Genetic inhibition of IL-12β suppresses systolic overload-induced cardiac oxidative stress, inflammation, and heart failure development

Abstract

Inflammation is a key factor in the development of heart failure (HF), with interleukin-12 (IL-12) and interleukin-23 (IL-23) acting as significant pro-inflammatory cytokines, both of which are simultaneously reduced by inhibiting IL-12β. This study utilized IL-12β knockout (KO) mice to investigate whether genetically inhibiting IL-12β could lessen transverse aortic constriction (TAC)-induced cardiac inflammation, hypertrophy, and dysfunction, as well as associated lung remodeling. We found that IL-12β KO significantly improved TAC-induced cardiac dysfunction in both male and female mice, evidenced by better left ventricular (LV) ejection fraction and fractional shortening. Additionally, IL-12β KO substantially reduced the TAC-induced increases in the weight of the LV, left atrium, lung, and right ventricle (RV), and their ratios to body weight or tibial length in both sexes. Furthermore, IL-12β KO markedly attenuated TAC-induced LV leukocyte infiltration, cardiomyocyte hypertrophy, fibrosis, and subsequent lung inflammation and remodeling. Bulk LV RNA sequencing demonstrated that IL-12β KO also mitigated TAC-induced changes in LV gene profiles linked to inflammation and fibrosis. We also found that IL-12β KO significantly reduced TAC-induced LV accumulation of various immune cell subsets, activation of CD4⁺ and CD8⁺ T cells, and the percentage of central memory CD4⁺ and CD8⁺ T cells within the cardiac drainage lymph nodes. Moreover, IL-12β KO mice exhibited a significant reduction in IFNγ⁺CD8⁺ and CXCR3⁺CD8⁺ T cells in the drainage lymph nodes compared to wild-type mice after TAC. Finally, IL-12β KO and IL-12β blocking antibody significantly decreased TAC-induced LV production of reactive oxygen species (ROS) and the expression of 3-nitrotyrosine (3-NT) and 4-hydroxynonenal (4-HNE). Collectively, these findings underscore the critical role of IL-12β in systolic overload-induced LV inflammation and HF, likely through mediating cardiac immune cell accumulation, oxidative stress, and fibrosis.

Graphical abstract

Highlights

• •IL-12β knockout attenuated systolic overload-induced cardiac hypertrophy, fibrosis, and HF in mice of both sexes.
• •IL-12β knockout attenuated systolic overload-induced cardiac oxidative stress.
• •IL-12β knockout reduced systolic overload-induced cardiac accumulation of multiple immune cell subsets.
• •IL-12β knockout reduced cardiac gene profile associated with inflammation and fibrosis.
• •IL-12β KO attenuated TAC-induced T cell activation and their expression of CXCR3 in cardiac drainage lymph nodes and cardiac tissues.

1. Introduction

Chronic heart failure (HF), often resulting from left ventricular (LV) failure, is a medical condition in which the LV cannot pump sufficient blood to meet the body's needs. HF remains a major public health problem and often leads to significant cardiovascular morbidity and mortality [1]. HF can result from several different diseases or medical conditions, such as coronary artery disease, systemic hypertension, cardiac valve diseases, diabetes, congenital heart disease, or chronic inflammatory diseases. Even under the current standard clinical care, HF often progresses to WHO class-2 pulmonary hypertension, lung remodeling, and eventual right ventricular (RV) failure [[2], [3], [4], [5], [6]]. Patients with HF-induced pulmonary hypertension and RV failure generally have very poor clinical outcomes; thus, identifying novel therapeutic targets for HF treatment is needed.

While the causes or risk factors for HF development are varied, chronic mild or moderate inflammation is often a shared phenomenon among different HF etiologies [[7], [8], [9], [10], [11], [12], [13]]. For example, previous studies have demonstrated that pro-inflammatory cytokines such as interleukin-1β (IL-1β) are increased in the heart and circulation of HF patients [7,8]. Also, the Canakinumab Anti-Inflammatory Thrombosis Outcomes Study (CANTOS) trial, which investigated IL-1β blockade with canakinumab (an anti-IL-1β monoclonal antibody) in patients with prior myocardial infarction and evidence of active inflammation, showed a reduction in the risk of major cardiovascular events (non-fatal myocardial infarction, non-fatal stroke, or death) as compared to standard treatment [9]. A subsequent exploratory analysis of the CANTOS data showed that IL-1β inhibition reduced the rate of HF hospitalizations or HF–related death in a dose-dependent manner [10,11]. Moreover, experimental studies from us and others also showed that inhibition of IL-1β signaling significantly attenuated myocardial infarction-induced HF development in rats [12] and transverse aortic constriction (TAC)-induced HF development and/or progression in mice [13].

We and others have also demonstrated that cardiac inflammation and HF development are regulated by immune cell subsets such as CD4⁺ T cells [14], CD8⁺ T cells [15], CD11c⁺ dendritic cells (DCs) [16], NK1.1⁺ cells [17], and macrophages [18] in experimental animals. Macrophages are often the most abundant cardiac immune cells in TAC or myocardial infarction-induced HF animals [18,19] and a major source of proinflammatory cytokines such as IL-1β, IL-18, IL-12, and IL-23 [20]. These proinflammatory cytokines could enhance the effective function of other immune cells, such as T cells and NK cells.

IL-12β is an essential subunit for IL-12 and IL-23, two important proinflammatory cytokines mainly produced by activated macrophages and DCs [[21], [22], [23], [24], [25]]. Thus, inhibition of IL-12β can simultaneously suppress the proinflammatory effects of IL-12 and IL-23. Since IL-12 promotes IFN-γ production and IL-23 promotes IL-17 production, inhibition of IL-12β simultaneously attenuates the proinflammatory IL-12/IFN-γ and IL-23/IL-17 signaling pathways. Specifically, IL-12 plays an important role in promoting proliferation, activation, and mobilization of CD8⁺ T cells, CD4⁺Th1 cells, and NK cells [21,[26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36]], immune cell types that could promote cardiac inflammation and HF secondary to systolic pressure overload produced by TAC, at least in mice [14,15,17]. Meanwhile, IL-23 plays an important role in stimulating IL-17 production by Th17 cells or γδT17 cells through activating IL-23 receptors expressed on these immune cell subsets through retinoic acid orphan receptor gamma (RORγ)-dependent pathway [23,24,[37], [38], [39], [40], [41]]. Studies have shown that the IL-23/IL-17 signaling pathway also promotes cardiac inflammation and HF [42]. Accordingly, inhibition of IL-12β may be an effective approach to attenuate the IL-12/IFN-γ axis and the IL-23/IL-17 axis, two important pathways that promote HF development.

To test the central hypothesis that IL-12β signaling exerts a critical role in promoting systolic overload-induced cardiac inflammation and HF development, we investigated the role and the underlying mechanisms of IL-12β by using IL-12β knockout (IL-12β KO) mice in TAC-induced LV inflammation and HF.

2. Materials and methods

Mice and Experimental Protocols: Male and female IL-12β knockout (Jackson Laboratory, Strain #002693) and wild-type (WT) C57BL/6J (Jackson Laboratory, Strain #000664) mice were subjected to sham or transverse aortic constriction (TAC) surgery, a procedure used to replicate clinical conditions such as hypertension and aortic stenosis. To minimize the potential day-to-day variation of the TAC procedure, the TAC procedure was performed in female IL-12β KO and female body weight-matched control WT mice on the same day by the same surgeon using a 27-gauge needle (relatively more aortic constriction), while the TAC procedure was performed in male mice by the same surgeon on the same day using a 26-gauge needle (relatively less aortic constriction). Body weight was monitored weekly, and left ventricular (LV) function was assessed prior to and every two weeks post-TAC. Final studies were performed when LV ejection fraction of one group was reduced below 45%, a timepoint when HF-induced lung remodeling generally occurs in mice after TAC [2,4]. Samples were collected at 6 weeks after TAC in male mice, while samples were collected 4 weeks after TAC in female mice. Cardiac and pulmonary tissues were subsequently utilized for histological, immuno-histological, and biochemical analyses. All mice were housed in a temperature-controlled environment with a 12-h light/dark cycle. This study received approval from the Animal Care and Use Committee at the University of Mississippi Medical Center.

TAC Procedure: TAC surgery was performed using a 26-gauge blunt needle in male mice and 27-gauge blunt needle in female mice after anesthetizing the mice with an intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg) as previously described [43,44]. The reason for using more severe aortic restriction in female mice is that female mice generally “resistant” to TAC-induced LV hypertrophy and HF as compared with male mice [45,46].In addition, because female mice generally have smaller body weight as compared with age-matched male mice, we reasoned that using the same size of needle would cause relative less aortic restriction or LV afterload in female mice. Thus, a relatively smaller size needle (27-gauge needle) was used in female mice of C57BL/6J background, and a relatively larger size needle (26-gauge needle) was used to create aortic constriction in male mice of C57BL/6J background.

Anti-IL-12β antibody treatment: Pharmacological inhibition of IL-12β was achieved by anti-IL-12β antibody treatment (BioXCell, BE0051) in WT female mice. Briefly, after the TAC procedure, mice were randomly divided into two groups and intraperitoneally injected with either anti-IL-12β antibody or control IgG (250 μg/mouse) every 3 days for 4 weeks. The control mice used for this study were sham-operated mice without anti-IL-12β antibody treatment. Tissue samples were collected at 4 weeks after TAC and utilized for subsequent histological analysis.

Echocardiography: Echocardiography was performed using a VisualSonics Vevo 3100 imaging system (FUJIFILM VisualSonics Inc., Canada) as previously described [44]. Briefly, the mice were anesthetized by inhalation of 1-2% isoflurane mixed with 100% oxygen. M-mode echocardiographic images were taken and analyzed using Vevo LAB software (FUJIFILM VisualSonics Inc., Canada) to measure heart rate, LV ejection fraction, LV fractional shortening, LV end-systolic diameter, LV end-diastolic diameter, LV end-systolic volume, LV end-diastolic volume, LV anterior and posterior wall thickness at end-systole or end-diastole, stroke volume, and cardiac output. Doppler flow velocity in the right innominate artery and the left common carotid artery was measured by using pulsed-wave Doppler method.

Histological Staining: The histological and immunological staining were performed according to previous studies [46]. The detailed method for histological and immune staining is also provided in the supplemental data.

Bulk RNA Sequencing: LV tissue samples were collected from WT and IL-12β KO female mice 4 weeks after sham surgery or TAC and stored at −80 °C until RNA isolation. LV tissue samples were used for RNA extraction and sequencing. The HiSeq4000 platform was used for RNA sequencing and 150 bp paired-end reads were generated. Raw reads were quality-checked and filtered, and clean reads were aligned to the mouse reference genome GCF_000001635.27_GRCm39 using HISAT2. Differential expression analysis was performed using DESeq2 platform. Genes with an adjusted false discovery rate (FDR) < 0.05 were considered statistically differentially expressed genes.

Flow Cytometry Analyses: LV tissues were harvested, minced into small pieces, and digested in HBSS supplied with Deoxyribonuclease I (66.7 μg/mL, Sigma-Aldrich) and Liberase^TM (125 μg/mL, Roche Diagnostics, Germany) at 37 °C for 30 min using a tissue dissociator (Miltenyi Biotec), followed with cell purification and immune staining as we previously described [17,47]. Data were acquired on a BD FACSymphony™ A3 Cell Analyzer (BD Biosciences) and analyzed by using FlowJo-v10 (FlowJo, OR) software. The gating strategies used for LV flow cytometry analysis are presented in Supplementary Fig. 1.

For the flow cytometry from lymph nodes, the lymph nodes were mechanically dissociated by gently pressing them with the plunger of a syringe and filtered through a 100 μm cell strainer. The staining procedure is the same as described above. The gating strategies used for lymph node flow cytometry analysis are presented in Supplementary Fig. 2. For cytokine production assay, the isolated immune cells were stimulated with 1X Cell Stimulation Cocktail (Invitrogen, 00-4970-93) and 1X Protein Transport Inhibitor Cocktail (00-4980-93) in RPMI 1640 with 10% FBS at 37 °C for 2 h as we previously described [17,47]. The cells were then stained with antibodies for cell surface markers, permeabilized, and stained with antibodies against different intracellular cytokines (Supplementary Table 1).

Western Blot Analyses: Western blot analysis was performed as previously described [46], and the detailed method is provided in the supplemental data.

Statistical Analysis: Data are presented as Mean ± SEM. A two-way ANOVA followed by a Bonferroni post hoc test was used to determine statistical differences between WT and IL-12β KO mice under sham or TAC conditions. A one-way ANOVA followed by a Bonferroni post hoc test was used to determine the statistical differences among the three groups. An unpaired t-test was used to test statistical differences between two groups. The log-rank test was used to analyze survival curves. All the statistical tests were performed using GraphPad Prism 10 software. p < 0.05 was considered statistically significant.

3. Results

3.1. LV IL-12β protein expression was significantly increased in mice after TAC

We first examined whether cardiac IL-12β expression is altered in response to pressure overload produced by TAC. Western blot analysis of LV tissues showed that LV IL-12β protein expression was significantly increased in WT mice 1 week after TAC as compared with WT sham controls (Fig. 1A). Immunohistological analysis further demonstrated that LV IL-12β⁺ area was significantly increased in WT mice 1 week and 6 weeks after TAC, while IL-12β⁺ expression was higher in mice 1 week after TAC as compared with mice 6 weeks after TAC (Fig. 1B). Immunostaining further showed that all IL-12β⁺ cells are CD45⁺ leukocytes (Fig. 1C), while majority of CD68⁺ macrophages are also IL-12β⁺ cells (Supplementary Fig. 3).

Fig. 1.

Cardiac IL-12β expression was significantly increased after TAC. (A) Western blots and quantified data of LV IL-12β expression in WT sham and WT TAC mice. (B) Representative LV IL-12β staining images and quantified data of IL-12β⁺ area. (C) Representative images of IL-12β and CD45 co-immunostaining of WT mice LV tissue under sham and TAC conditions. The green stain in panel B is due to cardiac autofluorescence recorded at FITC channel. ∗p < 0.05; n = 5 per group.

3.2. IL-12β KO significantly attenuated TAC-induced LV dysfunction in both male and female mice

To elucidate the role of IL-12β in cardiac inflammation and HF induced by systolic overload, both male and female IL-12β KO and WT mice underwent sham or TAC surgery. IL-12β KO significantly reduced mortality associated with TAC in female mice (1/23) compared to WT female controls (15/31). Conversely, no statistically significant difference in TAC-induced mortality was observed between male IL-12β KO mice (5/19) and male WT controls (7/17) (Fig. 2B). Meanwhile, blood flow velocity differences across the TAC site were comparable between WT and KO mice, as depicted in Supplementary Fig. 4, confirming equivalent TAC induction. Under sham conditions, echocardiographic analysis revealed no significant differences in LV function or dimensions between WT and KO mice (Fig. 2A and Supplementary Fig. 5). Critically, IL-12β KO significantly ameliorated TAC-induced LV dysfunction and dilatation in both male and female mice, a finding detailed in Fig. 2A and Supplementary Fig. 5.

Fig. 2.

IL-12β KO significantly attenuated TAC-induced cardiac dysfunction, an increase in LV weight, LA weight, lung weight, and RV weight in male and female mice. (A) Representative M-mode echocardiographic images of WT and IL-12β KO male mice: pre-TAC, 2 weeks, 4 weeks, and 6 weeks after TAC, and Quantified data of echocardiographic measurements of LVEF, LVFS, LVESD, and LVEDD of both sexes. (B) Survival curves of WT TAC and IL-12β KO TAC mice of both sexes (log-rank test). (C–F) The ratio of LV weight, left atrial (LA) weight, lung weight, RV weight to tibial length (TL) of the indicated groups. (G) Representative LV WGA staining images and quantified data of cardiomyocyte cross-sectional areas. ∗p < 0.05; ^#p < 0.05 compared to WT sham; ^†p < 0.05 compared to WT TAC; ^$p < 0.05 compared to IL-12β KO sham; n = 7 to 22 per group for panels A-F and n = 4-5 per group for panel G; LVEF, LV ejection fraction; LVFS, LV fractional shortening; LVESD, LV end-systolic diameter; LVEDD, LV end-diastolic diameter.

3.3. IL-12β KO significantly attenuated TAC-induced LV hypertrophy, increases in lung weight, and LV myocyte hypertrophy in both male and female mice

LV weight, left atrial (LA) weight, lung weight, RV weight, and their ratios to tibial length or body weight were comparable between WT and IL-12β KO mice under sham conditions (Fig. 2C–F, Supplementary Tables 2 and 3, and Supplementary Fig. 5). TAC significantly increased LV weight, LA weight, lung weight, RV weight, and their ratios to tibial length or body weight in male and female WT mice, but these changes were significantly attenuated in IL-12β KO mice as compared with corresponding WT mice (Fig. 2C–F, Supplementary Tables 2 and 3, and Supplementary Fig. 5). Moreover, wheat germ agglutinin (WGA) staining was performed to measure LV cardiomyocyte cross-sectional area. The data showed that the TAC-induced increase in LV cardiomyocyte cross-sectional area was significantly reduced in IL-12β KO mice (Fig. 2G).

3.4. IL-12β KO protected hearts against TAC-induced changes in the LV gene profile

To determine the underlying mechanisms of IL-12β KO in protecting mice from TAC-induced LV hypertrophy and dysfunction, bulk RNA sequencing was performed in LV tissues from female WT and IL-12β KO mice (Fig. 3). A principal component analysis (PCA) of LV RNA-seq data revealed distinct clusters of LV gene profiles among the WT TAC group, IL-12β KO TAC group, and the sham groups (both WT and IL-12β KO mice). PCA also showed that sham WT and IL-12β KO mice are clustered together, while IL-12β KO TAC mice were clustered near the sham mice (Fig. 3A).

Fig. 3.

IL-12β KO attenuated TAC-induced alterations in the gene expression related to fibrosis and inflammation (A) Principal component analysis of WT and IL-12β KO mice, under sham or TAC conditions. (B) Venn diagram showing differentially expressed genes (DEGs) of WT and IL-12β KO mice under sham or TAC conditions, as well as the shared and uniquely changed genes. (C) DEGs cluster heatmap in WT mice compared to IL-12β KO mice, under sham or TAC conditions. (D) Volcano plots showing upregulated and downregulated genes among the groups. (E) Gene ontology (GO) biological processes enrichment bubble chart in WT TAC mice compared to IL-12β KO TAC. (F) GO cellular components enrichment histogram shows the enriched cellular components between KO and WT mice after TAC. n = 2 per group for sham and n = 3 per group for TAC.

We further determined the differentially expressed genes (DEGs) among the experimental groups. Under basal/sham conditions, there were only 60 DEGs (39 upregulated and 21 downregulated) in LV tissues between IL-12β KO mice and WT mice (Supplementary Fig. 6). As compared with corresponding control mice, TAC caused 1336 DEGs in WT mice (728 upregulated vs 608 downregulated), and 859 DEGs in IL-12β KO mice (395 upregulated vs 464 downregulated) (Supplementary Fig. 6). Comparison of LV gene profiles between IL-12β KO TAC and WT TAC mice revealed 658 DEGs (232 upregulated vs 426 downregulated) (Supplementary Fig. 6). Venn diagrams were generated showing the DEGs of WT and IL-12β KO mice after TAC as compared with their corresponding sham groups, as well as the uniquely expressed and shared DEGs among these four experimental groups (Fig. 3B).

A heatmap was also generated for the four experimental groups using the DEGs between WT and IL-12β KO mice after TAC (Fig. 3C). The results clearly showed that IL-12β KO drastically attenuated TAC-induced alterations of LV gene profiles as LV gene profiles of IL-12β KO mice are like the sham mice (Fig. 3C). Moreover, the DEGs that changed over 2-fold among these groups are also presented with volcano plots (Fig. 3D). Together, these findings clearly demonstrate that IL-12β KO mice are protected from TAC-induced LV remodeling.

3.5. IL-12β KO suppressed LV genetic pathways associated with extracellular matrix remodeling, inflammation, and LV fibrosis in mice after TAC

To reveal the major molecular signaling pathways changed between WT and KO mice after TAC, pathway analyses were further performed for LV DEGs of WT and IL-12β KO mice after TAC. The gene ontology (GO) enrichment analysis was used to categorize the major affected biological processes, molecular functions, and cellular components. Go biological processes enrichment analysis showed that the top enriched pathways were collagen fibril organization, extracellular matrix organization, cell adhesion, ossification, angiogenesis, and TGFβ signaling for DEGs between WT and KO mice after TAC (Fig. 3E). The GO cellular components enrichment analysis showed that extensively enriched cellular components were collagen-containing extracellular matrix, extracellular region, extracellular matrix, and other extracellular matrix-related cellular components (Fig. 3F). The top enriched GO molecular pathways were extracellular matrix structural constituent, heparin-binding, integrin binding, and protein binding, suggesting that IL-12β deficiency is associated with major impact on extracellular matrix organization and fibrosis, as well as inflammation under pressure overload conditions (Supplementary Fig. 6). In addition, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of LV DEGs was used to identify the major changed pathways such as metabolism, cellular processes, and human diseases of the different groups. KEGG enrichment analysis of LV DEGs between WT and KO mice after TAC shows the top enriched pathways are protein digestion and absorption, diabetic cardiomyopathy, AGE-RAGE signaling in diabetic complications, and ECM-receptor interaction pathways (Supplementary Fig. 6).

To further determine statistically upregulated and downregulated functions/pathways between the WT TAC group and the IL-12β KO TAC group, Gene Set Enrichment Analysis (GSEA) with ReactomeGSA was performed for DEGs of the WT TAC group and the IL-12β KO TAC group. ReactomeGSA showed that the top upregulated functions/pathways include mitochondrial complex I biogenesis, mitochondrial translation, mitochondrial translation elongation, mitochondrial fatty acid beta oxidation, and branched-chain amino acid catabolism (Fig. 4A), indicating that IL-12β KO mice were protected from TAC-induced LV mitochondrial and metabolic dysfunction in mice. The findings showed that the top downregulated functions/pathways of IL-12β KO and WT mice after TAC are collagen formation, extracellular matrix organization, collagen biosynthesis and modifying enzymes, assembly of collagen fibrins and other multimeric structure, collagen chain trimerization etc., as well as pathways related to immune response such as integrin cell surface interactions, immunomodulatory interactions between lymphatic and non-lymphatic cells, and DAP12 interactions [DAP12 is an immunoreceptor tyrosine-based activation motif (ITAM)-bearing adapter molecule that transduces activating signals in NK and myeloid cells]. (Fig. 4A and B). Consistent with the findings that IL-12β KO suppressed LV genetic pathways associated with extracellular matrix remodeling, histological staining showed that IL-12β KO significantly attenuated TAC-induced increase in LV interstitial and perivascular fibrosis (Fig. 4C and D).

Fig. 4.

IL-12β KO suppressed LV genetic pathways associated with extracellular matrix remodeling, inflammation, and LV fibrosis in mice after TAC. (A) The top 20 upregulated and downregulated pathways between KO and WT mice after TAC, identified by GSEA with ReactomeGSA. (B) Representative GSEA plots of collagen formation and extracellular matrix organization. (C, D) Representative images and quantified data of LV interstitial and perivascular fibrosis performed by Sirius Red/Fast Green staining in male and female mice. ^#p < 0.05 compared to WT sham; ^†p < 0.05 compared to WT TAC; ^$p < 0.05 compared to IL-12β KO sham; n = 3 per group for panels A-B and n = 4-5 per group for panel D.

Overall, GSEA by ReactomeGSA and KEGG pathway showed that IL-12β KO significantly downregulated LV pathways associated with extracellular matrix processing, immune and inflammatory responses, and the interactions between immune cells with nonimmune cells to reduce inflammation. IL-12β KO also upregulated LV pathways, enhancing mitochondrial biogenesis and metabolism in mice after TAC.

3.6. IL-12β KO suppressed TAC-induced LV genetic pathways associated with inflammation and LV leukocyte infiltration in mice

Since GSEA shows that IL-12β KO significantly suppressed TAC-induced LV genetic pathways associated with inflammatory responses and the interactions between immune cells and nonimmune cells (Fig. 4A), GSEA by KEGG pathway was also performed. The findings showed that downregulated pathways are associated with various pathogen infection (such as pathways Staphylococcus aureus infection, Malaria infection, Human papillomavirus infection, EB virus infection etc.), immune or inflammatory responses (such as ECM-receptor interaction, Phagosome, Complement and coagulation cascades, Lysosome, Focal adhesion, Chronic myeloid leukemia, Rheumatoid arthritis, and Antigen processing and presentation etc.), and metabolic pathways associated cell recruiting and injury repair (such as Protein digestion and absorption, Focal adhesion, Glycosaminoglycan biosynthesis, TGF-beta signaling pathway etc.) (Fig. 5A).

Fig. 5.

IL-12β KO significantly attenuated TAC-induced LV inflammation in mice. (A) The top 20 upregulated and downregulated pathways between KO and WT mice after TAC were identified by GSEA with KEGG pathway analysis. (B) Heatmap shows the top LV immune and/or inflammation-related genes in KO and WT mice after TAC. (C, D) Representative LV CD45⁺ immuno-staining images and quantified data of LV CD45⁺ of the indicated groups. (E) Percentage distribution of major immune cell subsets within LV CD45⁺ cells of the indicated groups determined by flow cytometry. The green stain in panel C is due to cardiac autofluorescence recorded at FITC channel. ^#p < 0.05 compared to WT sham; ^†p < 0.05 compared to WT TAC; ^$p < 0.05 compared to IL-12β KO sham; n = 2-3 per group for panels A-B and n = 4-5 per group for panels D-E.

We further analyzed DEGs associated with infection, immune diseases, and the immune system of WT and IL-12β KO mice after TAC based on the KEGG Pathway Classification analysis (Supplementary Fig. 7). We found a total of 105 DEGs, and the top upregulated genes include Gnb3, Camk2b, Gadd45a, IL15, Rpl3l, Camk2a, Bcl2L11 etc., and the top downregulated DEGs include Cyfip2, Col1a2, Vim, Mmp2, Tgfb2, Fn1, C4b, Ace, Thbs4, and Spp1. To reveal the relative expressions of these immune-related DEGs among the four experimental groups, a heatmap was generated for the top 56 DEGs (p < 0.01, and net Log2 fold change is greater than 0.5) (Fig. 5B). The top downregulated immune related DEGs include adhesion molecules recruiting immune cells (such as Cxadr, CD44, Icam1, Itgb1 or CD29), complements, a group of genes modulating the interactions between CD8 and other cells (B2m, and MHCI molecules such as H2-T23, H2–K1, H2-Q7), and a group of collagens (such as Col1a2, Col1a1, and Col3a1) that are implicated in the development of inflammation (Fig. 5B). These findings suggest that cardiac immune signaling pathways related to cardiac immune cell infiltration, particularly the signaling related to CD8 T cell activity, are reduced in IL-12β KO mice after TAC.

We further determined the effect of IL-12β KO on TAC-induced LV immune cell infiltration by using immuno-histology and flow cytometry. IL-12β KO had no detectable effects on LV CD45⁺ leukocyte infiltration under sham conditions, but IL-12β KO significantly attenuated TAC-induced LV leukocyte infiltration in male and female mice (Fig. 5C and D). However, to our surprise, the percentages of the major immune cell subsets within LV CD45⁺ cells (such as macrophages, B cells, T cells, and neutrophils) were largely unaffected in IL-12β KO mice after sham or TAC (Fig. 5E).

3.7. IL-12β KO significantly attenuated TAC-induced pulmonary immune cell infiltration, fibrosis, and vessel remodeling in male and female mice

Our previous studies demonstrated that LV failure causes pulmonary remodeling and the consequent RV hypertrophy [2,13,48]. Thus, we further determined the effect of IL-12β KO on TAC-induced pulmonary remodeling. Although IL-12β KO had no detectable effects on pulmonary CD45⁺ leukocyte infiltration, fibrosis, and arteriole muscularization as compared to WT mice under sham conditions (Fig. 6A–F), IL-12β KO significantly attenuated TAC-induced pulmonary CD45⁺ leukocyte infiltration, fibrosis, and vessel muscularization in both male and female mice (Fig. 6A–F). These findings indicate that IL-12β KO also significantly attenuated TAC-induced HF progression in mice of both sexes.

Fig. 6.

IL-12β KO attenuated TAC-induced pulmonary inflammation, fibrosis, and vessel remodeling in male and female mice. (A, B) Representative images and quantified data of infiltrated CD45⁺ leukocytes in the lung performed by immuno-histological staining. (C, D) Representative images and quantified data of lung fibrosis performed by Sirius Red/Fast Green staining. (E, F) Representative images and quantified data of lung vessel remodeling performed by immuno-histological staining. ^#p < 0.05 compared to WT sham; ^†p < 0.05 compared to WT TAC; ^$p < 0.05 compared to IL-12β KO sham; n = 4-5 per group.

3.8. IL-12β KO significantly attenuated TAC-induced early-phase LV dysfunction, cardiomyocyte hypertrophy, inflammation, and fibrosis

TAC-induced LV immune cell infiltration often peaks ∼7 days after TAC, a time point at which pulmonary remodeling is not observed. To understand the impact of IL-12β on early-stage LV inflammation and remodeling, we further determined the cardiac immune cell infiltration and the immune cell subsets in WT and IL-12β KO mice 7 days after TAC. Interestingly, IL-12β KO already significantly attenuated TAC-induced LV dysfunction 7 days after TAC (Fig. 7A). The TAC-induced increases in LV end-systolic diameter (LVESD) and end-systolic volume (LVESV) were significantly reduced in IL-12β KO mice as compared to WT mice, while LV end-diastolic diameter (LVEDD) and LV end-diastolic volume (LVEDV) were similar between WT and IL-12β KO mice (Fig. 7A and Supplementary Fig. 8). In addition, TAC-induced LV hypertrophy, LV cardiomyocyte hypertrophy, LV CD45⁺ infiltration, and LV fibrosis were significantly decreased in IL-12β KO mice as compared to WT mice (Fig. 7B–D and Supplementary Fig. 8).

Fig. 7.

IL-12β KO significantly attenuated TAC-induced early-phase LV dysfunction, cardiomyocyte hypertrophy, and inflammation. (A) Representative M-mode echocardiographic images and quantified data of echocardiographic measurements of LVEF, LVFS, LVESD, and LVEDD of the indicated groups. (B) The ratios of cardiac and pulmonary tissues to tibial length (TL) of the indicated groups. (C, D) Representative images and quantified data of LV cardiomyocyte size and LV CD45⁺ cells of the indicated groups. (E) Percentage of major immune cell subsets within LV CD45⁺ cells determined by flow cytometry. (F) Representative flow cytometry plots and quantified data of CD44⁺CD3⁺, CD44⁺CD4⁺, and CD44⁺CD8⁺ T cells within the corresponding T cells. The green stain in panel D is due to cardiac autofluorescence recorded at FITC channel. ∗p < 0.05; n = 5-10 per group.

We further performed flow cytometry to detect the major LV immune cell subsets. Again, the percentages of LV macrophages, B cells, T cells, and neutrophils were similar between WT and IL-12β KO mice 7 days after TAC (Fig. 7E). Given the marked increase in LV CD45⁺ cell infiltration in WT mice compared to IL-12β KO mice (Fig. 7C), it can be inferred that the major immune cell subsets increased proportionally following TAC, and this response was attenuated in IL-12β KO mice. CD44 plays an important role in tissue recruitment of immune cells and stem cells, and CD44 expression is often used as a marker of T cell activation. Interestingly, the percentages of LV CD44⁺CD3⁺ T cells, CD44⁺CD4⁺ T cells, and CD44⁺CD8⁺ T cells were significantly attenuated in IL-12β KO mice as compared to WT mice after TAC (Fig. 7F). These findings indicate that IL-12β KO significantly attenuated cardiac dysfunction and inflammation as early as 7 days after TAC.

3.9. IL-12β attenuated "stem cell-like" central memory T cell activation and IFNγ⁺CD8⁺ T cells in cardiac drainage lymph nodes

Up to now, we found that IL-12β KO significantly suppressed TAC-induced cardiac immune cell infiltration and T cell activation, while the LV gene profile also showed that IL-12β KO significantly suppressed cardiac immunomodulatory interactions between lymphatic and non-lymphatic cells. Since the drainage lymph nodes play a crucial role in immune surveillance by promoting T cell proliferation, activation, and enhancing effective function by increasing their tissue homing capacity and production of cytotoxic molecules such as IFNγ. Since TAC-induced cardiac inflammation generally peaks ∼7 days after TAC, a time point at which there is no detectable effect on pulmonary immune cell infiltration and dysfunction, we investigated the immune cell composition and their activation in cardiac drainage lymph nodes. Surprisingly, we found that the percentage of major immune cell subsets within CD45⁺ cells was similar between WT and IL-12β KO mice under control conditions and after TAC (Fig. 8A and B and Supplementary Table 5). However, we found that IL-12β KO abolished TAC-induced increase of IFNγ⁺CD8⁺ cells without affecting the percentage of TNFα⁺CD8⁺ T cells (Fig. 8C), while the percentages of IFNγ⁺CD4⁺ T cells and TNFα⁺CD4⁺ T cells were similar between KO and WT mice under control conditions or after TAC (Supplementary Fig. 9). Since CD8⁺ T cells generally have higher expression of chemokine receptor CXCR3, CXCR3 plays an important role in promoting the CD8⁺ T cells’ infiltration into injured tissues. Interestingly, the percentages of CXCR3⁺CD8⁺ cells and CXCR3⁺CD4⁺ T cells in the drainage lymph nodes were significantly reduced in IL-12β KO after TAC as compared with corresponding WT mice (Fig. 8D and E).

Fig. 8.

IL-12β KO significantly attenuated "stem cell-like" central memory T cell activation and their IFNγ production in the drainage lymph nodes. (A, B) Percentages of major immune cell subsets within CD45⁺ cells in the drainage lymph nodes determined by flow cytometry. (C) Percentages of IFNγ⁺CD8⁺ and TNFα⁺CD8⁺ within CD8⁺ T cells. (D, E) Percentages of CXCR3⁺CD4⁺ and CXCR3⁺CD8⁺ T cells within CD4⁺ and CD8⁺ T cells, respectively. (F) Percentage of effective memory (CD44⁺CD62L⁻), naïve (CD44⁻CD62L⁺), and central memory (CD44⁺CD62L⁺) T cells of CD8⁺ T cells. ∗p < 0.05; T_EM, Effective Memory T Cells; T_naïve, Naïve T Cells; T_CM, Central Memory T Cells; n = 4-5 per group.

"Stem cell-like" central Memory T cells (CMTs, generally characterized as CD44⁺CD62⁺ T cells) primarily reside in lymphatic tissues such as drainage lymph nodes. CMTs are characterized by their ability to rapidly proliferate and differentiate into effector T cells upon re-exposure to APC-presented antigens. Interestingly, TAC caused significant increases in the percentages of CD4⁺CD44⁺CD62⁺ and CD8⁺CD44⁺CD62⁺ CMTs within their corresponding T cell subsets in WT mice, while IL-12β KO abolished TAC-induced increases of these central memory T cells (Fig. 8F and Supplementary Fig. 9).

3.10. Inhibition of IL-12β by IL-12β gene KO and pharmacological anti-IL-12β antibody significantly attenuated TAC-induced LV oxidative stress

We found that IL-12β KO significantly attenuated cardiac and pulmonary immune cell accumulation in mice after TAC. Since oxidative stress, and the interaction between oxidative stress and immune cells contributes to HF development and progression [[49], [50], [51]], we further determined the effect of IL-12β deficiency on TAC-induced cardiac oxidative stress (Fig. 9). Dihydroethidium (DHE) staining showed that the TAC-induced increase in LV ROS production was significantly attenuated in IL-12β KO mice (Fig. 9A). In addition, IL-12β KO also abolished TAC-induced increases in expression of markers of oxidative stress, such as 3-nitrotyrosine (3-NT) and 4-hydroxynonenal (4-HNE) in the LV tissues (Fig. 9B).

Fig. 9.

IL-12β KO and anti-IL-12β antibody treatment significantly attenuated TAC-induced LV oxidative stress. (A) Representative dihydroethidium (DHE) staining images and quantified data of DHE intensity of the indicated groups. (B) Representative western blots and quantification of LV 3-nitrotyrosine (3-NT) and 4-hydroxynonenal (4-HNE) of the indicated groups. (C-E) Representative DHE, 3-NT, and 4-HNE staining images, and quantified data of relative DHE, 3-NT, and 4-HNE intensity of the indicated groups. ∗p < 0.05; n = 5 to 6 per group.

Moreover, the TAC-induced increases in cardiac ROS production and the markers of oxidative stress, such as 3-NT and 4-HNE were also significantly attenuated with anti-IL-12β antibody treatment (Fig. 9C–E). These findings demonstrated that both genetic and pharmacological inhibition of IL-12β can attenuate TAC-induced LV oxidative stress.

4. Discussion

The provided study reveals several significant findings regarding the role of IL-12β in TAC-induced cardiac and pulmonary pathology. Key observations include: (i) LV IL-12β protein expression was significantly elevated in WT TAC mice compared to sham controls. (ii) CD45⁺ Leukocytes are the major IL-12β producers because all IL-12β⁺ cells are also CD45⁺, indicating that cardiac IL-12β is produced by infiltrated leukocytes. (iii) IL-12β KO markedly reduced TAC-induced mortality in female mice and mitigated LV inflammation, hypertrophy, fibrosis, and dysfunction in both sexes. (iv) IL-12β KO and antibody treatment effectively abolished TAC-induced LV oxidative stress. (v) Bulk RNA sequencing of LV tissues demonstrated that IL-12β KO protected hearts from TAC-induced gene expression alterations and pathways associated with extracellular collagen formation, organization, and remodeling. (vi) IL-12β KO significantly attenuated TAC-induced upregulation of genes and signaling pathways promoting inflammatory responses, immune cell recruitment, and lymphatic-non-lymphatic cell interactions in the LV. (vii) IL-12β KO attenuated TAC-induced LV inflammation and dysfunction as early as 7 days post-TAC, correlating with reduced LV T cell activation, central memory T cell activation, and IFNγ⁺CD8⁺ cells in cardiac drainage lymph nodes.

One of the major findings of this study is that IL-12β KO significantly attenuated TAC-induced cardiac dysfunction in male and female mice, as evidenced by the following changes. First, IL-12β KO significantly attenuated the TAC-induced decrease of LV ejection fraction, LV fractional shortening, and increase of LV end-systolic and end-diastolic diameter and volume. In addition, IL-12β KO significantly ameliorated TAC-induced LV inflammation, hypertrophy, and fibrosis. Moreover, the LV gene profile also demonstrated that IL-12β KO effectively protected the heart against TAC-induced LV remodeling. These findings are conceptually consistent with our recent report that inhibition of IL-12β by pharmacological blocking antibody effectively suppressed TAC-induced cardiac inflammation, fibrosis and HF [52].

The findings of significant cardiac dysfunction, LV inflammation, hypertrophy, and fibrosis in WT mice after TAC are consistent with the notion that inflammation plays an important role in pressure overload-induced HF development. Previous studies from us and others have demonstrated the important role of immune cells such as CD4⁺ T cells [14], CD8⁺ T cells [15], CD11c⁺ dendritic cells [16], and NK1.1⁺ cells [17] in pressure overload-induced cardiac inflammation and dysfunction. The finding that IL-12β KO significantly attenuates TAC-induced HF not only reinforces the critical role of inflammation in HF development [2,53] but also highlights the specific contribution of IL-12β to cardiac inflammation and HF development. However, all the above parameters were comparable between WT and IL-12β KO mice under sham conditions, suggesting that IL-12β KO does not significantly impact cardiac structure and function under unstressed conditions.

Previous studies reported that heart failure is associated with an increase in oxidative stress, and targeting oxidative stress attenuated TAC-induced HF development and progression [[49], [50], [51],54,55]. In line with these studies, the TAC-induced increases in relative LV ROS production and expression of oxidative stress markers such as 3-NT and 4-HNE were significantly attenuated in IL-12β KO mice and/or in mice treated with anti-IL-12β antibody. These findings indicate that reduced LV oxidative stress may have contributed to the cardioprotective effect against pressure overload seen in IL-12β KO mice. A previous study showed that exogenous injection of anti-inflammatory cytokine IL-35 delayed hindlimb ischemia-induced angiogenesis by reducing ROS production [56]. Similarly, we found that genetic and pharmacological inhibition of IL-12β, a shared subunit of pro-inflammatory cytokines IL-12 and IL-23, attenuated pressure overload-induced cardiac oxidative stress. Taken together, these findings suggest that ROS may be involved in IL-12 cytokine family-related tissue injury.

In addition, IL-12β KO significantly attenuated the TAC-induced increase in lung weight, RV weight, and their ratio to tibial length or body weight, indicating that IL-12β plays a critical role not only in pressure overload-induced HF development, but also in HF progression. This is further supported by findings showing that IL-12β KO significantly reduced pulmonary inflammation, fibrosis, and arteriole muscularization. Since LV failure could contribute to pulmonary inflammation, structural remodeling, and RV hypertrophy, the reduced lung inflammation, remodeling, and RV hypertrophy in IL-12β KO mice after TAC could be partially due to improved cardiac function. However, given that lung inflammation can drive pulmonary remodeling and RV hypertrophy independently of LV function in mice with pre-existing HF [48], IL-12β deficiency could directly attenuate lung inflammation and remodeling in HF mice. Inflammation not only contributes to the onset of HF but also plays a crucial role in its progression [2,48,53]. We have previously demonstrated that HF is associated with a substantial accumulation and activation of macrophages, CD11c⁺ dendritic cells, and T cell activation in the lungs, while modulating inflammatory response effectively attenuated chronic HF-induced pulmonary inflammation, remodeling, and RV hypertrophy in mice with pre-existing HF [13, 48, 57]. For example, aggravating lung inflammation by exposing mice to PM2.5 worsened HF-induced lung inflammation and remodeling in mice with pre-existing HF [46], whereas promoting endogenous induction of regulatory T cells (Tregs) [48] or inhibiting IL-1β [13] attenuated HF progression in mice with pre-existing HF. These findings support the concept that inflammation not only plays a critical role in pressure overload-induced HF development but also in HF progression [16,53,57].

Our study showed that multiple immune cell subsets of the drainage lymph nodes were changed in wild type and KO mice after TAC. In response to acute cardiac injury induced by pressure overload, innate immune cells—specifically neutrophils and monocytes—rapidly infiltrate the myocardial tissue to clear injured or dead cardiomyocytes. Subsequently, antigen-presenting cells that have phagocytosed the damaged tissue migrate from the myocardium to the cardiac drainage lymph nodes. Within the drainage lymph nodes, these innate immune cells present cardiac antigens to resident adaptive immune cells. Upon activation by the presented antigens, adaptive immune cells undergo proliferation within the drainage lymph nodes and subsequently traffic to the site of injury in the cardiac tissue. The degree of adaptive immune cell activation directly correlates with the extent of their infiltration and the subsequent inflammatory response observed in the heart. Thus, the cardiac drainage lymph nodes play a critical and central role as indispensable sites for antigen presentation and the activation of the adaptive immune system following myocardial injury.

Another important finding of our study is that IL-12β KO suppressed TAC-induced LV gene expression pathways associated with inflammation and LV leukocyte infiltration in mice. HF-induced cardiac dysfunction in WT mice is also corroborated by bulk RNA sequencing data from LV tissues showing significant upregulation of genes and enrichment of many pathways related to inflammation and fibrosis. KEGG pathway analysis of DEGs revealed that the major enriched pathways were related to protein digestion and absorption, diabetic cardiomyopathy, AGE-RAGE signaling in diabetes, proteoglycans in cancer, and ECM-receptor interaction.

The GO biological pathways enrichment analysis also revealed that the major enriched biological pathways in WT mice after TAC were related to collagen fibril organization, extracellular matrix organization, cell adhesion, ossification, angiogenesis, and TGF-β signaling. The findings that IL-12β KO significantly suppressed the above LV gene profiles, particularly those associated with collagen formation, extracellular matrix remodeling, and inflammation, demonstrate that IL-12β exerts important and broad roles in promoting systolic overload-induced LV inflammation, fibrosis, and dysfunction. The drastic reductions of multiple LV signaling pathways of extracellular matrix remodeling and immune-related pathways in IL-12β KO mice after TAC not only indicate a critical role of IL-12β in promoting LV fibrosis and inflammation but also support the notion that there is significant crosstalk/interaction between extracellular matrix remodeling and immune cells during TAC-induced LV inflammation and dysfunction.

Several findings from this study also suggest that T cells, particular CD8⁺ T cells, might play an important role in the reduced LV inflammation and dysfunction in IL-12β KO mice after TAC: (i) Both LV CD8⁺ cell activation and CD4⁺ cell activation were reduced in IL-12β KO mice after TAC; (ii) LV IFNγ⁺CD8⁺ cells in cardiac drainage lymph nodes were significantly reduced in IL-12β KO mice after TAC; (iii) IL-12β KO significantly reduced TAC-induced expression of multiple adhesion molecules (such as Cxadr, CD44, Icam1, and CD29) for recruiting immune cells and beyond. Cxadr is a protein that modulates both MHCI molecule processing and immune cell infiltration; (iv) Expressions of CXCR3 and CD44 in CD8⁺ T cells and CD4⁺ T cells in drainage lymph nodes were significantly reduced in IL-12β KO as compared with WT mice after TAC. (v) IL-12β KO significantly decreased the expression of LV genes that could promote CD8 T cells’ effective function (such as B2m, MHCI molecules). Thus, these findings suggest that IL-12β KO protects hearts against TAC-induced inflammation, at least partially, by restraining the overactive immune cells, particularly CD8⁺ T cells. While the reduced cardiac inflammation in IL-12β KO is likely a collective effect of multiple immune cells and nonimmune cells, the reduction of IL-12 and IL-23 by IL-12β KO likely play a central role in protecting hearts against TAC-induced LV inflammation and HF.

This study has several limitations. First, we found that IL-12β KO attenuated TAC-induced cardiac inflammation, cardiomyocyte hypertrophy, and fibrosis. As each of these above factors could independently contribute to HF development, our study is unable to fully determine the relative role of IL-12β in modulating LV inflammation, cardiomyocyte hypertrophy, and fibrosis. Second, HF could progress to lung inflammation and remodeling, as well as RV hypertrophy, while lung inflammation can also directly promote lung remodeling and RV hypertrophy independent of LV dysfunction. Thus, we could not identify whether the reduced lung inflammation and remodeling is partially due to the direct impact on lung inflammation. Nevertheless, this study underscores an important role for IL-12β inhibition to attenuate pressure overload-induced LV inflammation and HF development. Third, since TAC was performed in IL-12β KO mice, this current study did not detect whether genetic inhibition of IL-12β is equally effective in attenuating HF progression in mice with existing HF. Thus, future studies assessing the effects of IL-12β inhibition in mice with pre-existing HF are warranted. Fourth, in the present study, TAC surgery was performed using a 26-gauge blunt needle in male mice and 27-gauge blunt needle in female mice. The reason for using more severe aortic restriction in female mice was to quickly generate significant TAC-induced HF in female mice which are generally “resistant” to TAC-induced LV hypertrophy and HF as compared with age-matched male mice [45,46]. Specifically, because female mice generally have smaller body weight as compared with age-matched male mice, aortic restriction created with the same size of needle likely would cause relatively less aortic restriction or LV afterload in female mice. Thus, to effectively generate TAC-induced HF female mice, we used a relatively small size needle (a 27-gauge needle) in female mice, while a relatively large size needle (a 26-gauge needle) was used to create aortic constriction in age-matched male mice. Unfortunately, using different sizes of needles for male and female mice does not allow us to directly determine potential sex differences in TAC-induced LV hypertrophy and HF in our study. Nevertheless, because of the relatively smaller body weight and body weight gain in young adult female mice as compared with male mice, even using the same size of needle in male and female mice for the TAC procedure would make it difficult to assess sex differences in the responses to TAC; the findings would still be unavoidably affected by the smaller body weight and weight gain in female mice compared with male mice. However, our findings clearly showed that IL-12β KO is effective in protecting both male and female mice from TAC-induced HF and lung remodeling. Finally, since the overall immune cell infiltration and the relative percentages of distinct immune cell subsets fluctuate significantly following TAC induction, and since we only assessed the immune cell alterations at sham, seven-day, and six-week post-TAC timepoints, a more detailed temporal analysis of specific cell types is warranted.

In summary, our data clearly demonstrate that IL-12β plays an important role in promoting pressure overload-induced cardiac dysfunction, hypertrophy, inflammation, fibrosis, and the consequent pulmonary inflammation and remodeling, and RV hypertrophy. These findings suggest that inhibition of IL-12β signaling may be a viable approach to treat systolic overload-induced cardiac inflammation and dysfunction.

Funding sources

This work was supported by research grants R01HL161085, R01HL139797, P20GM104357, and P30GM149404 from the NIH.

CRediT authorship contribution statement

Umesh Bhattarai: Conceptualization, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. Ziru Niu: Formal analysis, Methodology. Lihong Pan: Formal analysis, Methodology. Xiaochen He: Formal analysis, Methodology, Writing – review & editing. Dongzhi Wang: Formal analysis, Methodology. Tristan C. Orman: Formal analysis, Methodology. Hao Wang: Formal analysis, Methodology. Heng Zeng: Writing – review & editing. Jian-Xiong Chen: Writing – review & editing. Xiaojiang Xu: Writing – review & editing. Joshua S. Speed: Writing – review & editing. John S. Clemmer: Writing – review & editing. John E. Hall: Writing – review & editing. Yingjie Chen: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.

Declaration of competing interest

The authors have no conflicts of interest to disclose.

Appendix A. Supplementary data

The following is/are the supplementary data to this article.

Data availability

Data will be made available on request.