle diameters of TSA-MSC exo were in the range of 50-150 nm and the average particle size was 93.64 nm (Fig. 1B). To evaluate the internalization of exosomes by cells, the PKH26-labeled TSA-MSC exo were cocultured with H9C2 cardiomyocytes for 6 h and observed using uorescence microscopy (Fig. 1C). Western blotting was used to con rm that both MSC exo and TSA-MSC exo expressed exosome-speci c marker proteins CD9, CD63, and Alix (Fig. 1D). PKH26-labeled TSA-MSC exo was injected into rats through the myocardium in situ to evaluate its residence time in the heart. We found that TSA-MSC exo could be detected even after 72 h of injection. The radiation e ciency of TSA-MSC exo was found to decrease gradually within 72 h (Fig. 1E). These results indicated that the vesicles extracted by ultracentrifugation were indeed exosomes, and whether in vivo or in vitro, exosomes could be internalized by cardiomyocytes, thereby laying the foundation for their therapeutic effects.
Intramyocardial delivery of TSA-MSC exo improved cardiac function and reduced infarct size after myocardial I/R A rat model of I/R was used to evaluate the cardioprotective effects of exosomes in vivo. After 3 days of reperfusion, M-mode echocardiographic analysis showed that MSC exo and TSA-MSC exo treatment improved the wall motion amplitude of the LV (Fig. 2A). The LV contractility and function of rats with I/R treated with MSC exo and TSA-MSC exo were found to improve based on the signi cant increase in LVEF and LVFS 3 days after injection. The maximal and minimal left ventricular pressure derivative (dP/dt) was assessed using invasive hemodynamic measurement of pressure-volume (PV) loops. MSC exo and TSA-MSC exo treatment in rats markedly increased dP/dt max and decreased dP/dt min compared with those in the I/R group. Importantly, treatment with TSA-MSC exo further improved the LVEF, LVFS, dP/dt max, and dP/dt min after I/R compared with treatment using MSC exo (Fig, 2B-E). TTC staining was performed to evaluate the outcome of myocardial infarction after 3 days of injection and to quantify the area of the myocardial infarct. The results showed that the infarct size in rats with I/R treated with MSC exo and TSA-MSC exo was remarkably reduced (Fig. 2F, G). HE and TUNELstaining demonstrated that in ammatory cell in ltration and apoptosis in the myocardial tissue of rats with I/R were markedly worsened, respectively, and treatment with MSC exo and TSA-MSC exo signi cantly alleviated the occurrence of these conditions. Consistently, treatment with TSA-MSC exo was found to have signi cant e cacy ( Fig. 2H-K). Taken together, the above ndings suggested that TSA-MSC exo exerted a higher cardioprotective effect in rats after I/R than MSC exo .
Intramyocardial delivery of TSA-MSC exo inhibited CCR2 activation, reduced monocyte in ltration, and promoted angiogenesis after I/R Studies have con rmed that interrupting the recruitment of CCR2 signals plays a cardioprotective effect in animal models of myocardial injury [29]. Therefore, we investigated the effects of MSC exo and TSA-MSC exo on CCR2 after I/R. The results indicated that the number of CCR2 + cardiac-resident macrophages and the expression of CCR2 mRNA in rat hearts were signi cantly increased after I/R injury. Treatment with MSC exo and TSA-MSC exo markedly reduced the number of CCR2 + cardiac-resident macrophages and the expression of CCR2 mRNA (Fig. 3A-C). To elucidate the role of MSC exo and TSA-MSC exo in the in ltration of monocytes in the heart of rats with I/R injury, we determined changes in the monocyte chemokines in the left ventricular tissue of rats after 3 days of reperfusion. The results suggested that the mRNA expression of CCL2, CCL3, CXCL2, and CXCL3 in rats in the I/R group was signi cantly increased compared with that in the sham group, and the mRNA expression of CCL2, CCL3, CXCL2, and CXCL3 in rats treated with MSC exo and TSA-MSC exo declined markedly (Fig. 3D-G). Monocytes, which are the circulating precursors of macrophages [30]. On entry into tissues, monocytes give rise to macrophages [31]. Therefore, CD68 was used as a marker of macrophages to assess the in ltration of monocytes. Immuno uorescence studies were conducted using tissue sections of rat hearts 3 days after reperfusion. The expression of CCR2 and CD68 in the I/R group increased signi cantly compared with that in the sham group. Both MSC exo and TSA-MSC exo treatments distinctly inhibited CCR2 and CD68 expression ( Fig. 3H-J). After myocardial I/R occurs, the circulating monocytes are recruited to the myocardium to induce in ammation, thereby enhancing protease activity and triggering angiogenesis. The continuous development of in ammation hinders myocardial tissue repair. However, angiogenesis is a phenotype of tissue repair [31]. Immuno uorescence co-staining results of CCR2/CD68 and CCR2/α-SMA indicated that CCR2 may have a regulatory effect on monocyte in ltration and angiogenesis.
Overexpression of CCR2 increases monocyte in ltration, inhibits angiogenesis, and weakens the cardioprotective effect of TSA-MSC exo after I/R injury To better understand whether CCR2 has a regulatory effect on monocyte in ltration and angiogenesis after myocardial I/R, AAV9 carrying a speci c sequence targeting CCR2 was applied to increase CCR2 expression in rat hearts. Eight weeks after AAV-CCR2 (CCR2 +/+ ) or AAV negative control (NC) was injected into the myocardium of rats, western blotting was performed to determine CCR2 expression in the heart tissues. The results showed that the protein expression of CCR2 in the CCR2 +/+ group was markedly higher than that in the NC group (Fig. 4A, B), which veri ed that the myocardial in situ injection of AAV-CCR2 successfully upregulated the expression of CCR2 in the rat heart, thereby justifying the development of subsequent experiments. Next, a myocardial I/R model was constructed on the basis of CCR2 +/+ and NC rats. Echocardiographic results showed that the LVEF and LVFS of the CCR2 +/+ group were not different from those in the NC group. Treatment with TSA-MSC exo still signi cantly improved the LVEF and LVFS after I/R (Fig. 4C, D), suggesting that CCR2 overexpression did not aggravate the deterioration of cardiac function. Results from HE and Sirius Red staining showed that collagen deposition and in ammatory cell in ltration in the CCR2 +/+ group were obviously increased compared with those in the NC group. Moreover, in CCR2 +/+ rats, there were no marked differences in in ammatory cell in ltration and collagen deposition between the TSA-MSC exo and I/R groups (Fig. 4E, F).
Immuno uorescence co-staining results showed that the expression of both CCR2 and CD68 was signi cantly increased in the CCR2 +/+ group compared with that in the NC group. In CCR2 +/+ rats, there were no obvious changes in the expression of CCR2 and CD68 in the TSA-MSC exo group compared with the I/R group (Supplementary Fig. 2A-C). CCR2/α-SMA and CCR2/VEGF immuno uorescence co-staining results indicated that the expression of CCR2 in the CCR2 +/+ group was obviously higher than that in the NC group, and the expression of α-SMA and VEGF was signi cantly decreased. In CCR2 +/+ rats, the expression of CCR2, α-SMA, and VEGF did not change remarkably in the TSA-MSC exo group compared with that in the I/R group (Fig. 4G, H). These ndings suggested that the overexpression of CCR2 increased monocyte in ltration and inhibited angiogenesis after I/R injury such that TSA-MSC exo almost lost its effect of reducing monocyte in ltration and promoting angiogenesis. miR-223-5p was a candidate effector for TSA-MSC exomediated improvement of cardiac function and myocardial injury The biological effects of exosomes depend on the function of miRNAs. Exosomes can exert therapeutic effects on cardiac I/R injury by delivering speci c miRNAs to regulate receptor cell function [32][33][34]. To determine how TSA-MSC exo improves myocardial I/R injury in rats, high-throughput sequencing was conducted to identify the differentially expressed miRNAs between MSC exo and TSA-MSC exo (Fig. 5A), in combination with the ndings from previous studies that reported that miR-223-5p can regulate in ammation and inhibit I/R injury in mice to promote tissue repair [35][36][37][38]. We further veri ed that the content of miR-223-5p in TSA-MSC exo was signi cantly higher than that in MSC exo (Fig. 5B). Hence, in subsequent experiments, we explored the role of miR-223-5p in improving I/R injury in rats. The miR-223-5p-inhibitory lentivirus (with GFP label) was transfected into MSCs to inhibit miR-223-5p (i223) expression, and the lentivirus was used as a vector to transfect into MSCs as an NC. Results from qPCR showed that the lentivirus was successfully transferred into MSCs and remarkably decreased miR-223-5p content in MSCs (Fig. 5C). After successful transfection, the MSCs were treated with TSA and the exosomes were subsequently extracted (TSA-MSC-NC exo and TSA-MSC-i223 exo ). Using qPCR, we determined that miR-223-5p content in TSA-MSC-i223 exo was signi cantly reduced than that in TSA-MSC-NC exo (Fig. 5D). After treatment with TSA-MSC-NC exo and TSA-MSC-i223 exo , the ultrasound results suggested that TSA-MSC-NC exo distinctly increased the LVEF and LVFS of rats compared with those in the I/R group. However, the LVEF and LVFS of rats treated with TSA-MSC-i223 exo were not signi cantly different from those of rats in the I/R group (Fig. 5G, H). TTC staining results suggested that TSA-MSC-NC exo signi cantly limited the infarct size of rat heart, and the cardiac infarction area treated with TSA-MSC-i223 exo were not signi cantly different from that of rats in the I/R group (Fig. 5I, J). miR-223-5p shuttling by TSA-MSC exo modulated monocyte in ltration and angiogenesis by targeting CCR2 The interaction between miR-223-5p and the CCR2 gene was predicted using bioinformatics analysis. The results revealed that the free energy of binding between the gene sequence of miR-223-5p derived from humans and the gene sequence of CCR2 derived from different species was lower than -10 kcal/mol ( Supplementary Fig. 3A), revealing that they could bind stably and interact with each other. Furthermore, in I/R rats injected with PKH26-labeled MSC exo , TSA-MSC-NC exo , and TSA-MSC-i223 exo , immuno uorescence results of the three labeled exosomes and CCR2 indicated that CCR2 expression was lower in the regions where exosomes were present after treatment with MSC exo and TSA-MSC-NC exo , whereas CCR2 expression in the regions where exosomes were present increased signi cantly after treatment with TSA-MSC-i223 exo (Fig. 6A-C), suggesting the key role of miR-223-5p in inhibiting CCR2 activation. Furthermore, immuno uorescence results of CCR2 and CD68 co-staining showed that the positive expression of CCR2 and CD68 in the myocardium of rats treated with TSA-MSC-NC exo was distinctly decreased compared with that of rats in the I/R group, and the positive expression of CCR2 and CD68 in the myocardium of rats treated with TSA-MSC-i223 exo did not change signi cantly compared with those in the I/R group ( Supplementary Fig. 3B-D). Immuno uorescence results of CCR2/α-SMA and CCR2/VEGF co-staining demonstrated that the positive expression of CCR2 in the myocardium of rats treated with TSA-MSC-NC exo was signi cantly reduced compared with that in the I/R group, and the positive expression of α-SMA and VEGF was signi cantly increased. However, the positive expression of CCR2, α-SMA, and VEGF in the myocardium of rats treated with TSA-MSC-i223 exo did not change signi cantly compared with that in the I/R group (Fig. 6D-I). Results from Sirius Red staining showed that collagen deposition from TSA-MSC-NC exo treatment was signi cantly decreased compared with that in the I/R group. Moreover, collagen deposition and from TSA-MSC-i223 exo treatment was not signi cantly different from that in the I/R group (Fig. 6J, K). These results indicated that diminishing miR-223-5p weakens or even almost eliminates the regulatory effect of TSA-MSC exo on monocyte in ltration and angiogenesis after I/R injury, suggesting that miR-223-5p encapsulated in TSA-MSC exo modulated monocyte in ltration and angiogenesis after I/R by inhibiting the activation of CCR2.

Discussion
In recent years, the development of stem cell regenerative medicine has brought new hope for the therapy of cardiovascular diseases. Several studies have shown that stem cells secrete paracrine factors, including exosomes, to exert cardioprotective effects [33]. MSC-derived exosomes can promote the repair of damaged myocardium and improve cardiac function after myocardial injury, suggesting the potential of exosome-based approaches as novel cell-free therapy in cardiac repair [34,39,40]. In this study, we found that the intramyocardial injection of MSC exo and TSA-MSC exo improved cardiac function, inhibited CCR2 activation, reduced monocyte in ltration, and promoted angiogenesis after myocardial I/R injury. In addition, CCR2 was found to have a regulatory effect on monocyte in ltration and angiogenesis after I/R.
It was worth noting that TSA-MSC exo was more effective in improving cardiac function and myocardial injury. The e cacy of TSA-MSC exo partially mediated by miR-223-5p. TSA-MSC exo -encapsulated miR-223-5p was transported to the injured heart tissue to inhibit CCR2 activation, thereby reducing monocyte in ltration and promoting angiogenesis to exert a cardioprotective effect (Fig. 7). To our knowledge, this is the rst study to determine the cardioprotective effects of TSA-pretreated MSC-derived exosomes.
Therapy with stem cell-derived exosomes shows low immunogenicity and low carcinogenic risk and similar effects as those observed with stem cell transplantation in the treatment of myocardial I/R injury.
In recent years, the modi cation of stem cell-derived exosomes has shown excellent outcomes in repairing myocardial injury. Studies have con rmed that the expression of anti-apoptotic miRNAs in exosomes derived from GATA4-overexpressing MSCs increases and, thus, enhances cardioprotection in MI rats [41]. Another study con rmed that the Akt-modi ed MSCs-derived exosomes enhanced angiogenesis by activating growth factor D [42]. Although these modi cations may increase the expression of cytokines or miRNA and enhance the therapeutic activity of exosomes, these outcomes are currently di cult to achieve in clinical practice. Pharmacological studies have shown that TSA can protect the heart from ischemic damage and may be a promising therapeutic agent for the treatment of myocardial I/R injury and arrhythmia; however, the poor bioavailability of TSA limits its e cacy [43]. In this study, the development of an alternative therapy for TSA-pretreated MSC-derived exosomes not only improved the shortcomings of TSA but also increased the therapeutic activity of exosomes. Our ndings revealed that TSA-MSC exo is more effective in enhancing EF value, limiting the infarct size, and improving myocardial damage after I/R compared with that observed using MSC exo treatment. Therefore, our ndings may provide a prospective strategy for the treatment of cardiovascular diseases and create the possibility for the use of TSA in a clinical setting.
Proin ammatory monocytes are recruited to the damaged myocardium after myocardial I/R to induce an in ammatory response, which is accompanied by the subsequent release of proin ammatory mediators and enhanced protease activity to exacerbate the in ammatory response. The continuous development of in ammation hinders myocardial tissue repair [39]. However, stimulating the growth of small blood vessels and the formation of collateral circulation in the ischemic regions of the myocardium is bene cial for tissue repair after myocardial injury [44]. Recent studies have demonstrated that exosomes isolated from stem cells, such as MSCs and cardiomyocyte-derived cells (CDCs), have the ability to enhance angiogenesis and reduce in ammation. Consistently, our results con rmed that MSC exo had the ability to inhibit the in ltration of monocytes and promote angiogenesis. Furthermore, TSA-MSC exo showed greater potential with respect to these aspects. Studies show that CCR2 activation leads to the recruitment of monocytes to in ltrate the infarct area and cause abnormal remodeling of coronary blood vessels, thereby exacerbating myocardial I/R injury [14,15]. The results from our study indicated that TSA-MSC exo signi cantly reduced CCR2 expression. Based on these ndings, we injected AAV9-CCR2 into the rat heart to increase CCR2 expression and successfully established the myocardial I/R model. We discovered that the overexpression of CCR2 promoted the in ltration of monocytes into the injured heart and interfered with angiogenesis after I/R, thus exacerbating myocardial injury and weakening the cardioprotective effect of TSA-MSC exo .
Increasing evidence reveals that the biological effects of exosomes depend on the function of miRNAs, which means that exosomes regulate the function of cellular receptors and interfere with the pathophysiological processes of the body by delivering speci c miRNAs [32,33]. Certain exosomal miRNAs, such as miR-21, miR-223, miR-146a, and miR-181b, derived from stem cells have important immunomodulatory properties [27,[45][46][47]. miR-223-5p is an effective negative regulator of cardiac in ammation, which inhibits monocyte in ltration and prevents polarization of M1 macrophages by interfering with one or more targets including NF-κB, IKKα, and STAT3, thereby playing a role in cardiovascular protection [35,48,49]. To date, only a few studies have explored the relationship between miR-223-5p and myocardial I/R injury. Based on the ndings from our sequencing data, supported by the previous literature, we identi ed miR-223-5p as the preferred research object. In this study, the expression of miR-223-5p in TSA-pretreated MSC-derived exosomes was remarkably higher in TSA-MSC exo than in MSC exo . Moreover, when miR-223-5p in TSA-MSC exo was inhibited, neither LVEF nor LVFS nor the infarct size of rats with I/R injury were signi cantly recovered. Our ndings prove, for the rst time, that the decline of miR-223-5p expression in TSA-MSC exo greatly weakened the regulatory effects of TSA-MSC exo on the cardiac function in rats after I/R injury. Furthermore, when the expression of miR-223-5p in TSA-MSC exo was suppressed, CCR2 expression was not signi cantly downregulated. Thus, bioinformatics analysis was used to identify the interaction between miR-223-5p and CCR2. We found that when the miR-223-5p level in TSA-MSC exo decreased, its e cacy in inhibiting CCR2 activation was partially reversed and was accompanied by an increase in monocyte in ltration and inhibition of angiogenesis. To the best of our knowledge, the relationship between miR-223-5p and CCR2 has not been reported previously. Our ndings con rmed that TSA-MSC exo inhibited CCR2 by transporting miR-223-5p into the heart to inhibit monocyte in ltration and promote angiogenesis, thus alleviating I/R injury.
Despite conclusive evidence indicating that miR-223-5p in exosomes is essential for cardioprotective effects after I/R, many other components in exosomes also have biological activity and can synergistically exhibit overall functional advantages. Therefore, we cannot ignore the contribution of other exosomal contents. Although there is evidence that CCR2 may be the downstream target protein of exosomal miR-223-5p, it is clear that exosomes and their miRNAs have effects on proteins. Thus, the possibility of their synergistic effects cannot be ruled out, and merely targeting CCR2 as a research object may oversimplify its actual biology. Lastly, although TSA-MSC exo had a signi cantly improved therapeutic affection compared with MSC exo , further preclinical studies are needed to ensure its e cacy and safety, prior to this strategy being developed into a potential cell-free therapy for myocardial repair.

Conclusion
Overall, we found that TSA-MSC exo could ameliorate cardiac function, limit infarct size, inhibit the activation of CCR2, reduce monocyte in ltration, and promote angiogenesis to attenuate myocardial I/R injury in rats. And miR-223-5p encapsulated in TSA-MSC exo participates in the regulation of monocyte  J, K Representative images of TUNELstaining of a cross-section of the heart 3 days after reperfusion (scale bar = 100 μm). Data are expressed as mean ± SD. **, p < 0.01 compared with the sham group; ##, p < 0.01 compared with the I/R group.   Overexpression of CCR2 regulates monocyte in ltration and angiogenesis after I/R injury. A Representative western blot of CCR2 from the heart 8 weeks after transfection of AAV9-CCR2 into the rat myocardium. B Quantitative analysis of CCR2 protein expression in the rat myocardium (n = 3). C, D Quantitative analysis of LVEF and LVFS 3 days after reperfusion (n = 6). E, F Representative images of HE (scale bar = 50 μm) and Sirius Red (scale bar = 100 μm) staining of a cross-section of the heart 3 days after reperfusion. G Representative images of CCR2 and α-SMA immuno uorescence of heart tissue sections 3 days after reperfusion (scale bar = 100 μm). H Representative images of CCR2 and VEGF immuno uorescence of heart tissue sections 3 days after reperfusion (scale bar = 100 μm). I-N Quanti cation of in ammatory cell in ltration, collagen deposition, CCR2 and α-SMA immuno uorescence, CCR2 and VEGF immuno uorescence (n = 3). Data are expressed as mean ± SD. **, p < 0.01 compared with the sham group; #, p < 0.05, ##, p < 0.01 compared with the I/R group.  = 100 μm). G-I Representative images and quanti cation of CCR2 and VEGF immuno uorescence of heart tissue sections 3 days after reperfusion (scale bar = 100 μm). J, K Representative images and quanti cation of Sirius Red staining of a cross-section of the heart 3 days after reperfusion (scale bar = 100 μm). Data are presented from three independent experiments. *, p < 0.05, **, p < 0.01 compared with the sham group; ##, p < 0.01 compared with the I/R group.

Figure 7
Proposed mechanisms of intramyocardial injection of pretreated with TSA-pretreated MSC-derived exosomes to protect the heart from myocardial I/R injury. TSA increases the production of miR-223-5p in MSCs. miR-223-5p, cargoed in and delivered by TSA-MSCexo to the heart, reduces the in ltration of monocytes and promotes angiogenesis to attenuate in ammation and promote myocardial tissue repair thus ameliorating myocardial I/R injury by inhibiting the activation of CCR2.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.