IL-17A-knockout (KO)²⁴ and IL-23p19-KO (IL-23-KO)²⁰ mice were described previously. TCRγδ-KO mice were purchased from Jackson Laboratories. The TLR2-KO, TLR4-KO, and TLR2/4 double-knockout mice were a generous gift from Dr Shizuo Akira (Osaka University) and described previously.²⁵ All mice were bred on the C57BL/6 background, and 10- to 16-week-old male mice were used in this study. All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee at the Keio University School of Medicine.

Mice were subjected to a permanent (MI) ligation of the left anterior descending artery or to a sham operation without ligation as described previously.²⁶ In brief, mice were lightly anesthetized with diethyl ether, intubated, and then fully anesthetized with 1.0% to 1.5% isoflurane gas while being mechanically ventilated with a rodent respirator. The chest cavity was opened via left thoracotomy to expose the heart such that the left anterior descending coronary could be visualized by microscopy and permanently ligated with a 7-0 silk suture at the site of its emergence from the left atrium. Complete occlusion of the vessel was confirmed by the presence of myocardial blanching in the perfusion bed. Mice that died during recovery from anesthesia were excluded from the analysis. Sham-operated animals underwent the same procedure without coronary artery ligation. In the functional experiments, the sham operation was done and waited for 28 days; however, in the immune cells infiltration experiments and qPCR experiments, the sham operation was done and sacrifice was on day 2 or day 7. In addition, we separately verified that the numbers of macrophages, T cells, and neutrophils in the heart remained constant on days 1, 4, 7, and 14 after the sham operation. To evaluate the infarct size on day 1 after MI, hearts were weighed and frozen at −80°C. The frozen hearts were cut transversely into 1-mm-thick slices using a Mouse Heart Slicer Matrix and stained with 2% triphenyltetrazolium chloride (TTC) in PBS (pH 7.4) for 20 minutes in a 37°C water bath. After fixation for 4 to 6 hours in 10% neutral buffered formaldehyde, both sides of each slice were photographed. Viable myocardium stained brick red, and infarct tissues appeared pale white. Infarct and LV area were measured by automated planimetry using Image J software (version 1.43u, National Institutes of Health), with the infarct size expressed as a percentage of the total LV area.

At each time, mice were deeply anesthetized and intracardially perfused with 40 mL of ice-cold PBS to exclude blood cells. The heart was dissected, minced with fine scissors, and then enzymatically digested with a cocktail of type II collagenase (Worthington Biochemical Corporation, Lakewood, NJ), elastase (Worthington Biochemical Corporation), and DNaseI (Sigma, St. Louis, MO) for 1.5 hours at 37°C with gentle agitation. After digestion, the tissue was triturated and passed through a 70-μm cell strainer. Leukocyte-enriched fractions were isolated by a 37% to 70% Percoll (GE Healthcare) density gradient centrifugation as described elsewhere.²⁷ Cells were removed from the interface and washed with RPMI-1640 cell culture medium for further analysis. Spleens were removed, homogenized, and then passed through a 70-μm nylon mesh in PBS. After the addition of red blood cell lysis buffer (eBioscience) to exclude erythrocytes, the single-cell suspension in PBS was refiltered through a 70-μm nylon mesh to remove connective tissue.

Cell suspensions isolated from spleen and leukocyte-enriched fractions from heart were analyzed by flow cytometry. To block nonspecific binding of antibodies to Fcγ receptors, isolated cells were first incubated with anti-CD16/32 antibody (2.4G2; BD Biosciences) at 4°C for 5 minutes. Subsequently, the cells were stained with a mixture of antibodies at 4°C for 20 minutes. Results were expressed as cell number per heart. Flow-cytometric analysis and sorting were performed on a FACSAria instrument (BD Biosciences) and analyzed using FlowJo software (Tree Star).

Anti-CD45-FITC (30F11.1; eBioscience), anti-CD45-PE (30-F11; BD Biosciences), anti-CD11b-PerCP-Cy5.5, anti-CD11b FITC (M1/70; eBioscience), anti-CD11b-PE (M1/70; BD Biosciences), anti-CD3e-FITC, anti-CD3e-APC (145-2C11; eBioscience), anti-CD3e-PE (145-2C11; BD Biosciences), anti-CD19-PE (1D3; BD Biosciences), anti-CD4-PE (GK1.5; eBioscience), anti-CD8a-PE (53-6.7; eBioscience), anti-TCR-γδ-PE (GL3; Biolegend), anti-NK1.1-PE (PK136; eBioscience), anti-CD11c-APC, anti-CD11c-PE (N418; Biolegend), anti-MHC-II (I-A/I-E)-PE (M5/114.15.2; eBioscience), anti-Gr-1-APC, anti-Gr-1-Alexa Fluor488 (RB6-8C5; eBioscience), anti-F4/80-FITC, anti-F4/80-PE (BM8; Biolegend), anti-F4/80-APC (BM8; eBioscience), anti-Ly-6G-PE (1A8; BD Biosciences), anti-CD206 (MMR)-AlexaFluor647 (MR5D3; Biolegend), anti-CCR6-Alexa Fluor(R) 647 (140706; BD Biosciences), anti-CXCR2-APC (FAB2164A; RD system), anti-Thy1.2-APC (53-2.1; eBioscience), anti-CD31-PerCP-eFluor710 (390; eBioscience), anti-IL-17A-APC (eBio17B7; eBioscience), and anti-IFN-γ-APC (XMG1.2; eBioscience) antibodies were used for flow-cytometric analysis in this study.

For surface and intracellular cytokine staining, single cells prepared from spleen and heart were restimulated for 4.5 hours with 50 ng/mL phorbol 12-myristate 13-acetate (PMA; Sigma-Aldrich) and 1 μg/mL ionomycin (Sigma-Aldrich) in the presence of Golgistop (Cytofix/Cytoperm Plus Kit with Golgistop, BD Biosciences). Surface staining was performed for 20 minutes with the corresponding mixture of fluorescently labeled antibodies. After fixation and permeabilization, the cells were incubated for 30 minutes at 4°C with anti-IL-17A-APC and anti-IFN-γ-APC (eBioscience).

Mouse heart cells were prepared from infarcted heart, and CD45 MicroBeads were used to enrich for heart-derived leukocytes. Cells were stimulated with 10 ng/mL rmIL-23, 10 ng/mL rmIL-1β (R&D Systems), 100 ng/mL LPS (Sigma-Aldrich), and 1 μg/mL Pam3CSK4 (InvivoGen) for 3 days in the presence or absence of 3 μg/mL IL-1RI (IL-1 receptor I) antibody. The supernatants were harvested and assayed for IL-17A levels by ELISA (R&D Systems). After 3 days stimulation, cells were also restimulated for 4.5 hours with PMA and ionomycin in the presence of Golgistop for intracellular IL-17A staining. For the γδT cell proliferation assay, cardiac cells were stained with 5(6)-carboxyfluorescein diacetate N-succinimidyl ester (CFSE; Sigma-Aldrich) according to the manufacturer's recommendations. Briefly, cells were stained in 10 μmol/L CFSE in PBS at 37°C for 5 minutes and then washed 3 times with cold PBS. Labeled cells were stimulated for 3 days, and then stained for surface markers CD3 and TCRγδ prior to flow cytometry.

Neonatal ventricles from 1-day-old C57BL6/J mice were minced and digested with collagenase type II (Worthington) solution.²⁸ To enrich for cardiomyocytes, the cells were preplated for 2 hours to remove nonmyocytes. The unattached viable cells, which were rich in cardiomyocytes, were plated on gelatin-coated plastic dishes and treated with Ara C (Sigma) to inhibit nonmyocyte proliferation. Using this protocol, we consistently obtained cell populations containing ≥90% to 95% cardiomyocytes. Nonmyocyte cells that attached to the dishes were cultured in DMEM supplemented with 10% FBS and allowed to grow to confluence; these were then trypsinized and passaged at 1 in 4. This procedure yielded cell cultures that were almost exclusively fibroblasts by the first passage. Experiments were carried out after 2 passages.

Adult cardiomyocytes were isolated using the Langendorff perfusion method as previously described.²⁸ For sorting of fibroblasts, macrophages, endothelial cells, and lymphocytes, single cells prepared from infarcted heart were incubated with APC-conjugated anti-Thy1 antibody (eBioscience), PerCP-eFluor710-conjugated anti-CD31 antibody (eBioscience), FITC-conjugated anti-CD45 antibody (eBioscience), and PE-conjugated anti-F4/80 antibody (eBioscience), after which they were analyzed and sorted using a FACSAria (BD Biosciences) and FlowJo software.

Total RNA samples from sorted cells, cultured cells, and heart tissue were prepared using using an RNeasy Mini Kit (Qiagen) or Trizol reagent (Invitrogen), according to the manufacturer's instructions. A First-strand cDNA synthesis kit (Invitrogen) was used for cDNA synthesis. Quantitative real-time PCR was performed using the ABI Prism 7700 sequence detection system (Applied Biosystems). Predesigned gene-specific primer and probe sets (Taqman Gene Expression Assays, Applied Biosystems) were used. The 18S ribosomal RNA was amplified as an internal control.

LVs were homogenized in PBS containing protease inhibitors (Sigma-Aldrich). Supernatants were collected after centrifugation and stored at −80°C. The concentrations of IL-23, IL-1β, and IL-17A in LV lysates and culture supernatants were measured by Quantikine ELISA kits (R&D Systems).

To evaluate the activity of gelatinase, matrix metalloproteinase 9 (MMP9), and MMP2, gelatin zymography was performed. Equal volumes containing 35 μg of protein were loaded into each lane of 10% gelatin zymogram gels (Novex, Invitrogen). After running at 125 V for 90 minutes, the gels were incubated in zymogram renaturing buffer for 30 minutes at room temperature with gentle agitation, equilibrated with zymogram developing buffer for 30 minutes, and then further incubated in zymogram developing buffer at 37°C overnight with gentle agitation. After washing with deionized water, the gels were stained with Coomassie blue for 90 minutes followed by destaining with deionized water. The presence of different MMPs was identified on the basis of their molecular weight. The gels were photographed using MF-ChemiBIS (DNR Bio-Imaging Systems) and analyzed using Image J software (version 1.43u, National Institutes of Health).

An anaerobic jar containing an Anaero Pack (Mitsubishi Gas Chemical) was used to expose the cells to hypoxic stress.²⁹ Cultured cardiomyocytes were serum starved in DMEM with 0.5% FBS, and then exposed to hypoxic stress and/or IL-17A stimulation (421-ML, R&D systems). Neutralizing anti-IL-17A antibody (AF-421-NA) was administered 2 hours before hypoxic stress. After 12 hours of exposure to hypoxia, the medium was replaced with 10% FBS-containing DMEM (reoxygenation medium). Cell viability was determined by a LIVE/DEAD Viability/Cytotoxicity Assay Kit (Invitrogen) on the basis of the simultaneous determination of live and dead cells with the calcein AM and ethidium homodimer-1 probes, which are specific for intracellular esterase activity and membrane integrity, respectively. The cells were imaged with a fluorescence microscope (BZ-9000; Keyence): live cells were labeled green, whereas nuclei of dead cells were labeled red.

Mouse fibroblasts were cultivated in DMEM supplemented with 10% fetal bovine serum (FBS), penicillin, and streptomycin in a cell incubator with 5% CO₂ at 37°C. Cell proliferation was analyzed by the CCK-8 assay (Dojindo Molecular Technologies, Japan) as directed by the manufacturer. Fibroblasts were seeded at 4000 cells per well in 96-well culture plates and cultured for 72 hours for cell proliferation analysis.

Heart tissue was fixed in formalin, embedded in paraffin, and cut into 5-μm-thick sections. Hematoxylin and eosin (H&E) and Azan staining were performed on paraffin-embedded sections to determine the morphological effects, infarct size, and extent of cardiac fibrosis. The infarct size was calculated as total infarct circumference divided by total LV circumference×100, as described previously.²⁶ In addition, for each Azan-stained section, 20 microscopic fields (×400 magnification) were randomly chosen (BZ-9000; Keyence, Osaka, Japan), and the area of myocardial fibrosis in noninfarct and infarct areas was measured and analyzed using analysis software (BZ image analyzer II; Keyence).

Transthoracic echocardiography was performed with a Vevo 2100 instrument (VisualSonics) equipped with an MS-400 imaging transducer. Mice were kept awake without anesthesia during the echocardiographic examination to minimize data deviation, and heart rate was maintained at ≍550 to 650 bpm in all mice. M-mode tracings were recorded through the anterior and posterior LV walls at the papillary muscle level to measure LV end-diastolic dimension (LVEDD) and LV end-systolic dimension (LVESD). LV fractional shortening (FS) was calculated according to the following formula: LV FS=([LVEDD−LVESD]/LVEDD)×100.

Cardiac catheterization studies were performed using a 1.4 French microtip catheter (SPR-671, Millar Instruments, Houston, TX) under sedation using 1.5% isoflurane inhalation with spontaneous respiration. LV end-systolic pressure (LVESP), maximum rate of isovolumic pressure development, and minimum rate of isovolumic pressure decay were measured using analysis software (PowerLab, AD Instruments). Ten sequential beats were averaged for each measurement.

For immunostaining of γδT lymphocytes in the murine ischemic heart tissue, we used frozen sections as described elsewhere.²⁷ Cryostat sections (6 μm) were air-dried and then fixed in cold acetone at room temperature for 10 minutes. Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide (Sigma-Aldrich) in PBS for 20 minutes. Blocking with normal goat serum was applied for 1 hour at room temperature. Sections were washed with PBS and then incubated overnight at 4°C with hamster-anti-mouse TCRγδ antibody (5 μg/mL; clone GL-3, BD Biosciences/Pharmingen). After 3 washes with PBS, the sections were incubated with HRP-conjugated secondary antibody (mouse anti-hamster IgG-HRP, Santa Cruz Biotechnology) at room temperature for 4 hours in the dark. After washing with PBS, staining was developed with 3, 3′-diaminobenzidine tetrahydrochloride (Histofine DAB Substrate Kit, Nichirei). To visualize the cardiomyocytes, a sequential section was stained with anti-α-actinin antibody (Sigma-Aldrich).

Frozen sections of the heart samples fixed by 4% paraformaldehyde were subjected to TUNEL staining using a commercially available kit (In Situ Apoptosis Detection Kit; Takara Biomedicals) as directed by the manufacturer. Anti-α-actinin (Sigma-Aldrich) was used to identify cardiomyocytes. TUNEL-positive nuclei were counted, and the data were normalized per total nuclei identified by DAPI staining (Invitrogen) in the same sections.

FTY720 was dissolved at 5 mg/mL in DMSO, and then diluted with distilled water. FTY720 (1 mg/kg body weight) was administered intravenously from day 0 to day 5 post-MI, and leukocyte infiltration into the heart was analyzed by flow cytometry on day 6 post-MI.

Values are presented as mean±SEM. Comparisons between groups were made using a Mann–Whitney U test, whereas data among multiple groups were compared using either the Kruskall–Wallis test with Dunn's multiple comparisons test or 2-way ANOVA followed by Tukey's post hoc analysis, as appropriate. Survival distributions were estimated by the Kaplan–Meier method and compared by the log-rank test. A value of P<0.05 was considered statistically significant. Statistical analysis was performed with GraphPad Prism 5.0 (Graph Pad Prism Software Inc, San Diego, CA) and SPSS 15.0 for Windows (SPSS, Inc, Chicago, IL).

We first investigated the dynamics of IL-23 and IL-17A in post-MI heart. Expression of IL-23p19 subunit mRNA increased rapidly in infarcted heart on day 1 post-MI, returning to near-baseline levels by 3 days post-MI (Figure 1A). In contrast, expression of IL-23 receptor (IL-23R), IL-12/IL-23p40 subunit, RORγt, and IL-17A mRNA increased more gradually, peaking 7 days post-MI. Expression of IL-17 receptor A (IL-17RA) mRNA increased on day 1 post-MI and remained elevated until 14 days post-MI. ELISA analysis showed that the levels of IL-23 protein in post-MI heart increased rapidly to a peak at 24 hours, and then remained substantially elevated until at least day 4 post-MI (Figure 1B).

We next investigated whether IL-23 acts as an upstream regulator of IL-17A in infarcted heart. The leukocyte-enriched fraction was collected from sham-operated and infarcted hearts on day 6 post-MI, and cells were cultured with either vehicle or IL-23 (10 ng/mL) for 24 hours (Figure 2). Expression of IL-17A and IL-21 mRNA was markedly increased by IL-23 treatment in both leukocyte-enriched fractions, whereas expression of tumor necrosis factor (TNF)-α, IL-6, and IL-1β mRNA was unaffected. IL-23 also did not affect RORγt expression, indicating that RORγt is constitutively expressed in a subset of leukocytes.

We then examined the functional significance of the IL-23/IL-17A axis in post-MI cardiac remodeling. On day 1 post-MI, infarct size (determined by TTC staining) and FS and LVEDD (assessed by echocardiography) in IL-17A-KO and IL-23-KO mice were comparable to those of wild-type (WT) mice (Figure 3A through 3C).

Severe anterior MI induced by proximal left coronary artery ligation in mice leads to high mortality because of severe LV dysfunction. Some of mice die from cardiac rupture. The protective effect of IL-23 and IL-17A deficiency on survival became obvious after 7 days. The survival rate on day 28 post-MI was 34.7% (25/72) in WT mice, 65.5% (19/29) in IL-17A-KO mice, and 62.5% (15/24) in IL-23-KO mice (Figure 4A).

Survivors were evaluated for cardiac remodeling on day 28 post-MI. Echocardiographic examination revealed a markedly enlarged heart (LVEDD 6.47±0.25 mm, n=16) with reduced LV systolic function (FS 5.2±1.1%, n=16) in WT mice following MI (Figure 4B), whereas IL-17A-KO and IL-23-KO mice showed significantly less LV enlargement (LVEDD 5.64±0.21 mm, n=16, and 5.61±0.20 mm, n=9, respectively) and less severe LV dysfunction (FS 11.0±1.7%, n=16, and 10.5±1.7%, n=9, respectively). LVESP and maximum and minimum dP/dt, the index of contractility, were higher in IL-17A-KO and IL-23-KO mice compared with WT mice, whereas the ratio of heart to body weight was lower in both IL-17A-KO and IL-23-KO mice than in WT mice (Table 1). The infarct size (infarct circumference/LV circumference) was significantly smaller in IL-17A-KO (35.7±2.0%, n=16) and IL-23-KO (36.3±2.7%, n=9) mice compared with WT mice (46.3±2.3%, n=16) (Figure 4C and 4D). The area of myocardial fibrosis in noninfarcted heart was significantly smaller in IL-17A-KO (0.65±0.22%, n=11) and IL-23-KO (0.70±0.18%, n=9) mice than in WT mice (1.72±0.27%, n=10) (Figure 4E and 4F).

We investigated the major cellular source of the respective cytokines in the infarcted hearts. The leukocyte-rich fraction collected from day 1 post-MI hearts was separated into CD45⁺CD11b⁺F4/80⁺ macrophages, CD45⁺CD11b⁺Ly-6G⁺ neutrophils, and other cells. The macrophages were further divided into 2 groups, CD206^low classically (M₁) and CD206^high alternatively activated (M₂). IL-23p19 mRNA expression was detected in the neutrophils and macrophages (M₁>M₂), but it was much lower in the other cell subsets (Figure 5A).

Intracellular cytokine staining revealed that 90.1±1.2% of IL-17A-expressing cells in the infarcted hearts on day 7 post-MI were CD3⁺ T lymphocytes (Figure 5B). IL-23-KO mice had a significantly smaller proportion of IL-17A-producing cells among CD3⁺ T lymphocytes compared with WT mice (13.6±0.4% versus 2.2±0.2%, n=4, P<0.001), whereas the proportion of IFN-γ-producing cells among CD3⁺ T lymphocytes was not affected in IL-17A-KO or IL-23-KO mice (Figure 5C). More than 90% of IL-17A-producing T cells in this study were CD4⁻ TCRγδ⁺ (γδT) cells, but not CD4⁺ T cells (Th17) (Figure 5D and 5F). By contrast, both CD4⁻ TCRγδ⁺ cells (65.2±2.6%) and CD4⁺ T cells (29.6±1.7%) were predominant cellular sources of IL-17A in the spleen (Figure 5E and 5G). After MI, the number of IL-17A-expressing cells gradually increased to a peak 7 days post-MI and then remained high up to 14 days post-MI (Figure 5H). The γδT cells started to increase on day 1 and peaked on day 7 after MI (Figure 5I). Immunohistochemical examination revealed that γδT cells were mainly in the infarct and border areas, but not in the noninfarct area (Figure 5J and 5K). In addition, TCRγδ⁺ cells were not major sources of IFN-γ in the infarcted heart, but showed a relatively higher contribution of IFN-γ production in spleen (5% in heart versus 10% in spleen), thus splenic γδT cells produced both IL-17A and IFN-γ (Figure 6A through 6E).

We analyzed the impact of deficiency in IL-23 or IL-17A on the cellular infiltrate in the infarcted hearts on day 7 post-MI. The total number of infiltrating CD45⁺ immune cells was not altered in IL-23-KO and IL-17A-KO mice compared with WT mice (Figure 7A). However, when we looked at leukocyte subsets, the number of infiltrating neutrophils was significantly reduced in both IL-23-KO and IL-17A-KO mice compared with WT mice. The number of infiltrating macrophages, DCs, NK cells, and NKT cells was not altered in IL-23-KO and IL-17A-KO mice compared with those in WT mice. Unexpectedly, the number of infiltrating CD4⁺ and CD8⁺ cells in the infarcted heart was higher in IL-23-KO and IL-17A-KO than in WT mice. Notably, the number of infiltrating γδT cells was markedly reduced in IL-23-KO mice but was not altered in IL-17A-KO mice compared with wild types (Figure 7A and 7B).

On day 2 post MI, expression of matrix metalloproteinases (MMP) 1, MMP3, MMP9, CCL2, IL-6, and IL-1β mRNA was not altered in IL-23-KO and IL-17A-KO mice compared with WT mice, whereas TNF-α expression was slightly higher in IL-23-KO mice (Figure 8A). In contrast, on day 7 post MI, expression of MMP1, MMP3, and MMP9 mRNA was significantly lower in both IL-23-KO and IL-17A-KO mice compared with that in WT mice, as was the mRNA expression of the fibrosis-related genes collagen 1, periostin, and TGF-β (Figure 8B). Consistent with this, MMP9 activity as assessed by gelatin zymography was significantly suppressed in both IL-23-KO and IL-17A-KO mice compared with WT mice (Figure 8C and 8D). mRNA expression of CCL2, a chemokine that mediates monocyte/macrophage recruitment, was significantly lower in the IL-23-KO and IL-17A-KO than in the WT mice. Expression of TNF-α, IL-6, and IL-1β mRNA was not altered in IL-23-KO and IL-17A-KO mice compared with that in WT mice (Figure 8B).

We investigated the pathogenic importance of the demonstrated γδT cell response in post-MI remodeling. When we subjected TCRγδ-deficient (TCRγδ-KO) mice to MI, infarct size and LV dysfunction on day 1 post MI were similar to those of WT mice (Figure 3A through 3C). However, the survival rate on day 28 post-MI was significantly improved in TCRγδ-KO mice (63.2% [12/19]) compared with that in WT mice (34.7% [25/72]) (Figure 9A), and this protective effect of γδT cell deficiency on survival became obvious after 7 days.

The post-MI LV remodeling in TCRγδ-KO mice on day 28 post-MI was significantly attenuated compared with that in WT mice, as was LV enlargement and the severity of LV dysfunction (LVEDD 6.56±0.24 versus 5.71±0.21 mm, FS 5.2±1.1% versus 10.7±1.8%, n=10 to 16) (Figure 9B). LVESP and maximum and minimum dP/dt were higher, whereas the heart weight/body weight ratio was lower in TCRγδ-KO mice compared with WT mice (Table 1). Azan staining revealed a reduced infarct size (infarct circumference/LV circumference) in TCRγδ-KO hearts compared with WT mice (46.3±2.3% versus 36.7±2.8%, n=10 to 16) (Figure 9C and 9D), and the area of myocardial fibrosis in noninfarcted heart was significantly smaller in TCRγδ-KO compared with WT mice (0.71±0.21%, n=9, versus 1.72±0.27%, n=10) (Figure 9E and 9F).

The TCRγδ-KO mice showed markedly lower numbers of infiltrating IL-17A-expressing T cells in infarcted hearts on day 7 post-MI compared with WT mice (12.1±0.3% versus 2.5±0.6%, n=4) (Figure 9G and 9I). In contrast, the number of infiltrating γδT cells in IL-17A-KO mice was comparable to that in WT mice in day 7 post-MI heart (Figure 9H and 9J), but there were no IL-17A-expressing cells. Furthermore, TCRγδ-KO mice had a markedly decreased number of CD45⁺-leukocytes including macrophages, T cells, and neutrophils than WT mice (Figure 10A). MMP1, MMP3, MMP9, collagen 1, periostin, TGF-β, and CCL2 were markedly attenuated on day 7 in TCRγδ-KO mice compared with WT mice (Figure 10B), whereas MMP1, MMP3, MMP9, TNF-α, IL-1β, IL-6, and CCL2 expression was comparable between the groups on day 2 post-MI (Figure 10C). Expression of IFN-γ mRNA on day 7 post-MI did not differ between TCRγδ-KO mice and WT mice (Figure 10D).

The immunomodulatory drug FTY720 interferes with sphingosine-1-phosphate (S1P) receptor signaling, leading to sequestration of lymphocytes in lymph nodes but not spleen.³⁰ Infusion of FTY720 significantly suppressed the number of infiltrating T cells, and particularly γδT cells, in the postinfarct heart (Figure 11A through 11C). These findings indicated that cardiac γδT cells migrated from lymph nodes, at least in part via S1P receptor signaling.

Chemokine receptor 6 (CCR6) contributes to the migration of Th17 cells to a particular site of inflammation.³¹ We found that almost all cardiac γδT cells constitutively expressed CCR6, whereas only 27% of spleen γδT cells expressed CCR6 (Figure 11D). Expression of the CCR6 ligand CCL20 was significantly increased on day 1 post-MI, before peaking on day 3 post-MI, and then remaining high to day 14 post-MI (Figure 11E). These findings suggested an important role for the CCL20–CCR6 signaling axis in recruiting γδT cells into infarcted heart.

Engagement of these receptors with their respective ligands could stimulate NF-κB and IL-23 production. TLR1 and TLR2 were also expressed in γδT cells.³² To examine whether TLR2 and TLR4 operate upstream of IL-17A production in the infarcted heart, we analyzed the number of IL-17A-expressing T cells in day 7 post-MI heart from WT, TLR2-KO, TLR4-KO, or TLR2/4-DKO mice using flow cytometry. TLR2-KO, TLR4-KO, and TLR2/4-DKO mice had a significantly smaller proportion of IL-17A-producing cells among CD3⁺ T lymphocytes than did WT mice (15.9±1.4% versus 6.2±0.9%, 5.2±0.5%, and 4.8±1.0%, respectively, n=4 to 7; P<0.001) (Figure 12A and 12B). IL-1β mRNA and protein levels in the heart were also dramatically increased as early as 24 hours after MI and remained significantly elevated above baseline levels thereafter (Figure 2C and 2D). Therefore, we examined the synergistic effect of TLRs ligands and IL-1β acting with IL-23 on the proliferation of and IL-17A production from γδT cells. Cardiac cells prepared from day 7 post-MI hearts were stimulated using lipopolysaccharide (LPS; TLR4 ligand), Pam3CSK4 (TLR1/2 ligand), and IL-1β, each alone and in combination with IL-23. A synergistic effect on IL-17A production was observed when cells were exposed to the combination of IL-23 and LPS or of Pam3CSK4 and IL-1β (Figure 12E and 12F). The synergistic effect of TLR ligands and IL-1β on IL-17A production was further examined using CD45⁺ cells sorted from day 7 post-MI hearts. LPS, Pam3CSK4, or IL-1β alone could not stimulate IL-17A production; however, IL-23 alone was sufficient to drive IL-17A production. A synergistic effect on IL-17A production was observed when CD45⁺ cells were exposed to the combination of IL-23 and IL-1β or of IL-23 and Pam3CSK4, but not to the combination of IL-23 and LPS (Figure 12G). Similarly, IL-17A production from cardiac cells or CD45⁺ cells was significantly suppressed when cells were pretreated with IL-1 receptor I antibody (IL-1RI) (Figure 12F and 12G), suggesting that the IL-1R signaling pathway is essential for IL-17A production in heart.

Examination of the proliferation-promoting effect of LPS, Pam3CSK4, and IL-1β on γδT cells by the CFSE dilution assay showed that the combination of IL-23 and IL-1β had the most pronounced effect on cardiac γδT cell proliferation (Figure 12H and 12I).

Finally, we investigated the mechanisms behind the pathogenic role of IL-17A in post-MI cardiac remodeling. To identify the important cellular targets for IL-17A in vivo, we separated macrophages, lymphocytes, fibroblasts, endothelial cells, and cardiomyocytes from the infarcted hearts on day 7 post-MI. IL-17RA was highly expressed in cardiomyocytes, fibroblasts, and macrophages (Figure 13A and 13B).

To examine the impact of IL-17A on cardiomyocyte survival, we performed TUNEL staining combined with α-actinin staining on the infarcted hearts of WT and IL-17A-KO mice on both day 2 and day 7 post-MI. Ablation of IL-17A had no effect on the number of TUNEL-positive cardiomyocytes in the border regions on day 2 (9.0±2.1% and 6.8±1.2%, respectively, n=6). Although there were markedly fewer TUNEL-positive cardiomyocytes on day 7 post-MI than on day 2 in WT mice, ablation of IL-17A further decreased the number of survivors (WT, 1.2±0.2%, versus IL-17A-KO, 0.5±0.1%, n=6) (Figure 13C). Consistent with previous reports,^22,23 IL-17A significantly augmented low serum/hypoxia-induced cell death in cultured murine cardiomyocytes, and this cytotoxic effect of IL-17A could be completely rescued by neutralizing anti-IL-17A antibody (Figure 14A).

Next we investigated the effect of IL-17A on cardiac fibroblasts. Real-time PCR analysis revealed that sorted fibroblasts from the infarcted hearts on day 7 post-MI had significantly lower mRNA levels of MMP3, MMP9, profibrotic genes (TGF-β, collagen 1, collagen 3, and periostin), and chemokines (CCL2, CXCL1) in IL-17A-KO mice compared with WT mice (Figure 13D). Expression of TNF-α was not altered. In culture, IL-17A increased murine cardiac fibroblast proliferation in a dose-dependent manner, and this proliferative effect of IL-17A was completely abolished by neutralizing anti-IL-17A antibody (Figure 14B). Furthermore, IL-17A stimulated the expression of proinflammatory cytokines (TNF-α, IL-6, and IL-β), MMPs (MMP1, -3, -9), profibrotic genes (TGF-β, collagen 1, collagen 3, and periostin), and chemokines (CCL2, CXCL1) in cultured cardiac fibroblasts, whereas almost all this IL-17A-stimulated gene induction was significantly abrogated by neutralizing anti-IL-17A antibody (Figure 14C).

We also examined the gene expression profiles in macrophages isolated from the infarcted hearts on day 7 post-MI (Figure 13E). Expression of M₁ macrophage signature genes such as TNF-α, IL-6, IL-1β, CCL2, and CXCL1 in the sorted macrophages were dramatically suppressed in the IL-17A-KO mice compared with WT mice, while those of M₂ signature gene YM-1 was not altered. The mRNA levels of MMP1, MMP9, and TGF-β in the sorted macrophages were also lower in the IL-17A-KO mice than in WT cells. In the cultured monocyte/macrophage cell line RAW264.7, IL-17A significantly increased mRNA expression of M₁ macrophage signature genes, such as TNF-α, IL-6, IL-1β, CCL2, MMP9, and CXCL1, but did not alter those of the M₂ signature gene arginase-1 (Figure 14D), and these gene inductions in RAW264.7 cells were blocked by neutralizing anti-IL-17A antibody. The above results strongly suggested that IL-17A has proapoptotic, profibrotic, and proinflammatory properties that combine to promote LV remodeling.