TY - JOUR
AU - Fujio,, Yasushi
AB - Abstract Aims Accumulating evidence demonstrates that cardiomyocyte death contributes to the onset and progression of heart failure (HF) after myocardial injury. Recent studies revealed that immune/inflammatory reactions play important roles in cardiovascular diseases. However, it remains unclear whether immunosurveillance system, which eliminates cytopathic cells, including infected or malignant cancer cells, is involved in cardiomyocyte death, though cardiomyocytes are exposed to pathological stresses during post-infarct remodelling. The aim of this study is to clarify the pathophysiological significance of Natural Killer Group 2 member D (NKG2D)/NKG2D ligand (NKG2DL)-mediated cell death in HF after myocardial infarction (MI). Methods and results MI was generated by ligating left anterior descending artery in mice. The expression of NKG2D, NKG2DLs, especially Retinoic acid early induced transcript-1ɛ (Rae-1ɛ), perforin and granzyme B was concomitantly up-regulated after MI. Immunohistological analysis revealed that Rae-1 was expressed on the membranes of injured cardiomyocytes in the infarct and border area. The MI-induced increase of Rae-1 expression was suppressed in p53−/− mice and Rae-1 was induced by the overexpression of p53. We identified p53-binding sites in Rae-1ɛ gene promoter, by chromatin immunoprecipitation assay, indicating that Rae-1 expression was mediated partially through p53. Flow cytometric analysis indicated that NKG2D-expressing immune cells in the post-infarct myocardium were mainly γδT cells. The co-culture with γδT cells increased the frequency of apoptotic cells in the cultured cardiomyocytes. The blockade of NKG2D/NKG2DL interaction by intraperitoneal injection of anti-Rae-1ɛ antibody after MI reduced the frequency of apoptotic cardiomyocytes, accompanied by suppression of cardiac fibrosis, attenuating cardiac dysfunction. Finally, tamoxifen-inducible cardiomyocyte-specific Rae-1ɛ overexpressing mice exhibited the susceptibility to post-infarct remodelling with increased cardiomyocyte apoptosis and severer cardiac dysfunction. Conclusion The interaction between immune cells and cardiomyocytes via NKG2D/NKG2DL induces cardiomyocyte death, exacerbating cardiac remodelling after MI. The blockade of NKG2D/NKG2DL interaction could be a promising therapeutic strategy against HF. Myocardial infarction, Cardiac remodelling, Immunosurveillance system, NKG2D/NKG2DL, Cell death 1. Introduction Myocardial infarction (MI) is a leading cause of heart failure (HF). Despite the remarkable advances in the interventional devices, the development of novel therapeutic strategies against HF after MI is an urgent necessity because of the poor prognosis. Although cardiomyocyte death is recognized as one of the provocative factors of HF, the mechanism is still unclear. Generally, it has been accepted that cardiomyocyte death is induced in a cell-autonomous manner by reactive oxygen species (ROS) production or Ca2+ overload, which activates intracellular death signal cascades.1–3 Recently, it has been reported that immune/inflammatory reactions contribute to cardiovascular diseases, proposing that regulating immune system may be a promising therapeutic strategy against HF. At the acute phase after MI, macrophages and neutrophils infiltrate into myocardium and evoke inflammatory reactions, leading to cardiac remodelling at the chronic phase.4,5 However, it remains to be elucidated how immune cells induce cardiomyocyte death as a cell non-autonomous process. Immunosurveillance was originally conceptualized as an anti-cancer system. This system works to eliminate cytopathic cells, such as infected or malignant cancer cells via the interaction between Natural Killer Group 2 member D (NKG2D) and its ligands (NKG2DLs). NKG2D is an activating receptor on cytotoxic immune cells, including NK cells, NKT cells, or γδT cells. Regarding the ligands, eight human ligands and nine mouse ligands have been identified and these are up-regulated under stressed conditions. When NKG2D binds to its ligands, immune cells are activated and cause apoptotic cell death in the target cells by releasing perforin and granzyme B.6–8 So far, the involvement of this system in HF after MI has not been addressed, though cardiomyocytes are exposed to pathological stresses in the post-infarct myocardium. In this study, we pursued a novel mechanism of cardiomyocyte death in terms of the interaction between injured cardiomyocytes and immune cells, focusing on immunosurveillance system. To clarify the pathophysiological significance of NKG2D/NKG2DL system in post-infarct cardiac remodelling, we employed murine coronary ligation model. Rae-1ɛ, one of NKG2DLs, was remarkably induced through p53 on the membranes of injured cardiomyocytes. NKG2D-expressing γδT cells infiltrated into the post-infarct myocardium. The inhibition of NKG2D/NKG2DL interaction by anti-Rae-1ɛ antibody administration showed cardioprotective effects, such as suppression of cardiomyocyte apoptosis and attenuation of cardiac fibrosis and dysfunction. Moreover, we confirmed the deterioration of pathological states in conditional cardiomyocyte-specific Rae-1ɛ overexpressing mice. This is the first demonstration that NKG2D/NKG2DL system contributes to cardiomyocyte death, proposing that the blockade of NKG2D/NKG2DL interaction could have the potential as a novel therapeutic strategy against HF. 2. Methods 2.1 Animal experiments Animal care was performed according to animal care guidelines of Osaka University and RIKEN Kobe Branch. All the animal experiments in this study conformed to the Guide for the Care and Use of Laboratory Animals, Eighth Edition, updated by the US National Research Council Committee in 2011 and were approved by Animal Experimentation Committee of Osaka University and Institutional Animal Care and Use Committee of RIKEN Kobe Branch. C57BL/6J male mice (8–12-week-old, 25–30 g body weight) were purchased from Japan SLC. p53−/− mice were purchased from RIKEN BRC.9 At the endpoints of all experiments, mice were anaesthetized by isoflurane and the hearts were harvested. All efforts were made to minimize suffering. Cardiomyocyte-specific transgenic (TG) mice were established by generating CAG/CATZ/Rae-1ɛ mice (Accession No. CDB0531T: http://www2.clst.riken.jp/arg/TG%20mutant%20mice%20list.html).10 We analysed two lines of transgenic mice with similar results. CAG/CATZ/Rae-1ɛ mice were mated with α-MHC-MerCreMer mice, generously gifted by Dr Jeffery Molkentin, Cincinnati Children’s Hospital Medical Center and Howard Hughes Medical Institute. Cardiomyocyte-specific Rae-1ɛ conditional overexpression was achieved by intraperitoneal injection of tamoxifen (8 µg/g/day) for 14 consecutive days before MI operation. MI was generated in mice by the occlusion of left anterior descending artery as described previously.11 The investigators were blinded to the identity of the mice in the experiments using TG mice. In the experiments using Rae-1ɛ blocking antibody, mice subjected to MI were randomly assigned to three groups, non-treatment group, control IgG group and blocking antibody group, by the researcher who were blinded to the process of MI operation. Phosphate buffered saline (PBS), rat control IgG or anti-Rae-1ɛ antibody (100 µg) was intraperitoneally administered 1 day after MI for TUNEL staining, 1 and 7 days after MI for other analyses. 2.2 Quantitative RT-PCR Quantitative RT-PCR was performed according to manufacturer’s protocol. Total RNA samples were isolated from homogenized heart tissues at various time points after MI or cultured cardiomyocytes, using QIAzol Lysis Reagent (QIAGEN). After complementary DNA was synthesized from the total RNA, the expression of Rae-1, NKG2D, p53, perforin, and granzyme B mRNA was measured by real-time RT-PCR using the SYBR green system (Applied Biosystems). The expression of GAPDH mRNA was also measured as an internal standard. The primers used in this study are listed (Supplementary material online, Table S1). 2.3 Immunohistochemistry Hearts were harvested 3 days after MI and frozen in O.C.T. Compound (Sakura Finetek Japan). The frozen blocks were sliced into sections at 5 μm-thickness using Leica CM 1950 (Leica). After fixation of the sections with 4% paraformaldehyde (PFA) for 15 min and blocking with 3% bovine serum albumin (BSA) in PBS containing 0.1% Triton X-100, the sections were stained with Avidin-Biotin-Complex (ABC) method. Nuclei and cytoplasm were also stained with haematoxylin and eosin, respectively. In immunofluorescence staining, isolated cardiomyocytes were fixed with 4% PFA for 15 min, followed by permeabilization. Primary antibodies and Alexa Fluor 488- or 546-conjugated secondary antibodies were used at 1/400. Nuclei were stained with Hoechst. Images were taken by FSX-100 (Olympus). The antibodies used in this study are listed (Supplementary material online, Table S2). 2.4 Histopathology Frozen heart sections (5 µm thickness) from mice 14 days after MI were fixed with acetone for 15 min, followed by Masson’s trichrome staining. The area or circumference ratio of the fibrotic region to left ventricle was measured with Image J software (National Institutes of Health). Infarct wall thickness was measured perpendicular to the infarcted wall at three different regions and averaged. 2.5 Cardiomyocyte isolation from post-MI mice Cardiomyocytes were isolated as described previously.12 Briefly, hearts were perfused and digested with the solution containing 0.05% collagenase B (Roche), 0.05% collagenase D (Roche) and 0.04% proteinase XIV (Sigma). Cell suspension was filtrated through a 200 μm cell strainer, followed by the filtration through a 40 μm cell strainer. Cells trapped on the latter filter were used as isolated cardiomyocytes. Purity of the cardiomyocytes was estimated as 70–80%. 2.6 Western blotting Western blotting was performed as described previously.13 The antibodies used in this study are listed (Supplementary material online, Table S2). 2.7 Chromatin immunoprecipitation (ChIP) assay Heart tissue harvested from mice 3 days after MI was used for ChIP assay. ChIP assay was performed with Simple ChIP Kit (Cell Signaling Technology) according to manufacturer’s protocol. The primers used for ChIP assay are listed (Supplementary material online, Table S3).14 2.8 Flow cytometric analysis On Day 4 and 7 after MI, mice were intraperitoneally injected with heparin sodium (50U/mouse) (WAKO). Then the infarcted hearts were harvested and perfused with PBS and 0.025% collagenase solution [0.0125% collagenase B, 0.0125% collagenase D and 0.002% proteinase XIV]. After the treatment, the hearts were minced into pieces and digested with 0.1% collagenase solution [0.05% collagenase B, 0.05% collagenase D, and 0.002% proteinase XIV] at 37°C. The cells were filtered through a 70 μm cell strainer, suspended in PBS supplemented with 3% foetal bovine serum (FBS) and filtered through a 40 μm cell strainer. The harvested cells were stained with PE or FITC-conjugated antibodies for flow cytometry. Flow cytometric analysis was performed on FACS Aria II (BD Biosciences) and analysed using BD FACS Diva software (BD Biosciences). The antibodies used in this study are listed (Supplementary material online, Table S2). 2.9 Cell culture Neonatal rat cardiomyocytes were prepared as previously described.13 Twenty-four hours after the primary culture, cardiomyocytes were incubated with adenovirus vector with the overexpression of p53 at 10 multiplicity of infection (MOI) for 24 h to analyse the mechanisms of the up-regulation of Rae-1 expression. In in vitro cell death assay, cardiomyocytes were co-cultured with isolated γδT cells for 24 h after they were infected with adenovirus vector expressing Rae-1ɛ or β-gal. 2.10 Adenovirus Adenovirus vector expressing Rae-1ɛ was generated as previously described.13 Cultured neonatal rat cardiomyocytes were infected with adenovirus vector expressing Rae-1ɛ or β-gal, a control, at 10 MOI. Adenovirus vector expressing p53 was a generous gift from Dr Yasuko Bando.15 2.11 Percoll method and magnetic activated cell sorting To isolate γδT cells, we utilized percoll method and magnetic activated cell sorting (MACS) with a Cell Separation Magnet (BD Biosciences). Spleens were crushed and passed through a 70 μm cell strainer in PBS containing 3% FBS. Cells were treated with Lysing buffer (BD Bioscience) to remove erythrocytes and suspended on 40/80% percoll buffer layers (GE Healthcare). After centrifugation for 25 min, the leucocyte-rich fraction was collected from the interface between 40% and 80% percoll buffer. After the blocking of Fc receptors, cells were incubated with a biotinylated antibody, resuspended in BD IMag™ Buffer (BD Biosciences), and incubated with Streptavidin Particles Plus-DM (BD Biosciences). Beads-tagged γδT cells were magnetically collected. 2.12 TUNEL staining Apoptotic cardiomyocytes were detected with In Situ Apoptosis Detection Kit (Takara) according to manufacturer’s protocol. In brief, after frozen heart sections were fixed with 4% PFA for 15 min and blocked with 3% BSA, DNA nicks were labelled. To identify cardiomyocytes, sections were also co-stained with anti-α-MHC antibody. Alexa Fluor 546-conjugated goat anti-rabbit IgG were used as a secondary antibody. The cells positively stained with TUNEL and α-MHC were counted in number as apoptotic cardiomyocytes and the number of apoptotic cardiomyocytes per mm2 was estimated. 2.13 Echocardiography Cardiac function was assessed with an iE33 model equipped with a 15-MHz transducer (Philips) under isoflurane anaesthesia. Echocardiography was performed in M-mode. Left ventricular (LV) internal dimension at diastole (LVIDd), LV internal dimension at systole (LVIDs), fractional shortening (FS), and heart rate of mice were estimated. The investigator was blinded to the identity of the mice for analysis. 2.14 Statistics Data were shown as mean ± standard deviation. Comparison between two groups was performed with Student’s t-test. One-way ANOVA followed by Dunnet or Tukey–Kramer test was used for multiple comparisons. P < 0.05 was considered to achieve statistical significance. 3. Results 3.1 The expression of Rae-1 was up-regulated in injured cardiomyocytes after MI After C57BL/6J mice were subjected to MI operation, the myocardial expression of murine NKG2DL transcripts was analysed. Among three NKG2DL families, including Rae-1, MULT1, and H60, the up-regulation of Rae-1 expression was the most remarkable (Supplementary material online, Figure S1). The expression of Rae-1 mRNA was measured by quantitative RT-PCR at various time points. Rae-1 mRNA was up-regulated by 24.7 ± 8.6 folds with its peak 3–4 days after MI (Figure 1A). Rae-1 family consists of five isoforms, Rae-1α to Rae-1ɛ. Among them, Rae-1δ and Rae-1ɛ are expressed in C57BL/6J mice, while Rae-1α, β, and γ in Balb/c mice. To identify which isoform was induced in the post-infarct myocardium, we performed RT-PCR using ligand-specific primers and found that both isoforms were increased after MI (Figure 1B). In further study, we focused on Rae-1ɛ because Rae-1ɛ has a much higher affinity for NKG2D.8 Then, we stained cardiac tissue with anti-Rae-1 antibody. Immunohistochemical staining revealed that Rae-1 protein was localized on the edge of cracked tissue in border and infarct area (Figure 1C). Furthermore, the triple-staining experiments with anti-Rae-1 and anti-α-actinin antibodies, and Hoechst revealed that cardiomyocytes, isolated from the infarcted hearts, expressed Rae-1 (Figure 1D). The proportion of cardiomyocytes expressing Rae-1 was as follows; non-MI: 0%, MI3d: 13.6%. Figure 1 View largeDownload slide Rae-1 expression was up-regulated in injured cardiomyocytes after MI. (A and B) The up-regulation of Rae-1 mRNA expression after MI. Mice were subjected to MI operation, and total RNA was prepared from the murine post-infarct hearts. (A) The expression of Rae-1 transcripts was measured by quantitative RT-PCR and normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI3d, n = 6 mice for MI1d, 14d, n = 7 mice for non-MI, n = 8 mice for MI4d, n = 9 mice for MI7d). *P < 0.05 vs. non-MI by one-way ANOVA followed by Dunnet test. (B) The expression of Rae-1 isoforms, Rae-1δ, and Rae-1ε, was analysed by RT-PCR. Representative data are shown (n = 5 mice for each time point). (C and D) Immunohistochemical analysis of Rae-1 expression. (C) Murine heart sections were prepared 3 days after MI. The sections were stained with anti-Rae-1 antibody. Representative images are shown (Bar; 50 μm) (n = 5 mice for non-MI and MI3d). (D) Cardiomyocytes were isolated from infarcted heart tissue by Langendorff reflux apparatus, followed by co-staining with anti-Rae-1 and anti-α-actinin antibodies, and Hoechst. Representative images are shown (Bar; 20 μm) (n = 5 mice for non-MI and MI3d). Figure 1 View largeDownload slide Rae-1 expression was up-regulated in injured cardiomyocytes after MI. (A and B) The up-regulation of Rae-1 mRNA expression after MI. Mice were subjected to MI operation, and total RNA was prepared from the murine post-infarct hearts. (A) The expression of Rae-1 transcripts was measured by quantitative RT-PCR and normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI3d, n = 6 mice for MI1d, 14d, n = 7 mice for non-MI, n = 8 mice for MI4d, n = 9 mice for MI7d). *P < 0.05 vs. non-MI by one-way ANOVA followed by Dunnet test. (B) The expression of Rae-1 isoforms, Rae-1δ, and Rae-1ε, was analysed by RT-PCR. Representative data are shown (n = 5 mice for each time point). (C and D) Immunohistochemical analysis of Rae-1 expression. (C) Murine heart sections were prepared 3 days after MI. The sections were stained with anti-Rae-1 antibody. Representative images are shown (Bar; 50 μm) (n = 5 mice for non-MI and MI3d). (D) Cardiomyocytes were isolated from infarcted heart tissue by Langendorff reflux apparatus, followed by co-staining with anti-Rae-1 and anti-α-actinin antibodies, and Hoechst. Representative images are shown (Bar; 20 μm) (n = 5 mice for non-MI and MI3d). 3.2 Rae-1 was induced by p53 activation Since inflammation is closely associated with cardiac remodelling, we investigated whether proinflammatory cytokines up-regulated Rae-1 expression. Cultured rat neonatal cardiomyocytes were stimulated with proinflammatory cytokines, such as TNF-α, IL-1β, or LPS. Though cultured rat neonatal cardiomyocytes expressed Rae-1 mRNA even under unstimulated condition, the expression level of Rae-1 was not altered by these stimuli (Supplementary material online, Figure S2). As pharmacological activation of p53 up-regulates the expression of ULPB2, one of the human NKG2D ligands,16 we examined whether p53 activation is essential for Rae-1 induction, using p53−/− mice. As previously reported,17 myocardial expression of p53 transcript was increased by 5.7 ± 2.7 folds after MI (Figure 2A). Importantly, Rae-1 up-regulation after MI was significantly suppressed by 54.7 ± 16.9% in p53−/− mice, compared with wild-type mice (Figure 2B and C). Next, cultured rat neonatal cardiomyocytes were infected with adenovirus vector expressing p53 and Rae-1 expression was analysed. Interestingly, p53 overexpression resulted in the up-regulation of Rae-1, as is the case with that of p21, a well-known downstream molecule of p53 (Figure 2D–F). Furthermore, to address whether p53 could directly activate Rae-1 transcription, we searched for p53-binding consensus sequences in the Rae-1ɛ promoter region and found three candidate sites, designated as site 1–3. ChIP assay demonstrated that p53 directly bound to sites 1 and 2 in the post-infarct myocardium (Figure 2G). Figure 2 View largeDownload slide Rae-1 expression was regulated by p53. (A) The induction of p53 transcripts after MI. Post-infarct hearts were collected and quantitative RT-PCR was performed to measure p53 mRNA level. The expression of the transcripts was normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI14d, n = 6 mice for non-MI, MI1d, MI3d, n = 7 mice for MI4d, MI7d). **P < 0.01 vs. non-MI by one-way ANOVA followed by Dunnet test. (B and C) The reduced expression of Rae-1 mRNA in p53−/− mice. Wild type (WT) mice or p53−/− mice were exposed to MI operation. (B) The lysates from infarcted hearts 3 days after MI were immunoblotted with anti-p53 and anti-GAPDH antibodies. Representative results are shown (n = 6 for p53−/− mice, n = 7 for WT mice). (C) Three days after MI operation, total RNA was prepared from WT or p53−/− mice and Rae-1 mRNA level was measured by quantitative RT-PCR. The expression of Rae-1 transcripts was normalized with that of GAPDH. Data are shown as mean ± SD (n = 6 for p53−/− mice, n = 7 for WT mice) (right) *P < 0.05 vs. WT by Student’s t-test. (D–F) The up-regulation of Rae-1 mRNA by p53 adenovirus infection. (D) The lysates from cultured cardiomyocytes were immunoblotted with anti-p53 and anti-GAPDH antibodies. Representative results are shown (n = 3 for control and p53 adenovirus group). (E, F) The expression of Rae-1 and p21 transcripts was measured by quantitative RT-PCR and normalized with that of GAPDH. Data are shown as mean ± SD (n = 3 for each condition). *P < 0.05, **P < 0.01 by one-way ANOVA followed by Tukey Kramer test. (G) Identification of p53-binding sites in the promoter region of Rae-1ε gene. Database search pointed out 3 putative binding sites of p53 in Rae-1ε gene. Heart tissue was harvested from mice 3 days after MI and ChIP assay was performed using anti-p53 and anti-phospho-histone antibodies and non-immune IgG in triplicate. PCR data are shown. The primers for p21 gene were used as positive control (n = 3 for each condition). Figure 2 View largeDownload slide Rae-1 expression was regulated by p53. (A) The induction of p53 transcripts after MI. Post-infarct hearts were collected and quantitative RT-PCR was performed to measure p53 mRNA level. The expression of the transcripts was normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI14d, n = 6 mice for non-MI, MI1d, MI3d, n = 7 mice for MI4d, MI7d). **P < 0.01 vs. non-MI by one-way ANOVA followed by Dunnet test. (B and C) The reduced expression of Rae-1 mRNA in p53−/− mice. Wild type (WT) mice or p53−/− mice were exposed to MI operation. (B) The lysates from infarcted hearts 3 days after MI were immunoblotted with anti-p53 and anti-GAPDH antibodies. Representative results are shown (n = 6 for p53−/− mice, n = 7 for WT mice). (C) Three days after MI operation, total RNA was prepared from WT or p53−/− mice and Rae-1 mRNA level was measured by quantitative RT-PCR. The expression of Rae-1 transcripts was normalized with that of GAPDH. Data are shown as mean ± SD (n = 6 for p53−/− mice, n = 7 for WT mice) (right) *P < 0.05 vs. WT by Student’s t-test. (D–F) The up-regulation of Rae-1 mRNA by p53 adenovirus infection. (D) The lysates from cultured cardiomyocytes were immunoblotted with anti-p53 and anti-GAPDH antibodies. Representative results are shown (n = 3 for control and p53 adenovirus group). (E, F) The expression of Rae-1 and p21 transcripts was measured by quantitative RT-PCR and normalized with that of GAPDH. Data are shown as mean ± SD (n = 3 for each condition). *P < 0.05, **P < 0.01 by one-way ANOVA followed by Tukey Kramer test. (G) Identification of p53-binding sites in the promoter region of Rae-1ε gene. Database search pointed out 3 putative binding sites of p53 in Rae-1ε gene. Heart tissue was harvested from mice 3 days after MI and ChIP assay was performed using anti-p53 and anti-phospho-histone antibodies and non-immune IgG in triplicate. PCR data are shown. The primers for p21 gene were used as positive control (n = 3 for each condition). 3.3 NKG2D-expressing immune cells infiltrated into heart tissue after MI Since NKG2DLs were induced in cardiomyocytes after MI, we next analysed the cellular dynamics of NKG2D-expressing immune cells. Quantitative RT-PCR analysis revealed that NKG2D mRNA was increased by 12.2 ± 11.0 folds in heart tissue with its peak 7 days after MI (Figure 3A). Since NKG2D is expressed on four types of immune cells, NK, NKT, γδT, or CD8+ NKT-like cells, we explored the immune cells that expressed NKG2D in the post-infarct myocardium by flow cytometry. Consequently, γδT cells, but not NK, NKT, or CD8+ NKT-like cells that express CD49b as a cell surface marker, were identified as NKG2D-expressing immune cells (Figure 3B). We also performed quantitative RT-PCR to measure perforin and granzyme B mRNA because they are released from activated NKG2D-expressing immune cells in immunosurveillance system. The transcripts of perforin and granzyme B were up-regulated by 2.2 ± 0.5 folds and 23.9 ± 12.5 folds, respectively, with the similar time-course with Rae-1 and NKG2D mRNA in the post-infarct myocardium (Figure 3C and D). Moreover, by RT-PCR analysis using γδT cells purified with percoll method and MACS, we showed that the transcripts of perforin and granzyme B were expressed in γδT cell fraction after MI, while only negligibly, in samples before purification (Figure 3E). These results suggest that NKG2D-expressing cells in the post-infarct myocardium are mainly composed of γδT cells. Figure 3 View largeDownload slide NKG2D-expressing γδT cells infiltrated into the post-infarct heart. (A) The expression of NKG2D mRNA in the post-infarct myocardium. Total RNA was prepared from the murine post-infarct hearts. The expression of NKG2D transcripts was measured by quantitative RT-PCR and normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI3d, n = 6 mice for MI1d, n = 7 mice for non-MI, MI14d, n = 8 mice for MI4d, MI7d). *P < 0.05 vs. non-MI by one-way ANOVA followed by Dunnet test. (B) Identification of NKG2D-expressing cells by flow cytometry. The infiltrated cells were isolated from infarcted hearts of mice 4 and 7 days after MI. The harvested cells were stained with PE-conjugated anti-CD49b or PE-conjugated anti-γδTCR and FITC-conjugated anti-NKG2D antibodies for flow cytometry. Representative images are shown (n = 3 for each time point). (C and D) The increased expression of perforin and granzyme B. Quantitative RT-PCR was performed for perforin and granzyme B at indicated time points after MI. The expression of the transcripts was normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI14d, n = 6 mice for non-MI, MI1d, MI3d, MI4d, MI7d). *P < 0.05, **P < 0.01 vs. non-MI by one-way ANOVA followed by Dunnet test. (E) Perforin and granzyme B were produced in γδT cells. Four days after MI, γδT cells were prepared from hearts. The expression of perforin and granzyme B transcripts in the samples before and after γδT cell purification was analysed by RT-PCR. Representative data are shown (n = 4 for each condition). Figure 3 View largeDownload slide NKG2D-expressing γδT cells infiltrated into the post-infarct heart. (A) The expression of NKG2D mRNA in the post-infarct myocardium. Total RNA was prepared from the murine post-infarct hearts. The expression of NKG2D transcripts was measured by quantitative RT-PCR and normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI3d, n = 6 mice for MI1d, n = 7 mice for non-MI, MI14d, n = 8 mice for MI4d, MI7d). *P < 0.05 vs. non-MI by one-way ANOVA followed by Dunnet test. (B) Identification of NKG2D-expressing cells by flow cytometry. The infiltrated cells were isolated from infarcted hearts of mice 4 and 7 days after MI. The harvested cells were stained with PE-conjugated anti-CD49b or PE-conjugated anti-γδTCR and FITC-conjugated anti-NKG2D antibodies for flow cytometry. Representative images are shown (n = 3 for each time point). (C and D) The increased expression of perforin and granzyme B. Quantitative RT-PCR was performed for perforin and granzyme B at indicated time points after MI. The expression of the transcripts was normalized with that of GAPDH. Data are shown as mean ± SD (n = 5 mice for MI14d, n = 6 mice for non-MI, MI1d, MI3d, MI4d, MI7d). *P < 0.05, **P < 0.01 vs. non-MI by one-way ANOVA followed by Dunnet test. (E) Perforin and granzyme B were produced in γδT cells. Four days after MI, γδT cells were prepared from hearts. The expression of perforin and granzyme B transcripts in the samples before and after γδT cell purification was analysed by RT-PCR. Representative data are shown (n = 4 for each condition). 3.4 NKG2D-expressing γδT cells induced cardiomyocyte apoptosis To address the biological significance of the interaction between cardiomyocytes and γδT cells, we co-cultured neonatal rat cardiomyocytes with γδT cells for 24 h and estimated apoptotic cardiomyocytes. Overexpression of Rae-1ɛ with adenovirus vector was confirmed by immunostaining (Figure 4A). Using percoll method and MACS, γδT cells were successfully purified, estimated by flow cytometry (Figure 4B). The co-culture with γδT cells increased the frequency of apoptotic cardiomyocytes by 6.1 ± 5.7% in non-transfected cardiomyocytes, because neonatal rat cultured cardiomyocytes expressed Rae-1 (Supplementary material online, Figure S2). Importantly, the induction of apoptosis by γδT cells were remarkably enhanced by 16.6 ± 6.9% in the cardiomyocytes infected with adenovirus vector expressing Rae-1ɛ, compared with those with β-galactosidase adenovirus, a control vector (Figure 4C and D). These findings indicated that interaction between cardiomyocytes and γδT cells via NKG2D/NKG2DL could induce cardiomyocyte apoptosis. Figure 4 View largeDownload slide Apoptotic cell death was induced in cultured rat cardiomyocytes by co-culture with γδT cells. (A) Adenoviral overexpression of Rae-1ε. Cultured neonatal rat cardiomyocytes were infected with adenovirus vector expressing Rae-1ε or β-gal, a control, at 10 MOI. Virus-infected cardiomyocytes were immunostained with anti-Rae-1ε antibody. Representative images are shown (Bar; 200 μm) (n = 4 for each condition). (B) Preparation of γδT cells by percoll method and MACS. γδT cells were isolated by percoll method and MACS. Purity of γδT cells was confirmed by flow cytometry. Representative images are shown (n = 4 for each condition). (C and D) Induction of cardiomyocyte apoptosis by co-culture with γδT cells. Neonatal rat cardiomyocytes transfected with adenovirus vector expressing Rae-1ε or β-gal, a control, followed by co-culture with γδT cells for 24 h. TUNEL staining was performed. Cells were also co-stained with anti-α-MHC antibody and Hoechst to identify cardiomyocyte nuclei. (C) Representative images are shown (Bar; 50 μm). (D) TUNEL positive cardiomyocytes were counted (α-MHC, Hoechst, and TUNEL merged cells). Data are shown as mean ± SD (n = 5 for each condition). **P < 0.01 by one-way ANOVA followed by Tukey Kramer test. Figure 4 View largeDownload slide Apoptotic cell death was induced in cultured rat cardiomyocytes by co-culture with γδT cells. (A) Adenoviral overexpression of Rae-1ε. Cultured neonatal rat cardiomyocytes were infected with adenovirus vector expressing Rae-1ε or β-gal, a control, at 10 MOI. Virus-infected cardiomyocytes were immunostained with anti-Rae-1ε antibody. Representative images are shown (Bar; 200 μm) (n = 4 for each condition). (B) Preparation of γδT cells by percoll method and MACS. γδT cells were isolated by percoll method and MACS. Purity of γδT cells was confirmed by flow cytometry. Representative images are shown (n = 4 for each condition). (C and D) Induction of cardiomyocyte apoptosis by co-culture with γδT cells. Neonatal rat cardiomyocytes transfected with adenovirus vector expressing Rae-1ε or β-gal, a control, followed by co-culture with γδT cells for 24 h. TUNEL staining was performed. Cells were also co-stained with anti-α-MHC antibody and Hoechst to identify cardiomyocyte nuclei. (C) Representative images are shown (Bar; 50 μm). (D) TUNEL positive cardiomyocytes were counted (α-MHC, Hoechst, and TUNEL merged cells). Data are shown as mean ± SD (n = 5 for each condition). **P < 0.01 by one-way ANOVA followed by Tukey Kramer test. 3.5 Blockade of NKG2D/NKG2DL interaction suppressed cardiomyocyte apoptosis after MI To assess the significance of NKG2D/NKG2DL interaction in adverse cardiac remodelling in vivo, we inhibited the interaction between NKG2D and Rae-1ɛ using anti-Rae-1ɛ blocking antibody. At Day 1 after MI operation, anti-Rae-1ɛ blocking antibody was intraperitoneally injected into mice and the hearts were harvested on the third post-operative day. Rat IgG was used as a control and PBS was injected into mice in non-treatment group. To evaluate the proportion of apoptotic cardiomyocytes, TUNEL staining was performed. Immunofluorescent micrographic analysis showed that TUNEL positive, apoptotic cardiomyocytes were mainly detected in border area.18 Importantly, the number of apoptotic cardiomyocytes was decreased in anti-Rae-1ɛ antibody-treated group, compared with control IgG group (control IgG 9.1 ± 2.9 cells/mm2, anti-Rae-1ɛ antibody 3.5 ± 2.1 cells/mm2) (Figure 5A and B). Figure 5 View largeDownload slide Blockade of NKG2D/NKG2DL interaction showed cardioprotective effects after MI. (A and B) Inhibitory effects of anti-Rae-1ε antibody on cardiomyocyte apoptosis after MI. Anti-Rae-1ε antibody or rat IgG (100 µg) was intraperitoneally administered 1 day after MI. Murine heart sections were prepared 3 days after MI. TUNEL staining was performed. (A) Representative images are shown (Bar; 50 μm). (B) TUNEL positive cardiomyocytes were counted (α-MHC, Hoechst, and TUNEL merged cells). Data are shown as mean ± SD (n = 5 for non-treatment, rat IgG, and anti-Rae-1ε antibody group). *P < 0.05 vs. rat IgG by one-way ANOVA followed by Tukey Kramer test. (C–F) Inhibitory effects of anti-Rae-1ε antibody on post-infarct cardiac remodelling. Anti-Rae-1ε antibody or rat IgG (100 µg) was intraperitoneally administered 1 and 7 days after MI. Murine heart sections were prepared and Masson’s trichrome staining was performed 14 days after MI. (C) Representative images are shown (Bar; 1 mm). The ratio of fibrotic area to LV area (D), the ratio of fibrotic circumference to LV circumference (E) and infarct wall thickness (F) were quantitatively estimated. Data are shown as mean ± SD (n = 8 mice for anti-Rae-1ε antibody administration group, n = 11 mice for non-treatment group, n = 12 mice for rat IgG administration group). *P < 0.05, **P < 0.01 vs. rat IgG by one-way ANOVA followed by Tukey Kramer test. Figure 5 View largeDownload slide Blockade of NKG2D/NKG2DL interaction showed cardioprotective effects after MI. (A and B) Inhibitory effects of anti-Rae-1ε antibody on cardiomyocyte apoptosis after MI. Anti-Rae-1ε antibody or rat IgG (100 µg) was intraperitoneally administered 1 day after MI. Murine heart sections were prepared 3 days after MI. TUNEL staining was performed. (A) Representative images are shown (Bar; 50 μm). (B) TUNEL positive cardiomyocytes were counted (α-MHC, Hoechst, and TUNEL merged cells). Data are shown as mean ± SD (n = 5 for non-treatment, rat IgG, and anti-Rae-1ε antibody group). *P < 0.05 vs. rat IgG by one-way ANOVA followed by Tukey Kramer test. (C–F) Inhibitory effects of anti-Rae-1ε antibody on post-infarct cardiac remodelling. Anti-Rae-1ε antibody or rat IgG (100 µg) was intraperitoneally administered 1 and 7 days after MI. Murine heart sections were prepared and Masson’s trichrome staining was performed 14 days after MI. (C) Representative images are shown (Bar; 1 mm). The ratio of fibrotic area to LV area (D), the ratio of fibrotic circumference to LV circumference (E) and infarct wall thickness (F) were quantitatively estimated. Data are shown as mean ± SD (n = 8 mice for anti-Rae-1ε antibody administration group, n = 11 mice for non-treatment group, n = 12 mice for rat IgG administration group). *P < 0.05, **P < 0.01 vs. rat IgG by one-way ANOVA followed by Tukey Kramer test. We also examined the effects of anti-Rae-1ɛ blocking antibody on cardiac fibrosis and function. Masson’s trichrome staining revealed that both the ratio of fibrotic area to LV area and fibrotic circumference to LV circumference were ameliorated in anti-Rae-1ɛ antibody group at Day 14 after MI (fibrotic area: control IgG 43.1 ± 10.0%, anti-Rae-1ɛ antibody 30.8 ± 8.7%, fibrotic circumference: control IgG 57.3 ± 13.3%, anti-Rae-1ɛ antibody 42.9 ± 11.3%) (Figure 5C–E). The ventricular wall thickness was sustained by anti-Rae-1ɛ antibody treatment, compared with control IgG group (control IgG 0.76 ± 0.21 mm, anti-Rae-1ɛ antibody 1.08 ± 0.23 mm) (Figure 5F). Consistent with the histological data, cardiac dysfunction was also attenuated in the same group (Table 1). Table 1 Cardiac function in antibody administration experiments Days after MI 2 7 Non-treatment Non-immune IgG Anti-Rae-1ε Ab Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 3.4 ± 0.7 3.7 ± 0.4 3.7 ± 0.4 4.7 ± 0.5 4.7 ± 0.5 4.1 ± 0.3 LVIDs (mm) 1.9 ± 0.5 2.0 ± 0.3 2.1 ± 0.3 3.0 ± 0.5 2.9 ± 0.3 2.3 ± 1.2* FS (%) 43.9 ± 5.8 44.2 ± 5.5 44.0 ± 2.8 36.2 ± 4.3 37.7 ± 2.2 43.7 ± 2.4* HR (b.p.m.) 565.2 ± 98.9 557.0 ± 105.2 587.6 ± 46.4 599.2 ± 60.0 565.0 ± 63.0 582.2 ± 64.3 Days after MI 2 7 Non-treatment Non-immune IgG Anti-Rae-1ε Ab Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 3.4 ± 0.7 3.7 ± 0.4 3.7 ± 0.4 4.7 ± 0.5 4.7 ± 0.5 4.1 ± 0.3 LVIDs (mm) 1.9 ± 0.5 2.0 ± 0.3 2.1 ± 0.3 3.0 ± 0.5 2.9 ± 0.3 2.3 ± 1.2* FS (%) 43.9 ± 5.8 44.2 ± 5.5 44.0 ± 2.8 36.2 ± 4.3 37.7 ± 2.2 43.7 ± 2.4* HR (b.p.m.) 565.2 ± 98.9 557.0 ± 105.2 587.6 ± 46.4 599.2 ± 60.0 565.0 ± 63.0 582.2 ± 64.3 Days after MI 14 Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 5.1 ± 0.6 5.4 ± 0.7 3.7 ± 0.5** LVIDs (mm) 3.5 ± 0.5 3.8 ± 0.7 2.2 ± 0.4** FS (%) 32.0 ± 3.2 30.6 ± 4.3 41.9 ± 2.0** HR (b.p.m.) 576.8 ± 56.6 604.0 ± 39.0 561.8 ± 45.3 Days after MI 14 Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 5.1 ± 0.6 5.4 ± 0.7 3.7 ± 0.5** LVIDs (mm) 3.5 ± 0.5 3.8 ± 0.7 2.2 ± 0.4** FS (%) 32.0 ± 3.2 30.6 ± 4.3 41.9 ± 2.0** HR (b.p.m.) 576.8 ± 56.6 604.0 ± 39.0 561.8 ± 45.3 Data are shown as mean ± SD. n = 5 mice for non-treatment and anti-Rae-1ε group, n = 6 mice for non-immune IgG group. * P < 0.05 vs. day 7-rat IgG administration group by Student’s t-test. ** P < 0.01 vs. day 14-rat IgG administration group by one-way ANOVA followed by Tukey–Kramer test. FS, fractional shortening; HR, heart rate; LVIDd, left ventricular internal dimension in diastole; LVIDs, left ventricular internal dimension in systole. Table 1 Cardiac function in antibody administration experiments Days after MI 2 7 Non-treatment Non-immune IgG Anti-Rae-1ε Ab Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 3.4 ± 0.7 3.7 ± 0.4 3.7 ± 0.4 4.7 ± 0.5 4.7 ± 0.5 4.1 ± 0.3 LVIDs (mm) 1.9 ± 0.5 2.0 ± 0.3 2.1 ± 0.3 3.0 ± 0.5 2.9 ± 0.3 2.3 ± 1.2* FS (%) 43.9 ± 5.8 44.2 ± 5.5 44.0 ± 2.8 36.2 ± 4.3 37.7 ± 2.2 43.7 ± 2.4* HR (b.p.m.) 565.2 ± 98.9 557.0 ± 105.2 587.6 ± 46.4 599.2 ± 60.0 565.0 ± 63.0 582.2 ± 64.3 Days after MI 2 7 Non-treatment Non-immune IgG Anti-Rae-1ε Ab Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 3.4 ± 0.7 3.7 ± 0.4 3.7 ± 0.4 4.7 ± 0.5 4.7 ± 0.5 4.1 ± 0.3 LVIDs (mm) 1.9 ± 0.5 2.0 ± 0.3 2.1 ± 0.3 3.0 ± 0.5 2.9 ± 0.3 2.3 ± 1.2* FS (%) 43.9 ± 5.8 44.2 ± 5.5 44.0 ± 2.8 36.2 ± 4.3 37.7 ± 2.2 43.7 ± 2.4* HR (b.p.m.) 565.2 ± 98.9 557.0 ± 105.2 587.6 ± 46.4 599.2 ± 60.0 565.0 ± 63.0 582.2 ± 64.3 Days after MI 14 Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 5.1 ± 0.6 5.4 ± 0.7 3.7 ± 0.5** LVIDs (mm) 3.5 ± 0.5 3.8 ± 0.7 2.2 ± 0.4** FS (%) 32.0 ± 3.2 30.6 ± 4.3 41.9 ± 2.0** HR (b.p.m.) 576.8 ± 56.6 604.0 ± 39.0 561.8 ± 45.3 Days after MI 14 Non-treatment Non-immune IgG Anti-Rae-1ε Ab LVIDd (mm) 5.1 ± 0.6 5.4 ± 0.7 3.7 ± 0.5** LVIDs (mm) 3.5 ± 0.5 3.8 ± 0.7 2.2 ± 0.4** FS (%) 32.0 ± 3.2 30.6 ± 4.3 41.9 ± 2.0** HR (b.p.m.) 576.8 ± 56.6 604.0 ± 39.0 561.8 ± 45.3 Data are shown as mean ± SD. n = 5 mice for non-treatment and anti-Rae-1ε group, n = 6 mice for non-immune IgG group. * P < 0.05 vs. day 7-rat IgG administration group by Student’s t-test. ** P < 0.01 vs. day 14-rat IgG administration group by one-way ANOVA followed by Tukey–Kramer test. FS, fractional shortening; HR, heart rate; LVIDd, left ventricular internal dimension in diastole; LVIDs, left ventricular internal dimension in systole. 3.6 Cardiomyocyte-specific Rae-1ɛ overexpressing transgenic mice showed deteriorated pathological condition after MI As described above, the blockade of NKG2D/NKG2DL interaction attenuated adverse cardiac remodelling. However, intraperitoneal administration of anti-Rae-1ɛ antibody might exhibit systemic effects. Therefore, we focused on NKG2D/NKG2DL interaction on cardiomyocytes by generating cardiomyocyte-specific TG mice expressing Rae-1ɛ. First, Rae-1ɛ overexpression was confirmed by western blotting (Figure 6A). α-MHC-MerCreMer mice were used as a control. TG mice normally developed and exhibited no remarkable phenotype under non-MI condition. To examine whether the overexpression of Rae-1ɛ influenced the post-infarct inflammation, we analysed the expression of proinflammatory cytokines, such as IL-6, IL-1β, and TNF-α; however, there were no significant differences in the expression of these cytokines between control and Rae-1ɛ TG mice (data not shown). TUNEL staining showed the significant increase of apoptotic cardiomyocytes in TG mice at Day 3 after MI, compared with Cre-transgenic mice (Cre 11.2 ± 1.5 cells/mm2, TG 27.9 ± 10.8 cells/mm2) (Figure 6B and C). Activation of apoptotic signalling pathway was also confirmed by the detection of cleaved caspase 3 fragments (Figure 6D and E). Consistently, post-infarct cardiac fibrosis was exacerbated in TG mice (Cre 35.2 ± 10.6%, TG 46.5 ± 14.1%), analysed by Masson’s trichrome staining (Figure 6F and G). Echocardiographic examination showed that FS was significantly reduced with hypertrophy at Day 7 after MI, suggesting that cardiac function was impaired more severely in TG mice. The detrimental effects of Rae-1ɛ overexpression continued to be observed at Day 14 after MI, though statistical significance was reduced probably because of the large standard deviation at Day 14 (Table 2). Taken together, the enhancement of NKG2D/NKG2DL interaction on cardiomyocytes aggravated post-infarct cardiac remodelling. Table 2 Cardiac function in experiments using TG mice Days after MI 0 2 7 14 Non-TG TG Non-TG TG Non-TG TG Non-TG TG LVIDd (mm) 3.8 ± 0.3 3.8 ± 0.5 4.2 ± 0.2 4.4 ± 0.3 4.4 ± 0.4 5.0 ± 0.5** 4.8 ± 0.6 5.0 ± 0.5 LVIDs (mm) 1.8 ± 0.2 1.9 ± 0.3 2.4 ± 0.2 2.6 ± 0.3* 2.6 ± 0.4 3.3 ± 0.4** 3.1 ± 0.5 3.4 ± 0.5 FS (%) 52.3 ± 3.3 52.2 ± 2.8 44.3 ± 2.7 40.9 ± 4.5* 40.2 ± 4.2 33.5 ± 3.3** 35.2 ± 2.4 33.0 ± 3.7 HR (b.p.m.) 486.1 ± 73.0 537.4 ± 45.5 519.9 ± 74.9 505.8 ± 54.7 523.5 ± 61.1 524.6 ± 58.1 498.2 ± 54.9 529.6 ± 58.7 Days after MI 0 2 7 14 Non-TG TG Non-TG TG Non-TG TG Non-TG TG LVIDd (mm) 3.8 ± 0.3 3.8 ± 0.5 4.2 ± 0.2 4.4 ± 0.3 4.4 ± 0.4 5.0 ± 0.5** 4.8 ± 0.6 5.0 ± 0.5 LVIDs (mm) 1.8 ± 0.2 1.9 ± 0.3 2.4 ± 0.2 2.6 ± 0.3* 2.6 ± 0.4 3.3 ± 0.4** 3.1 ± 0.5 3.4 ± 0.5 FS (%) 52.3 ± 3.3 52.2 ± 2.8 44.3 ± 2.7 40.9 ± 4.5* 40.2 ± 4.2 33.5 ± 3.3** 35.2 ± 2.4 33.0 ± 3.7 HR (b.p.m.) 486.1 ± 73.0 537.4 ± 45.5 519.9 ± 74.9 505.8 ± 54.7 523.5 ± 61.1 524.6 ± 58.1 498.2 ± 54.9 529.6 ± 58.7 n = 10 for Cre mice, n = 14 for Rae-1ε TG mice. * P < 0.05 vs. day 2-Cre mice. ** P < 0.05 vs. day 7-Cre mice by Student’s t-test. Table 2 Cardiac function in experiments using TG mice Days after MI 0 2 7 14 Non-TG TG Non-TG TG Non-TG TG Non-TG TG LVIDd (mm) 3.8 ± 0.3 3.8 ± 0.5 4.2 ± 0.2 4.4 ± 0.3 4.4 ± 0.4 5.0 ± 0.5** 4.8 ± 0.6 5.0 ± 0.5 LVIDs (mm) 1.8 ± 0.2 1.9 ± 0.3 2.4 ± 0.2 2.6 ± 0.3* 2.6 ± 0.4 3.3 ± 0.4** 3.1 ± 0.5 3.4 ± 0.5 FS (%) 52.3 ± 3.3 52.2 ± 2.8 44.3 ± 2.7 40.9 ± 4.5* 40.2 ± 4.2 33.5 ± 3.3** 35.2 ± 2.4 33.0 ± 3.7 HR (b.p.m.) 486.1 ± 73.0 537.4 ± 45.5 519.9 ± 74.9 505.8 ± 54.7 523.5 ± 61.1 524.6 ± 58.1 498.2 ± 54.9 529.6 ± 58.7 Days after MI 0 2 7 14 Non-TG TG Non-TG TG Non-TG TG Non-TG TG LVIDd (mm) 3.8 ± 0.3 3.8 ± 0.5 4.2 ± 0.2 4.4 ± 0.3 4.4 ± 0.4 5.0 ± 0.5** 4.8 ± 0.6 5.0 ± 0.5 LVIDs (mm) 1.8 ± 0.2 1.9 ± 0.3 2.4 ± 0.2 2.6 ± 0.3* 2.6 ± 0.4 3.3 ± 0.4** 3.1 ± 0.5 3.4 ± 0.5 FS (%) 52.3 ± 3.3 52.2 ± 2.8 44.3 ± 2.7 40.9 ± 4.5* 40.2 ± 4.2 33.5 ± 3.3** 35.2 ± 2.4 33.0 ± 3.7 HR (b.p.m.) 486.1 ± 73.0 537.4 ± 45.5 519.9 ± 74.9 505.8 ± 54.7 523.5 ± 61.1 524.6 ± 58.1 498.2 ± 54.9 529.6 ± 58.7 n = 10 for Cre mice, n = 14 for Rae-1ε TG mice. * P < 0.05 vs. day 2-Cre mice. ** P < 0.05 vs. day 7-Cre mice by Student’s t-test. Figure 6 View largeDownload slide Post-infarct cardiac remodelling was aggravated in cardiomyocyte-specific transgenic mice expressing Rae-1ε. (A) Generation of tamoxifen-inducible cardiomyocyte-specific Rae-1ε overexpressing mice. Transgenic strategy is shown (left). The lysates from the hearts of Cre, control, and Rae-1ε TG mice were immunoblotted with anti-Rae-1ε and anti-GAPDH antibodies. Representative results are shown (right) (n = 3 for Cre and Rae-1ε TG mice). (B and C) Measurement of apoptotic cardiomyocytes in TG hearts after MI. Murine heart sections were prepared 3 days after MI. TUNEL staining was performed. (B) Representative images are shown (Bar; 50 μm). (C) TUNEL positive cardiomyocytes were counted (α-MHC, Hoechst, and TUNEL merged cells). Data are shown as mean ± SD (n = 5 for Cre and Rae-1ε TG mice). **P < 0.01 vs. Cre by Student’s t-test. (D and E) Detection of cleaved caspase 3 in TG hearts after MI. (D) The lysates from the hearts of Cre and Rae-1ε TG mice were immunoblotted with anti-cleaved caspase 3 and anti-GAPDH antibodies. Representative results are shown. (E) The ratio of cleaved caspase 3 to GAPDH was quantitatively estimated. Data are shown as mean ± SD (n = 5 for Cre and Rae-1ε TG mice). **P < 0.01 vs. Cre by Student’s t-test. (F and G) Estimation of post-infarct fibrosis in TG hearts. Murine heart sections were prepared, and Masson’s trichrome staining was performed 14 days after MI. (F) Representative images are shown (Bar; 1 mm). (G) The ratio of fibrotic area to LV area was quantitatively estimated. Data are shown as mean ± SD (n = 10 for Cre mice, n = 14 for Rae-1ε TG mice). *P < 0.05 vs. Cre by Student’s t-test. Figure 6 View largeDownload slide Post-infarct cardiac remodelling was aggravated in cardiomyocyte-specific transgenic mice expressing Rae-1ε. (A) Generation of tamoxifen-inducible cardiomyocyte-specific Rae-1ε overexpressing mice. Transgenic strategy is shown (left). The lysates from the hearts of Cre, control, and Rae-1ε TG mice were immunoblotted with anti-Rae-1ε and anti-GAPDH antibodies. Representative results are shown (right) (n = 3 for Cre and Rae-1ε TG mice). (B and C) Measurement of apoptotic cardiomyocytes in TG hearts after MI. Murine heart sections were prepared 3 days after MI. TUNEL staining was performed. (B) Representative images are shown (Bar; 50 μm). (C) TUNEL positive cardiomyocytes were counted (α-MHC, Hoechst, and TUNEL merged cells). Data are shown as mean ± SD (n = 5 for Cre and Rae-1ε TG mice). **P < 0.01 vs. Cre by Student’s t-test. (D and E) Detection of cleaved caspase 3 in TG hearts after MI. (D) The lysates from the hearts of Cre and Rae-1ε TG mice were immunoblotted with anti-cleaved caspase 3 and anti-GAPDH antibodies. Representative results are shown. (E) The ratio of cleaved caspase 3 to GAPDH was quantitatively estimated. Data are shown as mean ± SD (n = 5 for Cre and Rae-1ε TG mice). **P < 0.01 vs. Cre by Student’s t-test. (F and G) Estimation of post-infarct fibrosis in TG hearts. Murine heart sections were prepared, and Masson’s trichrome staining was performed 14 days after MI. (F) Representative images are shown (Bar; 1 mm). (G) The ratio of fibrotic area to LV area was quantitatively estimated. Data are shown as mean ± SD (n = 10 for Cre mice, n = 14 for Rae-1ε TG mice). *P < 0.05 vs. Cre by Student’s t-test. 4. Discussion Cardiomyocyte death is a critical event in the development of HF after MI.19,20 Here, we examined whether immunosurveillance system is involved in cardiomyocyte death in the process of post-infarct cardiac remodelling. Rae-1ɛ expression was up-regulated downstream of p53 in injured cardiomyocytes under ischaemic condition. NKG2D-expressing γδT cells, which infiltrated into the post-infarct heart, caused apoptotic death in Rae-1ɛ-expressing cardiomyocytes. Importantly, cardiomyocyte apoptosis and adverse cardiac remodelling were significantly suppressed by the blockade of NKG2D/NKG2DL interaction, leading to the attenuation of cardiac dysfunction. Moreover, adverse cardiac remodelling was deteriorated in cardiomyocyte-specific TG mice overexpressing Rae-1ɛ after MI. Thus, the activation of NKG2D/NKG2DL interaction could be a novel signalling pathway that mediates cardiomyocyte death, leading to cardiac remodelling. Accumulating evidence has revealed that a variety of immune cells modulate post-infarct cardiac remodelling.5,21,22 It is widely accepted that ROS production and/or Ca2+ overload play important roles in cardiomyocyte death after MI by activating the downstream death signals in a cell-autonomous manner.1–3 Following cardiomyocyte loss, cardiac fibroblasts are activated by inflammatory cells and, as a result, cardiac fibrosis expands into interstitial space in the absence of myocardium23,24; however, it remains to be elucidated whether or how immune/inflammatory cells directly induce cardiomyocyte death. Here, we focused on immunosurveillance system and, for the first time, demonstrated that NKG2D/NKG2DL system is involved in apoptotic cardiomyocyte death after MI as a cell non-autonomous process. NKG2D-expressing immune cells, which infiltrated into heart tissue after MI, were mainly composed of γδT cells, suggesting that NKG2D-expressing γδT cells had cytotoxic effects and led to the aggravation of HF. Previously, it was reported that the ablation of γδTCR gene significantly improved survival rate after MI by modulating inflammatory reactions.25 Combined with our data presented here, it is suggested that γδT cells exhibit pathophysiological functions in post-infarct cardiac remodelling not only as inflammatory regulators but also as effector cells that induce cell death in injured cardiomyocytes. However, we should not conclude that heart-infiltrating γδT cells are uniformly detrimental, because biological functions of γδT cells are complicated. For example, RORγt+ cells that infiltrated into the post-infarct myocardium also expressed γδTCR but had cardioprotective effects after MI.26 Unfortunately, we cannot compare the gene expression profile between NKG2D-expressing γδT cells and RORγt+ cells at the present time, because RORγt+ cells have not been successfully prepared from the post-infarct myocardium due to the technical limitation. However, NKG2D-expressing γδT cells are likely to be distinct from RORγt+ γδT cells because NKG2D-expressing γδT cells did not express RORγt (Supplementary material online, Figure S3). Therefore, γδT cells are diverse, containing at least two subtypes, cytotoxic cells and cardioprotective cells. NKG2DLs are expressed in cytopathic cells, including virus-infected cells and cancer cells. We demonstrated that NKG2DLs were expressed in the post-infarct myocardium. Interestingly, p53, a tumour suppressor, is essential for the myocardial induction of NKG2DL. Previously, it was reported that p53 inhibited hypoxia-inducible factor-1-dependent induction of angiogenic factors, impairing angiogenesis and systolic function.27 Therefore, we examined the capillary density in Rae-1ɛ TG hearts after MI. However, there was no difference in capillary density (Supplementary material online, Figure S4). Thus, p53 could deteriorate cardiac remodelling through NKG2DL induction, independently of angiogenesis, proposing a novel role of p53 in cardiac remodelling. Blockade of NKG2D/NKG2DL by anti-Rae-1ɛ antibody attenuated cardiac remodelling. Therefore, the treatment with blocking antibody against NKG2DL could be a promising strategy against HF. However, it should be noted that the structures of NKG2DLs are not conservative among animal species. Therefore, we need to identify the counterpart of Rae-1ɛ in human ligands, considering clinical application. Human ligands consist of two families and eight ligands have been identified so far. Among these, four ligands, UL16-binding protein (ULBP) 1, 2, 3, and 6, have the similar structure to Rae-1ɛ, in that they are attached to the cell surface membrane via glycosylphosphatidylinositol anchors,6–8 suggesting that some of these four ligands might be the candidates as therapeutic targets. Interestingly, adenoviral overexpression of p53 in human iPS cells-derived cardiomyocytes resulted in the up-regulation of ULBP1 mRNA about by 200 folds (Supplementary material online, Figure S5). Thus, ULBP1 could be a candidate of the counterpart of Rae-1ɛ among human NKG2DLs, though more detailed analyses would be required. Recent advances in oncocardiology have revealed that cardiomyocytes share the survival signals with cancer cells,28–34 in spite of a large difference in cell proliferative activities between these two cell types. For example, Trastuzumab, a molecularly-targeted antibody that inhibits the proliferation of HER2-expressing cancer cells, induces cardiomyopathy in the patients after doxorubicin treatment,35 because HER2 signal promotes cardiomyocyte survival. Similarly, though angiogenesis inhibitors exhibit remarkable anti-tumour effects, they cause cardiac ischaemia and dysfunction as adverse drug reactions,36,37 because angiogenic factors contribute to the maintenance of cardiac homeostasis. In addition to the commonality of survival signals, our data presented here could provide a novel concept that the pathological status of cardiomyocytes is checked by the immunosurveillance system as a common cell-quality control system. 5. Conclusion In conclusion, NKG2D/NKG2DL system-mediated cardiomyocyte death worsened the pathological condition after MI. Suppression of immunosurvaillance system by blocking NKG2D/NKG2DL interaction may be a novel therapeutic strategy against HF. Footnotes Time for primary review: 25 days Acknowledgements We greatly thank Dr Yasuko Bando for giving us the adenovirus vector expressing p53. Conflict of interest: Y.F.: Bristol-Myers Squibb. All remaining authors have nothing to declare. Funding This work was supported by MEXT/JSPS KAKENHI (25670193, 23390057, and 18H02603 to Y.F., 15K18987 to M.O., 15K08232 to M.M.) and Suzuken Memorial Foundation. This work was partially supported by Platform Project for Supporting Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research (BINDS)) from AMED under Grant Number JP18am0101084. References 1 Hori M , Nishida K. Oxidative stress and left ventricular remodelling after myocardial infarction . Cardiovasc Res 2009 ; 81 : 457 – 464 . Google Scholar Crossref Search ADS PubMed 2 Talukder MA , Zweier JL , Periasamy M. Targeting calcium transport in ischaemic heart disease . Cardiovasc Res 2009 ; 84 : 345 – 352 . Google Scholar Crossref Search ADS PubMed 3 Takano H , Zou Y , Hasegawa H , Akazawa H , Nagai T , Komuro I. Oxidative stress-induced signal transduction pathways in cardiac myocytes: involvement of ROS in heart diseases . Antioxid Redox Signal 2003 ; 5 : 789 – 794 . Google Scholar Crossref Search ADS PubMed 4 Ma Y , Yabluchanskiy A , Lindsey ML. Neutrophil roles in left ventricular remodeling following myocardial infarction . Fibrogenesis Tissue Repair 2013 ; 6 : 11 . Google Scholar Crossref Search ADS PubMed 5 Nahrendorf M , Swirski FK. Innate immune cells in ischaemic heart disease: does myocardial infarction beget myocardial infarction? Eur Heart J 2016 ; 3 . 868 – 872 . Google Scholar Crossref Search ADS 6 Raulet DH , Gasser S , Gowen BG , Deng W , Jung H. Regulation of ligands for the NKG2D activating receptor . Annu Rev Immunol 2013 ; 31 : 413 – 441 . Google Scholar Crossref Search ADS PubMed 7 Raulet DH. Roles of the NKG2D immunoreceptor and its ligands . Nat Rev Immunol 2003 ; 3 : 781 – 790 . Google Scholar Crossref Search ADS PubMed 8 Champsaur M , Lanier LL. Effect of NKG2D ligand expression on host immune responses . Immunol Rev 2010 ; 235 : 267 – 285 . Google Scholar Crossref Search ADS PubMed 9 Tsukada T , Tomooka Y , Takai S , Ueda Y , Nishikawa S , Yagi T , Tokunaga T , Takeda N , Suda Y , Abe S. Enhanced proliferative potential in culture of cells from p53-deficient mice . Oncogene 1993 ; 8 : 3313 – 3322 . Google Scholar PubMed 10 Agah R , Frenkel PA , French BA , Michael LH , Overbeek PA , Schneider MD. Gene recombination in postmitotic cells. Targeted expression of Cre recombinase provokes cardiac-restricted, site-specific rearrangement in adult ventricular muscle in vivo . J Clin Invest 1997 ; 100 : 169 – 179 . Google Scholar Crossref Search ADS PubMed 11 Obana M , Maeda M , Takeda K , Hayama A , Mohri T , Yamashita T , Nakaoka Y , Komuro I , Takeda K , Matsumiya G , Azuma J , Fujio Y. Therapeutic activation of signal transducer and activator of transcription 3 by interleukin-11 ameliorates cardiac fibrosis after myocardial infarction . Circulation 2010 ; 121 : 684 – 691 . Google Scholar Crossref Search ADS PubMed 12 Miyawaki A , Obana M , Mitsuhara Y , Orimoto A , Nakayasu Y , Yamashita T , Fukada SI , Maeda M , Nakayama H , Fujio Y. Adult murine cardiomyocytes exhibit regenerative activity with cell cycle reentry through STAT3 in the healing process of myocarditis . Sci Rep 2017 ; 7 : 1407 . Google Scholar Crossref Search ADS PubMed 13 Tonegawa K , Otsuka W , Kumagai S , Matsunami S , Hayamizu N , Tanaka S , Moriwaki K , Obana M , Maeda M , Asahi M , Kiyonari H , Fujio Y , Nakayama H. Caveolae-specific activation loop between CaMKII and L-type Ca(2+) channel aggravates cardiac hypertrophy in alpha1-adrenergic stimulation . Am J Physiol Heart Circ Physiol 2017 ; 312 : H501 – h514 . Google Scholar Crossref Search ADS PubMed 14 Zhang T , Gunther S , Looso M , Kunne C , Kruger M , Kim J , Zhou Y , Braun T. Prmt5 is a regulator of muscle stem cell expansion in adult mice . Nat Commun 2015 ; 6 : 7140 . Google Scholar Crossref Search ADS PubMed 15 Morimoto Y , Bando YK , Shigeta T , Monji A , Murohara T. Atorvastatin prevents ischemic limb loss in type 2 diabetes: role of p53 . J Atheroscler Thromb 2011 ; 18 : 200 – 208 . Google Scholar Crossref Search ADS PubMed 16 Li H , Lakshmikanth T , Garofalo C , Enge M , Spinnler C , Anichini A , Szekely L , Karre K , Carbone E , Selivanova G. Pharmacological activation of p53 triggers anticancer innate immune response through induction of ULBP2 . Cell Cycle 2011 ; 10 : 3346 – 3358 . Google Scholar Crossref Search ADS PubMed 17 Kossmehl P , Kurth E , Faramarzi S , Habighorst B , Shakibaei M , Wehland M , Kreutz R , Infanger M , Ah JD , Grosse J , Paul M , Grimm D. Mechanisms of apoptosis after ischemia and reperfusion: role of the renin-angiotensin system . Apoptosis 2006 ; 11 : 347 – 358 . Google Scholar Crossref Search ADS PubMed 18 Yaoita H , Ogawa K , Maehara K , Maruyama Y. Apoptosis in relevant clinical situations: contribution of apoptosis in myocardial infarction . Cardiovasc Res 2000 ; 45 : 630 – 641 . Google Scholar Crossref Search ADS PubMed 19 Moe GW , Marín-García J. Role of cell death in the progression of heart failure . Heart Fail Rev 2016 ; 21 : 157 – 167 . Google Scholar Crossref Search ADS PubMed 20 Marunouchi T , Tanonaka K. Cell death in the cardiac myocyte . Biol Pharm Bull 2015 ; 38 : 1094 – 1097 . Google Scholar Crossref Search ADS PubMed 21 Swirski FK , Nahrendorf M. Leukocyte behavior in atherosclerosis, myocardial infarction, and heart failure . Science 2013 ; 3 . 161 – 166 . Google Scholar Crossref Search ADS 22 Nian M , Lee P , Khaper N , Liu P. Inflammatory cytokines and postmyocardial infarction remodeling . Circ Res 2004 ; 94 : 1543 – 1553 . Google Scholar Crossref Search ADS PubMed 23 Van Linthout S , Miteva K , Tschope C. Crosstalk between fibroblasts and inflammatory cells . Cardiovasc Res 2014 ; 102 : 258 – 269 . Google Scholar Crossref Search ADS PubMed 24 Hartupee J , Mann DL. Role of inflammatory cells in fibroblast activation . J Mol Cell Cardiol 2016 ; 93 : 143 – 148 . Google Scholar Crossref Search ADS PubMed 25 Yan X , Shichita T , Katsumata Y , Matsuhashi T , Ito H , Ito K , Anzai A , Endo J , Tamura Y , Kimura K , Fujita J , Shinmura K , Shen W , Yoshimura A , Fukuda K , Sano M. Deleterious effect of the IL-23/IL-17A axis and gammadeltaT cells on left ventricular remodeling after myocardial infarction . J Am Heart Assoc 2012 ; 1 : e004408 . Google Scholar PubMed 26 Enomoto D , Matsumoto K , Yamashita T , Kobayashi A , Maeda M , Nakayama H , Obana M , Fujio Y. RORgammat-expressing cells attenuate cardiac remodeling after myocardial infarction . PLoS One 2017 ; 12 : e0183584 . Google Scholar Crossref Search ADS PubMed 27 Sano M , Minamino T , Toko H , Miyauchi H , Orimo M , Qin Y , Akazawa H , Tateno K , Kayama Y , Harada M , Shimizu I , Asahara T , Hamada H , Tomita S , Molkentin JD , Zou Y , Komuro I. p53-induced inhibition of Hif-1 causes cardiac dysfunction during pressure overload . Nature 2007 ; 446 : 444 – 448 . Google Scholar Crossref Search ADS PubMed 28 Lee CL , Moding EJ , Cuneo KC , Li Y , Sullivan JM , Mao L , Washington I , Jeffords LB , Rodrigues RC , Ma Y , Das S , Kontos CD , Kim Y , Rockman HA , Kirsch DG. p53 functions in endothelial cells to prevent radiation-induced myocardial injury in mice . Sci Signal 2012 ; 5 : ra52 . Google Scholar Crossref Search ADS PubMed 29 Knight RA , Scarabelli TM , Stephanou A. STAT transcription in the ischemic heart . JAKSTAT 2012 ; 1 : 111 – 117 . Google Scholar PubMed 30 Kitamura H , Ohno Y , Toyoshima Y , Ohtake J , Homma S , Kawamura H , Takahashi N , Taketomi A. Interleukin-6/STAT3 signaling as a promising target to improve the efficacy of cancer immunotherapy . Cancer Sci 2017 ; 108 : 1947 – 1952 . Google Scholar Crossref Search ADS PubMed 31 Fujio Y , Maeda M , Mohri T , Obana M , Iwakura T , Hayama A , Yamashita T , Nakayama H , Azuma J. Glycoprotein 130 cytokine signal as a therapeutic target against cardiovascular diseases . J Pharmacol Sci 2011 ; 117 : 213 – 222 . Google Scholar Crossref Search ADS PubMed 32 Chen Z , Han ZC. STAT3: a critical transcription activator in angiogenesis . Med Res Rev 2008 ; 28 : 185 – 200 . Google Scholar Crossref Search ADS PubMed 33 Carmeliet P , Jain RK. Angiogenesis in cancer and other diseases . Nature 2000 ; 407 : 249 – 257 . Google Scholar Crossref Search ADS PubMed 34 Bykov VJN , Eriksson SE , Bianchi J , Wiman KG. Targeting mutant p53 for efficient cancer therapy . Nat Rev Cancer 2018 ; 18 : 89 – 102 . Google Scholar Crossref Search ADS PubMed 35 Seidman A , Hudis C , Pierri MK , Shak S , Paton V , Ashby M , Murphy M , Stewart SJ , Keefe D. Cardiac dysfunction in the trastuzumab clinical trials experience . J Clin Oncol 2002 ; 20 : 1215 – 1221 . Google Scholar Crossref Search ADS PubMed 36 Qi WX , Fu S , Zhang Q , Guo XM. Bevacizumab increases the risk of severe congestive heart failure in cancer patients: an up-to-date meta-analysis with a focus on different subgroups . Clin Drug Invest 2014 ; 34 : 681 – 690 . Google Scholar Crossref Search ADS 37 Penumetsa SC , Hoque MZ , Giugliano G. Reversible myocardial dysfunction following intraocular bevacizumab admninistration . J Cardiovasc Dis Res 2013 ; 4 : 58 – 60 . Google Scholar Crossref Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Blockade of NKG2D/NKG2D ligand interaction attenuated cardiac remodelling after myocardial infarction
JO - Cardiovascular Research
DO - 10.1093/cvr/cvy254
DA - 2019-03-15
UR - https://www.deepdyve.com/lp/oxford-university-press/blockade-of-nkg2d-nkg2d-ligand-interaction-attenuated-cardiac-3fPDUDgChi
SP - 765
VL - 115
IS - 4
DP - DeepDyve
ER -