Role of HIF-1α-Activated IL-22/IL-22R1/Bmi1 Signaling Modulates the Self-Renewal of Cardiac Stem Cells in Acute Myocardial Ischemia

Abstract

Impaired tissue regeneration negatively impacts on left ventricular (LV) function and remodeling after acute myocardial infarction (AMI). Little is known about the intrinsic regulatory machinery of ischemia-induced endogenous cardiac stem cells (eCSCs) self-renewing divisions after AMI. The interleukin 22 (IL-22)/IL-22 receptor 1 (IL-22R1) pathway has emerged as an important regulator of several cellular processes, including the self-renewal and proliferation of stem cells. However, whether the hypoxic environment could trigger the self-renewal of eCSCs via IL-22/IL-22R1 activation remains unknown. In this study, the upregulation of IL-22R1 occurred due to activation of hypoxia-inducible factor-1α (HIF-1α) under hypoxic and ischemic conditions. Systemic IL-22 administration not only attenuated cardiac remodeling, inflammatory responses, but also promoted eCSC-mediated cardiac repair after AMI. Unbiased RNA microarray analysis showed that the downstream mediator Bmi1 regulated the activation of CSCs. Therefore, the HIF-1α-induced IL-22/IL-22R1/Bmi1 cascade can modulate the proliferation and activation of eCSCs in vitro and in vivo. Collectively, investigating the HIF-1α-activated IL-22/IL-22R1/Bmi1 signaling pathway might offer a new therapeutic strategy for AMI via eCSC-induced cardiac repair.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s12015-024-10774-8.

Introduction

Acute myocardial ischemia (AMI) leads to cardiac remodeling and cardiomyocyte necrosis [66], which have emerged as unmet medical needs for the introduction of cardiac tissue repair after AMI [32]. Cardiac stem cells (CSCs), the promising cells in regenerative therapy, primarily act through paracrine mechanisms in post-AMI patients or hypoplastic left ventricle (LV) or dilated cardiomyopathy [55]. One type of CSCs population has been reported to display a mesenchymal cell-surface profile (CD34⁻ and CD105⁺) [58], and it has been shown to express Bmi1 [76], capable of differentiating into cardiomyocytes, endothelial, and smooth muscle-like cells. When CSCs were injected into the infarcted myocardium of mice, rats, or pigs, the transplanted cells exerted the recovery of heart function [26, 37, 43, 52, 53, 88]. Their benefits include anti-fibrotic [89], anti-apoptotic [45], angiogenic [89], and anti-inflammatory effects [3]. With over 70 labs worldwide [2, 23, 59, 84, 101] and ten completed clinical trials [56], attest to their safety and potential disease-modifying effects in humans [50, 51, 54].

A common method in the field involves the ex vivo expansion of resident CSCs, followed by their delivery to the heart, with cardiosphere (CS) being frequently utilized [75]. Although CS exhibit readily isolated properties and can be expanded ex vivo [75], the time-consuming nature of their isolation and expansion process for clinical implantation compromises their therapeutic benefits, potentially missing the optimal repair window for patients in need. Therefore, in addition to implantation, the induction of endogenous CSC (eCSC) activities could be a more efficient and protective strategy to improve therapeutic benefits by shorting the time for CSCs to initiate repair after AMI.

Interleukin 22 (IL-22), belonging to the IL-10 cytokine family, interacts with the IL-22 receptor complex, including IL-22 receptor alpha 1 (IL-22R1 or IL-22RA1), to exert anti-inflammatory, tissue homeostatic, cell proliferative, and cytoprotective effects [47, 78, 79, 86, 98] through the cell surface receptors IL-10R2 and IL-22R1 [16, 40]. To date, evidence suggests that IL-22 promotes the proliferation and survival of several tissue stem cells. For example, the IL-22/IL-22R1 cascade exerts cytoprotective and anti-inflammatory effects to provide a signal for the survival and proliferation of intestinal stem cells and protects intestinal epithelium completeness during inflammatory bowel disease and experimental colitis [27, 78]. Increased expression of IL-22 is correlated with increased proliferation of liver progenitor cells (LPCs). IL-22 binding to IL-22R1 activates the signal transducer of STAT3 to regulate the proliferation and survival of LPCs to protect mice against various types of liver injuries by directly attenuating hepatocyte necrosis [21, 102]. Further, studies have shown that the IL-22/IL-22R1 pathway acts as a myocardium-protective cytokine in patients with viral myocarditis [39]. Based on the findings of these studies, we hypothesize that the IL-22/IL-22R1 axis exerts its myocardial repair function by enhancing eCSC activation after AMI.

Recent studies have reported that IL-22 can ameliorate AMI [12, 82, 86] by activating the IL-22R1/STAT3 signaling pathway [82], blocking the decrease in mitochondrial membrane potential, and inhibiting ROS production as well as cytochrome C release [12] after ischemia–reperfusion injury. However, the regulatory effects of hypoxia on IL-22/IL-22R1 expression remains unknown. Previous studies have reported increased expression of IL-10 and IL-20, which is regulated by hypoxia-inducible factor-1α (HIF-1α) activation, in patients with AMI [9, 13, 22]. Because IL-22 is a member of the IL-10 family, we hypothesize that AMI induces the upregulation of IL-22 and IL-22R1 by activating HIF-1α in the myocardium.

This study investigated whether HIF-1α activation could activate the downstream biological response of the IL-22/IL-22R1 signaling pathway, with particularly focus on the self-renewal ability of eCSCs to attenuate left ventricular (LV) dysfunction, remodeling, and inflammation after AMI. RNA microarray analysis was performed to identify the candidate gene(s), such as Bmi1, involved in IL-22/IL-22R1 signaling, along with bioinformatics analysis. We speculated that the HIF-1α-activated IL-22/IL-22R1/Bmi1 cascade in eCSCs might offer a new therapeutic strategy for AMI via eCSC-induced cardiac repair. Such global and comprehensive investigations on eCSC activities for cardiac repair after AMI are urgently warranted for improving the clinical efficacy of AMI treatment strategies.

Methods

Animal Care and Use

Both 6–8 week-old Sprague Dawley rats and wild type 6–8 week-old C57BL/6 mice were purchased from the National Laboratory Animal Center (Taiwan). The Institutional Ethical Committee for Animal Research in China Medical University reviewed and approved all animal experiments (CMUIACUC-2021-394). All animal work was performed according to protocols approved by the Institutional Animal Care and Use Committee (IACUC). For knockout (KO) mice, the colony of each mouse line was maintained in the animal facility of the China Medical University, Taiwan, according to the Institutional Animal Care guidelines.

HIF-1α Knockout Mouse Lines

HIF-1α conditional KO mice (HIF‐1α^–/–), in C57BL/6 background carrying a loxP-flanked allele of HIF-1α, were induced by feeding doxycycline at a dose of 2 mg/ml in 5% (wt/vol) sucrose solution from embryonic day 15 to postnatal day 1 [44, 69]. No non-inclusion or exclusion parameters were used in our studies.

IL-22R1 Knockout Mouse Lines

For generations of KO mice, according to the design principle and program of CRISPR/Cas9, we designed two target sgRNAs (IL22ra1 5’-1sgRNA and IL22ra1 3’-1sgRNA, Transgenic Mouse model Core Facility, Taiwan) that targeted exons 2–4 of IL-22R1 in C57BL/6 background mouse genomes for deleting the mouse IL-22R1 gene, as previously described [96]. To detect off-target mutations, genome-wide unbiased identification of double stranded breaks (DSBs; Genome-wide Unbiased Identifications of DSBs Evaluated by Sequencing, Joung Lab, USA) based on global capturing of DSBs introduced by RNA-guided endonucleases was applied to enable genome-wide profiling of off-target cleavage by CRISPR-Cas9 nucleases (Thermo Fisher Scientific, USA) and mitigate off-target effects [87]. Homozygote normal littermates and heterozygote newborns (IL-22R1^+/–) were genotyped through polymerase chain reaction (PCR) using primers [96]. IL-22R1^+/– mice were crossed with each other to yield IL-22R1^–/– animals. No non-inclusion or exclusion parameters were used in our studies.

Bmi1 Knockout Mouse Lines

Mice heterozygous of Bmi1^+/– for the null mutation (Bmi1^–/–) in 129/Ola background were a kind gift from Dr. Van Lohuizen [6, 62]. Timed embryos were obtained by examining for vaginal plugs in mated females. The morning of plug detection was designated E0.5. The homozygote normal littermate and heterozygote newborns were genotyped according to the procedure established by PCR using primers as described [62]. No non-inclusion or exclusion parameters were used in our studies.

Animal Model of Acute Myocardial Infarction (AMI)

Adult male Sprague-Dawley rats (SD, 200–250 g) or C57BL/6 mice (25–35 g) were subjected to AMI by ligation of left anterior descending (LAD) coronary artery as described previously [35]. In brief, after induction of anaesthesia with 2% isoflurane in 100% oxygen, rats or mice received artificial ventilation using a respirator (SN-480-7, Japan) with a tidal volume of 1mL/100 g and respiratory rate 80/min. A left thoracotomy was performed in the 4-5th intercostal space using a rib retractor (MY-9454 S, Japan). An intra-myocardial ligature was placed 1–2 mm below the atrioventricular groove using a 6-O polyethylene suture needle with thread (Ethicon, UK) before the thorax was closed. Sham rats or mice underwent the same protocol with the exception of the ligation of the coronary artery. After AMI at certain days, IL-22 (R&D, USA) and 2-Methoxyestradiol (2-ME2, Thermo Fisher Scientific, USA) will be injected by subcutaneously and intraperitoneal injection, respectively.

Rat Model of IL-22 or rIL-22BP-Fc Treatment Protocol

The rats with AMI were subdivided into 4 treated groups and injected subcutaneously with mouse recombinant IL-22 (R&D, USA) of 0.5 µg/kg, 5 µg/kg, 50 µg/kg or saline control group on 1 day post-AMI [41]. The rats in the sham group underwent the same procedure except for the LAD coronary artery ligation. Experimental design included that inflammatory response was assessed after 3 days, LV functional changes were assessed on 28 days, and structural remodeling was assessed at 28 days post-AMI. In the blocking experiments, rIL-22BP-Fc (10 µg/kg, i.p., R&D) was administered 4 h before the treatment [97].

Primary Culture of Cardiac Cells (PCC)

Primary cultures of mouse fetuses (E12.5) cardiac cells (PCCs) were prepared as previously described [8]. In brief, the hearts were removed and washed three times with Ca²⁺ and Mg²⁺-free PBS (D-PBS, Thermo Fisher Scientific, USA). They were mechanically cut into pieces smaller than 0.5 cm³, treated with collagenase type II (Gibco, Thermo Fisher Scientific) and 0.05% trypsin-EDTA (Gibco, Thermo Fisher Scientific) incubated for 5 min at 37 °C in a 95% air/5% CO₂ humidified atmosphere. The dissociated cells were pre-plated for 1 h to enrich the culture with myocytes and loose attached cells. The nonadherent cells were then plated at a density of 1200 cells/mm² in plating medium consisting of DMEM (Gibco, Thermo Fisher Scientific) supplemented with 5% FBS, penicillin (100 U/mL), streptomycin (100 µg/mL), and 2 mg/mL vitamin B12. The cardiac myocytes were detected by immunostaining with α-sarcomeric actin antibody (α-SA) (all from Thermo Fisher Scientific).

Isolation of Cardiospheres and Self-Renewal Assay

The collected isolated myocardial tissues form mouse fetuses (E12.5) were prepared as previously described with modification [14, 75, 83]. In brief, the hearts were washed three times with PBS. They were mechanically cut by scissors cut extensively into pieces smaller than 0.5 cm³, treated with collagenase type II (Gibco, Thermo Fisher Scientific, USA) and 0.05% trypsin-EDTA (Gibco, Thermo Fisher Scientific) incubated for 5 min hour at 37 °C in a 95% air/5% CO₂ humidified atmosphere. The explants then were cultured in complete explant medium (CEM) (DMEM containing 10% fetal calf serum (FCS), 100 U/mL penicillin G, 100 µg/mL streptomycin, 2 mmol/L L-glutamine, and 0.1 mmol/L 2-mercaptoethanol) at 37 °C with a 95% air/5% CO₂ humidified atmosphere. They were left undisturbed for 7–10 days to allow for migration of the cells from the explants and a layer of fibroblast-like cells was generated from adherent explants over which small, phase-bright cells migrated. These loose attached phase-bright cells obtained were seeded at 5 × 10⁴ cells/mL in ultra-low attachment plates (Corning, Merck, Germany) in cardiosphere growth medium (CGM) (35% complete IMDM/65% DMEM–Ham F-12 mix containing 2% B27, 0.1 mmol/L 2-mercaptoethanol, 10 ng/mL epidermal growth factor (EGF), 20 ng/mL basic fibroblast growth factor (bFGF), 40 nmol/L cardiotrophin-1, 40 nmol/L thrombin, and antibiotics) to form primary cardiospheres (CSs) by sphere formation. The frequency of primary CSs formation was calculated by dividing the number of CSs by the number of total 1 × 10⁴ seeding cells per 24-well each. The CS diameters were measured using a bright-field with confocal microscopy (Zeiss LSM510) and image analysis software (minimum diameter > 50 μm, the images of five fields per well were obtained; Microcomputer Imaging Device Program, Canada). To form secondary CSs, primary CSs were dissociated into single cells by 0.05% trypsin/EDTA and re-seeded in ultra-low attachment plates with CGM. We quantified self-renewal potential by dividing the number of secondary CSs by the number of total 1 × 10⁴ single-seeding cells dissociated from primary CSs.

Experimental Treatments in Cardiosphere Cultures

CSs were exposed to the various dosage of IL-22 (1, 10 and 100 ng/mL) [77] as indicated in detail in the figure legends and in the text.

Hypoxia Procedure

PCC or CSs cultured at 37 °C in 5% CO₂-humidified incubators were treated in normoxic (21% O₂) or various hypoxic conditions (1%, 3% and 5% O₂) for different time point as previously described [33]. Hypoxic cultures were cultivated in a two-gas incubator (Jouan SA, France) equipped with an O₂ probe to regulate N₂ gas levels. Cell number and viability were evaluated using trypan blue exclusion assay.

2-Methoxyestradiol Treatment in Vitro and in Vivo

2-Methoxyestradiol (2-ME2, Merck, Germany) was dissolved in DMSO to obtain a 10 mmol/L stock solution. For in vivo experiments, the whole procedure was performed as previously described [67]. Experimental rats were treated with an intraperitoneal injection of a liposomal preparation (di-oleoyl-phosphotidylcholine; Avanti Polar Lipids, USA) of 2-ME2 (20 mg/mL) in a different concentrations (50, 100 or 150 mg/kg) for 5–10 consecutive days pre- and after the onset of AMI. For in vitro experiments with 2-ME2 treatment, PCCs were pretreated with different concentrations of 2-ME2 (0.1 µM, 1 µM and 10 µM) for 16 h as previously described [15].

Lentiviral Constructs of IL-22R1, HIF-1α and HIF-2α, and Transduction in Vitro

The lentiviral constructs were generated by co-transfection of human kidney derived 293T cells with three plasmids using the calcium phosphate method as previously described with modification [70]. In the transducing vector, an expression cassette with the Rev responsive element and the EF-1α promoter are used to direct the expression of IL-22R1 (IL-22R1 cDNA, NP 067081, OriGene, USA), mouse HIF-1α and HIF-2α [49] and GFP (GFP cDNA; Clontech, TaKaRa Bio, Japan). Lentiviral vector particles were generated by transient co-transfection of 293T cells with the lentiviral shuttle plasmid from TRIP GFP plasmid vector [103], an HIV-1-derived packaging plasmid, and a VSV-G envelope expressing plasmid. Two days after transfection, lentiviral constructs (LV-IL-22R1, LV-HIF-1α, LV-HIF-2α or LV-GFP) were harvested in the culture medium and concentrated by ultra-centrifugation. Viral titers were quantified by using HIV-1 p24 antigen assay (Beckman Coulter, USA) according to the manufacturer’s instructions. The p24 concentration was used to determine the vector dose (expressed in nanograms) administered in the various in vitro and in vivo experiments. The lentiviral titers were determined by infection of 293T cells seeded in six-well plates at 1 × 10⁵ cells per well the day before infection with serial dilution of the concentrated viral stock. After overnight incubation, the culture medium was changed and the cells incubated for two more days. GFP fluorescent cells were identified by a fluorescent activated cell sorter. Titers ranged from 10⁸ to10⁹ infectious units (IU)/mL.

In in vitro lentiviral vectors transduction, PCCs was plated in 10-cm dishes with 5 ml medium per dish. Transductions were carried out in the presence of 8 µg/ml polybrene at m.o.i. of 25 for all vectors. After incubation for 24 h, the transduction medium was replaced with fresh original medium for each cell. Western blot was performed to assess for transgene expression after lentiviral vector administration by quantification of IL-22R1, HIF-1α and HIF-2α production in vitro.

Gene Silencing with RNA Interference

Specific knockdown was achieved by lentiviral delivery of shRNA for IL-22 (LV-IL-22-sh; sc-39664-V, Santa Cruz, USA), shRNA for IL-22R1 (LV-IL-22R1-sh; sc-88174-V, Santa Cruz), shRNA for HIF-1α (LV-HIF-1α-sh; sc-35562-V, Santa Cruz), shRNA for HIF-2α (LV-HIF-2α-sh; sc-35316-V, Santa Cruz), and the control shRNA (LV-control-sh; sc-108080, Santa Cruz) under manufacture’s instruction.

Immunocytochemistry

The CSs or PCCs were plated on glass chamber slides (BD Falcon, USA) for 2 days and fixed with 4% PFA for 20 min. Cells were permeabilized using 0.1% Triton X-100, immunostained with the primary antibody including HIF-1α (1:200; Millipore, Merck, Germany), IL-22R1 (1:200; Millipore), α-sarcomeric actin (α-SA, Merck, Germany), CD105, and NKX2.5 (R&D Systems, USA); c-kit (AbD Serotec, UK); Bmi1 (Millipore) overnight at 4 °C, washed 3 times and then incubated with a Alexa Fluor 488- or 594-conjugated secondary antibody (Thermo Fisher Scientific, USA) for 1 h at 37 °C, which was visualized with confocal microscopy (Zeiss LSM510, Germany).

Chromatin Immunoprecipitation (ChIP) Assay

PCCs subjected to 4-hours hypoxia (3% O₂) were fixed with 1% formaldehyde (added directly to the culture medium) for 20 min at 37˚C to allow for reversible cross-linkage [65]. The binding of HIF-1α to the promoter of IL-22R1 (NCBI Accession number: NC_000001.11) was examined using a commercial kit for ChIP assay (Upstate Biotechnology, Millipore, Merck, USA) in accordance to the manufacturer’s protocol with minor modifications. DNA–protein complexes were immunoprecipitated with primary antibody against HIF-1α linked to protein A agarose beads, and eluted with 1% SDS, and 0.1 M NaHCO₃. The cross-links were reversed by incubation at 65⁰C for 5 h, and the proteins were removed with proteinase K. Isolated DNA was extracted with phenol/chloroform, re-dissolved and PCR-amplified with IL-22R1 promoter primers (PCR product: 140 bp; sense: 5’-AGAAAACCATTTCCTAAACA-3’; and antisense: 5’-GGGGCTAACTTCATGAATG-3’).

Transthoracic Echocardiography

Transthoracic echocardiographic studies were performed before (baseline) as well as at 28 days post-AMI on rats anesthetized with a mixture of 1.5% isoflurane and oxygen (1 L/min) using two-dimensional (2D) echocardiogram (M-mode) (Logiq Book, General Electric, USA) equipped with a 30-MHz transducer as previously described [74]. In brief, 2D images of the left ventricle (LV) were obtained both in long and short axes at a frame rate of > 200/s. LV wall thickness, cavity dimensions both in systole (LV end-systolic diameter, LVESD), diastole (LV end-diastolic diameter, LVEDD), percentage of fractional shortening (FS) and percentage of ejection fraction (EF) were measured through the largest diameter of the ventricle by M-mode tracings with 2D guidance.

Cell Proliferation Assays

To measure the percentage of bromodeoxyuridine (BrdU) incorporation in CSs were incubated with BrdU (10 µM, Merck, Germany) for 24 h, and cytocentrifuged onto object glasses (Cytospin 2) for immunocytochemical (and immunohistochemical) analysis according to instructions of the BrdU labeling kit (Roche, Switzerland). The percentage of proliferating cells was counted by dividing the number of BrdU⁺ cells with the number of total nuclei. Moreover, cell proliferation ability measured by the Ki-67 immunoreactivity (Abcam, UK) was also counted as the percentage of Ki-67⁺ cells out of all nuclei.

Analyses of RNA Microarray

Microarray assay and RNA sample preparation were performed according to the Affymetrix Clariom™ Expression Analysis Technical Manual (Affymetrix, USA). Total RNA was isolated from primary CSs after each treatment using RNeasy Mini Kits (Qiagen, Hilden, Germany). The RNA samples were used for cDNA synthesis with the A260/280 ratio greater than 1.9. The Affymetrix cDNA microarray hybridization and analysis were performed on mouse Clariom™ S Assay (whole-transcript expression analysis) by the NHRI Microarray Core (Zhunan, Taiwan).

Statistics and Reproducibility

All measurements in this study were performed in a blinded design. Statistical analysis was performed using GraphPad Prism software (La Jolla, USA), and data are represented as mean ± standard deviation (SD). Error bars represent SD of three or more independent experiments. Two-tailed Student’s t tests were used to evaluate significance of mean differences between the control and the treated groups. Differences between groups were evaluated by one-way ANOVA with the Tukey’s multiple comparisons test unless otherwise noted. A P value < 0.05 was considered significant. All experiments in Figures were repeated for at least three times, and the performed numbers were provided in the regarding figure legends.

Results

Hypoxia/Ischemia Increased the Expression of IL-22/IL-22R1 in PCCs and Hearts

To determine whether the expression of IL-22/IL-22R1 increased after AMI, we evaluated IL-22 and IL-22R1 levels in rat heart samples. Ischemic rat hearts were collected from the peri-infarcted region (border zone of the LV infarct). Samples from similar heart areas without left anterior descending coronary artery ligation were used as normal controls. ELISA revealed a significant increase in IL-22 levels in the ischemic heart compared with nonischemic controls (Fig. 1A). We observed that AMI increased the level of HIF-1α and IL-22R1 (Fig. 1B) in a time-dependent manner. Of note, the expression of IL-22R1 was observed not only in α-SA⁺ cardiomyocytes but also in stromal cells in heart by the results of immunohistochemical (IHC) analysis of the ischemic heart (Fig. 1C).

Fig. 1.

Hypoxia/ischemia increased the expression of IL-22/IL-22R1 in PCCs and hearts. (A) The enhancement of IL-22 production was presented in the ischemic rat hearts in comparison with non-ischemic controls by ELISA. (B) Higher IL-22R1 and HIF-1α expression was observed in the tissue samples of the peri-infarcted region than in controls in a time-dependent manner by western blotting. (C) AMI induced an increase in IL-22R1⁺ expression in the peri-infarcted area (yellow square) via IHC analysis. (D-E) Compared with controls without hypoxia, IL-22R1 expression was observed in a dose-dependent (D, higher at 3% O₂ level) and time-dependent (E, higher at 8 h) manner in hypoxic PCCs by western blotting. n = 5 in (A), (C), and (D). n = 4 in (B) and (E). Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control. Bar = 50 μm

Next, to further verify whether hypoxia could enhance the expression of IL-22R1 in PCCs, PCCs were subjected to hypoxic treatment at various different oxygen levels (21%, 1%, 3%, and 5%) and times (1, 4, 8, 16, and 24 h) to identify the optimal conditions for investigating the association between hypoxia and IL-22R1 in vitro. The hypoxic concentrations were not aimed to simulate the oxygen levels of in vivo ischemia, but rather to determine the optimal hypoxic time and concentration for subsequent in vitro experiments regarding the molecular mechanism of IL-22R1 as previously described [33]. Compared with the control without hypoxia, the expression of IL-22R1 were significantly upregulated after hypoxia with higher at 3% O₂ level (Fig. 1D) and with higher at 8 h (Fig. 1E), in a dose-dependent and time-dependent manner, respectively. Moreover, IL-22R1⁺ PCCs were co-expressed with specific cardiomyocyte markers for CD105 at 3% O₂ level for 16 h (Figure S1). These findings indicate that hypoxia/ischemia can increase the expression of IL-22R1 in PCCs, and IL-22/IL-22R1 in hearts.

HIF-1α Activation Upregulated IL-22R1 Expression

To investigate whether the upregulation of IL-22R1 after AMI was mediated by HIF-1α activation, we used 2-methoxyestradiol (2-ME2), an HIF-1α inhibitor, and HIF-1α^−/− mice to assess IL-22R1 expression in the hearts of ischemic rats. Western blotting revealed a 2-ME2 dose-dependent decrease in the expression of IL-22R1 after AMI (Fig. 2A). The same phenomena were observed in the hearts of HIF-1α^−/− mice, which did not exhibit IL-22R1 upregulation after induction of ischemia (Fig. 2B).

Fig. 2.

HIF-1α activation regulated the upregulation of IL-22R1 expression in vitro and in vivo. (A) 2-ME2 inhibited the upregulation of IL-22R1 in a dose-dependent manner after AMI by western blotting. (B) In HIF-1α^−/− mice, AMI was not able to increase IL-22R1 expression compared with the littermate (HIF-1α-NL) heart samples. (C) Both HIF-1α and IL-22R1 upregulation required hypoxia-induced HIF-1α activation, which was demonstrated by 2-ME2 (1 µM) or HIF-1α knockdown using LV-HIF-1α-sh in PCCs. (D) Administration of 2-ME2 inhibited HIF-1α-induced IL-22R1 expression. However, 2-ME2 did not block LV-IL-22R1-induced IL-22R1 overexpression. (E-F) Treatment of LV-HIF-1α-sh with hypoxia and LV-HIF-1α, but not LV-HIF-2α-sh with hypoxia and LV-HIF-2α, significantly modulated IL-22R1 expression in PCCs. (G) Schematic representation of a putative HIF-1α-binding site (HRE) in the IL-22R1 promoter sequence from − 826 to − 821 (5′-ACGTG-3′). (H) The recruitment and direct binding of HIF-1α to the IL-22R1 promoter were demonstrated in PCCs subjected to 4 h of hypoxia but not in PCCs subjected to 4 h of normoxia or PCCs treated with LV-HIF-1α-sh with 4 h of hypoxia by the chromatin immunoprecipitation (ChIP) assay. (I) In the luciferase reporter assay, compared with a control construct (pIL22R1-D4) and an HRE-mutant construct (pIL22R1-D2mut), pIL22R1-D2 activity was significantly increased under hypoxic conditions. n = 5 in (A) to (F). n = 3 in (G) to (I). Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control

To further explore the effects of HIF-1α activation on IL-22R1 expression, we examined the subcellular localization of HIF-1α and IL-22R1 in PCCs under hypoxic conditions. Under normoxic conditions, HIF-1α was localized both in the cytosol and nucleus and was coexpressed with IL-22R1. In contrast, under 3% O₂ hypoxic conditions for 16 h, HIF-1α translocated to the nuclei or perinuclear areas (Figure S2). Pretreatment of PCCs with 2-ME2 (1 µM) for 16 h inhibited the translocation of HIF-1α to the nucleus (Figure S2). Interestingly, upregulation of HIF-1α and IL-22R1 required HIF-1α activation because both were inhibited after 2-ME2 pretreatment (pretreated for 16 h) of PCCs or HIF-1α knockdown using LV-HIF-1α-shRNA (Fig. 2C). We also observed that 2-ME2 specifically inhibited HIF-1α-induced IL-22R1 expression. However, 2-ME2 had no inhibitory effect on lentiviral-induced IL-22R1 (LV-IL-22R1) overexpression (Fig. 2D). Taken together, these findings suggest that inhibition of HIF-1α activity, by blocking the nuclear translocation of HIF-1α, can downregulate the expression of IL-22R1 in ischemic hearts and PCCs.

HIF-1α is Recruited to the IL-22R1 Gene Promoter in Response to Hypoxia

To prove that hypoxia-induced IL-22R1 upregulation occurs only via HIF-1α, we overexpressed HIF-1α (LV-HIF-1α) and HIF-2α (LV-HIF-2α) or knocked down HIF-1α (LV-HIF-1α-sh) and HIF-2α (LV-HIF-2α-sh) via lentiviral transduction. Knockdown of HIF-1α (LV-HIF-1α-sh) (Fig. 2E) or overexpression of HIF-1α (LV-HIF-1α) (Fig. 2F), but not knockdown of HIF-2α (LV-HIF-2α-sh) (Fig. 2E) after hypoxic treatment for 8 h [44] or overexpression of HIF-2α (LV-HIF-2α) (Fig. 2F), significantly modulated the expression of IL-22R1 in PCCs. To obtain direct evidence for the interaction between HIF-1α and the IL-22R1 promoter, we performed bioinformatics analysis to predict the molecular mechanism by which hypoxia induces IL-22R1 expression. We identified one putative HIF-1α-binding site (hypoxia response element, HRE) in the IL-22R1 promoter sequence from − 826 to − 821 (5′-ACGTG-3′) (Fig. 2G), suggesting that HIF-1α regulates IL-22R1 expression by directly binding to its promoter; this was further confirmed using the chromatin immunoprecipitation (ChIP) assay. The ChIP assay revealed the recruitment and direct binding of HIF-1α to the IL-22R1 promoter in PCCs subjected to 8 h of hypoxic treatment but not in those subjected to 4 h of normoxia and in those treated with LV-HIF-1α-shRNA and 4 h of hypoxia (Fig. 2H).

To further determine whether the upregulation of IL-22R1 expression was a result of activated HIF-1α binding to the HRE on the IL-22R1 promoter, the luciferase promoter assay was performed using PCCs subjected to hypoxia. The activity of the luciferase reporter gene construct (pIL22R1-D2) containing HRE from the IL-22R1 gene promoter coupled to an SV40 promoter was much higher than that of a control construct (pIL22R1-D4) and an HRE-mutant construct (pIL22R1-D2mut) under hypoxia (Fig. 2I). Overall, our results indicate that hypoxia-induced IL-22R1 upregulation is mediated by HIF-1α-induced transcriptional activity.

IL-22 Administration Attenuated LV Dysfunction, Reduced Infarct Size, and Diminished Fibrosis by Activating IL-22/IL-22R1 Signaling after AMI

To determine that the IL-22/IL-22R1 signaling cascade plays a significant role in rescuing the heart from ischemic injury, we examined infarct size and LV function after AMI. First, greater infarct size was observed after AMI in IL-22R1^−/− mice than in littermate (IL-22R1-NL) mice (Fig. 3A), indicating that loss of IL-22R1 signaling induced detrimental injury to the heart. Thus, to select the most effective treatment dosage of IL-22, rats were divided into four groups (rats with AMI treated with 0.5, 5, and 50 µg/kg IL-22 and saline control group). The infarct size of the 5 and 50 µg/kg IL-22-treated groups 28 days after AMI was much smaller than that of the control group (Fig. 3B-C). Further, the 5 and 50 µg/kg IL-22-treated groups showed significantly increased infarct wall thickness compared with the thin wall in the control group (Fig. 3B, D). M-mode echocardiography was performed to assess LV function 28 days after AMI. The 5 and 50 µg/kg IL-22-treated groups showed improved cardiac function with lower LVESD–LVEDD and higher FS–EF than the control group (Fig. 3E-F). There was no influence on the heart rates among the groups (data not shown). However, pretreatment of the 50 µg/kg IL-22-treated group with rIL-22BP-Fc abolished all the recovery parameters of LV function after AMI (Fig. 3E-F).

Fig. 3.

IL-22 administration attenuated left ventricular (LV) dysfunction, reduced infarct size, and diminished fibrosis and apoptotic cell death via IL-22/IL-22R1 signaling activation after AMI. (A) Representative images of the gross view and T–M stain of mice hearts revealed a larger infarct size and thinner infarct wall thickness in IL-22R1^−/− mice than in IL-22R1-NL mice 3 days after AMI. (B-D) Representative images of the gross view and T–M stain of rats’ heart revealed that IL-22 (5 and 50 µg/kg)-treated groups had lower (C) infarct volumes and higher (D) infarct wall thickness than the control groups 28 days after AMI. IL-22-mediated decrease in infarct volumes and increase in thickness of infarct wall were inhibited with rIL-22BP-Fc pretreatment 28 days after AMI. (E-F) M-mode echocardiography of rat hearts revealed that the IL-22 (5 and 50 µg/kg)-treated groups had significantly decreased (E) LVESD–LVEDD and increased (F) EF–FS after IL-22 injection compared with the control groups 28 days after AMI. IL-22-induced lower LVESD–LVEDD and higher FS–EF were blocked with rIL-22BP-Fc pre-treatment 28 days after AMI. (G) T–M staining from (B) revealed significant attenuation of fibrosis in the IL-22 (5 and 50 µg/kg)-treated groups compared with the control groups 28 days after AMI. (H) Gelatin in-gel zymography revealed that significantly reduced MMP-9 activity was observed in rat hearts in the IL-22 (5 and 50 µg/kg)-treated groups than in those in the control groups 3 days after AMI. However, rIL-22BP-Fc treatment inhibited IL-22-mediated reduction in MMP-9 activity in the left ventricle after AMI. (I) Representative TUNEL images of the border zone of the LV infarct revealed a significant reduction in AMI-induced apoptotic cell death in the IL-22 (5 and 50 µg/kg)-treated groups than in the control groups 28 days after AMI. IL-22-induced reduction in AMI-induced apoptosis was blocked with rIL-22BP-Fc pre-treatment. n = 8 in each experimental group. Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control. Bar = 50 μm

Next, we observed that the grade of fibrosis was significantly increased in the left ventricle in trichrome–Masson (T–M)-stained sections compared with the sham 28 days after AMI (Fig. 3B, G; Figure S2A). Interestingly, compared with the saline control group, fibrosis was significantly reduced after AMI in the 5 and 50 µg/kg IL-22-treated groups (Fig. 3B, G). Because metalloprotease (MMP)-9 is a protein related to cardiac fibrosis after AMI [104], we performed gelatin in-gel zymography and found that MMP-9 activity was higher after AMI (Fig. 3H). Importantly, the 5 and 50 µg/kg IL-22-treated groups had a greater reduction in MMP-9 activity after AMI than the saline control group (Fig. 3H). In contrast, inhibition of the IL-22/IL-22R1 signaling cascade by rIL-22BP-Fc in the 50 µg/kg IL-22-treated group abolished the IL-22-mediated reduction in fibrosis and MMP-9 activity in the LV after AMI (Fig. 3H).

IL-22/IL-22R1 Signaling Reduces AMI-Induced Apoptotic Cell Death

To investigate whether IL-22/IL-22R1 signaling induces antiapoptotic effects, we performed the TUNEL assay to detect cardiac cell death 3 days (Figure S3) after AMI. Compared with treatment with saline, IL-22 treatment at dosages of 5 and 50 µg/kg significantly attenuated AMI-induced apoptotic cell death in the border zone of the LV infarct (Fig. 3I). In contrast, pretreatment of the 50 µg/kg IL-22-treated group with rIL-22BP-Fc reversed the IL-22-induced antiapoptotic effects 3 days after AMI (Fig. 3I).

Anti-Inflammatory Effects of the IL-22/IL-22R1 Cascade on the Ischemic Myocardium

To further examine whether IL-22 treatment can suppress the inflammatory response after AMI, we performed IHC analysis to study inflammatory CD68⁺ cell infiltration and RTq-PCR to study the expression of various inflammation-related factors 3 days after AMI. H&E staining revealed a significant reduction in inflammatory cell infiltration in the 5 and 50 µg/kg IL-22-treated groups compared with the saline control group (Figure S3C, Figure S4A). The 5 and 50 µg/kg IL-22-treated groups showed significantly fewer infiltrated CD68⁺ cells in the peri-infarct area 3 days after AMI than the saline control group (Figure S4B). In addition, for proinflammatory cytokines, the 50 µg/kg IL-22-treated groups showed a significant reduction in the mRNA expression of IL-1β and IL-6 in the peri-infarct area 3 days after AMI than the saline control group (Figure S4C). Further, the 5 and 50 µg/kg IL-22-treated group showed a significant reduction in the mRNA expression of IFN-γ and TNF-α in the peri-infarct area 3 days after AMI than the saline control group (Figure S4C). For anti-inflammatory cytokines, there was a significant increase in IL-10 expression in the 5 and 50 µg/kg IL-22-treated groups (Figure S4D). However, pharmacological blockade by rIL-22R1-Fc in the 50 µg/kg IL-22-treated group inhibited the anti-inflammatory effect of IL-22 on inflammatory CD68⁺ cell infiltration and proinflammatory factor expression after AMI (Figure S4A-D).

HIF-1α-Activated IL-22/IL-22R1 Signaling Promoted the Proliferation and Self-Renewal of CSCs in Vitro

To determine whether the hypoxia cascade regulates the proliferation of CSCs, we examined the effect of the HIF-1α/IL-22R1 pathway on the self-renewal ability of CSCs. Hypoxia induction (3% O₂ level for 8 h) increased the expression of IL-22R1 in CSCs (Figure S5). Immunocytochemical (ICC) analysis revealed that IL-22R1 was coexpressed with progenitor cell markers CD105 and proliferation markers (BrdU and Ki-67) in CSCs (Fig. 4A). Hypoxia with O₂ level at 3% stimulated a significantly higher frequency of sphere formation (also in 5% O₂ level), sphere size, and self-renewal potential than that with ordinary O₂ level (21%) (Fig. 4B-C). However, knockdown using LV-HIF-1α-sh or LV-IL-22R1-sh (Figure S6) or using IL-22R1^−/− mice abolished the proliferation and self-renewal effects of hypoxia on sphere formation in CSCs (Fig. 4C). In addition, hypoxic conditions significantly increased BrdU incorporation and Ki-67 immunostaining in the spheres compared with normoxic conditions (Fig. 4D). In contrast, enhancement of BrdU incorporation and Ki-67 immunostaining in CSCs was inhibited by the administration of LV-HIF-1α-sh or LV-IL-22R1-sh or by using the IL-22R1^−/− mice (Fig. 4D). Taken together, these findings indicate that hypoxia can promote the proliferation and self-renewal of CSCs by regulating the HIF-1α/IL-22R1 pathway in vitro.

Fig. 4.

HIF-1α-activated IL-22/IL-22R1 pathway stimulated the proliferation and self-renewal of CSCs in vitro. (A) Representative immunocytochemical (ICC) images of CSCs and IL-22R1⁺ CSCs colocalized with the progenitor cell markers (upper panel, CD105) and the proliferation markers (lower panel, BrdU and Ki-67). (B) Representative images of CSCs under normoxic and hypoxic conditions (1%, 3%, and 5% O₂ level) for 8 h. (C) Hypoxia with 3% O₂ for 8 h promoted a higher frequency of sphere formation (also in 5% O₂ level), sphere size (the average sphere size was obtained of five fields per sample), and self-renewal potential than ordinary O₂ level (21%). In contrast, the proliferative effect on the sphere formation by hypoxia was abolished by LV-HIF-1α-sh or LV-IL-22R1-sh (left panel) administration into CSCs and IL-22R1^−/− mice (right panel). (D) Hypoxia with 3% O₂ for 8 h induced more enhancement of BrdU incorporation and Ki-67 immunostaining in CSCs than in normal cells. However, a significant increase in BrdU incorporation and Ki-67 immunostaining was blocked by the administration of LV-HIF-1α-sh or LV-IL-22R1-sh (left panel) in CSCs and IL-22R1^−/− mice (right panel). (E) After 8 h of hypoxic conditions at 3% oxygen, representative images of CSCs were captured following IL-22 administration (1, 10, 100 ng/ml). (F) IL-22 treatment after hypoxia induced a higher frequency of cardiac sphere formation, sphere size, and self-renewal potential than vehicle control in a dose-dependent manner. However, the proliferative effect on the CSCs was abolished by gene knockdown by LV-IL-22R1-sh or 2-ME2 treatment. (G) Administration of IL-22 after hypoxia triggered a significant enhancement of BrdU incorporation and Ki-67 immunostaining in CSCs compared with control cells in a dose-dependent manner. Conversely, the promotion of BrdU incorporation and Ki-67 immunostaining was inhibited by treating CSCs with LV-IL-22R1-sh or 2-ME2 in. n = 3 in each experimental group. Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control. Bar = 50 μm in (A) and 100 μm in (B) and (E)

To further determine whether IL-22 modulates the self-renewal of CSCs, we examined the effect of IL-22/IL-22R1 pathway modulation by hypoxia on the self-renewal potential of CSCs. ICC analysis revealed that IL-22 administration stimulated a significantly higher frequency of cardiac sphere formation, sphere size, and self-renewal potential than the vehicle control in a dose-dependent manner under 3% O₂ (Fig. 4E-F). In contrast, gene knockdown using LV-IL-22R1-sh or pharmacological inhibition using 2-ME2 in the culture inhibited the proliferation effects on CSCs (Fig. 4F). In addition, significant enhancement of BrdU incorporation and Ki-67 immunostaining in the sphere was observed after IL-22 treatment compared with the vehicle control in a dose-dependent manner under 3% O₂ (Fig. 4G). However, administration of LV-IL-22R1-sh or pharmacological inhibition using 2-ME2 abolished the increase in BrdU incorporation and Ki-67 immunostaining in CSCs (Fig. 4G). Taken together, these data suggest that the IL-22/IL-22R1 signaling cascade regulated by hypoxia can augment the proliferation and self-renewal potential of CSCs in vitro.

Systemic Activation of IL-22/IL-22R1 Signaling Enhanced the Proliferation of eCSCs in the Ischemic Myocardium

To study the self-renewal potential of eCSCs after IL-22 treatment in vivo, we performed IHC analysis of CSCs in the peri-infarcted area of the rat heart. Three days after AMI, administration of IL-22 (50 µg/kg) resulted in a higher number of CD105⁺/BrdU⁺ (Fig. 5A), and CD105⁺/Ki-67⁺ (Fig. 5B) CSCs compared with treatment with the vehicle control in a time-dependent manner. Since Nkx2.5, abundantly expressed in CSCs [10], serves as the earliest marker of the cardiac lineage [10], it is crucial to further investigate whether the IL-22/IL-22R1 signaling cascade, regulating the proliferation of CD105⁺/Nkx2.5⁺ CSCs, significantly contributes to cardiac repair. We found that rIL-22BP-Fc injection (10 µg/kg, i.p.) abolished the IL-22-induced increase in the number of CD105⁺/Nkx2.5⁺ (Fig. 5C) CSCs in the ischemic heart 3 days after AMI. These data suggest that IL-22/IL-22R1 signaling can augment the activation and proliferation of CSCs in the ischemic heart.

Fig. 5.

Systemic activation of the IL-22/IL-22R1 cascade enhanced eCSC proliferation after AMI in the ischemic myocardium. After AMI, IHC analysis revealed that compared with sham control, a significant increase in the cell numbers of (A) CD105⁺/BrdU⁺ and (B) CD105⁺/Ki-67⁺ CSCs was observed with a peak at 3 days in a time-dependent manner. (C) CD105⁺/Nkx2.5⁺ CSCs, which was blocked using the rIL-22BP-Fc injection (10 µg/kg, i.p., was administered 4 h before the AMI), in the ischemic heart 3 days after AMI. n = 3 mice for each group, and the images were representative of 6 images for each group. (A)-(B) represented images after 3 days of AMI; (C) represented an image after 3 days of AMI with a subcutaneous injection of mouse recombinant IL-22 (50 µg/kg, s.c.) on 1 day after AMI. Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control. Bar = 50 μm

Bmi1 Enhanced the Activation of CSCs via IL-22/IL-22R1 Signaling

To explore the downstream signaling mediators that regulate CSC proliferation, unbiased RNA microarray analysis was performed in CSCs for screening transcriptional expression, mimicked pathological AMI condition (IL-22 treatment under 3% O₂ hypoxia for 8 h, recognized as hypoxia-IL-22, called H22) versus normal physiological condition (IL-22 treatment under normoxia, recognized as normoxia-IL-22, called N22). Under the screening criteria of |log₂FC| ≥ 2 and p-value ≤ 0.05, 1000 transcripts were identified as differentially expressed genes (DEGs), including 646 upregulated genes and 354 downregulated genes between the hypoxia-IL-22 versus normoxia-IL-22 CSC groups (Fig. 6A). We observed that compared with the normoxia-IL-22 group, the hypoxia-IL-22 group has increased expression of progenitor genes, such as CD105 (endoglin), and Sca-1 (Ly6a) (Fig. 6B). Downstream signaling mediator analysis revealed that the gene expression of CSC-related transcription factors, including Bmi1, Oct4 (Pou5f1), and Nkx2-5, was enhanced after hypoxia-IL-22 treatment than after normoxia-IL-22 treatment (Fig. 6C). Song et al. reported that Bmi1^hi cardiac cells were enriched in cardiac stem/progenitor cells after AMI [76, 93]. However, whether CSCs express Bmi1 and if cardiac Bmi1 expression plays a role in the IL-22/IL-22R1 axis after AMI remains unclear. We found that the transcriptional expression of Bmi1 was significantly increased in the 100 µg/ml IL-22-treated CSCs after 8 h of hypoxia than in the normoxia-IL-22-treated CSCs (Fig. 6D). To study the role of Bmi1 in the self-renewal potential of eCSCs after IL-22 treatment in vivo, we performed IHC analysis of eCSCs in the peri-infarcted area of rat hearts. Analysis revealed that IL-22 administration (50 µg/kg) increased the numbers of CD105⁺/Bmi1⁺ eCSCs (Fig. 6E), which were reduced after rIL-22BP-Fc injection (10 µg/kg, i.p.), in the ischemic heart 3 days after AMI. Taken together, these data suggested that IL-22 enhanced CSC activation through the downstream Bmi-1 via IL-22R1.

Fig. 6.

Bmi1 enhanced CSCs activation via the IL-22/IL-22R1 signaling pathway. (A) Upregulated and downregulated gene analyses of the differentially expressed genes (DEGs) with screening criteria of |log₂FC| ≥ 2 and p-value ≤ 0.05. (B-C) The heatmap after hierarchical clustering of samples treated with 8 h of hypoxia-IL-22 (H22) versus normoxia-IL-22 (N22) based on the mRNA expression data obtained from RNA microarray analysis. The heatmap revealed the fold change of DEGs (log₂FC of DEGs and p-value ≤ 0.05) across the two treatment groups of (B) cardiac progenitor genes and (C) cardiac progenitor cell-related transcription factors (TFs). (D) The mRNA expression of Bmi1 was significantly increased after treating CSCs with mouse 100 µg/ml IL-22 by RTq-PCR (n = 3 in each experimental group). (E) Pharmacological IL-22 treatment (50 µg/kg) increased the cell numbers of CD105⁺/Bmi1⁺ eCSCs, which was blocked using the rIL-22BP-Fc injection (10 µg/kg, i.p., was administered 4 h before the AMI), in the ischemic heart 3 days after AMI. The representative image after 3 days of AMI with a subcutaneous injection of mouse recombinant IL-22 (50 µg/kg, s.c.) on 1 day after AMI. n = 3 mice for each group, and the images were representative of 6 images for each group. Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control. Bar = 50 μm

Bmi1 Promoted CSC-Mediated Cardiomyocyte Repair After AMI

To further verify whether Bmi1 is the downstream mediator of IL-22/IL-22R1 signaling to modulate CSCs self-renewal, we used Bmi1 knockout mice to examine the effect of the modulation of the IL-22/IL-22R1 pathway on CSCs activation with IL-22 administration after 8 h of hypoxia. A higher frequency of cardiac sphere formation (Fig. 7A-B), sphere size (Fig. 7A, C), and self-renewal potential (Fig. 7A, D) was observed in CSCs derived from Bmi1-NL and stimulated with IL-22 treatment under hypoxic conditions; these effects were inhibited after rIL-22BP-Fc (1 µg/ml) treatment (Fig. 7A-D). In contrast, these positive effects after IL-22 treatment were inhibited in CSCs derived from Bmi1^−/− mice (Fig. 7A-D). These results suggest that Bmi1 enhances the activation and proliferation of CSCs via the IL-22/IL-22R1 axis.

Fig. 7.

Bmi1 promoted CSC-mediated cardiomyocyte repair after AMI. (A) Representative images of primary and secondary CSCs from Bmi1-NL and Bmi1^−/− mice after IL-22 administration (100 ng/ml) before 3% O₂ hypoxia for 8 h. (B-D) IL-22 treatment significantly increased the frequency of (B) cardiac sphere formation, (C) sphere size (the average sphere size was obtained of five fields per sample), and (D) self-renewal potential in CSCs derived from Bmi1-NL, which were inhibited via rIL-22BP-Fc treatment (1 µg/mL). However, these effects of IL-22 treatment were abolished in CSCs derived from Bmi1^−/− mice. The trophic factors produced by CSCs, such as (E) stromal cell-derived factor 1 (SDF-1), (F) hepatocyte growth factor (HGF), and (G) insulin-like growth factor 1 (IGF-1), were assayed in CSCs derived from Bmi1-NL and Bmi1^−/− mice after hypoxia with IL-22 treatment by RTq-PCR. n = 3 in each experimental group. Data are shown as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 vs. control. Bar = 50 μm

To further ascertain the role of Bmi1 in CSC-related cardiac repair, we used CSCs derived form Bmi1 knockout mice to evaluate the production of trophic factors, such as stromal cell-derived factor 1 (SDF-1) [34, 36], hepatocyte growth factor (HGF), and insulin-like growth factor 1 (IGF-1) [68, 91], after IL-22 treatment under 3% O₂ hypoxia for 8 h. We found that IL-22 triggered a significant enhancement in the production of SDF-1 (Fig. 7E), but not HGF (Fig. 7F) or IGF-1 (Fig. 7G). However, the production of SDF-1 from CSCs after IL-22 treatment was abolished in Bmi1^−/− mice (Fig. 7E). Taken together, the results suggest that the HIF-1α-activated IL-22/IL-22R1/Bmi1 axis induces SDF-1 expression to repair cardiac damage after AMI.

Discussion

Therapeutic strategies targeting inflammatory responses, which are associated with proinflammatory cytokines such as TNF-α, IL-1β, and IL-6 [63], promote cardiac remodeling after AMI [95]. In this study, we used recombinant IL-22 to inhibit inflammatory responses in the heart to study LV function and cardiac remodeling after AMI. We demonstrated that IL-22 administration can reduce myocardial inflammation and MMP-9 expression 3 days after AMI as well as attenuate LV dysfunction and remodeling, with its effects on fibrosis, 28 days after AMI. Additionally, 3 days after AMI, IL-22 blocked the increase in the number of TUNEL-positive cells followed by significant infiltration of CD68-positive monocytes/macrophages in the border zone of the myocardium [81], which is associated with an increase in the mRNA expression of various proinflammatory cytokines (IL-1β, IL-6, TNF-α, and IFN-γ). The abovementioned effects after IL-22 treatment were abolished by the administration of rIL-22BP-Fc, an IL-22R1 inhibitor. Overall, these findings indicate that IL-22/IL-22R1 signaling markedly improves LV function and attenuates LV remodeling after AMI.

IL-22, together with IL-10, IL-19, IL-20, IL-24, and IL-26, belongs to the IL-10 family of cytokines [17]. Because monocytes, B cells, T cells, NK cells, monocyte-derived macrophages, and dendritic cells do not express IL-22R1, these immune cells are not the target of IL-22 [98]. Thus, IL-22 is an unusual interleukin; it activates cells of the skin and digestive and respiratory systems as well as those of the pancreas, liver, kidneys, and joints [64]. In our study, we identified the heart as another new target organ, which expresses IL-22R1 and could be regulated by IL-22 in vitro and in vivo. Regarding physiological function, besides the enhancement of innate defense mechanism [24, 80], induction of specific chemokines [4], and promotion of cellular mobility [5], IL-22 could also augment cellular proliferation [42] and inhibit differentiation [99]. Therefore, tissue-specific endogenous stem/progenitor cells probably contribute to the strong protective properties of IL-22 against damage in selected tissues such as the liver and intestine [21, 27]. Because of the high IL-22R1 expression in the liver and intestine, IL-22 strongly promotes liver stem cell proliferation and protects intestinal stem cells from inflammatory insults [21, 27]. In this study, we provided in vivo and in vitro evidence that IL-22 is also an important factor that promotes CSCs proliferation to repair ischemic injury after AMI. Cardiac IL-22 and IL-22R1 levels were markedly elevated in ischemic cardiomyocytes after AMI. Pharmacological inhibition using rIL-22BP-Fc and genetic knockdown of IL-22R1 via RNA interference blocked ischemia–hypoxia-induced proliferation and self-renewal of CSCs. These results indicate that IL-22/IL-22R1 signaling is essential for the self-renewal of CSCs in the ischemic heart.

Various markers have been used to identify different populations of CSCs. Distinct groups have been classified based on stem cell markers and functional assays, such as Sca-1⁺ CSCs [57, 94], Islet-1⁺ CSCs [7, 61], and cardiosphere-derived cells [46, 75]. Previous studies have demonstrated that it was revealed that the adult heart contains a resident population of Bmi1-expressing progenitor cells (B-CPCs, Sca1⁺/c-kit⁻/Bmi1⁺) within the non-cardiomyocyte Sca-1⁺ fraction using a validated lineage tracing strategy to monitor Bmi1 locus activity [92]. These B-CPCs exhibit higher expression levels of various multipotency and stemness markers, and their numbers increase as the mouse ages. Throughout the whole life, B-CPCs significantly contribute to the formation of new cardiomyocytes, endothelial cells, and smooth muscle cells. Additionally, the research indicates that B-CPCs possess a stemness genetic profile, become activated, and supply the new generation of cardiomyocytes following AMI [93]. In the future, further research focusing on Bmi1CreER/⁺;Rosa26YFP/⁺ (Bmi1-YFP) mice would be worthwhile to provide in vivo evidence of the HIF-1α-activated IL-22/IL-22R signaling pathway in repairing ischemic heart after AMI by regulating and expanding Sca-1⁺ eCSC or B-CPC activation, as well as subsequent cardiomyocyte differentiation.

According to previous studies on B-CPCs [92, 93], Bmi1 regulates cardiac progenitor differentiation through its antioxidant- and anticlastogenic-related functions [31]. They reported that oxidative stress disturbed Bmi1 function to induce deregulation of its downstream genes and leading to cardiac progenitor differentiation in vivo. Reduction of pathology-related ROS levels restored Bmi1’s epigenetic regulation ability in vivo. This correlation between oxidative stress and progenitor cell differentiation underscored ROS as a key regulator of adult cardiac progenitor turnover in vivo. In order to keep the homeostasis, there is an inverse relationship between levels of expression of Bmi1 and ROS level for governing the B-CPCs self-renewal. In response to oxidative damage in vivo, Bmi1 activity could be modified by derepressing canonical target genes in favor of it’s antioxidant and anticlastogenic functions to promote differentiation of cardiac progenitors even in a steady state. Therefore, ROS-associated differentiation of cardiac progenitors to the three main cardiac cell lineages would be robustly enhanced by severe ischemic damage (i.e., acute infarct, etc.). In this study, in-deep investigation would be need to understand how Bmi1 regulates cardiac progenitor differentiation through its antioxidant- and anticlastogenic-related functions within the HIF-1α-activated IL-22/IL-22R1 signaling pathway to regulate eCSC activation, protection, expansion, and their balanced differentiation.

Self-renewal of adult stem cells is necessary for maintaining tissue development and resupplying the stem cell pool after organ injury [29]. Because HIFs activated after hypoxia may induce the expression of numerous gene products, including pluripotency-associated transcription factors (Oct-3/4, Nanog, and Sox-2) [25], hypoxia is known to facilitate an undifferentiated state in several stem and progenitor cell populations [1, 48, 60]. The signaling cascade and transcriptome activation in hypoxic conditions have been observed after different biological responses among specific stem/progenitor cell lineages; however, the direct effect of hypoxia on stem/progenitor cells has received little attention [73]. Regarding the proliferation machinery, the molecular mechanisms of self-renewal that allow stem/progenitor cells to maintain their pluripotency involve cytokines and growth factors such as SDF-1, IGF-1, and HGF [18, 85]. These factors are required for the modulation of self-renewal in several types of stem cells, including neural [38], hematopoietic [90], and cancer stem cells [30]. Importantly, these proliferation-inducing cytokines and growth factors are regulated by hypoxia via HIF-1α activation [11, 19, 20, 28, 72]. Moreover, some transcriptional regulators, such as Bmi1, have been demonstrated to play important roles in the self-renewal and maintenance of stemness of many tissue-specific stem/progenitor cells, including cardiac progenitor cells [31], neural progenitor cells [100], and intestinal stem cells [71]. Taken together, the relationship between HIF-1α activation during hypoxia and the expression of cytokines, growth factors, and transcriptional regulators facilitates the regulation of the stem/progenitor cell pool. In our study, we first demonstrated that the IL-22/IL22R1 signaling pathway was activated in cultured CSCs during hypoxia and in the ischemic heart. Activated HIF-1α under hypoxic conditions directly binds to the HRE in the promoter of IL-22R1 to activate its expression. Further, we observed that hypoxia-induced IL-22R1 overexpression stimulated CSC proliferation in cardiac sphere cultures and ischemic hearts. Moreover, activation of Bmi1 after the IL-22/IL22R1 cascade induced CSC self-renewal; this effect could be abolished in Bmi1 knockout mice. In summary, the HIF-1α-IL-22/IL-22R1/Bmi1 cascade might play an important role in the modulation of the CSC pool in the ischemic heart.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Abbreviations

AMI

Acute myocardial ischemia

CSCs

Cardiac stem cells

CSs

Cardiospheres

ChIP

Chromatin immunoprecipitation

eCSCs

Endogenous cardiac stem cells

IL-22

Interleukin 22

IL-22R1

IL-22 receptor alpha 1

LPCs

Liver progenitor cells

LV

Left ventricular

MMP

Metalloprotease

PCCs

Primary cardiac cells

Author Contributions

WL designed the research, performed experiments, analyzed the data, and wrote the manuscript; S-LL, J-YC, T-YC, W-WC, and S-RH conducted experiments and analyzed the data; C-SC, M-CH, and B-LY contributed to manuscript preparation and provided feedback; H-TL and K-CC supervised experiments and oversaw the research; L-BJ, and W-CS conceived the idea, oversaw the research, revised the manuscript, and provided funding.

Funding

This work was supported by a research grant from “New Drug Development Center, China Medical University” from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan and Taiwan’s Ministry of Science and Technology (NSTC 113-2321-B-039-007).

Data Availability

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

Code Availability

Not applicable.

Declarations

Ethics Approval

All animal studies were conducted in accordance with guidelines using protocols approved by the Institutional Animal Care and Use Committees of China Medical University (CMUIACUC-2021-394).

Consent to Participate

Not applicable: the study did not involve human participants.

Consent for Publication

Not applicable.

Competing Interests

The authors declare no conflict of interest.