The hiPSC cell line PBMC-5 (at passage 10) was donated by Dr Tiancheng Zhou. hiPSCs were prepared from peripheral blood mononuclear cells of a healthy Han Chinese female donor (age, 31 years) in CAS Key Laboratory of Regenerative Biology, Guangzhou Institutes of Biomedicine and Health (Guangzhou, China). Cells were separated using a Ficoll-Paque gradient (Shenzhen DAKEWE Bio-Engineering Co., Ltd.) and reprogramed using the CTS™ CytoTune™-iPS Sendai virus reprograming kit (Thermo Fisher Scientific, Inc.). The hiPSCs were maintained as large cell colonies on Matrigel-coated plates (BD Biosciences) in serum free mTeSR1 medium (STEMCELL Technologies) at 37˚C, in 5% CO₂.

Dispersed single hiPSC cells at passage 15 were seeded onto a Matrigel-coated 12-well plate in mTeSR1 medium at a density of 2x10⁴ cells/cm² 2 days before induction. Cells were treated with the following different MSC differentiation mediums for 6 days: i) DMEM basal medium supplemented with 10% knockout serum replacement (Invitrogen; Thermo Fisher Scientific, Inc.), 1 mM L-glutamine, 10 mM non-essential amino acids, 50 U/ml penicillin/streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.) and 0.5 µl/ml DMSO (MilliporeSigma; Merck KGaA); ii) DMEM containing 5 or 10 µM CHIR99021 (Stemgent, Inc.; REPROCELL), iii) DMEM containing 10 or 20 µg/ml TGF-β (Creative BioMart); or iv) DMEM medium containing 5 µM CHIR99021 and 10 µg/ml TGF-β. The selected concentrations of CHIR99021 and TGF-β used in the present study were selected on the basis of previous reports (20,30-33) and the induction medium was changed every other day.

At 6 days after cell induction, the differentiated cells were dispersed into single cells through TrypleSelect (Invitrogen; Thermo Fisher Scientific, Inc.) digestion at 37˚C for 5 min and collected by centrifugation at 500 x g for 5 min. The collected cells were suspended in MSC medium [DMEM basal medium supplemented with 10% FBS (HyClone; Cytiva), 1 mM L-glutamine, 50 U/ml penicillin/streptomycin and 10 mM non-essential amino acids (Invitrogen; Themo Fisher Scientific, Inc.)] and seeded at a density of 4x10⁴ cells/cm². In subsequent passages, the cell concentration was reduced to 2x10⁴ cells/cm². The cell induction experiment was repeated thrice.

Total RNA isolation from the induced cells was conducted using the RNase mini kit (Qiagen, Inc.) and a high-capacity RNA-to-cDNA kit (Applied Biosystems; Thermo Fisher Scientific, Inc.) was used to reverse transcribe RNA into cDNA according to the manufacturer's instructions (incubation at 85˚C for 5 sec and 37˚C for 1 h). For quantification of gene expression, RT-qPCR was performed using specific primers (Table I) and SYBR-Green RT-PCR reagents (Applied Biosystems; Thermo Fisher Scientific, Inc.) under the following thermocycling conditions: denaturation at 94˚C for 30 sec; annealing at 60˚C for 35 sec and extension at 72˚C for 40 sec, 30 cycles. The 2^-ΔΔCq method was used to analyse relative gene expression levels and GAPDH was used as the normalisation control (34). The quantification of gene transcription was repeated three times.

The EpiQuik whole cell extraction kit (AmyJet Scientific, Inc.) was used to extract total protein from the induced cells and protein concentration was measured using the Bradford DC protein assay (Bio-Rad Laboratories). Protein samples (30 µg) were separated by electrophoresis on 12% Bis-Tris polyacrylamide gel and then transferred to nitrocellulose membranes (MilliporeSigma; Merck KGaA). The membranes were blocked with 5% BSA (Wuhan Servicebio Technology Co., Ltd.) at room temperature for 1.5 h and incubated overnight at 4˚C with the following primary antibodies: SMAD 2/3 (cat. no. #3102S), ERK1/2 (cat. no. #4695T) (Cell Signalling Technology, Inc.), GSK-3 (cat. no. #PA532440; Thermo Fisher Scientific, Inc.) and SNAIL (cat. no. #ABD38; MilliporeSigma) primary antibodies at 1:1,000 dilution at 4˚C for 12 h, followed by incubation with goat anti-rabbit polyclonal HRP conjugated secondary antibodies (1:10,000 dilution in 1% milk/TBS; cat. no. #A16110; Thermo Fisher Scientific, Inc.) for 2 h at room temperature. Protein bands were visualized using a chemiluminescence detection kit (Amersham; Cytiva) and analyzed using Tanon 5220S image analysis system (Tanon Image software version 1.0; Tanon Science and Technology Co., Ltd.). Analysis of protein expression levels was repeated thrice.

At six passages after the initial MSC induction from hiPSCs, the adherent cells with a spindle-like morphology were collected by 0.25% trypsin treatment at 37˚C for 5 min and centrifugation at 500 x g for 5 min. Then the cells were resuspended in PBS at 1x10⁶ cells/ml for determining the expression levels of typical MSC surface markers (CD29, CD44, CD73, CD90 and CD105) and hematopoietic cell lineages markers [CD34, CD45 and human leukocyte antigen-DR isotype (HLA-DR)] by fluorescence-activated cell sorting. The antibodies used were as follows: fluorescent conjugated anti-CD29, -CD44, -CD73, -CD90 and -CD105 antibodies (cat. nos. #354603, #338807, #344003, #328109 and #323205, respectively; BioLegend, Inc.); primary anti-CD45, -CD34, -HLA-DR antibody and CY3 conjugated goat anti-rabbit IgG secondary antibody (cat. nos. #A0055-3, #BM4082, #BM4546 and #BA1032, respectively; Wuhan Boster Biological Technology, Ltd.). The primary antibodies (1:200 dilution) and secondary antibody (1:5,000 dilution) were incubated respectively with the cells for 30 min at room temperature, and then the cell surface marker expression was measured by BD-FACScan (Becton, Dickinson and Company), and the data were analysed using FlowJo (V10; FlowJo LLC). Measurements were repeated three times.

The induction of osteogenic, chondrogenic and adipocytic differentiation and subsequent evaluation of the induced cells was conducted as previously described (33). Osteogenic differentiation was initiated by culturing cells with osteo-inductive medium [DMEM basal medium supplemented with 10% FBS (HyClone; Cytiva), 10 mM glycerol phosphate, 50 µM ascorbate phosphate, 10^-7 M dexamethasone (MilliporeSigma; Merck KGaA) and 100 ng/ml recombinant human bone morphogenic protein-2 (Invitrogen; Thermo Fisher Scientific, Inc.)] for 2 weeks. For chondrogenic differentiation, cells were cultured in chondrogenic induction medium [DMEM basal medium supplemented with 10% FBS, 6.25 µg/ml of insulin (Fisher Scientific, Thermo Fisher Scientific, Inc.), 50 nM of ascorbate phosphate and 10 ng/ml of TGF-β] for 2 weeks. The adipocytic differentiation potential was analysed using adipogenic induction medium (DMEM supplemented with 10% FBS, 10^-7 M of dexamethasone and 10 µg/ml of insulin) to culture the cells for ≥2 weeks.

The cells growing on 60 mm culture dish (Corning) were fixed by 4% formaldehyde for 10 min at room temperature, and then stained respectively by 0.1% Alizarin red, 1% Alcian blue and 0.5% Oil-Red-O at room temperature for 30 min to visualize calcium nodule formation, sulphated proteoglycan formation and intracellular lipid accumulation in the cells under light microscope at magnification x200 (Olympus).

A total of 30 male Wister rats (age, 4 weeks; weight, 230-270 g) were purchased from The Medical Experimental Animal Centre of Guangdong Province. The rats were housed under constant 20˚C temperature, 45% humidity and a 12 h light-dark cycle and had free access to standard laboratory chow and tap water.

Approval for the animal experimentation was granted by The Animal Ethics Committee of The Peking University & Hong Kong Science and Technology University Medical Centre (approval no. 2020-027). Chinese laws relating to the care and treatment of animals and the Animal Research: Reporting of In Vivo Experiments guidelines were followed throughout the study (35).

The rats were randomly separated into MSC injection group, PBS injection group and sham group, with 10 rats in each group. In order to construct the AMI model, rats were first anaesthetised by pentobarbital sodium (40 mg/kg) intraperitoneal injection and then the heart was exposed via a left thoracotomy between the 4th and 5th ribs of the animal. A tight suture was made around the left anterior descending coronary artery and the resulted ischemia was maintained for 40 min and then the suture was removed to restore blood flow. In the sham control group, the suture was immediately removed without any further maintaining.

Before sacrifice, the rats were administered an IP injection of pentobarbital sodium (40 mg/kg). When the rats were anaesthetised and unresponsive to stimuli, cervical dislocation was used to sacrifice each animal. In order to confirm death, the absence of breath sounds and heart beat was monitored by observation and echocardiography.

Immediately after AMI, model rats were divided randomly into hiPSC-MSC injection group (n=10) and PBS injection control groups (n=10). An IP injection of 1x10⁷ hiPSC-MSCs (passage 6 after the initial MSC induction from hiPSC) was administered to rats in the hiPSC-MSC treated group each week for 4 weeks, with PBS injections used as a control.

To determine the distribution of the IP injected hiPSC-MSCs, 1x10⁷ hiPSC-MSCs were labelled with Xeno Light Dir (PerkinElmer, Inc.) and injected into the peritoneal cavity of rats (n=3). The signals of the injected cells were detected using the Lumina IVIS II System (PerkinElmer, Inc.) and the fluorescence images of the peritoneal cavity and individual colicomenta were obtained using the Living Image Software (version 4.1; PerkinElmer, Inc.) immediately after the injection, 3 days after the injection, 1 week after the injection and 3 weeks after the injection.

For in vivo cytokine assay analysis, 1 week after the IP cell injection, 10 ml of PBS was injected into the peritoneal cavity of the anesthetised (40 mg/kg pentobarbital sodium) hiPSC-MSC injected (n=3) and control rats (n=3) with a 10-gauge syringe and the peritoneal lavage was collected. The concentrations of insulin-like growth factor (IGF), VEGF, hepatocyte growth factor (HGF), prostaglandin E2 (PGE2), IL-4 and IL-10 in the lavage were measured using ELISA kits following the manufacturer's instructions. The kits used were as follows: Human IGF-1 ELISA kit (cat. no. #EK1131-48), PGE2 Competitive ELISA kit (cat. no. #EK8103/2-48) (Hangzhou Lianke Biotechnology Co., Ltd.), human VEGF ELISA Kit (cat. no. #ab100663), hepatocyte growth factor ELISA kit (cat. no. #ab275901) (Abcam), human IL-10 ELISA kit (cat. no. KIT10947A), human IL-4 ELISA kit (cat. no. KIT11846) (Sinopharm Chemical Reagent Co., Ltd.). A 1510 spectrophotometer (Thermo Fisher Scientific, Inc.) was used to analyse the samples and the measurements were repeated three times.

A 2D echocardiography system (Vevo 2100; Visual Sonics) with a MS250 transducer (13-24 MHz) was used to evaluate the cardiac function of the AMI model rats. The parameter changes pertaining to left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS) were measured and analysed.

The area of myocardial scars of the AMI model rats was assessed using Masson's trichrome staining 11 weeks after AMI induction. Briefly, tissue sections of the infarcted area were fixed in 4% formaldehyde, embedded in paraffin and cut into 5 µm slices. The slices were stained with Masson's trichrome to detect fibrosis and three tissue slices taken from each rat were assessed by a pathologist who was blinded to the samples. The scar area percentage of total tissue area was quantified using a computerised morphometry system with a light microscope (Cell Sens; Ver.1.16; Olympus Corporation).

Heart tissue was fixed with 4% paraformaldehyde (Beyotime Institute of Biotechnology) for 72 h at room temperature, embedded in paraffin and cut into sections (4 µm). Tissue slices (n=3) from each rat were deparaffinized, rehydrated and pressure-cooked at 121˚C for 4 min in citrate buffer (10 mM, pH 6.0) for antigen retrieval. The sections were incubated in 1% H₂O₂/methanol at room temperature for 10 min and then blocked in 4% bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.) for 30 min at room temperature. Immunohistochemistry detection of α-SMA expressing cells was conducted with primary mouse anti-rat α-SMA antibodies (1:50 dilution; incubation 2 h at 4˚C; cat. no. #614852; BioLegend, Inc.) and secondary HRP conjugated anti-mouse IgG antibodies (1:100 dilution, incubation 30 min at room temperature, Wuhan Boster Biological Technology, Ltd. Cat.# BA1051). α-SMA positive capillaries in the infarcted regions were manually counted under light microscope using high-power fields (hpf) of magnification (x200) in each tissue section.

To quantify the extent of cardiomyocyte apoptosis, a TUNEL kit (Nanjing KeyGen BioTech Co., Ltd.) was used to stain apoptotic cells present in heart tissue sections (prepared the same way as in the scar size quantification) according to the manufacturer's recommendations. 0.5% Hematoxylin counterstaining was conducted to better visualize normal nuclei. Apoptotic cells were counted in 5 randomly selected hpf of magnification (x400) in the infarcted area of three tissue sections from each rat and the mean count calculated.

Data were presented as mean ± standard deviation and analysed using GraphPad Prism (version 5; GraphPad; Dotmatics). A one-way ANOVA with Dunnett's test was used for multiple comparisons and an unpaired t-test was used for comparisons between two groups. P<0.05 was considered to indicate a statistically significant difference.

To examine whether CHIR99021 or TGF-β treatment could initiate mesoderm commitment, hiPSCs were treated with CHIR99021, TGF-β or a CHIR99021 + TGF-β combination for 6 days. The mRNA expression levels of EMT-related genes N-cadherin and Vimentin, mesenchymal to epithelial transition (MET)-related gene E-cadherin and pluripotency-related transcription factors OCT4, SOX2 and NANOG were evaluated using RT-qPCR (Fig. 1A).

After treatment with CHIR99021, TGF-β or the CHIR99021 + TGF-β combination for 6 days, compared with the untreated control cells, the mRNA expression levels of EMT related genes N-cadherin and Vimentin in the CHIR99021 or TGF-β treated cells was markedly upregulated and significant upregulation was demonstrated in the 5 µM CHIR99021 +10 µg/ml TGF-β treated cells (N-cadherin, P<0.05; Vimentin, P<0.01). A decrease in the cell growth rate and apoptosis were observed when the concentration of CHIR99021 and TGF-β was increased to 10 µM and 20 µg/ml, respectively (data not shown). As a result, the combination of 5 µM CHIR99021 +10 µg/ml TGF-β was used in the subsequent experiments. The mRNA expression levels of OCT4, SOX2, NANOG and E-cadherin in the CHIR99021 or TGF-β treated cells were not significantly altered compared with that of the untreated control cells.

To further investigate the potential mechanism involved in the CHIR99021 and TGF-β induced EMT of hiPSCs, analysis of SNAIL, SMAD2/3, ERK1/2 and GSK-3 protein expression levels was conducted by western blotting in the control, CHIR99021, TGF-β and CHIR99021 + TGF-β treated hiPSCs (Fig. 1B and C). These results demonstrated that after 6 days of induction, the protein expression levels of SNAIL and ERK1/2 in the 5 µM CHIR99021 +10 µg/ml TGF-β treated cells were significantly increased compared with that of the DMEM or single reagent treated cells. There was no significant difference in the protein expression levels of SMAD2/3 in control and treated cells. Furthermore, CHIR99021 + TGF-β treatment significantly increased the GSK-3 protein expression levels compared with the control and single reagent treatments.

The morphology of hiPSC-MSCs was analysed (Fig. 2A). At passage 3, the spontaneously differentiating cells mainly had a pleomorphic morphology. In the cells treated with 10 or 20 µg/ml TGF-β, <20% of cells exhibited a short spindle and olive-shaped morphology. In the cells treated with 5 µM CHIR99021, >30% of the cells exhibited a short spindle shape, whereas in the 10 µM CHIR99021 treated cells, the percentage of spindle shaped cells was close to 50%. In cells treated with 5 µM CHIR99021 +10 µg/ml TGF-β, ≥70% of the adherent cells exhibited a spindle-shaped morphology.

MSC surface marker expression and osteogenic, chondrogenic and adipogenic differentiation analysis was conducted on the cells at passage 6. MSC surface marker expression analysis demonstrated that >90% of cells expressed typical MSC surface markers (CD29, CD44, CD73, CD90 and CD105), while the percentage of cells expressing hematopoietic cell lineage markers (CD34, CD45 and HLA-DR) was ≤0.6% (Fig. 2B).

After 2 weeks osteogenic, chondrogenic and adipogenic differentiation, alizarin red staining positive calcium nodules and alcian blue staining positive sulphated proteoglycans were detected in osteogenic and chondrogenic differentiating cells respectively. In adipogenic differentiated cells, intercellular lipid vacuoles were identified by Oil-Red-O staining. No typical positive stained cell was found among the three corresponding un-induced control cells (Fig. 2C).

The distribution of hiPSC-MSCs was monitored after IP injection (Fig. 3A). At 1 h after the labelled hiPSC-MSCs injection, the fluorescent signal was detected around the injection site and 3 days after cell injection, the fluorescent signal moved from the injection site to colicomentum. Fluorescent signals could be detected for up to 3 weeks on the colicomentum and no fluorescent signals were detected in the liver, spleen or kidneys. There was no obvious evidence of an associated inflammatory response or tumour formation in the peritoneal cavity.

Angiogenic and immune regulatory factors were measured 1 week after hiPSC-MSC injections using the peritoneal lavage of hiPSC-MSC injected rats and control rats (Fig. 3B). The levels of IGF, VEGF, HGF, PGE2, IL-4 and IL-10 in the lavage of hiPSC-MSC injected rats were significantly increased compared with that in the control rats.

Cardiac function was monitored by measuring LVEF and LVFS at 2-week intervals from the initial hiPSC-MSC injection to the end of the experiment. Representative echocardiogram images taken 11 weeks after AMI was induced demonstrated that the parameters reflecting the structure and function of heart such as anterior and posterior wall thickness, ventricular dimension enlargement and ventricular contraction of the sham group remained normal (Fig. 4A-a). In the PBS injection group, anterior and posterior wall thickness, ventricular dimension enlargement and ventricular contraction decrease were demonstrated (Fig. 4A-b), whereas in the hiPSC-MSC injection group, anterior and posterior wall thickness, ventricular dimension enlargement and ventricular contraction decrease were improved compared with that of the PBS injection group (Fig. 4A-c).

Following AMI induction, the LVEF of the sham group was stable from 74.50±1.46% at week 1 to 75.19±1.93% at week 11 (Fig. 4B-a). In the PBS injection control group, LVEF declined from 44.14±4.02% at week 1 to 41.35±4.16% at week 11. In the cell injection group, LVEF increased from 46.94±2.89% at week 1 to 57.70±6.76% at week 11. A significant increase in LVEF was demonstrated when comparing the cell injection group with the PBS control group at the 11-week time point (Fig. 4B-b).

The LVFS of the sham group was stable from 46.30±4.38% at week 1 to 44±2.10% at week 11 (Fig. 4C-a). In the PBS injection control group, LVFS declined from 30.35±6.23% in week 1 to 21.27±4.73% in week 11. In the cell injection group, LVFS increased from 28.68±6.73% in week 1 to 34.45±5.79% in week 11. A significant increase was demonstrated in the LVFS of the cell injection group compared with the PBS control group at week 11 (Fig. 4C-b).

Over the 11-week experiment, four rats died in the PBS injection control group (death rate, 40%) and two rats died in the cell injection group (death rate, 20%).

Histological analysis was conducted 11 weeks after AMI to evaluate the percentage of scar tissue formation in the infarct area (Fig. 5A). The scar area in the PBS control group was 5.25±0.96%, which was significantly higher compared with the scar area in the cell injection group (3.33±0.88%) (Fig. 5B).

Immunohistology staining demonstrated that hiPSC-MSC injections significantly increased the number of α-SMA positive arteries in the infarcted area of the hiPSC-MSC injection group (9.42±2.45) compared with that of the PBS injection group (5.26±1.84) (Fig. 6A and B).

The results of the TUNEL assay demonstrated that the apoptotic cell number in the infarcted area was significantly reduced in the cell injection group (55.5±11.98 cells/hpf) compared with the PBS injection control group (87.26±19.68 cells/hpf) (Fig. 7A and B).