Activation of Invariant Natural Killer T Cells Further Ameliorates Post-Infarct Cardiac Remodeling in Mice With MSC Transplantation

The purpose of this study was to examine the effect of MSCs on the inltration of iNKT cells and further observe whether the activation of iNKT cells could assist the therapeutic action of MSCs on ventricular remodeling.

in ammatory and reparative phases is subtle and requires proper ne-tuning to prevent excessive in ammation or inadequate stimulation of repair (4). The in ammatory response to MI plays a critical role in determining MI size. A persistent proin ammatory reaction can contribute to adverse postinfarction remodeling; ultimately, excessive in ammation results in limited capacity for self-renewal and adverse remodeling, which nally leads to depressed left ventricular (LV) function (5). All of these factors make in ammation an important therapeutic target for improving outcomes following AMI.
Administration of mesenchymal stem cells (MSCs) in an animal model of MI yields variable degrees of functional improvement and is a promising tool for the treatment of a range of human degenerative and in ammatory diseases. MSCs are applied as a form of regenerative medicine, mainly based on their capacity to differentiate into speci c cell types and to act as bioreactors of soluble factors that promote tissue regeneration (6). In addition to these regenerative properties, the discovery that MSCs hold an immunoregulatory capacity was made over a decade ago, when it was observed that MSCs abrogated Tcell proliferation in vitro (7). The immune system has a critical role in the pathogenesis and progression of a number of degenerative diseases, raising the possibility that MSCs may be effective in repairing damaged organs by promoting cell formation and modulating the associated immune response. Based on this premise, the immunogenicity and immunomodulatory properties of MSCs have been thoroughly characterized to evaluate their potential clinical application (8-11). MSCs were reported to interact with several cell types of the immune system and may be an excellent means to reduce detrimental in ammatory reactions (12).
As the regulation of the in ammatory reaction appears to be ine cient after massive necrosis, the interest of current research has turned towards the induction of anti-in ammatory or regulatory subsets of immune cells to reduce apoptosis and brosis. Invariant natural killer T (iNKT or type 1 NKT) cells are innate-like T lymphocytes that coexpress αβT-cell receptor and NK cell markers and recognize glycolipid antigens (13). Moreover, iNKT cells have a protective role against autoimmune and neoplastic diseases, including acute liver injury (14) and rheumatic disease (15). As reported by Sobirin et al. (16) and Homma et al. (17), activation of iNKT cells by α-galactosylceramide (α-GC), a speci c activator of iNKT cells (18), ameliorates myocardial ischemia/reperfusion injury and postinfarct cardiac remodeling in mice.
In 2009, Prigione et al. (8) demonstrated that human bone marrow MSCs abolished the proliferation and interferon (IFN)-γ production of iNKT cells in vitro through the release of soluble factors. While the effects of MSCs on iNKT cells appear to be ideal in an in vitro scenario, in ammation in vivo is complex, which may have unpredictable in uences on MSCs, and further exploration and validation with an animal model for the interaction between MSCs and iNKT cells are required. If in vivo MSCs inhibit the in ltration of iNKT cells in the postinfarction heart, we wondered whether activating iNKT cells before MSC transplantation can have a synergistic effect on LV remodeling. The present study aims to demonstrate the effect of MSCs on iNKT cells in vivo and to investigate further the potential therapeutic value of MSC transplantation together with iNKT cell activation on LV remodeling.

Methods
Animals. All experiments were performed in accordance with the ethics code for animal experimentation. Ethical approval for all work was received from the animal research committee of Zhengzhou University (Zhengzhou, China). A total of 400 male C57BL/6J mice were purchased from Beijing HFK Bioscience Co., Ltd. (Beijing, China). The speci c criteria for animal euthanasia included absence of food or water intake, low or no mobility, weak or absent heartbeat and absence of respiratory movement during the ongoing study, and all animals were euthanized at the end of the study. The mice were divided into different groups by sortition randomization method to control the cofounding bias. After intraperitoneal anesthesia (ketamine/midazolam 75 mg/kg and 5 mg/kg, respectively), mice were euthanized by cervical dislocation to minimize suffering. The method to con rm the death of mice included the absence of movement and heartbeat. All efforts were made to minimize suffering, including gaseous anesthesia with iso urane (2-3%). Animals at 6-8 weeks of age (weight 20-25 g) were used for the experiments. Mice were kept in a pathogen-free environment at an invariable temperature and under a 12-h light-dark cycle. Pathogen-free chow and water sterilized using a high-pressure steam sterilizer were provided ad libitum.
MSC culture and expansion. MSCs isolated from the bone marrow of C57BL/6 mice were obtained from Cyagen Biosciences Inc. (Guangzhou, China). The identi cation of cells according to multipotency and the cell surface phenotypes was performed by the provider. The cell surface phenotypes were CD29 + , CD44 + , Sca-1 + , CD117and CD31 -, characterized by uorescence-activated cell sorting analysis Fig.1 .
The cells had the potential for differentiation into the osteogenic and adipogenic lineages as determined by staining with Alizarin red and oil red O, respectively. This evidence veri ed their identity as MSCs.
The mouse MSCs were cultured in Dulbecco's modi ed Eagle's medium supplemented with 10% fetal bovine serum (Gibco, Thermo Fisher Scienti c, Inc., Waltham, MA, USA), penicillin (100 U/ml), streptomycin (100 mg/ml) and 2 mM L-glutamine at 37°C in a humidi ed atmosphere containing 5% CO 2 . To obtain MSC clones, cells at 80-90% con uence were harvested and seeded into T25 asks at 2x10 4 cells/cm 2 . Each clone was then picked and expanded. Cells at the 8 th to 10 th passages cultured in serum-free medium were used in the experiments.
Myocardial infarction and MSC transplantation. Mice were anesthetized by intraperitoneal injection of ketamine/medazolam (75 mg/kg and 5 mg/kg, respectively). With the application of a rodent ventilator, tracheal intubation was performed on the mice, a small thoracotomy was performed, and left anterior descending coronary artery (LAD) ligation was created as previously described (19), with a slight modi cation. A 10/0 Prolene suture was passed under the LAD at 1-1.5 mm distal to the left atrial appendage, immediately following the bifurcation of the major left coronary artery. At 1 h after surgical intervention, a total of 1x10 6 C57BL6/J MSCs in 25 µl complete medium were injected into ve different points of the peri-infarct LV region in the MI group or the LV region in the sham group (20). Five points were separately injected with 2x10 5 MSCs in 5 µl medium. After 7, 14 and 28 days, TTC staining was performed to determine the infarct area of the LV. The remaining area of the LV was con rmed as the periinfarct LV.
Experimental design RT-qPCR for iNKT cells in post-MI hearts.
A total of 18 MI and 6 sham mice were created as described above. To observe time-dependent changes in iNKT cells, mice were sacri ced at 7, 14, and 28 days, and the LVs of their hearts were excised for reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis. As in a previous study (21), RT-qPCR for Vα14Jα18 (a speci c marker of iNKT cells in C57BL/6 mice) was performed.
RT-qPCR for iNKT cells in post-MI hearts with MSC transplantation.
At 1 h after surgery, MI and sham mice were randomly assigned to one of the following groups: i) 25 µl serum-free medium (Me) alone, ii) 25 µl complete cell-free medium after cell culture (cMe), or iii) 25 µl complete medium with 1x10 6 MSCs injected into the myocardium of LV, with the suspension infused within 30 sec (22). Thus, there were six groups: Sham+Me (n=6), Sham+MSCs (n=6), Sham+cMe (n=6), MI+Me (n=18), MI+MSCs (n=18) and MI+cMe (n=18). After 7, 14 or 28 days, RT-qPCR analysis was performed to observe time-dependent changes in iNKT cells in the PLV area of the MI groups compared with those in the LV area of the sham groups.
RT-qPCR for iNKT cells, Masson staining and TUNEL staining in post-MI hearts with MSC treatment and iNKT cell activation.
Mice were randomly divided into ve groups. At 1 h after MI, one group was injected with serum-free medium (Me) as a control, and another two groups randomly received an injection of 1x10 6 MSCs in 25 µl cMe with or without indome. In the fourth and fth groups, α-galactosylceramide (α-GC; 0.1 µg/g body weight; Avanti Polar Lipids, Inc., Alabaster, AL, USA) was administered 30 minutes before MI speci cally to activate iNKT cells while 1x10 6 MSCs with 25 µl complete medium were also injected into the sixth group. Therefore, there were six groups: Sham+Me (n=6), MI+Me (n=15), MI+MSCs (n=18), MI+MSCs+indome (n=18), MI+α-GC (n=18) and MI+MSCs+α-GC (n=18). After 7 and 28 days, RT-qPCR analysis of Vα14Jα18 was performed. After 28 days, the mice were euthanized, and the hearts were obtained. Tissue brosis of the LV was assessed by Masson's trichrome staining, and apoptosis was evaluated by terminal deoxynucleotidyl transferase deoxyuridine triphosphate nick end labeling (TUNEL) and western blotting analysis for cleaved caspase-3 protein. Fibrotic and total areas of each image were measured by computerized planimetry (Image ProPlus 5.0, Media Cybernetics, Silver Spring, MD, USA), and the percentage of the brotic area was calculated as follows: (Fibrotic area/total area) x100%.

Methods of western blotting and TUNEL
Western blot analysis. Total protein was extracted from LV samples using IP lysis buffer after pulverizing in liquid nitrogen. The concentration of the protein was quantitatively measured by a BCA protein kit (Pierce, USA). The protein was mixed with loading buffer and boiled at 100°C for 10 min to denature it. Total protein (20 μg) was separated by 12% SDS-PAGE and transferred to 0.45 µm polyvinylidene uoride (PVDF) membranes. After blocking with 5% nonfat-dried milk at room temperature for 1 h, membranes were incubated with primary antibodies [anti-cleaved caspase-3 antibody (Cat. No. ab49822; Abcam; USA) and anti-GAPDH antibody (Cat. ab181602 Abcam; USA)] overnight at 4°C and then incubated with horseradish peroxide-conjugated secondary antibodies at room temperature for 1 h. The blots were developed with ECL reagent (Millipore, Germany). GAPDH was used for normalization. The sum density of each protein band was measured by ImageJ software.
TUNEL. The mouse heart LV samples were dehydrated by standard graded alcohol solutions and embedded in para n as described previously (PMID: 31419728). LV samples in para n were sectioned longitudinally into 5 μm thin sections. After three washes with PBS (pH = 7.4), the sections were incubated with proteinase K for 30 min at 37°C. After treatment with cell permeable uid (Servicebio, G1204), the sections were assayed using an in situ death detection kit (Roche, 11684817910) according to the manufacturer's instructions. All of the sections were counterstained with 4′,6-diamidino-2phenylindole (DAPI). Finally, the numbers of total cells and TUNEL-positive cells were counted in at least three noncontinuous elds of each specimen using an inverted uorescence microscope (Zeiss, Germany). The percentage of positive cells (positive cells/total cells ×100) was determined as the apoptotic rate (%). Cell counting was performed by a pathologist blinded to the experimental conditions. Statistical analysis. Data analysis was performed using SPSS software 18.0 for Windows (SPSS, Inc., Chicago, IL, USA). Values are expressed as the mean ± standard deviation. Each experiment was performed at least three times independently. Analysis of variance and ANOVA were performed for multiple and two-group comparisons. P<0.05 was considered to indicate a signi cant difference.

Results
Elevation of iNKT cell receptors in post-MI hearts.
The quanti cation of iNKT cells was veri ed by gene expression of cell surface receptor-Vα14Jα18 (Fig. 2). Similar to those in a previous study (16), the results indicated that the in ltration of iNKT cells into the peri-infarct LV (PLV) signi cantly increased at 7 days after MI (1.72-fold change vs. sham group; P < 0.05) but almost returned to baseline levels at 14 and 28 days after MI. A persistent elevation of Vα14Jα18 gene expression was observed in the infarct LV (ILV) at 7, 14 and 28 days after MI (1.79-, 1.77and 1.72-fold change vs. sham group, respectively; P < 0.05).
Suppressive effect of MSCs on iNKT cells in the post-MI heart.
As presented in Fig. 3a, the values are all presented as the ratio of the Sham + Me group. RT-qPCR demonstrated that the gene expression of Vα14Jα18 was signi cantly suppressed at day 7 in the MI + MSCs (1.25 ± 0.29) and MI + cMe groups (1.19 ± 0.25) compared with the MI + Me group (1.76 ± 0.20; P < 0.05). However, the gene expression of Vα14Jα18 at day 7 in the MI + MSCs and MI + cMe groups was still higher than that in the corresponding sham groups (1.25 vs. 1.00; 1.19 vs. 1.01; P > 0.05), while the difference between the two treated groups was not signi cant. In addition, there was no signi cant difference between the MI + Me, MI + MSCs and MI + cMe groups at days 14 and 28.
Although the mechanisms of immunoregulation of MSCs are unknown, direct cell-to-cell contact and/or soluble factors, such as TGF-β1, NO and PGE 2 , may mediate the effect. The results may indicate that soluble factors in complete medium after MSCs are cultured may have a major role in immunoregulation. The effect of the administration of antagonists and neutralizing antibodies of certain soluble factors on the gene expression of Vα14Jα18 was measured, and the results are shown in Fig. 3b. The speci c inhibition of iNOS and TGF-β1 had no signi cant in uence on the in ltration of iNKT cells, while a signi cant effect was observed with indome, a speci c antagonist of PGE 2 . As presented in Fig. 3b, in the MI + cMe + indome group, the quanti ed expression of Vα14Jα18 was 0.54-fold higher than that in the MI + cMe group (1.74-vs. 1.20-fold change; P < 0.05). To verify further the effect of PGE 2 , 25 µl complete medium with added PGE 2 was injected into post-MI mice. The results indicated that the gene expression of Vα14Jα18 was obviously suppressed, with a signi cant difference compared with the MI + cMe + indome group (1.04-vs. 1.74-fold change; P < 0.05).

Synergistic effect of iNKT cells and MSC transplantation on ventricular remodeling
In mice treated with soluble factors and MSCs, quanti cation of Vα14Jα18 was performed at days 7 and 28 after MI, with values expressed as the fold-change of the Sham + Me group (Fig. 4). As indome is an antagonist of PGE 2, counteracting the effect of PGE 2 , a signi cant difference between the MI + MSCs and MI + MSCs + indome groups was observed after 7 days (1.74-vs. 1.19-fold; P < 0.05) but not at 28 days. In the MI + α-GC and MI + MSCs + α-GC groups, the gene expression of Vα14Jα18 was signi cantly elevated in the PLV at day 7; however, it remained signi cantly increased at day 28 (6.7-vs. 2.6-fold in the MI + MSCs + α-GC group, 7.1-vs 2.7-fold in the MI + α-GC group), which was in accordance with the result of previous study (16), the gene expression of Vα14Jα18 in the MI + MSCs + α-GC group tended to be less than in the MI + α-GC group, but with no signi cant difference.
To investigate the underlying molecular mechanisms of the antiapoptotic effects of α-GC injection, the protein expression of caspase-3 in PLV heart tissue was analyzed by western blot analysis. The results indicated that the proapoptotic cleaved protein, caspase-3, was signi cantly attenuated in the MI + MSC group compared with the MI + Me group. Compared with the MI + MSC group, the level of caspase-3 was not signi cantly different in the MI + MSCs + indome and MI + α-GC groups, while it was signi cantly decreased in the MI + MSCs + α-GC group (Fig. 5b).
To observe the effect of MSCs on cardiac brosis, the brotic areas in the PLV and ILV sections were quanti ed. The PLV and ILV sections of MSC-treated hearts exhibited less brosis than those of Metreated hearts (P < 0.05; Fig. 5c; Table 1). The MI + MSCs + indome group did not exhibit a signi cant difference in brotic area compared with the MI + MSC group, while compared with the MI + MSC and MI + α-GC groups, α-GC + MSCs signi cantly attenuated brosis (P < 0.05; Fig. 5c; Table 1).

Discussion
The results of the present study demonstrated that MSC transplantation suppressed the in ltration of iNKT cells in the peri-infarct areas of C57BL/6 mice and that PGE 2 secreted by MSCs was involved.
Previous studies have demonstrated the protective effect of iNKT cells on the postinfarct heart (16-17). In the present study, activation of iNKT cells by α-GC before MSC transplantation further ameliorated ILV and PLV remodeling after MI in mice, accompanied by decreases in interstitial brosis and apoptosis. The present study provided direct evidence for the suppressive effects of MSCs on iNKT cells in vivo and that iNKT cell activation further improved the e cacy of MSCs in the treatment of post-MI hearts.
MI causes sterile in ammation, which is characterized by the recruitment and activation of immune cells of the innate and adaptive immune systems. The tasks of these in ammatory cells, including neutrophils, macrophages and lymphocytes, involve clearance of dead tissues, the reparative response and adverse remodeling (23)(24)(25)(26)(27). However, excessive chronic in ammation in the peri-infarct heart may cause further cell apoptosis and myocardial brosis, nally leading to impairment of cardiac function (28). The results of the present study are consistent with those of a previous study (16) in terms of iNKT cells (Vα14Jα18) in ltrating into the heart after MI. Previous studies have demonstrated that anti-in ammatory therapy attenuates LV dilation and contractile dysfunction, which are associated with a decrease in white blood cell in ltration (29,30). Chronic in ammation in the post-MI heart, particularly in peri-infarct areas, has a crucial role in cardiac remodeling and failure after MI, while the precise role of in ammatory cells and chemokines in this process remains to be fully elucidated. Furthermore, iNKT cells rapidly mediate various functions by producing a mixture of T helper type 1 (TH1) and TH2 cell cytokines and a vast array of chemokines (13). Thus, iNKT cells may function as a bridge between the innate and adaptive immune systems and orchestrate tissue in ammation. Sobirin  Emerging evidence suggests that progenitor cells have immunomodulatory properties and speci cally suppress the proliferation or activation of T cells (31). MSCs are considered to be one of the most promising progenitor cell types for therapeutic applications. In addition, MSCs have an immunoregulatory capacity and elicit immunosuppressive effects in various settings. These cells have immense plasticity coupled with their ability to modulate the activity of immune cells (12). Previous studies have indicated that in vitro-expanded MSCs exert broad-spectrum immunoregulatory functions on innate immune cells, including dendritic cells, NK cells and adaptive immune cells (10,31). Prigione et al. (8) previously demonstrated that MSCs expanded from human bone marrow abolished the in vitro proliferation of resting peripheral blood Vα24 + Vβ11 + cells through the release of PGE 2 . The present study demonstrated that MSCs had a suppressive effect on iNKT cells in postinfarct mouse hearts by releasing PGE 2 , with a reduction in iNKT cell in ltration in peri-infarct areas, which may be unbene cial or even harmful. While previous studies (32-34) concluded that certain other molecules released by MSCs might be involved in the modulation of T cells, the results of the present study indicated that iNOS and TGF-β1 did not have any role in this modulation.
In the present study, the inhibitory effect of MSCs on iNKT cells was abolished by the nonselective cyclooxygenase inhibitor indomethacin, which is an inhibitor of PGE 2 , while α-GC, which is well known to activate iNKT cells (35), signi cantly increased the in ltration of iNKT cells in the peri-infarct area. The iNKT cell-surface receptors, including T-cell receptor (TCR) and NK1.1, become downregulated following MI, which renders iNKT cells invisible by ow cytometric detection (36, 37). The downregulation of TCR remains until at least 24 h; subsequently, iNKT cells rapidly proliferate and increase to the peak level at 72 h after α-GC administration (36)(37)(38)(39)(40). The present study demonstrated that the proportion of iNKT cells (Vα14Jα18) increased within the heart at day 7 after α-GC administration and decreased at day 28 but remained higher than that in the sham group. In a model of experimental autoimmune encephalomyelitis, early immunization with α-GC protected against this disease, whereas later immunization potentiated it (41). In the present study, pretreatment with α-GC injection signi cantly enhanced iNKT cell in ltration, and the most important nding was that the activation of iNKT cells (Vα14Jα18) by α-GC prior to MSC transplantation further attenuated LV remodeling after MI, which further improved the e cacy of MSCs on post-MI hearts compared with MSC transplantation. In 2009, Burch eld et al. (42) suggested that bone marrow mononuclear cells mediated cardiac protection (decreased T lymphocyte accumulation, reactive hypertrophy and myocardial collagen deposition) after MI, partly dependent on interleukin (IL)-10. In 2011, Dayan et al. (43) demonstrated that the mechanism of MSC-mediated enhancement of cardiac function possibly proceeded via an IL-10-mediated switch from in ltration of proin ammatory to antiin ammatory macrophages at the infarct site. In addition, previous studies have indicated that the therapeutic effects of α-GC against TH1-like autoimmune diseases are mediated via a shift from a TH1 pattern toward a TH2 pattern and the induction of the immunosuppressive cytokine IL-10 (44,45). Sobirin et al. (16) reported that iNKT cells have a protective role against post-MI remodeling and failure, partly through the enhanced expression of IL-10. In 2016, Tsutsui found that an anti-IL-10 receptor antibody abrogated the protective effects of α-GC on MI remodeling, suggesting that NKT cells play a protective role against post-MI LV remodeling and failure through the enhanced expression of cardioprotective cytokines such as IL-10 (46). It may be speculated that the potential mechanisms of iNKT cell activation after MSC transplantation against LV remodeling are partially mediated via enhanced expression of IL-10, a decrease in apoptotic cardiomyocytes, myocardial collagen deposition and LV failure; however, further study of the precise mechanism is required.
Of note, the present study has several limitations that must be acknowledged. First, previous studies have indicated the involvement of PGE 2 , TGF-β1 and iNOS in various immunoregulatory activities of MSCs, and the present study only assessed these three factors, with the results indicating that TGF-β1 and iNOS do not have any involvement, whereas certain other molecules may be involved in the modulation of iNKT activities. Further study regarding this nding is required. Second, it is not possible to demonstrate the location of iNKT cells in postinfarct hearts in situ by immunohistochemical analysis or in situ hybridization. Further study of the realization of in situ detection is required to clarify this important issue.

Conclusion
The present study suggested that MSCs inhibited iNKT cell in ltration in post-MI mouse hearts, and activating iNKT cells enhanced the cardioprotective effect of MSCs and attenuated cardiac remodeling. Therapies designed to activate iNKT cells before MSC transplantation might be bene cial to enhance the effectiveness of MSCs in the post-MI heart. This approach may become a new direction of immune treatment for infarcted hearts.

Declarations
Ethics approval and consent to participate No human studies were carried out by the authors for this article. Ethical approval for all work was received from the animal research committee of Zhengzhou University (Zhengzhou, China).

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