Deletion of CD28 Co-stimulatory Signals Exacerbates Left Ventricular Remodeling and Increases Cardiac Rupture After Myocardial Infarction

Abstract

Background: Inflammatory responses, especially by CD4⁺ T cells activated by dendritic cells, are known to be important in the pathophysiology of cardiac repair after myocardial infarction (MI). Although co-stimulatory signals through B7 (CD80/86) and CD28 are necessary for CD4⁺ T cell activation and survival, the roles of these signals in cardiac repair after MI are still unclear.

Methods and Results: C57BL/6 (Control) mice and CD28 knockout (CD28KO) mice were subjected to left coronary artery permanent ligation. The ratio of death by cardiac rupture within 5 days after MI was significantly higher in CD28KO mice compared with Control mice. Although there were no significant differences in the infarct size between the 2 groups, left ventricular end-diastolic and end-systolic diameters were significantly increased, and fractional shortening was significantly decreased in CD28KO mice compared with Control mice. Electron microscopic observation revealed that the extent of extracellular collagen fiber was significantly decreased in CD28KO mice compared with Control mice. The number of α-smooth muscle actin-positive myofibroblasts was significantly decreased, and matrix metalloproteinase-9 activity and the mRNA expression of interleukin-1β were significantly increased in CD28KO mice compared with Control mice.

Conclusions: Deletion of CD28 co-stimulatory signals exacerbates left ventricular remodeling and increases cardiac rupture after MI through prolongation of the inflammatory period and reduction of collagen fiber in the infarct scars.

Heart failure is the major cause of death after myocardial infarction (MI).^1,2 After MI, left ventricular remodeling, which includes dilatation of the left ventricle (LV) and an increase in interstitial fibrosis with histopathological changes in LV, is a key factor inducing cardiac pump dysfunction after MI with poor clinical outcomes. Cardiac rupture is another cause of fatal complications after MI. Although recent medical advances in pharmacological and interventional therapies, including emergent coronary artery reperfusion therapies, fortunately reduce LV remodeling and cardiac rupture, these complication are not fully prevented. Therefore, it is important to elucidate a novel approach to prevent LV remodeling and cardiac rupture after MI.

Recently, it has been reported that monocytes and lymphocytes infiltrated into ischemic heart tissue are major factors in the pathophysiology of tissue repair after MI. When the myocardium is damaged after MI, dendritic cells (DCs) are activated and become capable of productively activating cardiac myosin-specific naïve T cells, leading to differentiation of CD4⁺ T cells.³ Anzai et al demonstrated that deletion of CD11c⁺ DCs impaired cardiac function and tissue repair.⁴ In contrast, Hofmann et al reported that CD4⁺ T cells were activated and facilitated wound healing of the injured myocardium after MI.⁵ They also demonstrated by using OT-II mice (C57BL/6-Tg(TcraTcrb)425Cbn/Crl) that T cell receptor (TCR) signaling is important to activate CD4⁺ T cells after MI. Consequently, antigen presentation from DCs to CD4⁺ T cells is important in the pathology of MI. Meanwhile, regulatory T cells (Tregs), which are a subpopulation of T cells and suppress many types of immune cells, also have beneficial effects on wound healing process after MI.⁶

The CD28 molecule, which is the protein expressed on the cell membrane of T cells, is the receptor of B7 (CD80 [B7.1] and CD86 [B7.2]) and provides co-stimulatory signals required for T cell activation and survival. To activate T cells, it is necessary to stimulate simultaneously both TCR and co-stimulatory signals from antigen presenting cells (APCs) such as DCs. In contrast, the CD28 molecule is involved in the development and maintenance of Tregs. Tai et al reported that CD28 co-stimulatory signals directly activate thymocytes to express Foxp3 and to initiate a Tregs differentiation program.⁷ However, the role of the CD28 molecule is unclear in the pathophysiology of tissue repair after MI.

Therefore, the purpose of the present study was to elucidate the role of the CD28 molecule in tissue repair and LV remodeling after MI. In this study, we used CD28 knockout (CD28KO) mice to demonstrate the effects of the lack of the CD28 molecule on LV remodeling and cardiac rupture after MI in mice.

Methods

Animals

Wild-type male C57BL/6 mice (8-weeks-old) were purchased from Japan SLC (Shizuoka, Japan) and used in the present study. CD28KO mice, whose construction and characterization has been described previously, were kindly provided by Ryo Abe (Research Institute for Biological Sciences, Science University of Tokyo, Chiba, Japan).⁸ These mice were bred and kept under specific pathogen-free conditions. All experimental procedures were performed according to the guidelines established by Chiba University for experiments with animals, which conform to the “Guide for the Care and Use of Laboratory Animals” published by the US National Institutes of Health (NIH Publication, 8th Edition, 2011).

Induction of MI

Mice were anesthetized by inhalation of isoflurane (2.5%) and artificially ventilated with a respirator. Mice were subjected to ligation of the left anterior descending coronary artery (LAD) or to a sham operation, as described previously.^9,10 The heart was exposed through a thoracotomy, and LAD was permanently ligated with an 8-0 prolene suture to stimulate MI. A sham operation was performed by cutting only the pericardium. After examination by transthoracic echocardiography, the mice were sacrificed by cervical dislocation and the hearts were extirpated.

Echocardiography

Transthoracic echocardiography was performed on 3 and 5 days after MI with a Visual Sonics High-Resolution In Vivo Micro-Imaging System (Vevo 770; FujiFilm VisualSonics, Inc, Seattle, WA, USA) equipped with a 30-MHz imaging transducer. Mice were anesthetized by inhalation of isoflurane (0.5–1.0%) during the echocardiographic examination, and the heart rate was approximately 400–550 beats/min in both group mice. Measurements were performed by the same operator, blinded to the experimental groups, before and after MI in each mouse.

Histological Analysis

Mice hearts fixed in 10% formalin overnight were embedded in paraffin and sectioned at 5-μm thickness for Masson’s trichrome staining and picrosirius red staining to determine infarct size and cardiac fibrosis. Immunohistochemical staining was performed with anti α-smooth muscle actin (SMA) antibody (Dako, Glostrup, Denmark).

Gelatin Zymography

To assess the activity of gelatinase, matrix metalloproteinase (MMP)-2 and MMP-9, gelatin zymography was performed, as described previously.^4,11,12 The gels were stained with Quick-Coomassie Brilliant Blue (Wako Pure Chemical Industries, Inc, Osaka, Japan). The presence of different MMPs was identified on the basis of their molecular weights. Finally, the gelatinolytic bands were measured with ChemiDoc MP (Bio-Rad, Hercules, CA, USA) and analyzed by Image J.

Real-Time Reverse Transcriptase-Polymerase Chain Reaction

Mice hearts were incubated in RNA later solution (Invitrogen) overnight at 4℃ to allow thorough penetration of the tissue, then transfered to –20℃ for storage. Then, samples were thawed at room temperature, and RNA was extracted using ISOGEN II (NIPPON GENE, Tokyo, Japan). cDNA was synthesized from 1μg RNA with an iScript^TM cDNA Synthesis Kit (Bio-Rad). Quantitative real-time reverse transcriptase-polymerase chain reaction (PCR) was performed by using Step One Plus^TM with Fast SYBR^® Green Master Mix (Thermo Fisher Scientific, Waltham, MA, USA). The sequences of primer pairs are described in Table 1.

Table 1. Primers for Real-Time Polymerase Chain Reaction

Gene	Sequence
TNF-α	(F) CCCTCACACTCAGATCATCTTCT
	(R) GCTACGACGTGGGCTACAG
IL-1β	(F) GCAACTGTTCCTGAACTCAACT
	(R) ATCTTTTGGGGTCCGTCAACT
IFN-γ	(F) ATGAACGCTACACACTGCATC
	(R) CCATCCTTTTGCCAGTTCCTC
IL-10	(F) GCTCTTACTGACTGGCATGAG
	(R) CGCAGCTCTAGGAGCATGTG
TGF-β	(F) CTCCCGTGGCTTCTAGTGC
	(R) GCCTTAGTTTGGACAGGATCTG
iNOS	(F) GTTCTCAGCCCAACAATACAAGA
	(R) GTGGACGGGTCGATGTCAC
Arginase-1	(F) CTCCAAGCCAAAGTCCTTAGAG
	(R) AGGAGCTGTCATTAGGGACATC
Foxp3	(F) CCCATCCCCAGGAGTCTTG
	(R) ACCATGACTAGGGGCACTGTA
procollagen α‐1 (I)	(F) GCTCCTCTTAGGGGCCACT
	(R) CCACGTCTCACCATTGGGG
procollagen α‐1 (III)	(F) CTGTAACATGGAAACTGGGGAAA
	(R) CCATAGCTGAACTGAAAACCACC

F, forward; IFN-γ, interferon-γ; IL-1β, interleukin-1β; IL-10, interleukin-10; iNOS, inducible nitric oxide synthase; R, reverse; TGF-β, transforming growth factor-β; TNF-α, tumor necrosis factor-α.

Cell Isolation

At 5 days after MI, the mice were sacrificed by cervical dislocation and intracardially perfused with 25 ml of ice-cold phosphate-buffered saline to wash out blood cells. Whole hearts were soaked in 2.5 ml of RPMI 1640-medium (Thermo Scientific) containing 0.5 mg of Liberase TM (Roche, Basel, Switzerland) and 0.25 mg of DNase I (Roche) and enzymatically digested by using a gentleMACS^TM Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany). After digestion, the tissue was passed through a 50-μm cell strainer. Leukocyte-enriched fractions were isolated by density gradient centrifugation using Lympholyte (CEDARLANE, Burlington, NC, USA). Cells were removed from the interface and washed with RPMI 1640-medium for further analysis.

Flow Cytometric Analysis

Cells isolated from leukocyte-enriched fractions were incubated by using a Zombie Aqua^TM Fixable Viability Kit (Biolegend, San Diego, CA, USA) at room temperature for 15 min, and then anti-CD16/32 antibody (Abcam, United Kingdom) was added at room temperature for 15 min to block Fc receptors. After staining with a mixture of antibodies (Table 2) at 4℃ for 20 min, cells were analysed with flow cytometric analysis (Canto II; BD Biosciences, San Jose, CA, USA) and analyzed with FlowJo software (Tree Star, Ashland, OR, USA).

Table 2. Antibodies for Flow Cytometric Analysis

Antibodies	Manufacturer	Clone
CD11b	Biolegend	M1/70
Ly6G	Biolegend	1A8
CD64	Biolegend	X54‐5/7.1
MerTK	R&D Systems
Streptavidin	Biolegend
Ly6C	Biolegend	HK1.4
CD45	BD Bioscience	30-F11
CD4	BD Bioscience	RM4‐5
CD8	Biolegend	53‐6.7
TCRγδ	Biolegend	UC-13D5
B220	Biolegend	RA3‐6B2
Foxp3	eBioscience	FJK-16 s
IFN-γ	Biolegend	XMG1.2

Intracellular Cytokine Analysis

Collected cells from the mice hearts were stimulated with phorbol 12-myristate 13-acetate (PMA) (Merck Millipore, Darmstadt, Germany), ionomycin (Merck Millipore) and BD golgiplug (BD Biosciences) for 1 h. After the surface staining, cells were fixed and permeabilized with Perm/Wash buffer (eBiosciences, Santa Clara, CA, USA), and then incubated with each antibody (Table 2) at 4℃ for 30 min.

Statistical Analysis

All data are shown as mean±SEM. An unpaired Student’s t-test was used to detect significant differences when 2 groups were compared. The parameters of echocardiography were analyzed for differences by 2-way ANOVA with repeated measures, followed by the Bonferroni post-hoc test to assess individual differences. Values of P<0.05 were considered statistically significant.

Results

Deletion of CD28 Impairs Survival Rate and Cardiac Function

To identify the effects of CD28 deletion after MI, we assessed cumulative survival, cardiac function, and infarct size using CD28KO mice. The survival rate at 5 days after MI was significantly lower in CD28KO mice than in Control mice (Control: 86.21%, CD28KO: 69.84%, P<0.05) (Figure 1A). The rate of cardiac rupture was higher in CD28KO mice than in Control mice (Control: 10.34%, CD28KO: 30.16%, P<0.01) (Figures 1B,C). There were no significant differences in physiological parameters, such as body weight and heart rate, and echocardiographic parameters between Control mice and CD28KO mice before MI (Table 3). At 5 days after MI, LV end-diastolic diameter and LV end-systolic diameter were significantly increased, and the percentage fractional shortening (FS) was significantly decreased in CD28KO mice compared with Control mice. We performed Masson trichrome staining by using heart tissue at 5 days after MI to evaluate infarct size. There were no significant differences in infarct size between Control mice and CD28KO mice (Control: 61.59±4.37%, CD28KO: 64.01±3.13%; NS) (Figure 1D). These results suggest that the deletion of CD28 induces LV dilatation and increases cardiac rupture death independent of infarct size.

Figure 1.

Effects of CD28 deletion on survival rate and infarct size. (A) Kaplan-Meier survival analysis in Control mice (n=58) and CD28 knockout (CD28KO) mice (n=63). (B) Representative macroscopic and microscopic photographs of myocardial infarction (MI) with left ventricular (LV) myocardial rupture in CD28KO mice on day 3 after MI. (C) Analysis of cardiac rupture rate as determined by using the log-rank test. (D) Representative Masson trichrome staining of heart tissue in Control mice and CD28KO mice at 5 days after MI. Quantitative analysis of infarct size at 5 days after MI. *P<0.05 as determined by the 2-tailed Student’s t-test (n=8 per group). Scale bar=1 mm.

Table 3. Echocardiographic Data

	Control			CD28KO
	Pre-operation	3 days	5 days	Pre-operation	3 days	5 days
n	8	8	8	9	9	9
Heart rate (beats/min)	479±12	428±15	419±16*	467±9	441±12*	432±10*
LVEDD (mm)	3.41±0.01	4.51±0.05*	5.29±0.08*	3.40±0.01	4.64±0.06*	6.04±0.20*,†
LVESD (mm)	1.86±0.01	3.92±0.05*	4.77±0.08*	1.84±0.01	4.07±0.06*	5.60±0.20*,†
FS (%)	45.55±0.28	13.09±0.35*	9.87±0.44*	45.79±0.28	12.16±0.55*	7.32±0.37*,†
AWT (mm)	0.86±0.01	0.65±0.01*	0.58±0.01*	0.86±0.01	0.64±0.02*	0.53±0.02*
PWT (mm)	0.86±0.01	0.71±0.03*	0.67±0.03*	0.87±0.01	0.76±0.02*	0.60±0.02*

*P<0.05 vs. corresponding Pre-operation group. ^†P<0.05 vs. Control group mice at 5 days after MI. AWT, anterior wall thickness; FS, fractional shortening; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; MI, myocardial infarction; PWT, posterior wall thickness.

Deletion of CD28 Abolishes Scar Formation After MI

We suspected that the composition of the infarct area may be involved in the increases of LV dilatation and cardiac rupture death in CD28KO mice after MI. To confirm our hypothesis, we performed picrosirius red staining to assess collagen composition in the infarct area. At 5 days after MI, the extent of extracellular collagen in the infarct area was significantly decreased in CD28KO mice compared with Control mice (Control: 29.42±1.84%, CD28KO: 20.34±2.05%, P<0.05) (Figure 2A). There were no significant differences in cardiomyocyte size and myocardial fibrosis in the non-ischemic area between Control mice and CD28KO mice. Electron microscopic analysis demonstrated that there were few collagen fibrils in the myocardium of CD28KO mice compared with Control mice (Figures 2B–E). Furthermore, picrosirius red polarized microscopy of collagen fibers in the infarct area showed that yellow fibers, which are composed by collagen type I, the amount of yellow fibers low, and green fiber, which are composed by collagen type III, tended to be high in CD28KO mice compared with Control mice (Figures 2F–I). These results suggest that there was a delay of the wound healing in the myocardium of CD28KO mice compared with Control mice.

Figure 2.

Effects of CD28 deletion on cardiac fibrosis after myocardial infarction (MI). (A) Representative picrosirius red staining of the infarct area in Control mice and CD28 knockout (CD28KO) mice at 5 days after MI. Quantitative analysis of the fibrotic area in the infarct area at 5 days after MI. *P<0.05 as determined by the 2-tailed Student’s t-test (n=5 per group). Scale bar=100 μm. (B,C) Representative electron micrographs of the infarct area in Control mice and CD28KO mice at 5 days after MI. (D,E) Higher magnification of the boxed area in B,C. Scale bar=1 μm. *Collagen fibrils. (F–I) Representative picrosirius red staining of the infarct area in Control mice (F) and CD28KO mice (G), and polarized microscopic view of a corresponding picrosirius red-stained section (H,I). Scale bar=50 μm.

Deletion of CD28 Reduces the Production of Collagen Fibers and Accelerates the Degradation of Collagen Fibers

To investigate the reduction of collagen fibers in the infarct area in CD28KO mice, we evaluated the mRNA expression of procollagen α-1 (I) and procollagen α-1 (III). However, at 5 days after MI, there were no significant differences in the mRNA expression of procollagen α-1 (I) and procollagen α-1 (III) between Control mice and CD28KO mice (procollagen α-1 (I)/HPRT: Control: 1.51±0.19; CD28KO: 2.09±0.30; procollagen α-1 (III)/HPRT: Control: 29.58±4.34; CD28KO: 35.21±2.99). Thus, we also performed immunohistochemical staining and assessed the number of α-SMA-positive spindle-shaped myofibroblasts in the infarct area. At 5 days after MI, the number of myofibroblasts in the infarct area was significantly decreased in CD28KO mice compared with Control mice (Control: 731±96/mm²; CD28KO: 320±27/mm², P<0.01) (Figure 3A). There were only a few cardiac myofibroblasts before MI in both groups. Next, we analyzed the activities of MMP-2 and MMP-9 in the hearts after MI. There were no significant differences in the activities of MMP-2 and MMP-9 between Control mice and CD28KO mice at 3 days after MI. But at 5 days after MI, MMP-9 activity was significantly higher in CD28KO mice compared with Control mice (Control: 22.97±7.98; CD28KO: 99.41±17.05, P<0.01) (Figure 3B). These results suggest that the degradation of collagen fiber by MMP-9 is the main reason for the reduction of collagen fiber in the scar area of CD28KO mice after MI.

Figure 3.

Effects of CD28 deletion of collagen fiber production and degradation. (A) Representative photograph of immunohistochemical staining for α-smooth muscle actin (α-SMA; brown) in the infarcted heart of Control mice and CD28 knockout (CD28KO) mice at 5 days after myocardial infarction (MI). Quantitative analysis of the number of α-SMA-positive myofibroblasts in the infarct area in Control and CD28KO mice. **P<0.01 as determined by the 2-tailed Student’s t-test (n=4 per group). Scale bar=50 μm. (B) Representative photograph of a zymographic gel demonstrating the time-course of matrix metalloproteinase (MMP)-2 and MMP-9 activities in the infarcted heart in Control mice and CD28KO mice. Quantitative analysis of MMP-9 and MMP-2 activities in the infarcted heart. **P<0.01 vs. corresponding sham group. ^††P<0.01 vs. Control group mice at 5 days after MI as determined by 2-way ANOVA followed by the Bonferroni post-hoc test (n=4 per group).

Deletion of CD28 Enhances Inflammatory Cytokines

To determine the involvement of inflammation, we evaluated the mRNA expressions of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, interferon (IFN)-γ, IL-10, and transforming growth factor (TGF)-β. There were no significant differences in the mRNA expressions of IL-10 and TGF-β, which are associated with suppressing inflammation and wound healing, between Control mice and CD28KO mice at 5 days after MI. But the mRNA expressions of TNF-α, IL-1β, and IFN-γ, which are pro-inflammatory cytokines, were significantly higher in CD28KO mice than in Control mice at 5 days after MI (Figures 4A–E). These results suggest that deletion of CD28 extends the inflammatory period after MI and delays the wound healing in the infarct area.

Figure 4.

Quantitative analysis of gene expression in the infarct area. Time-course of mRNA expression of tumor necrosis factor (TNF)-α (A), interleukin (IL)-1β (B), interferon (IFN)-γ (C), IL-10 (D), and transforming growth factor (TGF)-β (E) in the infarct area of Control mice and CD28KO mice. *P<0.05 as determined by the 2-tailed Student’s t-test (n=8 per group). Quantitative analysis of mRNA expression of inducible nitric oxide synthase (iNOS) (F), Arginase-1 (G), and Foxp3 (H) in the infarct area at 5 days after myocardial infarction (MI). *P<0.05, **P<0.01 vs. the corresponding sham group. ^†P<0.05, ^††P<0.01 vs. the Control group mice at 3 or 5 days after MI as determined by 2-way ANOVA followed by the Bonferroni post-hoc test (n=8 per group).

Deletion of CD28 Extends Inflammatory Cell Infiltration

Next, we determined immune cell infiltration in the myocardium at 5 days after MI by flow-cytometric analysis. There were no significant differences in the total number of infiltrating CD45⁺ immune cells between Control mice and CD28KO mice. However, the percentage of infiltrating Ly6G⁺ neutrophils in CD11b⁺ myeloid cells was significantly higher in CD28KO mice than in Control mice (Control: 10.77±3.30%; CD28KO: 27.80±5.71%; P<0.05) (Figures 5A,B). Furthermore, we focused on CD11b⁺ Ly6G^– CD64⁺ myeloid-epithelial-reproductive tyrosine kinase (MerTK)⁺ macrophage subsets. The percentage of Ly6C^high M1 macrophages (classically activated macrophages) from the total macrophages was significantly higher (Control: 7.30±1.91%; CD28KO: 16.62±1.43%; P<0.05), and the percentage of Ly6C^low M2 macrophages (alternatively activated macrophages) from the total macrophages was significantly lower in CD28KO mice than in Control mice (Control: 92.70±1.91%; CD28KO: 83.38±1.43%; P<0.05) (Figures 5A,B). The mRNA expression of inducible nitric oxide synthase (iNOS) was significantly higher in CD28KO mice than in Control mice, but there was no significance differences in the mRNA expression of arginase-1 between the 2 groups (Figures 4F,G). These results suggest that deletion of CD28 extends the persistence of proinflammatory immune cells in the infarct area.

Figure 5.

Representative flow cytometric analysis of immune cells that infiltrated into the infarct heart. (A) Representative flow cytometric dot plots of neutrophil (CD11b⁺ Ly6G⁺), M1 macrophage (CD11b⁺ Ly6G^– CD64⁺ MerTK⁺ Ly6C^high), and M2 macrophage (CD11b⁺ Ly6G^– CD64⁺ MerTK⁺ Ly6C^low) in the infarcted heart in Control mice and CD28 knockout (CD28KO) mice. (B) Quantitative analysis of the ratio of neutrophil to total CD11b⁺ cells. Quantitative analysis of the ratio of M1 and M2 macrophages to the total macrophage population (CD11b⁺ Ly6G^– CD64⁺ MerTK⁺) in the infarcted heart in Control mice and CD28KO mice. **P<0.01 as determined by the 2-tailed Student’s t-test (n=6 per group). (C) Quantitative analysis of the ratio of CD4+ T cell, CD8+ T cell, TCRγδ T cell, and B cell to total CD11b^– cells in the infarcted heart in Control mice and CD28KO mice. Not significant (N.S.) as determined by the 2-tailed Student’s t-test (n=6 per group). (D) Representative flow cytometric dot plots of Tregs (CD11b^– CD4⁺ IFN-γ^– Foxp3⁺) in the infarcted heart in Control mice and CD28KO mice. Quantitative analysis of the ratio of Tregs to total CD4⁺ T cells. **P<0.01 as determined by the 2-tailed Student’s t-test (n=6 per group).

To further investigate the mechanisms of the prolongation of the inflammatory period in the infarct area after MI in CD28KO mice, we analyzed lymphocytes infiltrated into the infarct area. Although there were no significant differences in the percentages of infiltrating CD4⁺ T cells, CD8⁺ T cells, γδT cells, and B cells between Control mice and CD28KO mice (Figure 5C), the percentage of Foxp3⁺ Tregs in total CD4⁺ T cells was significantly lower in CD28KO mice than in Control mice (Figure 5D). The mRNA expression of Foxp3 was also lower in CD28KO mice than in Control mice (Figure 4H). These results suggest that insufficient recruitment of Tregs in the infarct area may induce prolongation of the inflammatory period and delay the wound healing process.

Discussion

In the present study, we confirmed that the deletion of CD28 impaired cardiac function and survival rate, and cause the progression of LV remodeling and cardiac rupture after MI. Deletion of CD28 decreased the number of α-SMA positive myofibroblasts and increased the activation of MMP-9, leading to a decrease in the extent of myocardial fibrosis in the infarct area by degradation of collagen fiber. Deletion of CD28 extended the inflammatory period and delayed the time-course of the wound healing process in the infarct area.

A recent study suggested that there is a relationship between the progression of LV remodeling and the wound healing process of infarct heart tissue in the mice MI model.¹³ Three phases of the wound healing process (inflammatory phase, reparative phase and maturation phase) are critical following MI.¹⁴ In the inflammatory phase, M1 macrophages clear dead myocyte debris through phagocytosis and proteolysis, and secrete inflammatory cytokines including IL-1β, IL-6 and TNF-α, as well as proteases including MMP.^15,16 In the reparative phase, M2 macrophages secrete the anti-inflammatory cytokines, IL-10 and TGF-β, which in turn recruit and activate reparative myofibroblasts. Myofibroblasts secrete large amounts of extracellular matrix (ECM) in order to replace lost ventricular tissue with a stable scar.^17,18 The maturation phase is marked by apoptosis of the majority of inflammatory and reparative cells, scar maturation and remodeling. During the replacement process of necrotic tissue to collagen fibers, both MMP-2 and MMP-9 play an important role in the rearrangement of collagen fibers; myofibroblasts also produce collagen fibers.^12,17–20 It has been reported that MMP-9 is produced by M1 macrophages.²¹ The activity of MMP-9 is increased at 3 days after MI and attenuated at 7 days after MI, while MMP-2 was increased during 3 days to 7 days after MI.^4,12,22 In our experiments, we confirmed a significant increase in the activity of MMP-9, with the increases in the M1 macrophage in CD28KO mice compared with Control mice at 5 days after MI. Therefore, we think that the increased in number M1 macrophage during the infarcted area in production of collagen fibers by myofibroblasts in the infarcted area.^17,18 The mRNA expression of TGF-β was higher in CD28KO mice compared with Control mice at 3 days after MI in our experiments. Some reports revealed that IL-1β inhibits the proliferation and differentiation of cardiac fibroblasts to myofibroblasts in the healing heart after MI, independent of TGF-β.^23,24 We confirmed the significant increases of the mRNA expression of IL-1β in CD28KO mice compared with Control mice. These results suggest that both the decrease of collagen production by myofibroblasts and degradation of collagen fiber by MMP-9 are the main reasons for the reduction of collagen fiber in the scar area of CD28KO mice after MI. The expression levels of IL-10 in the myocardium were increased at 3 and 5 days after MI in both Control mice and CD28KO mice in the present study. A previous paper also reported that IL-10 in the myocardium is increased at 7 day after MI.⁴ Because there were no significant differences in the expression levels of IL-10 between Control mice and CD28KO at 3 and 5 days after MI, we believe that IL-10 is not involved in cardiac repair in CD28KO mice. Another paper reported that the level of IL-10 does not affect the repair of ischemic heart after MI using IL-10 KO mice.²⁵ Both iNOS and Arginase-1 are known as the marker of M1 and M2 macrophages, respectively. We checked the expression levels of iNOS and Arginase-1 to confirm the M1 and M2 macrophage polarization, respectively. In fluorescence activated cell sorting (FACS) analysis, the M1 macrophage was increased and the M2 macrophage was decreased in the infarcted area after MI in CD28KO mice compared to Control mice (Figures 5A,B). Foxp3 is a transcriptional factor that is expressed specifically in Tregs and controls the differentiation of Tregs. We believe that the decrease in expression level of Foxp3 is associated with the decrease in number of Tregs in the infarcted area in CD28KO mice (Figure 5D). Although there were no significant histopathological changes in the non-infarcted area (data not shown), we evaluated the gene expression in the non-infarcted area in the heart. There were no significant differences in gene expression levels of TNF-α, IL-1β, IFN-γ, IL-10, TGF-β, iNOS, Arg-1, Foxp3, and Collagen I/III in the non-infarcted area at 5 days after MI (data not shown). Therefore, we believe that the protective response induced by CD28 does not occur in the non-ischemic area. By the prolongation of inflammation in the infarcted area in CD28KO mice, the increase in IL-1β released by the M1 macrophage and neutrophils in the heart might inhibit differentiation of cardiac fibroblasts to myofibroblasts, leading to cardiac rupture and LV remodeling after MI.

Cardiac rupture is a serious complication that occurs at several days after the onset of MI. An inadequate healing process after MI may result in cardiac rupture, LV aneurysm, and congestive heart failure by LV remodeling. Acute myocyte loss and breakdown of ECM promote early ventricular expansion, which is a trigger for subacute cardiac rupture after MI. Inhibition of collagen synthesis was shown to be associated with an increase in the risk of cardiac rupture in MI patients.²⁶ These results suggest that appropriate reparative fibrosis is important in preventing cardiac rupture after MI. As well as collagen synthesis, inflammation is also a crucial factor in the post-MI healing process. An excessive inflammatory response in the infarcted myocardium is related to adverse cardiac events, including cardiac rupture.^27,28 Although the detailed mechanisms of cardiac rupture is unclear, it is important to determine the factors that are associated with it’s pathophysiology. In the present study, the levels of cytokines such as TNF-α, IL-1β, IFN-γ, TGF-β and iNOS were increased in CD28KO mice. Moreover, the number of neutrophils and M1 macrophages were increased in CD28KO mice group. These changes in cytokine level and immune cell number seem to be related to the occurrence of cardiac rupture. Further experiments are needed to examine the relationship of these data and outcome in MI patients because these values may be useful as prognostic markers of cardiac rupture.

The CD28 molecule on the CD4⁺ T-cell membrane interacts with CD80/86 and provides the co-stimulatory signals required for T-cell activation and survival.^29,30 Shahinian et al demonstrated that the development of T and B cells appeared normal in the CD28KO mice, but T cells derived from CD28KO mice were impaired in the ability to produce IL-2.⁸ They also suggested that CD28 is not required for all T-cell responses in vivo, and instead proposed that alternative co-stimulatory pathways may exist. In fact, there are many co-stimulatory and co-inhibitory molecules, and these molecules have been investigated as new therapeutic targets.^31,32 Therefore, it is difficult to abolish the effects of co-stimulatory signals in vivo completely. There were 2 important studies about antigen presentation after MI. Anzai et al revealed that DC ablation, which means that the antigen presenter is absent, induced the increases of M1 macrophage infiltration and decreases of M2 macrophage infiltration, and deteriorated left ventricular function and remodeling.⁴ Hofmann et al showed that CD4⁺ helper T-cell ablation, which means the receiver is absent, induced a disturbance of collagen matrix formation in the infarct zone, and exacerbated left ventricular remodeling.⁵ They also demonstrated that a lack of TCR signals led to the increases of cardiac rupture death.⁵ This study, however, did not refer to the effects of co-stimulatory signals after MI. In this study, we demonstrated the effect of the lack of co-stimulatory signals after MI by using CD28KO mice, and our results were in accordance with those prior studies about antigen presentation after MI. We consider that prevention of co-stimulatory signals is one of the reasons for the delay of wound healing after MI in CD28KO mice. Currently, soluble fusion protein, composed of the Fc region of the immunoglobulin IgG1, fused to the extracellular domain of cytotoxic T-lymphocyte-associated protein-4, Abatacept, which blocks CD80/86–CD28 signals and inhibits the TCR signals in T cells, and has already been used for the treatment of rheumatoid arthritis patients.³³ Although we could not identify the detailed molecular mechanisms of CD28 deletion in the post-MI heart in the present study, we suggested the possibility of harmful effects of CD80/86–CD28 signal inhibition on the MI heart. There is a possibility that deletion of CD28 signaling may progress LV remodeling after MI. In another recently reported study found that the activation of CD28 by a superagonistic anti-CD28 antibody ameliorates the progression of LV remodeling.⁶ However, the use of a superagonistic anti-CD28 antibody may not be realistic because of the many serious side-effects, such as multi-organ failure and near-fatal cytokine release syndrome, which were found in another clinical study.³⁴ Further studies are necessary to confirm the effects of CD28 modification on LV remodeling after MI and elucidate the detailed mechanism of it.

Conclusions

Deletion of CD28 exacerbates LV remodeling and increases cardiac rupture after MI by the prolongation of the inflammatory period and reduction of collagen fiber in the infarct hearts in a mice MI model.

Acknowledgments

The authors thank Ryo Abe (Research Institute for Biological Sciences, Science University of Tokyo, Chiba, Japan) for the kind gift of CD28KO mice. The authors thank Y. Ohtsuki, M. Furuyama, and Y. Shimizu for their excellent technical assistance.

Funding

This work was supported by the grants from MSD K.K., Terumo Life Science Foundation, Takeda Science Foundation, and the Japan Diabetes Foundation.

Disclosures

H. Takano is supported by grants from MSD K.K., Kowa Pharmaceutical Co Ltd, Sumitomo Dainippon Pharma Co. Ltd, Takeda Pharmaceutical Co Ltd, Novartis Pharma K.K., and Astellas Pharma Inc. And Y. Kobayashi is supported by grants from Genzyme Japan K.K., MSD K.K., Sumitomo Dainippon Pharma Co Ltd, Shionogi & Co Ltd, Nihon Medi-Physics Co Ltd, Kowa Pharmaceutical Co Ltd, Nippon Boehringer Ingelheim Co Ltd, Phizer Japan, Inc, Bristol-Myers Squibb Company, Chugai Pharmaceutical Co Ltd, Takeda Pharmaceutical Co Ltd, Otsuka Pharmaceutical Co. Ltd, Mochida Pharmaceutical Co Ltd, St. Jude Medical, Inc, Japan Lifeline Co Ltd, Novartis Pharma K.K., Sanofi K.K., Astellas Pharma Inc, FUJIFILM RI Pharma Co Ltd, Daiichi Sankyo Co Ltd, Abbott Vascular Japan Co Ltd, and Biotronik Japan Co Ltd.

References

• 1.    Tomoike  H,  Yokoyama  H,  Sumita  Y,  Hanai  S,  Kada  A,  Okamura  T, et al. Nationwide distribution of cardiovascular practice in Japan: Results of Japanese circulation society 2010 annual survey. Circ J 2015; 79: 1058–1067.
• 2.    Ishihara  M,  Fujino  M,  Ogawa  H,  Yasuda  S,  Noguchi  T,  Nakao  K, et al. Clinical presentation, management and outcome of Japanese patients with acute myocardial infarction in the Troponin Era: Japanese Registry of Acute Myocardial Infarction Diagnosed by Universal Definition (J-MINUET). Circ J 2015; 79: 1255–1262.
• 3.    Lichtman  AH. The heart of the matter: Protection of the myocardium from T cells. J. J. Autoimmun 2013; 45: 90–96.
• 4.    Anzai  A,  Anzai  T,  Nagai  S,  Maekawa  Y,  Naito  K,  Kaneko  H, et al. Regulatory role of dendritic cells in postinfarction healing and left ventricular remodeling. Circulation 2012; 125: 1234–1245.
• 5.    Hofmann  U,  Beyersdorf  N,  Weirather  J,  Podolskaya  A,  Bauersachs  J,  Ertl  G, et al. Activation of CD4 + T lymphocytes improves wound healing and survival after experimental myocardial infarction in mice. Circulation 2012; 125: 1652–1663.
• 6.    Weirather  J,  Hofmann  UDW,  Beyersdorf  N,  Ramos  GC,  Vogel  B,  Frey  A, et al. Foxp3+ CD4+ T cells improve healing after myocardial infarction by modulating monocyte/macrophage differentiation. Circ Res 2014; 115: 55–67.
• 7.    Tai  X,  Cowan  M,  Feigenbaum  L,  Singer  A. CD28 costimulation of developing thymocytes induces Foxp3 expression and regulatory T cell differentiation independently of interleukin 2. Nat Immunol 2005; 6: 152–162.
• 8.    Shahinian  A,  Pfeffer  K,  Lee  KP,  Kündig  TM,  Kishihara  K,  Wakeham  A, et al. Differential T cell costimulatory requirements in CD28-deficient mice. Science 1993; 261: 609–612.
• 9.    Ueda  K,  Takano  H,  Niitsuma  Y,  Hasegawa  H,  Uchiyama  R,  Oka  T, et al. Sonic hedgehog is a critical mediator of erythropoietin-induced cardiac protection in mice. J Clin Invest 2010; 120: 2016–2029.
• 10.    Kubota  A,  Takano  H,  Wang  H,  Hasegawa  H,  Tadokoro  H,  Hirose  M, et al. DPP-4 inhibition has beneficial effects on the heart after myocardial infarction. J Mol Cell Cardiol 2016; 91: 72–80.
• 11.    Woessner  JF. Quantification of matrix metalloproteinases in tissue samples. Methods Enzymol 1995; 248: 510–528.
• 12.    Matsumura  S,  Iwanaga  S,  Mochizuki  S,  Okamoto  H,  Ogawa  S,  Okada  Y. Targeted deletion or pharmacological inhibition of MMP-2 prevents cardiac rupture after myocardial infarction in mice. J Clin Invest 2005; 115: 599–609.
• 13.    Frangogiannis  NG. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol 2014; 11: 255–265.
• 14.    Francis Stuart  SD,  De Jesus  NM,  Lindsey  ML,  Ripplinger  CM. The crossroads of inflammation, fibrosis, and arrhythmia following myocardial infarction. J Mol Cell Cardiol 2016; 91: 114–122.
• 15.    Frangogiannis  NG. Regulation of the inflammatory response in cardiac repair. Circ Res 2012; 110: 159–173.
• 16.    Hofmann  U,  Frantz  S. Role of lymphocytes in myocardial injury, healing, and remodeling after myocardial infarction. Circ Res 2015; 116: 354–367.
• 17.    van den Borne  SWM,  Diez  J,  Blankesteijn  WM,  Verjans  J,  Hofstra  L,  Narula  J. Myocardial remodeling after infarction: The role of myofibroblasts. Nat Rev Cardiol 2010; 7: 30–37.
• 18.    Shinde  AV,  Frangogiannis  NG. Fibroblasts in myocardial infarction: A role in inflammation and repair. J Mol Cell Cardiol 2014; 70: 74–82.
• 19.    Hayashidani  S,  Tsutsui  H,  Ikeuchi  M,  Shiomi  T,  Matsusaka  H,  Kubota  T, et al. Targeted deletion of MMP-2 attenuates early LV rupture and late remodeling after experimental myocardial infarction. Am J Physiol Heart Circ Physiol 2003; 285: H1229–H1235.
• 20.    Ducharme  A,  Frantz  S,  Aikawa  M,  Rabkin  E,  Lindsey  M,  Rohde  LE, et al. Targeted deletion of matrix metalloproteinase-9 attenuates left ventricular enlargement and collagen accumulation after experimental myocardial infarction. J Clin Invest 2000; 106: 55–62.
• 21.    Murray  PJ,  Allen  JE,  Biswas  SK,  Fisher  EA,  Gilroy  DW,  Goerdt  S, et al. Macrophage activation and polarization: Nomenclature and experimental guidelines. Immunity 2014; 41: 14–20.
• 22.    Sobirin  MA,  Kinugawa  S,  Takahashi  M,  Fukushima  A,  Homma  T,  Ono  T, et al. Activation of natural killer T cells ameliorates postinfarct cardiac remodeling and failure in mice. Circ Res 2012; 111: 1037–1047.
• 23.    Saxena  A,  Chen  W,  Su  Y,  Rai  V,  Uche  OU,  Li  N, et al. IL-1 induces proinflammatory leukocyte infiltration and regulates fibroblast phenotype in the infarcted myocardium. J Immunol 2013; 191: 4838–4848.
• 24.    Palmer  JN,  Hartogensis  WE,  Patten  M,  Fortuin  FD,  Long  CS. Interleukin-1 beta induces cardiac myocyte growth but inhibits cardiac fibroblast proliferation in culture. J Clin Invest 1995; 95: 2555–2564.
• 25.    Zymek  P,  Nah  DY,  Bujak  M,  Ren  G,  Koerting  A,  Leucker  T, et al. Interleukin-10 is not a critical regulator of infarct healing and left ventricular remodeling. Cardiovasc Res 2007; 74: 313–322.
• 26.    Peuhkurinen  K,  Risteli  L,  Jounela  A,  Risteli  J. Changes in interstitial collagen metabolism during acute myocardial infarction treated with streptokinase or tissue plasminogen activator. Am Heart J 1996; 131: 7–13.
• 27.    Anzai  T,  Yoshikawa  T,  Shiraki  H,  Asakura  Y,  Akaishi  M,  Mitamura  H, et al. C-reactive protein as a predictor of infarct expansion and cardiac rupture after a first Q-wave acute myocardial infarction. Circulation 1997; 96: 778–784.
• 28.    Takahashi  T,  Anzai  T,  Yoshikawa  T,  Maekawa  Y,  Asakura  Y,  Satoh  T, et al. Serum C-reactive protein elevation in left ventricular remodeling after acute myocardial infarction -- role of neurohormones and cytokines. Int J Cardiol 2003; 88: 257–265.
• 29.    Sperling  AI,  Bluestone  JA. The complexities of T-cell co-stimulation: CD28 and beyond. Immunol Rev 1996; 153: 155–182.
• 30.    Boussiotis  VA,  Freeman  GJ,  Gribben  JG,  Nadler  LM. The role of B7-1/B7-2:CD28/CLTA-4 pathways in the prevention of anergy, induction of productive immunity and down-regulation of the immune response. Immunol Rev 1996; 153: 5–26.
• 31.    Rhodes  DA,  Reith  W,  Trowsdale  J. Regulation of immunity by butyrophilins. Annu Rev Immunol 2016; 34: 151–172.
• 32.    Shanker  A,  Pellom  ST,  Dudimah  DF,  Thounaojam  MC,  de Kluyver  RL,  Brooks  AD, et al. Bortezomib improves adoptive T-cell therapy by sensitizing cancer cells to FasL cytotoxicity. Cancer Res 2015; 75: 5260–5272.
• 33.    Caporali  R,  Bugatti  S,  Cavagna  L,  Antivalle  M,  Sarzi-Puttini  P. Modulating the co-stimulatory signal for T cell activation in rheumatoid arthritis: Could it be the first step of the treatment? Autoimmun Rev 2014; 13: 49–53.
• 34.    Vessillier  S,  Eastwood  D,  Fox  B,  Sathish  J,  Sethu  S,  Dougall  T, et al. Cytokine release assays for the prediction of therapeutic mAb safety in first-in man trials -- whole blood cytokine release assays are poorly predictive for TGN1412 cytokine storm. J Immunol Methods 2015; 424: 43–52.