Deficiency of IL-12p35 improves cardiac repair after myocardial infarction by promoting angiogenesis

Abstract

Aims

IL-12p35 is a pro-inflammatory cytokine that participates in a variety of inflammatory diseases. This study aimed to determine whether IL-12 regulates cardiac injury and repair following acute myocardial infarction (AMI) and investigate the underlying mechanisms.

Methods and results

Mice with AMI showed a marked increase in IL-12p35 expression of ischaemic cardiac tissues. IL-12 was mainly produced by CD11b⁺ monocytes. Cardiac functions were significantly improved in IL-12p35 knockout (p35-KO) mice compared with wild-type (WT) littermates in response to AMI. IL-12p35 deficiency attenuated the infarct scar and hypertrophy compared with WT mice. RNA transcriptome sequencing and quantitative RT–PCR analysis of CD11b⁺ monocytes isolated from WT and p35-KO ischaemic hearts revealed a distinct transcriptional profile in p35-KO CD11b⁺ monocytes, displaying pro-angiogenesis and anti-inflammation properties. Angiogenesis was enhanced in p35-KO mice with AMI and hindlimb ischaemia. Moreover, tube formation assay and Matrigel plug analysis demonstrated that IL-12 inhibition of angiogenesis was dependent on monocytes. IL-12p35 deficiency inhibited inflammation by reducing chemokine production and monocyte infiltration into the heart. Finally, administration of an IL-12p35-neutralizing antibody limited AMI-induced inflammatory cell infiltration into the heart and improved angiogenesis and cardiac function.

Conclusions

Deficiency of IL-12p35 limited AMI-induced cardiac injury by promoting pro-angiogenesis and anti-inflammatory functions of monocytes.

IL-12, Monocyte, Angiogenesis, Myocardial infarction

1. Introduction

Acute myocardial infarction (AMI) remains a leading cause of death worldwide.¹ Despite numerous advances in therapeutic strategies, such as percutaneous coronary intervention and coronary artery bypass grafting, post-AMI mortality is still high (5-year survival rates are 30–70%).² Current therapies are unable to prevent cardiac dysfunction and transition to heart failure in myocardial infarction patients. Thus, there is a need for comprehensive understanding of the mechanisms of cardiac injury and repair after AMI.

Inflammatory responses play a critical role in myocardial infarction followed by left ventricular injury and repair mechanisms.³ Dying cardiomyocytes release their intracellular contents shortly after AMI, triggering an intense inflammatory reaction that is mainly characterized by leucocyte infiltration. This response clears dead cells from the infarcted area and extracellular matrix debris during the early phase, while extensive inflammation causes cardiac damage through reactive oxygen species or activation of the complement cascade.⁴ Inflammation has a significant role in reparative processes through promotion of angiogenesis. At ∼5 days after AMI, the remaining inflammatory cells, mostly monocytes/macrophages, coordinate with endothelial cells to regulate angiogenesis and enhance the blood supply to hypoxic cardiomyocytes.⁵ Therefore, the inflammatory response is involved in all stages of the pathological process of myocardial infarction and repair.

IL-12 is a key pro-inflammatory cytokine consisting of p40 and p35 subunits joined by disulphide bonds.⁶ It is produced by macrophages, dendritic cells, and B cells in response to microbial pathogens.⁷ IL-12 promotes the differentiation of naive CD4⁺ T cells into mature Th1 effector cells that secrete IFN-γ.⁸ IL-12 exerts pro-atherogenic effects in atherosclerosis, and IL-12 levels are positively correlated with arterial stiffness in atherosclerosis-prone individuals.⁹ There is also a positive correlation between serum IL-12 concentrations and atherosclerosis progression.¹⁰ Our previous study showed that IL-12p35 deletion induced production of transforming growth factor (TGF)-β in macrophages, leading to cardiac injury.¹¹ IL-12 may also be involved in the process of AMI, because serum IL-12 levels are significantly higher than those in healthy subjects.¹² However, the specific role of IL-12 in AMI is still unknown.

The aim of this study was to determine whether IL-12 affected cardiac injury/repair after AMI. We found that AMI markedly up-regulated IL-12p35 expression in monocytes. IL-12p35 deficiency induced by genetic or pharmacological approaches improved cardiac functions and repair after AMI. The deficiency in IL-12p35 promoted pro-angiogenesis and anti-inflammatory properties of monocytes during AMI. Thus, IL-12 is a potential therapeutic target for AMI.

2. Methods

Detailed descriptions of reagents, histopathology, immunohistochemistry, quantitative (q)RT–PCR, and tube formation assays are presented in the Supplementary material online, Methods.

2.1 Animals and AMI surgery

We used 10- to 12-week-old male IL-12p35 knockout (p35-KO) mice and littermate wild-type (WT) mice (all with a C57BL/6 background). p35-KO mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) and had been back-crossed with C57BL/6 mice for 10 generations. The genotype of p35-KO mice was confirmed by PCR as described previously.¹¹ Mice were maintained under specific pathogen-free conditions and provided with free access to food and water. The study was approved by the Institutional Animal Care and Use Committee of Capital Medical University, Beijing, China. All animal procedures conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

WT and p53-KO mice were anaesthetized by 2% isoflurane inhalation and underwent surgery to induce a myocardial infarction model by occlusion of the left anterior descending coronary artery (LAD) as described previously.^13,14 The sham group underwent the same surgical procedure without LAD occlusion. Post-operative analgesia [buprenorphine, 0.05 mg/kg/12 h, intraperitoneal (i.p.)] was administered for 48 h. Mice were sacrificed at 4 weeks post-surgery by carbon dioxide narcosis, and then heart tissues were harvested.

2.2 Flow cytometric analysis and sorting of cells for RNA sequencing

Flow cytometric analysis and sorting of cells were performed as described previously.^11,15 Mouse hearts were perfused with cold PBS for 4 min, quickly minced into small pieces, and then digested with 0.1% collagenase II and 2.4 U/mL dispase II in PBS at 37°C for 30 min. The cell suspension was filtered, centrifuged, and resuspended. Cells were then incubated for 30 min at 4°C with FITC-conjugated anti-CD31, Percp-Cy5.5-conjugated anti-CD45.2, PE-conjugated anti-CD11b, and FITC-conjugated anti-CD3 antibodies (all purchased from Biolegend, San Diego, CA, USA) diluted in PBS with 1% fetal bovine serum (FBS) while protected from light. Expression of surface molecules was analysed by an Epics XL flow cytometer (Beckman Coulter, Miami, FL, USA). For RNA sequencing, we used 10 WT and p35-KO mice that underwent AMI induction, which were sacrificed at 7 days post-surgery. CD45⁻CD31⁻, CD45⁻CD31⁺, CD45⁺CD3⁺, and CD45⁺ CD11b⁺ cells were isolated from pooled infarcted hearts by a MoFlo TM XDP (Beckman Coulter). Total RNA was extracted from these cells using TRIzol. The global gene expression profile was then examined by RNA sequencing analysis.

2.3 Western blot analysis

Western blot analysis was performed as described previously.¹¹ Briefly, protein extracts were prepared from ischaemic hearts below the ligature with cell lysis buffer [50 mmol/L Tris (pH 8), 250 mmol/L NaCl, 2 mmol/L EDTA, 1% IGEPAL, 200 mmol/L NaF, and 1 mmol/L Na3VO4, and complete protease inhibitor cocktail (Roche, Mannheim, Germany)]. Protein samples (50 µg) were separated by 10% SDS–polyacrylamide gel electrophoresis and then transferred to nitrocellulose membranes (Bio-Rad). The membranes were incubated with anti-CD31 (1:500), anti-IL12p35 (1:200), and anti-GAPDH (1 : 1000) primary antibodies at 4°C overnight and then infrared dye-conjugated secondary antibodies (1 : 5000; Rockland Immunochemicals, Gilbertsville, PA, USA) for 1 h at room temperature. Images were quantified using the Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA).

2.4 Matrigel plug assay

The Matrigel plug assay, a model for assessment of angiogenesis, was conducted as described previously¹⁶ with minor modifications. Mice were anaesthetized with one dose of ketamine (200 mg/kg, i.p.) and xylazine (10 mg/kg, i.p.) in 50 mL saline and then subcutaneously injected with 0.5 mL Matrigel plugs containing untreated WT or p35-KO monocytes, or those treated with recombinant mouse (rm)IL-12 at a site near the abdominal midline. Post-operative analgesia (buprenorphine, 0.05 mg/kg/12 h, i.p.) was induced for 48 h. At 7 days after injection, mice were sacrificed by carbon dioxide narcosis, and the Matrigel plugs were harvested, fixed with neutral buffered 10% formalin, and embedded in paraffin to prepare sections for immunohistochemical staining. Newly invaded endothelial cells were assessed by CD31 staining. A cross-section (5 µm) of the entire Matrigel plug was imaged at a magnification of ×100 under a Nikon microscope. The number of CD31⁺ cells was counted per field. Five fields were counted in each group. Three independent experiments were performed.

2.5 Hindlimb ischaemia induction

We also established a mouse hindlimb ischaemia (HLI) model as described previously.^17,18 Mice were anaesthetized with one dose of ketamine (200 mg/kg, i.p.) and xylazine (10 mg/kg, i.p.) in 50 mL saline and underwent surgery to induce HLI by resection of the femoral artery. Serial blood flow was then measured in the hindlimb for 3 days after surgery using a laser Doppler perfusion image (LDPI) analyzer (Moor Instruments, Wilmington, DE, USA). Mice were sacrificed at 3 days post-surgery by carbon dioxide narcosis, and gastrocnemius muscle tissues were harvested for RNA and immunohistochemical analyses.

2.6 Cell culture

Bone marrow-derived macrophages were prepared as described previously.^11,19 Briefly, WT and p35-KO mice were sacrificed by carbon dioxide narcosis, and then bone marrow cells were obtained by flushing their femurs and tibias. Suspensions were cleared of adipose and connective tissues by filtration and then subjected to Ficoll-gradient centrifugation to remove residual erythrocytes and non-lymphocytes. Myeloid-derived macrophages were cultured in DMEM (HyClone, Waltham, MA, USA) supplemented with 10% heat-inactivated FBS and 50 ng/mL macrophage colony-stimulating factor. Human umbilical vein endothelial cells (HUVECs; Lonza, Walkersville, MD, USA) were cultured to confluency at 37°C with 5% CO₂ in endothelial cell medium containing 5% FBS, 1% endothelial cell growth supplement, and 1% antibiotics (ScienCell).

2.7 Mouse treatments

Various doses of an anti-mouse IL-12p35 antibody (clone C18.2; eBioscience) or its respective isotype control were administered to mice as described previously,²⁰ including the optimal dose (100 μg/mouse), via the angular vein at 0, 2, 4, and 6 days after AMI surgery.

2.8 Statistics

All data are presented as the mean ± standard error of the mean. For all of the variables, the normality of distribution was checked by the Kolmogorov–Smirnov test. All data were normally distributed. Statistical analysis was performed using GraphPad Prism version 6.00 (GraphPad Software, Inc., San Diego, CA, USA) and SPSS 16.0 (SPSS Inc., Chicago, IL, USA) software. Two group comparisons were analysed by unpaired t-test. Multi-group comparisons were performed using one- or two-way ANOVA. The S-N-K post hoc test was used after the ANOVA analyses when appropriate. Survival data are shown as Kaplan–Meier curves; data were analysed by a log-rank test. Repeated-measures analysis of variance was used to evaluate the statistical significance of data acquired from the same animal over multiple time points. A P-value of <0.05 was considered statistically significant.

3. Results

3.1 IL-12p35 expression is increased in monocytes after AMI

To assess whether IL-12 has a role in AMI, we examined IL-12p35 expression in hearts of WT mice that underwent ligation of the LAD surgery. IL-12p35 mRNA expression was markedly up-regulated in heart tissue at 3 days after AMI (Figure 1A). Moreover, western blot analysis revealed that IL-12p35 protein expression in the infarction and border zones of the heart at 3 days after AMI was higher than that in sham-operated mice (Figure 1B). In addition, immunofluorescence revealed that the number of IL-12p35-positive cells in the infarction and border zones was greater than that in the remote zone of the heart at 3 days after AMI (Figure 1C). Double immunofluorescence staining for IL-12p35 and CD11b⁺ revealed that IL-12p35 mostly co-localized with CD11b⁺ monocytes in hearts at 3 days after AMI (Figure 1D). To determine the source of IL-12p35, we sorted CD45⁻CD31⁺ cells (endothelial cells), CD45⁻CD31⁻ cells (cardiac cells), CD45⁺CD3⁺ cells (T cells), and CD45⁺CD11b⁺ cells (monocytes) from infarcted hearts at 3 days after AMI. The RNA of these cells was extracted to detect the expression of IL-12p35 by qRT–PCR analysis. We found that IL-12p35 was mainly produced by monocytes (Figure 1E). These data demonstrated that AMI increases IL-12p35 at both mRNA and protein levels in the heart.

Figure 1

IL-12p35 expression is increased in monocytes after AMI. (A) Quantitative RT–PCR (qRT–PCR) analysis of IL-12p35 mRNA expression at 3 days after AMI (one-way ANOVA). *P < 0.05 vs. remote zone; n = 6. (B) Western blot analysis of IL-12p35 protein expression (left panel) and quantification (right panel) in mice at 3 days after AMI (one-way ANOVA). *P < 0.05 vs. sham operation; n = 5. (C) Immunofluorescence analysis of IL-12p35 protein expression in remote, infarction, and border zones at 3 days after AMI (n = 6). Scale bar = 50 μm. (D) Immunofluorescence analysis of IL-12p35 expression in CD11b⁺ monocytes of the infarction zone at 3 days after AMI (n = 6). Scale bar = 50 μm. (E) qRT–PCR analysis of IL-12p35 mRNA expression at 3 days after AMI (n = 4).

3.2 IL-12p35 deficiency improves cardiac function after AMI

To explore the role of IL-12p35 in the pathological processes of AMI, both WT and p35-KO mice underwent LAD occlusion. IL-12p35 deficiency significantly improved the post-AMI survival rate (Figure 2A). Cardiac functions were determined at 4 weeks after AMI by echocardiography. As shown in Figure 2B and C, IL-12p35 deficiency significantly augmented the left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS), and decreased the left ventricular end-diastolic diameter (LVEDD) and left ventricular end-systolic diameter (LVESD).

Figure 2

IL-12p35 deficiency improves cardiac function after AMI. (A) Survival curves of wild-type (WT) and p35-KO mice after AMI (log-rank test). (B) Representative M-mode echocardiograms of WT and p35-KO mice, including sham (top) and 4-week (bottom) graphs. (C) The LVEF, LVFS, LVEDD, and LVESD were measured (two-way ANOVA). (n = 6). *P < 0.05 vs. WT. (D and E) Masson trichrome-stained cardiac tissue from WT and p35-KO mice at 4 weeks after AMI, and quantification of interstitial fibrosis and the ratio of the infarct scar. Blue represents fibrosis (two-way ANOVA) (n = 6). *P < 0.05 vs. WT. Scale bar = 500 μm.

Because pathological remodelling is the primary cause of post-ischaemic cardiac dysfunction and heart failure,²¹ we investigated whether IL-12p35 deficiency might attenuate pathological remodelling. Over the course of 4 weeks, IL-12p35 deficiency significantly attenuated interstitial fibrosis and the infarct scar compared with WT mice (Figure 2D and E). Moreover, IL-12p35 deficiency reduced hypertrophy, decreased the heart to body weight ratio by 20% (P < 0.05, see Supplementary material online, Figure S1A and B), and reduced the cardiomyocyte size by 30% in remote non-ischaemic zones (P < 0.05, see Supplementary material online, Figure S1C and D). These data demonstrated that IL-12p35 deficiency improves cardiac functions of the left ventricle in AMI mice and attenuates adverse cardiac remodelling.

3.3 IL-12p35-KO CD11b⁺ monocytes exhibit a distinct transcriptome profile

After the onset of myocardial ischaemia, monocytes were recruited to the infarcted myocardium and regulated the injury/repair process.²¹ Because IL-12 was predominantly expressed by monocytes in the heart after AMI, CD11b⁺ monocytes were isolated from the infarcted hearts of WT and p35-KO mice at 7 days after AMI, and the global gene expression profile was examined by RNA sequence analysis. Compared with WT CD11b⁺ monocytes, differentially expressed gene analysis of the transcriptome revealed differential modulation of 1297 genes (1036 down-regulated and 261 up-regulated; false discovery rate: < 0.01) in p35-KO CD11b⁺ monocytes (Figure 3A). These differentially expressed genes were grouped using Gene Ontology (GO) analysis, which identified inflammatory response-related genes and angiogenesis-related genes as the most significantly regulated gene function groups in p35-KO CD11b⁺ monocytes (Figure 3B). As indicated in Table 1, p35-KO CD11b⁺ monocytes up-regulated several ‘pro-angiogenic’ genes, including angiogenin (ANG) and matrix metallopeptidase 8 (MMP8), which mediate key processes in angiogenesis.^22,23 p35-KO CD11b⁺ monocytes also down-regulated anti-angiogenic genes such as thrombospondin (THBS1) and arginase 1 (ARG-1). p35-KO CD11b⁺ monocytes down-regulated a large number of pro-inflammatory cytokine and chemokine genes including IL6, CCL4, and CCL5. p35-KO CD11b⁺ monocytes also up-regulated the gene encoding the anti-inflammatory factor SOCS3. Taken together, the transcriptome data indicated that p35-KO CD11b⁺ monocytes possess a distinct gene expression profile characterized by pro-angiogenic and anti-inflammatory properties. These results suggested that IL-12p35 regulates monocyte functions after AMI.

Table 1

Selected inflammatory genes and angiogenic genes in MI CD11b⁺ monocytes

Gene symbol	Full name	Fold change
Pro-angiogenic genes
ANG	Mus musculus angiogenin (ANG), mRNA.	1.44
MMP8	Mus musculus matrix metallopeptidase 8 (MMP10), mRNA.	2.87
MMP14	Mus musculus matrix metallopeptidase 14 (MMP10), mRNA.	1.22
Anti-angiogenic genes
MMP19	Mus musculus matrix metallopeptidase 9 (MMP10), mRNA.	−1.28
THBS1	Mus musculus thrombospondin 1(THBS1), mRNA.	−1.28
CXCL14	Mus musculus chemokine (C-X-C motif) ligand 14 (CXCL14), mRNA.	−1.97
COL18A1	Mus musculus collagen, type XVIII, alpha 1 (COL18A1), mRNA.	−2.14
TIMP1	Mus musculus tissue inhibitor of metalloproteinase 1 (TIMP1), mRNA.	−2.27
ARG1	Mus musculus arginase 1 (ARG1), mRNA.	−2.65
TIMP3	Mus musculus tissue inhibitor of metalloproteinase 3 (TIMP3), mRNA.	−2.76
COL15A1	Mus musculus collagen, type XV, alpha 1 (COL15A1), mRNA.	−1.6
COL4A1	Mus musculus collagen, type IV, alpha 1 (COL4A1), mRNA.	−3.34
COL4A2	Mus musculus collagen, type IV, alpha 2 (COL4A2), mRNA.	−3.11
Pro-inflammatory genes
CCL4	Mus musculus chemokine (C-C motif) ligand 4 (CCL4), mRNA.	−1.21
CCL6	Mus musculus chemokine (C-C motif) ligand 6 (CCL6), mRNA.	−1.24
CCL5	Mus musculus chemokine (C-C motif) ligand 5 (CCL5), mRNA.	−1.4
CCL7	Mus musculus chemokine (C-C motif) ligand 7 (CCL7), mRNA.	−1.57
CCL22	Mus musculus chemokine (C-C motif) ligand 22 (CCL22), mRNA.	−2.34
IL1R1	Mus musculus IL-1 receptor 1, (IL-1R1), mRNA.	−3.04
CXCL1	Mus musculus chemokine (C-X-C motif) ligand 1 (CXCL1), mRNA.	−1.83
CXCL14	Mus musculus chemokine (C-X-C motif) ligand 14 (CXCL14), mRNA.	−1.97
CXCL13	Mus musculus chemokine (C-X-C motif) ligand 13 (CXCL13), mRNA.	−2.53
CXCL12	Mus musculus chemokine (C-X-C motif) ligand 12 (CXCL12), mRNA.	−2.64
CXCL3	Mus musculus chemokine (C-X-C motif) ligand 3 (CXCL3), mRNA.	−3.07
IL-1R2	Mus musculus IL-1 receptor 2, (IL-1R2), mRNA.	−1.49
IL-1A	Mus musculus IL-1, alpha (IL-1A), mRNA.	−1.59
IL-6	Mus musculus IL-6, mRNA.	−2.31
IL-33	Mus musculus IL-6, (IL-33), mRNA.	−6.09
CXCL2	Mus musculus chemokine (C-X-C motif) ligand 2 (CXCL2), mRNA.	−1.24
Anti-inflammatory gene
SOCS3	Mus musculus suppressor of cytokine signalling 3 (SOCS3), mRNA.	1.26

Figure 3

IL-12p35-KO CD11b⁺ monocytes have a distinct transcriptome. (A) Volcano plot of RNA sequence data as generated by clustering based on probes that were enriched or depleted in p35 KO CD11b⁺ monocytes compared with WT CD11b⁺ monocytes at 7 days after AMI. (B) Gene Ontology (GO) analysis of differentially modulated genes classified by their biological functions and arranged according to their statistical significance (P-value indicated on the x-axis) from DAVID (database for annotation, visualization, and integrated discovery).

3.4 IL-12p35 deficiency promotes angiogenesis that is mediated by monocytes

We next confirmed the pro-angiogenic role of p35-KO CD11b⁺ monocytes. Figure 4A shows a heat map representation of selected angiogenesis-related genes that were differentially expressed in p35-KO CD11b⁺ monocytes compared with WT CD11b⁺ monocytes. Decreased Thbs1 and Arg-1 expression and increased Ang expression were found in infarcted hearts from p35-KO mice by qRT–PCR (Figure 4B). Flow cytometry revealed a higher ratio of endothelial cells (CD45⁻CD31⁺ cells) in the infarction zone of p35-KO hearts compared with that of WT hearts at 1 week after AMI (Figure 4C). Immunohistochemistry and western blotting showed that the expression of CD31 was increased in both infarction and border zones of hearts in p35-KO mice compared with WT mice (Figure 4D and E). In a mouse HLI model, we performed serial measurements by laser Doppler analysis. In p35-KO mice, the blood flow was significantly higher than that in WT mice over 3 days (Figure 4F). The number of CD31⁺ cells in the ischaemic gastrocnemius muscle was higher in p35-KO mice than in WT mice (see Supplementary material online, Figure S2A). Therefore, IL-12p35 deficiency promotes angiogenesis in response to ischaemia.

Figure 4

IL-12p35 deficiency promotes angiogenesis that is mediated by monocytes. (A) Heat map showing differential expression of the indicated genes in p35-KO CD11b⁺ monocytes compared with WT CD11b⁺ monocytes at 7 days after AMI. (B) mRNA expression levels of Thbs1, Arg-1, and Ang in p35-KO and wild-type (WT) hearts at 7 days after AMI (t-test). *P < 0.05 vs. WT; n = 6. (C) Flow cytometric analysis of CD45⁻CD31⁺ cells in p35-KO and WT hearts at 7 days after AMI (two-way ANOVA). *P < 0.05 vs. WT; n = 6. (D) Immunohistochemical staining of CD31 in remote, border, and infarct zones of p35-KO and WT hearts at 7 days after AMI (two-way ANOVA). *P < 0.05 vs. WT; n = 6. Scale bar = 50 μm. (E) Western blot analysis of CD31 protein in border and infarct zones of p35-KO and WT hearts at 7 days after AMI (two-way ANOVA). *P < 0.05 vs. WT; n = 6. (F) Laser Doppler analysis of blood flow in ischaemic and non-ischaemic limbs of p35-KO and WT mice during 3 days after HLI (repeated-measures analysis of variance). *P < 0.05 vs. WT; n = 6.

To demonstrate the anti-angiogenic effect of IL-12, we performed a series of experiments in vitro. First, qRT–PCR showed that rmIL-12 treatment inhibited the expression of Ang and promoted the expression of Thbs1 and Arg-1 in monocytes (Figure 5A). Second, we assessed the tube formation of endothelial cells stained with calcein in a three-dimensional assay analysed by fluorescence microscopy. We found that rmIL-12 had no direct effect on HUVECs (Figure 5B). WT monocytes promoted tube formation of HUVECs compared with the control group. This stimulating effect was however less prominent than that induced by p35-KO monocytes (Figure 5C). Expression of vWF was found in newly formed tubes by immunofluorescence staining and laser scanning confocal microscopy (Figure 5C). We also found that HUVECs formed tubes with lumens by laser scanning confocal microscopy (see Supplementary material online,Supplementary Data, arrow). Therefore, these tubes had lumens and expressed an endothelial marker vWF on the surface of endothelial cells (Figure 5C). In addition, we performed a Matrigel plug assay in mice. Matrigel plugs containing WT or p35-KO monocytes untreated or treated with rmIL-12 were subcutaneously injected into mice. At 7 days after the initial injection, Matrigel plugs were removed and photographed, and the newly invaded endothelial cells were assessed by CD31 staining. The number of CD31⁺ cells was statistically significantly higher in p35-KO plugs than that in WT plugs. In addition, IL-12 significantly reduced the number of CD31⁺ cells among WT and p35-KO monocytes compared with untreated monocytes (Figure 5D). Therefore, IL-12p35 deficiency promotes angiogenesis by regulating the pro-angiogenic properties of monocytes.

Figure 5

IL-12p35 deficiency promotes angiogenesis that is mediated by monocytes. (A) mRNA expression levels of Thbs1, Arg-1, and Ang in untreated monocytes and monocytes treated with rmIL-12 (t-test). *P < 0.05 vs. monocytes; n = 6. (B) Tube formation assay of HUVECs in control, bFGF, and bFGF + rmIL-12 groups (n = 6). (one-way ANOVA). Scale bar = 200 μm. (C) Tube formation assay of HUVECs and immunofluorescence of vWF in control, WT monocyte, and p35-KO monocyte groups. Expression of vWF was detected by laser scanning confocal microscopy (one-way ANOVA). *P < 0.05 vs. control; ^#P < 0.05 vs. WT monocyte; n = 6. Scale bar = 200 μm. (D)  Matrigel plug assays with WT monocyte, p35-KO monocyte, WT monocyte + rmIL-12, and p35-KO monocyte + rmIL-12 groups (two-way ANOVA). *P < 0.05; n = 6. Scale bar = 200 μm.

3.5 IL-12p35 deficiency decreases inflammatory cell infiltration and chemokine expression in hearts after AMI

To validate the inflammation-related gene expression profile of p35-KO CD11b⁺ monocytes, a panel of differentially modulated genes was chosen from Table 1 and assessed by qRT–PCR. The qRT–PCR analysis confirmed a significant decrease in the expression of pro-inflammatory chemokine genes (CCL4, CCL5, CXCL1, and CXCL2) in the infarction zone of p35-KO mice compared with WT mice (Figure 6A and B). As shown in Figure 6C, when treated with rmIL-12, the expression of these genes was elevated in monocytes compared with control monocytes. It has been reported that CCL4/5 chemokines have a chemoattractant effect on monocytes during AMI.²⁴ Our data showed that AMI markedly increased the infiltration of CD11b⁺ cells into WT hearts, but this effect was decreased in p35-KO hearts (Figure 6D). In addition, immunohistochemical analysis demonstrated that the number of Mac-2-positive cells (a marker of macrophages) was significantly lower in p35-KO hearts compared with WT hearts (Figure 6E). Similarly, in a mouse HLI model, haematoxylin–eosin (HE) and immunohistochemical staining of Mac-2 in the ischaemic gastrocnemius muscle showed that IL-12p35 deficiency decreased the inflammatory response compared with WT mice (see Supplementary material online,Supplementary Data). To provide evidence of the switch in macrophage phenotype, qRT–PCR analysis revealed a marked decrease in the mRNA level of inducible NO synthase (a marker of M1 macrophages) and an increase in that of Fizz1 (a marker of M2 macrophages) in the ischaemic gastrocnemius muscle of p35-KO mice compared with WT mice (see Supplementary material online,Supplementary Data). Therefore, IL-12 deficiency promotes macrophage differentiation into M2 macrophages and decreases the expression of chemokines and subsequent inflammatory cell infiltration into hearts after AMI.

Figure 6

IL-12p35 deficiency decreases inflammatory cell infiltration and chemokine expression in the heart after AMI. (A) Heat map showing differential expression of the indicated genes in p35-KO CD11b⁺ monocytes compared with wild-type (WT) CD11b⁺ monocytes at 7 days after AMI. (B) mRNA expression levels of chemokines (CCL4, CCL5, CXCL1, and CXCL2) in the hearts of WT and p35-KO mice subjected to either AMI or a sham operation (t-test). *P < 0.05 vs. WT; n = 6. (C) mRNA expression levels of chemokines (CCL4, CCL5, CXCL1, and CXCL2) in untreated monocytes and monocytes treated with rmIL-12 (t-test). *P < 0.05 vs. WT; n = 6. (D and E) Immunohistochemical staining of CD11b and Mac2 in p35-KO and WT hearts at 7 days after AMI (two-way ANOVA). *P < 0.05 vs. WT; n = 6. Scale bar = 50 μm.

3.6 Treatment with an IL-12p35-neutralizing antibody improves cardiac functions and repair after AMI

To determine the effect of IL-12 on AMI-induced cardiac injury, C57B/L6 mice that underwent AMI induction were administered an IL-12p35-neutralizing antibody or isotype control (100μg/mouse) via the angular vein for 0, 2, 4, and 6 days. Cardiac functions were determined at 1 week after AMI by echocardiography. IL-12p35 neutralization significantly augmented LVEF and LVFS (Figure 7A). Moreover, IL-12p35 antibody administration markedly attenuated the AMI-induced deposition of collagen and scar formation in left ventricular cardiac tissues (Figure 7B). Treatment with the anti-IL-12p35 antibody significantly increased the CD31⁺ area compared with isotype control treatment (Figure 7C). Moreover, IL-12 antibody administration markedly decreased the infiltration of CD11b⁺ and Mac-2⁺ cells compared with IgG-treated WT hearts (Figure 7D and E). Therefore, blocking IL-12p35 preserves cardiac functions after AMI.

Figure 7

Treatment with an IL-12p35-neutralizing antibody improves cardiac functions and repair after AMI. (A) The LVEF, LVFS, LVEDD, and LVESD were measured (t-test). *P < 0.05 vs. isotype; n = 4. (B) Quantification of interstitial fibrosis and the ratio of the infarct scar in wild-type (WT) and p35-KO mice at 1 week after AMI (t-test). *P < 0.05 vs. isotype; n = 4. (C–E) Immunohistochemical analysis of CD31, CD11b, and Mac2 expression in p35-KO and WT hearts at 7 days after AMI (t-test). *P < 0.05 vs. isotype; n = 4. Scale bar = 50 μm.

4. Discussion

In the present study, we found that AMI markedly up-regulated IL-12p35 expression in monocytes. Genetically or pharmacologically induced deficiency of IL-12p35 improved cardiac functions after AMI. The IL-12p35 deficiency promoted polarization of macrophages into pro-angiogenesis and anti-inflammation phenotypes during AMI. Thus, IL-12p35 is a potential therapeutic target for AMI treatment.

AMI triggers an intense inflammatory response that is mainly characterized by inflammatory cell infiltration, especially monocytes/macrophages.¹ In the current study, we demonstrated an increase in IL-12p35 expression of mainly CD11b⁺ cells in ischaemic hearts (Figure 1). AMI promoted both IL-12 mRNA and protein expression in ischaemic hearts (Figure 1). In addition, flow cytometric analysis and qRT–PCR confirmed that IL-12p35 was produced by monocytes (Figure 1), which is consistent with our previous study showing an increase in the production of IL-12 in monocytes/macrophages of hypertensive hearts. Moreover, damage-associated molecular pattern molecules released from ischaemic tissues simulate infiltrated monocytes/macrophages to secrete IL-12²⁵ and may be responsible for IL-12 production by monocytes in infarcted hearts.

Activation of the immune system occurs in the first few days after ischaemic injury, and pro-inflammatory activity dominates the injured heart tissues.^26–29 Inflammatory monocytes rapidly invade the wound to clear necrotic debris, allowing subsequent tissue regeneration.³⁰ During heart repair, monocytes perform less inflammatory functions that support tissue regeneration.^27,31 Recent studies have demonstrated that monocytes may perform different functions in various diseases. For example, it has been reported that monocyte subsets accelerate atherosclerosis progression by promoting angiogenesis in intra-plaques with a pivotal role of pro-inflammation in plaque progression, destabilization, rupture, and thrombus formation.³² Our previous study demonstrated that IL-12p35 deletion promotes differentiation of macrophages into M2 macrophages that are further categorized into M2a, M2b, and M2c cells. Among these cell types, M2c macrophages produce high levels of IL-10 and perform immunoregulatory functions.¹¹ We have previously shown that, during angiotensin II-induced hypertension, p35-KO promotes the M2 macrophage phenotype, and TGF-β secreted by M2 macrophages promotes activation of fibroblasts, leading to increases in collagen production and cardiac fibrosis. Consistent with this finding, in the present study, p35-KO promoted the M2 macrophage phenotype after ischaemia. In addition to their ability to secrete TGF-β, M2 macrophages promote angiogenesis and reduce inflammation.³³ During ischaemia, increases in angiogenic and anti-inflammatory factors can improve cardiac repair.³⁴ Activation of fibroblasts and synthesis of collagen in the early stage of AMI are important to prevent heart rupture and improve cardiac functions and repair.³⁵ Our present study demonstrated multifactorial roles of IL-12 in cardiac remodelling in response to different injuries. We analysed the transcriptome of CD11b⁺ monocytes isolated from WT and p35-KO mice in the context of AMI by RNA sequencing. The results showed that IL-12p35 deficiency induced pro-angiogenesis and anti-inflammatory monocyte characteristics compared with WT cells during AMI (Figures 3 and 4). p35-KO CD11b⁺ monocytes were pro-angiogenic macrophages with up-regulated expression of ANG and MMP8. p35-KO CD11b⁺ monocytes also exhibited anti-inflammatory properties with reduced expression of many pro-inflammatory cytokine and chemokine genes such as IL6 and CCL4. Furthermore, rmIL-12 treatment promoted the anti-angiogenesis and pro-inflammatory properties of monocytes. Thus, IL-12 regulates monocyte/macrophage functions.

Angiogenesis plays a pivotal role in preserving cardiac functions during AMI, which recovers oxygen, rescues cardiomyocytes from necrosis, and improves the prognosis of ischaemic hearts.³⁶ Flow cytometry, immunohistochemistry, and western blotting all showed enhanced angiogenesis in p35-KO mice (Figure 4). We also confirmed these results in a mouse HLI model by laser Doppler analysis to evaluate angiogenesis in vivo. As a result, p35-KO significantly promoted perfusion of the ischaemic muscle (Figure 4). The number of CD31⁺ cells was also increased in p35-KO muscle (see Supplementary material online, Figure S2). IL-12p35-neutralizing antibody treatment also increased the number of CD31⁺ cells in the ischaemic heart. Previous studies have shown that IL-12 promotes potent anti-angiogenic activity indirectly by promoting IFN-γ production in NK and T cells, which modulates induction of the angiostatic chemokine IFN-γ-induced protein 10 (IP-10).^37–39 However, we found that rmIL-12 treatment had no direct effect on tube formation by endothelial cells (Figure 5). Previous studies found that IL-12 exerts an anti-angiogenesis effect in tumours that is independent of IFN-γ production. IL12Rβ2 knockout mice have severe defects in IFN-γ production, display higher endogenous serum levels of IL-12, and show defective microvessel formation, suggesting that IL-12 inhibits tumour angiogenesis independently of the IFN-γ-IP-10 axis.⁴⁰ Our RNA sequence data revealed that p35-KO CD11b⁺ cells exhibited a distinct transcriptome profile of pro-angiogenesis, such as down-regulation of THBS1 and ARG-1 (Figure 4). THBS1 is a matricellular protein and major negative regulator of angiogenesis by mediating endothelial cell survival, migration, and responses to vascular endothelial growth factor.⁴¹ It also inhibits angiogenesis in tumours.¹⁶ Aginase I is a cytosolic enzyme expressed in blood vessels including the aorta, carotid and pulmonary artery, and coronary and gracilis muscle arterioles. It inhibits angiogenesis and endothelial functions by regulating the activity of NO synthase.¹⁷ The tube formation assay and Matrigel plug analysis revealed that the anti-angiogenic effect of IL-12 was dependent on monocytes (Figure 5). Monocytes from IL-12p35-KO mice promoted tube formation and invasion of endothelial cells compared with WT monocytes. Taken together, our data demonstrate that IL-12 suppresses angiogenesis by modulating the angiogenic functions of monocytes.

Furthermore, RNA sequencing demonstrated down-regulation of pro-inflammatory genes and up-regulation of anti-inflammatory genes in monocytes isolated from p35-KO mice after AMI (Figure 6). IL-12p35 deficiency or administration of the IL-12p35-neutralizing antibody decreased inflammatory cell infiltration and chemokine production in ischaemic hearts (Figures 6 and 7). In cardiovascular diseases, we previously found that infiltrated monocytes further stimulate cytokine production in fibroblasts,⁴² T cells,⁴³ and smooth muscle cells⁴⁴ leading to amplification of inflammation. IL-12p35 deficiency might lessen such an effect, i.e. fewer inflammatory cytokines recruit fewer monocytes, and fewer cytokines produced by fewer monocytes reduce the overall cytokine level and further inhibit recruitment of monocytes. Thus, our data demonstrated that IL-12p35 deficiency promoted a subset of monocytes characterized by pro-angiogenesis and anti-inflammatory properties.

Adverse ventricular remodelling is a major cause of contractile dysfunction and heart failure after AMI.¹² In this study, IL-12p35 deficiency improved the post-AMI survival rate, markedly preserved cardiac functions (Figure 2) and significantly reduced the occurrence of adverse ventricular remodelling including myocardial hypertrophy and myocardial infarction scar formation (Figure 2). Treatment with the IL-12p35-neutralizing antibody induced the same phenotype (Figure 7). The induction of angiogenesis after AMI increases blood supply for myocytes in danger to prevent secondary damage to the heart, and prolonged pro-inflammatory responses are associated with increases in adverse remodelling.^1,3 Much evidence from experimental models of AMI suggests that timely suppression, resolution, and containment of the post-infarction inflammatory response result in a good outcome for the infarcted heart.^45–47 We found that a deficiency in IL-12p35 promoted pro-angiogenesis and anti-inflammatory properties of monocytes in response to ischaemia. Decreases in pro-inflammatory chemokines and cytokines might reduce secondary myocardial cell damage. Additionally, timely increases in angiogenesis might provide a blood supply for myocytes under stress. Therefore, IL-12 deficiency improves cardiac functions and repair by regulating the inflammatory and angiogenetic functions of monocytes during AMI.

In conclusion, we demonstrated that IL-12p35 deficiency preserved cardiac functions by promoting pro-angiogenesis and anti-inflammatory properties of monocytes during AMI. Administration of an IL-12p35-neutralizing antibody improved cardiac function and repair, suggesting that blocking IL-12p35 might be a promising therapeutic approach to promote a beneficial microenvironment in post-infarcted hearts.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Conflict of interest: none declared.

Funding

This work was supported by the National Natural Science Foundation of China (grant numbers 81230006, 81470428); the Beijing Natural Science Foundation of China (grant number 7132043); The Key Laboratory of Remodeling-related Cardiovascular Diseases, Ministry of Education; and the Beijing Collaborative Innovative Research Centre for Cardiovascular Diseases (grant number PXM2014_014226_000002).