Injury polarized CD4⁺ T cells orchestrate redox and unfolded protein responses to promote cardiac repair after myocardial infarction

Abstract

Myocardial infarction (MI) initiates a wound-healing response where immune cells shape inflammation, tissue repair, and long-term remodeling. Although CD4⁺ T cells are increasingly recognized as contributors to post-MI healing, the transcriptional reprogramming defining their early pro-reparative functions remains incompletely resolved. Here, RNA sequencing of cardiac CD4⁺ T cells isolated 1 week after MI found that a substantial post-MI transcriptional fraction lay outside canonical cytokine-induced T helper subset polarization, including type 1 T helper cell (Th1), Th2, Th17, nature-occurring CD4⁺ regulatory T cell (nTreg), and peripherally induced Treg (iTreg) reference transcriptomic programs. Instead, this response was organized into a distinct CD4⁺ Th tissue injury-polarized (CD4^+/TIP) transcriptomic module enriched for extracellular matrix organization, adhesion, vascular, and developmental programs, with a coordinated downregulated arm involving RNA metabolism, chromatin regulation, and protein catabolic processes. Within the CD4^+/TIP population, MI induced and polarized at least 10 transcriptionally distinct CD4⁺ Th subsets at 1 week post-MI. Integration with curated transcription factors, epigenetic, reduction-oxidation (redox), and unfolded protein response (UPR) datasets identified a stress-adaptive architecture, in which regulatory subsets and the CD4^+/TIP population shared redox attenuation features, while the CD4^+/TIP state showed the strongest coupling to reparative tissue interaction programs, predominant Activating Transcription Factor 6 (ATF6)-aligned UPR structure, restrained proteostasis related outputs, and an innate-adjacent immune and secretome signature enriched for complement-associated, inflammatory recruitment, and extracellular communication genes. These findings identify a specific MI-associated CD4⁺ T-cell transcriptomic state during early MI inflammation and tissue repair. They establish a coordinated framework in which redox control, UPR, and tissue-injury-polarized CD4⁺ T-cell immune programs converge outside traditional Th and Treg lineages, offering new targets for CD4⁺ T-cell-mediated tissue repair after MI.

Graphical abstract

1. Introduction

Myocardial infarction (MI) initiates a complex wound healing process in which early inflammatory signals trigger a reparative program characterized by extracellular matrix remodeling, neovascularization, and scar stabilization. Precise temporal regulation of these inflammatory cascades is essential for effective healing; when these responses become excessive, prolonged, or mistimed, they may contribute to maladaptive remodeling and progression toward heart failure [1,2].

While innate immune cells such as neutrophils and monocyte-derived macrophages dominate the earliest inflammatory wave, CD4⁺ T cells are rapidly recruited to the injured myocardium and can act as major regulators of infarcted tissue organization and long-term remodeling [3,4]. Remarkably, cardiac CD4⁺ T cells, including regulatory subsets, appear within 24 h in the MI lesion, preceding conventional antigen-driven activation of T cells in draining lymph nodes [5,6].

This rapid innate immune engagement positions CD4⁺ T cells to respond to the same microenvironmental cues that shape innate inflammation after MI, including hypoxia, damage-associated molecular patterns (DAMPs) from necrotic cardiomyocytes, cytokine surges, and extracellular matrix injury [2,7], thereby revealing innate immune functions of CD4⁺ T cells [8]. Yet, how these early tissue-derived signals program CD4⁺ T cell states to influence cardiac repair remains poorly defined.

CD4⁺ T cells integrate environmental cues through “redox and stress sensing” pathways. Reactive oxygen species (ROS) couple metabolic and mitochondrial stress to T cell activation thresholds, transcriptional output, and cytokine production [[9], [10], [11], [12], [13]].

Concurrently, endoplasmic reticulum (ER) stress triggers the unfolded protein response (UPR), secretory capacity and effector differentiation, with Inositol-requiring enzyme 1α (IRE1α), X-box binding protein 1 (XBP1) and Activating Transcription Factor 6 (ATF6) pathways modulating proliferation and inflammatory programs [[14], [15], [16]]. Despite this mechanistic framework, whether redox and ER stress signals in the post-MI niche directly shape early CD4⁺ T cell transcriptomic identity and CD4⁺ T cell subset polarization remains unexplored.

In this study, we identify a novel, MI-specific CD4⁺ T cell transcriptomic state present 1-week post-infarction that is not captured by canonically polarized T helper cell 1 (Th1), Th2, Th17, naturally occurring thymus-derived CD4⁺ regulatory T cell (nTreg), or peripherally induced Treg (iTreg) subset gene programs. Instead, this response is dominated by a distinct CD4^+/TIP transcriptomic module comprising a reparative arm enriched for extracellular matrix organization, adhesion, and tissue interaction pathways, together with a coordinated suppression of genes involved in RNA metabolism, chromatin regulation, and ER and protein homeostasis. These cells exhibit transcriptomic reprogramming characterized by coordinated intracellular redox-regulatory features that intersect with innate-like immune and extracellular signaling programs, and by strong ATF6-aligned unfolded protein response associations, defining a stress-adaptive CD4⁺ T cell immune program within the cardiac infarct niche. Our findings identify a distinct, injury-imprinted, CD4^+/TIP transcriptomic program in MI that integrates early post-infarct immune remodeling with unfolded protein response and regulates the repair-associated tissue programs. These results identify novel therapeutic targets to enhance CD4⁺ T cell-mediated tissue repair after MI.

2. Results

2.1. CD4⁺ T cell transcriptomics is reprogrammed toward wound healing and the regulation of stress-associated metabolic activity 1 week after MI

To characterize the early post myocardial infarction (MI) transcriptional landscape in CD4⁺ T cells, RNA sequencing was conducted on CD4⁺ T cells isolated from 8-week C57BL/6J mice at 1-week post-MI (Fig. 1A). Differential gene expression analysis revealed a 4736 upregulated genes and 3436 downregulated genes at one-week post-MI (adj. p < .05, FC > 1.5), suggesting a comprehensive remodeling of CD4⁺ T cell states in response to the post-myocardial infarct environment (Fig. 1B1 and 1B2). This extent of reprogramming aligns with the understanding that inflammatory and extracellular matrix (ECM) remodeling signals associated with MI influence lymphocyte recruitment and functional polarization during the early stages of repair [2].

Fig. 1.

MI reprograms CD4⁺ T cell transcriptomics at 1 week post-infarction.

(A) Eight-week-old C57BL/6 mice were given MI injury via left anterior descending (LAD) Ligation. Single-stranded RNA (ssRNA) sequencing was performed on CD4⁺ T cells isolated from 1-week post-MI (1W PMI, n-2 per group) and compared against a bulk RNA sequencing atlas of CD4⁺ T cells (Stubbington et al., 2015, PMID: 25886751). Created in BioRender. Knight, C. (2026) https://BioRender.com/c5sj070.

(B.1-B.2) Distinct temporal gene expression of Top 20 Upregulated and Downregulated DEGs between 1W post-MI and Sham (false discovery rate (FDR) < 0.05, adjusted p-value (p-adj.) < 0.05, fold change (FC) > 1.5).

(C.1-C.2) Pathway enrichment analysis on up- and downregulated DEGs at 1W PMI for Gene Ontology, Reactome Gene Sets, and KEGG Pathway shows an upregulation of wound healing-related terms.

(D) K-means Clustering highlights temporal differences during injury in CD4⁺ T cells in the top 4000 DEGs (p-adj. < .0.05).

(E) Volcano plot of 1-week post-MI versus sham up- and downregulated genes illustrating broad transcriptional remodeling with prominent induced and repressed transcripts (Log2(FC) > 1.5 and (p-adj.) < 0.05. Genes in red are upregulated, and genes in green are downregulated. The top 10 DEGs are labeled.

Among the most strongly upregulated transcripts were genes linked to antigen presentation and immune activation (CD74 (Invariant Polypeptide Of Major Histocompatibility Complex, Class II Antigen-Associated) [log2 fold change (log2FC) = 7.19, adj. p = 7.45E-271]), injury associated lipid handling and immune signaling (Apolipoprotein E (ApoE) [log2FC = 8.23, adj. p = 8.99E-217], Transmembrane Immune Signaling Adaptor TYROBP (Tyrobp) [log2FC = 11.66, adj. p = 1.03E-118]), chemotactic recruitment (C-C Motif Chemokine Receptor 2 (Ccr2) [log2FC = 8.58, adj. p = 5.28E-132]), and extracellular matrix remodeling (Collagen Type IV Alpha 2 Chain (Col4a2) [log2FC = 10.04, adj. p = 4.56E-150], Collagen Type V Alpha 1 Chain (Col5a1) [log2FC = 11.40, adj. p = 5.67E-114], Fibronectin 1 (Fn1) [log2FC = 12.62, adj. p = 2.08E-110], Latent Transforming Growth Factor Beta Binding Protein 4 (Ltbp4) [log2FC = 9.55, adj. p = 3.21E-158]) (Fig. 1B1). Metascape was used for pathway enrichment of upregulated differentially expressed genes (DEGs), leveraging simultaneous analysis of Gene Ontology (GO), Reactome Gene Sets, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, which further revealed coordinated activation of leukocyte migration, extracellular matrix organization, cytoskeletal remodeling, angiogenesis, and wound response programs, consistent with phase-specific repair processes following myocardial infarction consistent with phase specific repair processes following infarction (Fig. 1C1) [1]. Additionally, there is an upregulation of Protein Transport Protein Sec61 Subunit Gamma (Sec61g) [log2FC = 6.15, adj. p = 4.65E-160], which is central to protein translation across the ER and turns off UPR signaling during ER stress [17,18]. Importantly, enrichment of the electron transport chain points to concurrent metabolic reprogramming, a key determinant of CD4⁺ T cell differentiation and function [19,20]. Collectively, these data define a transcriptional state in CD4⁺ T cells at one week post-MI, characterized by coordinated activation of tissue-targeting, stress-dampening, and repair-associated effector programs, coupled with a metabolically adaptive, redox-linked signaling network.

Conversely, the most downregulated transcripts were enriched for regulators of transcriptional and RNA processing and signaling pathways linked to lymphocyte differentiation (Fig. 1B2). Pathway analysis further highlighted suppression of RNA and mRNA metabolism, DNA templated transcriptional programs, chromatin-remodeling and macromolecular biosynthesis (Fig. 1C2). This coordinated downregulation indicates a reduction in baseline homeostatic and biosynthetic activity at 1-week post myocardial injury, occurring in parallel with the activation of repair associated programs. Together, these shifts suggest a functional reallocation of CD4⁺ T cell transcriptomes from steady-state maintenance toward stress-adaptive, tissue-directed responses within the injured myocardium [21].

To further distinguish MI induced transcriptional, we performed k means clustering of the top 4000 Differentially Expressed Genes (DEGs) identified distinct gene modules associated with cell surface receptor signaling, cell to cell and cell to matrix interactions, migration, extracellular matrix remodeling, and vascular morphogenesis (Fig. 1D). In contrast, Cluster D was selectively enriched for RNA and transcriptional processes (Fig. 1D), reinforcing the coordinated repression observed among downregulated DEGs. Consistent with these findings, the volcano plot highlights a dominant induction of repair and remodeling associated transcripts in post-MI CD4⁺ T cells (Fig. 1E). Collectively, these data establish that CD4⁺ T cells at 1-week post-MI adopt a specialized transcriptional state defined by activation of tissue remodeling, migratory and repair programs, coupled with suppression of transcriptional and RNA processing pathways that may otherwise contribute to cellular stress, including ER stress. Given the heterogeneity of CD4⁺ T cell subsets, we next sought to delineate subset-specific signaling programs in CD4 T helper cells within Th and Treg populations at this stage of injury.

2.2. A distinct CD4⁺ T cell population emerges at 1-week post-MI with a pro-reparative gene signature independent of canonical Th1, Th2, Th17, and Treg lineage

CD4⁺ T cells can adopt diverse effector functions such as cytokines and regulatory states defined by distinct lineage-specific transcriptional programs, including Th1 (T-Box Transcription Factor 21, TBX21, T-bet), Th2 (GATA Binding Protein 3, Gata3), Th17 (retinoid orphan receptor(ROR)-related orphan receptor gamma t, RORγt), and Forkhead Box P3 (Foxp3)⁺ regulatory T cell subsets that can be thymus-derived (nTreg) or peripherally induced (iTreg) [[22], [23], [24], [25]]. To determine whether the 1-week post-MI transcriptional signature reflects shifts among these canonical states, we compared DEGs between post-MI and sham CD4⁺ T cells with established subset-specific gene programs (Fig. 2A) [26].

Fig. 2.

Myocardial infarction distinctly reprograms canonical CD4⁺ T cell subset associated gene programs and reveals a large undefined, CD4^+/TIP gene module at 1 week post-injury with five top pathways such as extracellular matrix organization, cell-cell adhesion, blood vessel development, heart development and cytoskeleton in muscle cells.

(A) Experimental design and analysis workflow. CD4⁺ T cells were isolated from 1-week post-MI C57BL/6 mice, followed by RNA sequencing and DEG calling ((false discovery rate (FDR) < 0.05, adjusted p-value (p-adj.) < 0.05, fold change (FC) > 1.5). DEG lists were intersected with subset enriched gene targets for type 1 CD4⁺ T helper cell (Th1), CD4⁺ T helper cell 2 (Th2), CD4⁺ T helper cell 17 (Th17), natural CD4/Foxp3+ regulatory T cell (nTreg), and induced CD4/Foxp3+ regulatory T cell (iTreg) derived from a bulk RNA sequencing atlas (Stubbington et al., 2015, PMID: 25886751).Created in BioRender. Knight, C. (2026)https://BioRender.com/hlnfu8s.

(B.1) Pie chart summary of overlaps between 1-week post-MI upregulated genes (4736 genes) and subset gene lists. A large fraction of upregulated genes (2236 genes, 47.21%) did not overlap any subset list and were classified as an undefined, CD4^+/TIP upregulated module; additional genes overlapped multiple subsets (2016 genes, 42.57%) or uniquely overlapped Th1, Th2, Th17, nTreg, or iTreg programs (54 genes, 1.14%; 78 genes, 1.65%; 97 genes, 2.05%; 144 genes, 3.04%; 111 genes, 2.34%), as indicated.

(B.2) Pathway enrichment for distinct upregulated gene fractions assigned to nTreg, iTreg, or the CD4+/TIP module.

(C.1) Pie chart summary of overlaps between 1-week post-MI downregulated genes (3436 genes) and subset gene lists. A substantial fraction (1132 genes, 32.95%) did not overlap any subset list and was classified as an undefined, CD4^+/TIP downregulated module; an additional fraction overlapped multiple subsets (1731 genes, 50.38%), with smaller unique overlaps for Th1, Th2, Th17, nTreg, or iTreg programs (117 genes, 3.41%; 66 genes, 1.92%; 229 genes, 6.66%; 107 genes, 3.11%; 54 genes, 1.57%), as indicated.

(C.2) Pathway enrichment (top terms) for distinct downregulated gene fractions assigned to nTreg, iTreg, or the CD4^+/TIP module.

Among 4736 upregulated genes, 2016 (42.6%) overlapped with multiple subset signatures, while only small fractions mapped exclusively to individual lineages (Th1: 1.14%, Th2: 1.65%, Th17: 2.05%, nTreg: 3.04%, iTreg: 2.34%) (Fig. 2B1). Strikingly, nearly half of the upregulated genes (2236 genes; 47.21%) did not align with any canonical subset programs, which led to the proposal of a novel CD4⁺ T cell state – tissue injury polarization (TIP) subset, CD4⁺T^TIP. Pathway analysis of this CD4^+/TIP gene set revealed strong enrichment for pro-reparative processes, including extracellular matrix organization, cell-to-cell adhesion, and cardiovascular developmental programs. These data identify a previously unrecognized, pro-reparative CD4⁺ T cell transcriptional state of tissue injury polarization that is distinct from established Th- and Treg-polarized lineages (Fig. 2B2).

A similar pattern was observed among downregulated genes. Of the 3436 suppressed transcripts, 1132 (32.95%) were not associated with any canonical CD4⁺ T subsets (Fig. 2C1). This CD4^+/TIP downregulated program was enriched for RNA metabolism, chromatin regulation, and protein catabolic processes, indicating coordinated suppression of transcriptional and biosynthetic activity (Fig. 2C2). This repression suggests active restraint of cellular programs tightly linked to stress signaling and inflammatory amplification.

Mechanistically, RNA processing and protein turnover pathways are closely coupled to endoplasmic reticulum (ER) stress responses and the unfolded protein response (UPR), which can drive inflammation through NF-κB activation, Toll-like receptor (TLR) X-Box Binding Protein 1 (XBP1) signaling, and inflammasome engagement. In this context, the suppression of these pathways within the CD4^+/TIP population likely represents a novel regulatory mechanism that uncouples ER stress from proinflammatory signaling [16,21,27,28].

This regulatory state is further linked to redox control, as ER proteostasis and UPR signaling are intrinsically redox sensitive. Reactive oxygen species (ROS) and Nicotinamide Adenine Dinucleotide Phosphate (NADPH) oxidase-dependent pathways are known to amplify inflammatory outputs downstream of ER stress. Thus, coordinated repression of RNA metabolism and protein turnover programs in this subset may limit redox-dependent feed-forward inflammatory signaling [27,29,30].

Collectively, these findings define a distinct CD4⁺ T cell state at 1-week post-MI characterized by a pro-reparative transcriptional program coupled with suppression of stress-responsive and redox-amplifying pathways. This balance likely enables effective tissue repair while preventing transition into chronic canonical inflammatory effector states. We next sought to resolve the subset-specific contributions to redox signaling and further define the transcriptional identity of this CD4^+/TIP population.

2.3. Flow cytometry identifies expansion of a noncanonical CD4^+/TIP population at 1-week post-MI

To determine whether the CD4^+/TIP transcriptional module identified by RNA sequencing corresponded with an expanded CD4⁺ T cell population at 1-week post-MI, we performed flow cytometric analysis of cardiac CD4⁺ T cell subsets from sham heart controls and MI hearts (Fig. 3A). This strategy was designed to quantify both canonical CD4⁺ T cell lineages and the noncanonical CD4^+/TIP population defined by exclusion of major Th and Treg lineage-associated transcription factors-stained populations.

Fig. 3.

The CD4^+/TIP population expands during the early response to myocardial infarction.

(A) Experimental workflow for isolating cardiac CD4⁺ T-cell populations from sham and 1-week post-MI hearts by flow cytometry. Created in BioRender. Knight, C. (2026) https://BioRender.com/ll6v8uv.

(B) Representative gating strategy used to identify the CD4^+/TIPpopulation by excluding cells expressing the canonical lineage-defining transcription factors T-bet, GATA3, RORγt, and FOXP3.

(C.1–C.6) Quantification of cardiac CD4⁺ T-cell populations in sham and 1-week post-MI hearts, normalized to heart weight. MI significantly increased the number of noncanonical CD4^+/TIP cells, supporting the expansion of a distinct injury-associated CD4⁺ T-cell population during the early reparative phase after myocardial infarction (Welch's t-test).

Cardiac single-cell suspensions were first gated to identify the primary cell population, followed by enrichment of viable CD45⁺ lymphocytes and identification of CD3⁺CD4⁺ T cells (Fig. 3B). Within the CD4⁺ T cell compartment, canonical lineage-associated subsets were identified using Tbet, Gata3, RORγt, and Foxp3, corresponding to Th1-, Th2-, Th17-, and Treg-associated programs, respectively. The remaining CD4⁺ T cells lacking Tbet, Gata3, RORγt, and Foxp3 expression were classified as the flow cytometry-defined CD4^+/TIP population (Fig. 3B).

Consistent with the broad transcriptional remodeling observed by RNA sequencing, total CD45⁺CD3⁺CD4⁺ T cells were significantly increased in MI hearts compared with sham controls when normalized to heart weight (Fig. 3C1). Canonical CD4⁺ T cell subsets were also increased following MI, including CD4⁺Tbet⁺ cells, CD4⁺Gata3⁺ cells, CD4⁺RORγt⁺ cells, and CD4⁺Foxp3⁺ cells (Fig. 3C2-C5). These findings support that MI promotes expansion of multiple canonical CD4⁺ T cell subset populations during the early post-injury period.

Importantly, the CD4^+/TIP population was markedly increased at 1-week post-MI compared with sham control hearts (Fig. 3C6). This increase supports the conclusion that myocardial injury induces broad CD4⁺ T cell remodeling while also promoting a distinct tissue injury-polarized CD4⁺ T cell compartment outside the major canonical Th and Treg-associated lineage transcription factors.

Collectively, these data provide flow cytometry-identified protein expression results supporting the CD4^+/TIP population identified in the transcriptomic analysis. The expansion of CD4^+/TIP cells after MI strengthens the interpretation that this population represents a responsive injury-polarized CD4⁺ T cell state rather than a residual unassigned transcriptional fraction.

2.4. MI induces and polarizes CD4⁺ T cells into at least 10 new CD4⁺ Th subsets within the CD4^+/TIP population at 1-week post-MI, coordinating redox and oxidative stress regulation

To determine whether regulatory CD4⁺ T cell subsets at 1-week post-MI (nTreg, iTreg, and the CD4^+/TIP population) contribute to redox control and stress attenuation, we integrated subset-specific gene programs with curated transcription factor, epigenetic, and redox-associated gene sets.

nTregs displayed enrichment of transcriptional regulators linked to lineage stability and stress-responsive control, including IKAROS Family Zinc Finger 2 (IKZF2, Helios) and Activating Transcription Factor 3 (ATF3), alongside redox-associated genes involved in mitochondrial function and intracellular trafficking (Fig. 4A). Notably, components of mitochondrial electron transport (e.g., NADH:Ubiquinone Oxidoreductase Subunit B7, NDUFB7) and redox buffering systems (e.g., CDGSH Iron Sulfur Domain 3, CISD3) were coupled with vesicular trafficking regulators (Trafficking Protein Particle Complex Subunit 2L, TRAPPC2L), supporting a model in which nTregs maintain suppressive function through coordinated mitochondrial redox control and regulated secretory pathways (Fig. 3B) [[31], [32], [33], [34], [35]]. Consistent with this, the nTreg downregulated program showed reduced expression of inflammatory coactivators (e.g., SET Domain Containing 7, Histone Lysine Methyltransferase, SETD7) (Supp Fig. 3A) and glycolytic drivers (e.g., Phosphofructokinase, Platelet, PFKP) (Supp. Fig. 3B), alongside factors involved in RNA processing and protein folding, collectively reinforcing a stable, low-inflammatory, oxidative metabolic state [[36], [37], [38], [39], [40], [41]].

Fig. 4.

Distinct transcription factors, epigenetic regulators, and redox architecture of regulatory CD4⁺ T-cell programs are identified at 1 week post-myocardial infarction.

Differentially expressed genes (DEGs) from 1W MI vs Sham were intersected with curated transcription factor and epigenetic factor catalogs and with redox-associated gene sets to resolve regulatory circuitry within natural CD4/Foxp3+ regulatory T cell (nTreg), induced CD4/Foxp3+ regulatory T cell (iTreg), and the MI-specific CD4^+/TIP pool.

(A) Upregulated regulatory programs show subset-specific enrichment of transcriptional control nodes. nTreg intersections include IKZF2 (Helios) and ATF3. iTreg intersections are comparatively compact and center on ID1. The CD4^+/TIP program shows the broadest recovery of transcription factors and epigenetic regulators and prominently includes tissue-linked regulators such as TSC22D3 (GILZ) and GATA5, together with chromatin and RNA regulators including SMARCD3, KAT8, and RBM24.

(B). Rows classify ROS regulators by functional compartment and directionality within the ROS framework, including Pro-Mito ROS, Anti-Mito ROS, Pro-Cellular ROS, Anti-Cellular ROS, and ROS Upregulated Genes outlined in PMID: 38207075. The distribution highlights subset-specific deployment of mitochondrial and cellular ROS-promoting and ROS-buffering regulators, with the MI-specific CD4^+/TIP compartment exhibiting the highest density of interpretable ROS regulators across multiple categories.

In contrast, iTregs exhibited a more focused regulatory program defined by ID1 (Fig. 4A) and the mitochondrial complex I component NADH: Ubiquinone Oxidoreductase Core Subunit V1 (NDUFV1) (Fig. 4B). This signature suggests coupling of differentiation restraint with tight regulation of mitochondrial redox flux, positioning iTregs in a controlled, stress-adaptive state that limits effector differentiation while maintaining redox balance [[42], [43], [44], [45]].

Strikingly, the CD4^+/TIP population showed that MI induces a distinct eight-transcription-factor network indicative of an injury-imprinted, pro-reparative state. This program integrates cardiogenic and matrix-remodeling regulators (e.g., GATA Binding Protein 5 (GATA5), CAMP Responsive Element Binding Protein 5 (CREB5) (Fig. 4A) with factors governing metabolic adaptation, developmental stability, genome integrity, and inflammatory restraint (e.g., Nuclear Receptor Subfamily 1 Group H Member 4 (NR1H4), SRY-Box Transcription Factor 6 (SOX6), Zinc Finger And SCAN Domain Containing 5B (ZSCAN5B), Distal-Less Homeobox 3 (DLX3), Zic Family Zinc Finger 3 (ZIC3), and Mesenchyme Homeobox 2 (MEOX2) (Fig. 4A and B). These features collectively define a non-canonical regulatory state with strong pathophysiological relevance aligned with tissue repair rather than classical CD4⁺ T cell subset immune polarizations into Th subsets originally defined in vitro by cytokine inductions [[46], [47], [48], [49], [50], [51], [52], [53], [54]].

Concurrently, the CD4^+/TIP downregulated program revealed coordinated suppression of proteostasis, transcriptional stress responses, and redox-amplifying pathways. Reduced expression of regulators of proteasome recovery (NFE2 Like BZIP Transcription Factor 1, NFE2L1), antioxidant response modulation (MAF BZIP Transcription Factor G, MafG), and mitochondrial metabolic entry (Pyruvate Dehydrogenase E1 Subunit Alpha 1, PDHA1), along with signaling nodes such as Glycogen Synthase Kinase 3 Beta (GSK3β), suggests active limitation of ER stress–linked and redox-driven inflammatory amplification (Supp. Fig. 3A and 3B) [[55], [56], [57], [58], [59]]. If we follow classical principle that each transcription factor defines a CD4⁺ T helper subset, such as T-Bet defines Th1, GATA3 defines Th2, RORgt defines Th17, Foxp3 defines Treg, respectively, our identification of 10 transcription factors in the CD4^+/TIP population outside of Th1, Th2, Th17, nTreg and iTreg suggests that MI induces and polarizes the CD4⁺ T cells into at least ten new CD4⁺ Th subsets within the CD4^+/TIP population.

Collectively, these data support a unifying model in which CD4⁺ T cell subsets at 1 week post-MI engage in distinct yet convergent transcriptional strategies to restrain stress and redox signaling. While nTregs and iTregs stabilize canonical regulatory programs through mitochondrial and metabolic control, the CD4^+/TIP population extends this framework into a reparative, injury-adaptive state. This previously unrecognized CD4⁺ T cell program integrates redox regulation with tissue remodeling, positioning it as a key contributor to coordinated immune-mediated repair in the post-MI environment.

2.5. Loss-of-function models reveal that intersections of ER stress signatures indicate dampened UPR effector machinery at 1 week post-MI and predict altered redox buffering capacity

To mechanistically anchor the stress-adaptive profile inferred from redox signaling, we investigated whether 1-week post MI CD4⁺ T-cell subset programs correspond with canonical unfolded protein response (UPR) transcriptional outputs through a directional intersection strategy (Fig. 5A). Distinct gene lists for nTreg, iTreg, and CD4^+/TIP population were intersected with publicly available endoplasmic reticulum (ER) stress knock-out (KO) datasets representing genetic disruption of major UPR pathways (ATF4-KO, ATF6-KO, Eukaryotic Translation Initiation Factor 2 Alpha Kinase 3 (EIF2AK3, PRKR-Like Endoplasmic Reticulum Kinase, PERK)-KO, and IRE1-KO), as well as pharmacologic induction of ER stress by thapsigargin (a non-competitive inhibitor of sarco/ER-ATPase (SERCA) pumps) [28,60]. For genes upregulated in MI CD4⁺ T-cell subsets (Fig. 5), intersections were performed against genes downregulated in UPR-KO datasets to identify RNA transcripts expected to increase when the corresponding UPR pathway remains intact. For genes downregulated in MI CD4⁺ T-cell subsets (Supp. Fig. 2), intersections were performed against genes upregulated in UPR-KO datasets to identify UPR-linked transcripts that increase when a given arm is disrupted yet decrease in the post-MI setting. ER stress inducer Thapsigargin signatures were compared in the canonical upregulated-to-upregulated and downregulated-to-downregulated directions. This directional framework prioritized downstream stress-response and ER-handling genes that were lower at 1- week post-MI across subsets, consistent with restrained engagement of broad UPR effector outputs at this stage [[60], [61], [62]].

Fig. 5.

Endoplasmic reticulum (ER) stress intersection analysis of CD4⁺ subsets at 1-week post-MI reveals a dominant ATF6-linked proteostasis structure within the CD4^+/TIP population and more restricted, subset-specific PERK- and ATF6-associated programs in nTreg and iTreg cells.

(A) Experimental workflow outlining intersection analysis between distinct CD4⁺ subset gene lists (Th1, Th2, Th17, nTreg, iTreg, and CD4^+/TIP) and ER-stress perturbation datasets (ATF4-KO [GSE10470], ATF6-KO [GSE49646], PERK-KO [GSE29929], IRE1α-KO [GSE130952], and thapsigargin treatment [GSE200626]). Differentially expressed genes (DEGs) from 1-week post-MI vs sham (P adj ≤ 0.05, |log₂FC| ≥ 1.5) were used to identify overlapping genes within upregulated and downregulated modules. The table reports DEG counts and overlap sizes for each subset. Pathway enrichment for the top 10 terms was performed for intersected genes (LogP < −3). Created in BioRender. Knight, C. (2026) https://BioRender.com/mz1wc20.

(B.1-B.3) nTreg upregulated genes (26 total): ATF6-KO Down (69.23%, 18 genes), PERK-KO Down (7.69%, 2 genes), IRE1α-KO Down (3.85%, 1 gene), and Thapsigargin Up (19.23%, 5 genes). Enriched terms include extracellular matrix organization and cellular stress response, indicating reparative activation partially dependent on ATF6/PERK signaling. B.2 and B.3 had less than 10 enriched terms.

(C) iTreg upregulated genes (13 total): ATF6-KO Down (84.62%, 11 genes), IRE1α-KO Down (7.69%, 1 gene), and Thapsigargin Up (7.69%, 1 gene). No significant pathway enrichment.

(D.1-D.3) CD4^+/TIP upregulated genes (296 total): ATF6-KO Down (72.97%, 216 genes), PERK-KO Down (9.12%, 27 genes), IRE1α-KO Down (2.70%, 8 genes), and Thapsigargin Up (15.20%, 45 genes). These overlaps highlight robust ATF6-linked induction of ER folding and redox-associated genes, consistent with adaptive proteostasis and stress buffering.

In nTregs, the upregulated module (26 genes, Fig. 5B1) was predominantly characterized by overlap with ATF6-deficiency (knock-out, KO) downregulated genes (ATF6 promoted genes, 18 genes; 69.23%), with smaller contributions from thapsigargin-up (5 genes; 19.23%), PERK-KO down (PERK promoted genes, 2 genes; 7.69%), and IRE1-KO down (IRE1 promoted gene, 1 gene; 3.85%), indicating selective induction of an ATF6-linked adaptive transcriptional component. Consistent with this restricted engagement, the primary pathway summaries (Fig. 5B2) highlight immune regulatory functions and the negative regulation of activation. Additionally, terms were upregulated for muscle development in the Thapsigargin overlap (Fig. 5B3). Conversely, the downregulated module in nTregs (65 genes, Supp 2A.1) primarily overlapped with ATF6-KO upregulated genes (ATF6 suppressed genes, 61 genes; 93.85%), with a smaller PERK-KO upregulated component (PERK suppressed genes, 4 genes; 6.15%). The pathway summaries (Supp Fig. 2A2) focus on the governance of RNA and protein handling, consistent with a targeted limitation of biosynthetic and processing infrastructure under post-injury constraints shown in previous figures.

The iTreg upregulated set (13 genes, Fig. 5C) was similarly dominated by overlap with ATF6-KO downregulated genes (ATF6 promoted genes, 11 genes; 84.62%), with minor contributions from IRE1-KO downregulated (IRE1 promoted gene, 1 gene; 7.69%) and thapsigargin-upregulated (1 gene; 7.69%) sets, but did not yield significantly enriched pathways. Conversely, the downregulated iTreg module (29 genes; Supp Fig. 2B1) primarily aligned with ATF6-KO upregulated genes (ATF6 suppressed genes, 25 genes; 86.21%), with a secondary PERK-KO upregulated component (PERK suppressed genes, 4 genes; 13.79%). The pathway summaries (Supp Fig. 2B2) generally suggest restricted remodeling processes and transcriptional regulation. Across both Treg-lineage subsets, these findings support a selective configuration of the UPR axis mainly governed by ATF6, with a secondary influence from PERK.

In contrast to the confined Treg-lineage patterns, the CD4^+/TIP population exhibited a more pronounced ER stress response. The upregulated genes within the CD4^+/TIP group (296 genes, Fig. 5D1) were predominantly characterized by overlap with ATF6-KO downregulated genes (ATF6 promoted genes, 216 genes; 72.97%), with additional contributions from thapsigargin-up (45 genes; 15.20%), PERK-KO downregulated (PERK promoted genes, 27 genes; 9.12%), and IRE1-KO downregulated (IRE1 promoted genes, 8 genes; 2.70%). The corresponding pathway summaries (Fig. 5D2-D.3) reflect tissue interaction, extracellular communication, and categories related to inflammation and chemotaxis, consistent with an injury-polarized activation program.

The CD4^+/TIP populations-downregulated module (756 genes, Supp Fig. 2C1) showed a dominant overlap with ATF6-KO upregulated genes (ATF6 suppressed genes, 704 genes; 93.12%), with smaller contributions from PERK-KO upregulated (PERK suppressed genes, 38 genes; 5.03%), ATF4-KO upregulated (ATF4 suppressed genes, 7 genes; 0.93%), thapsigargin-downregulated (6 genes; 0.79%), and IRE1-KO upregulated (IRE1 suppressed gene, 1 gene; 0.13%). In the pathway summaries (Supp Figs. 2C2-C.3), the top downregulated annotations cluster around RNA metabolism and post-transcriptional control, proteolysis and catabolic proteostasis, translation-associated governance, and intracellular transport, with a strong representation of Golgi and vesicle trafficking.

A significant implication of this directional UPR intersection analysis is that the redox-adaptive attributes depicted in Fig. 3 support a targeted, subset-distinct set of regulatory mechanisms. In the CD4^+/TIP population, Fig. 4 identifies an upregulated program dominated by overlap with ATF6-KO downregulated genes, together with a smaller thapsigargin-associated component, while the downregulated program supports a broad RNA- and protein-regulatory module. Collectively, these findings position the CD4^+/TIP population as an injury-adaptive, lineage-independent CD4⁺ T cell state with restrained UPR-linked redox signaling at 1-week post-MI.

2.6. ER stress signature intersections identify an MI-specific CD4^+/TIP population enriched for innate-adjacent immune signaling and secretome programs

Building on the identification of the CD4^+/TIP population as a repair-associated and stress-adaptive CD4⁺ T cell state, we next examined whether the UPR intersection structure observed at 1-week post-MI aligned with transcription factor and epi(genetic)factor programs, immune-signaling genes, and secretome-associated gene lists within the MI-specific CD4^+/TIP population. As illustrated in Fig. 6A, CD4^+/TIP gene modules intersecting ATF4-KO, ATF6-KO, PERK-KO, IRE1α-KO, and thapsigargin-associated ER stress datasets were cross-referenced with curated transcription factor, epifactor, immune signaling, and secretome gene lists. This knowledge-based profiling strategy was designed to prioritize candidate regulatory and effector programs within the CD4^+/TIP population in the context of UPR-linked stress adaptation [63].

Fig. 6.

Unfolded protein response (UPR) intersection structure in the 1-week post-MI CD4^+/TIPprogram links constrained ER proteostasis with redox remodeling and innate-like immune signaling.

(A) Intersection workflow and summary of shared genes between the 1-week post-MI CD4^+/TIP population and the ER stress perturbation signatures shown in Fig. 4. The CD4^+/TIP MI gene modules were further cross-referenced with the transcription factor lists, immune signaling gene lists, and secretome gene lists used in Fig. 3 to prioritize regulatory and effector candidates that co-segregate with UPR-related signatures. Created in BioRender. Knight, C. (2026) https://BioRender.com/gs7wlmx.

(B) Stress-axis coupling at the level of transcriptional regulators and chromatin-associated factors within the CD4^+/TIP MI program. Overlaps between the CD4^+/TIP MI transcription factor set and ER stress perturbation intersections highlight regulatory candidates that align with constrained UPR states, including Creb5, Gata5, Rorb, and Bnc2, together with chromatin or epigenetic factors such as Smarcd3 and Aurkc.

(C) Innate-like immune signaling identity within the CD4^+/TIP MI compartment revealed by recurrent immune intersection overlap. Shared genes across immune-aligned categories include innate-associated receptors and signaling nodes (Csf1r, Ptafr, Cd300a, Tlr13, Trem1, Treml4, C5ar1, Cr2, Cfp), chemokine-associated outputs (Cxcl9, Cxcl14), The repeated appearance of these nodes across independent immune intersections supports their interpretation as stable characteristics of the CD4^+/TIP pool rather than artifacts of a single gene list.

(D) Secretome gene lists within the CD4^+/TIP Upregulated MI compartment. The overlaps in the MI-up CD4^+/TIP secretome highlight an extracellular communication and tissue-interaction program, including Apoe, matrix and structural components Col4a2, Lama4 and Thbs2, and immune recruitment and activation-associated genes Csf1r, Trem1, and Lcn2.

(E.1-E.15) CD4^+/TIP T cells isolated from experiments outlined in Fig. 3 to compare mRNA expression between Sham versus MI via qPCR. Fold change normalized to S18 (Welch's t-test).

Stress-axis coupling remained evident at the level of transcriptional regulators and epigenetic factors (Fig. 6B). The overlap between the 1-week MI-upregulated CD4^+/TIP transcription factor module and ATF6-KO downregulated genes identified RAR Related Orphan Receptor B (Rorb), CAMP Responsive Element Binding Protein 5 (Creb5), GATA Binding Protein 5 (Gata5), and Basonuclin Zinc Finger Protein 2 (Bnc2), while the PERK-KO downregulated intersection identified Nuclear Receptor Subfamily 2 Group E Member 1 (Nr2e1) (Fig. 6B). Epifactor intersections further identified SWI/SNF Related BAF Chromatin Remodeling Complex Subunit D3 (Smarcd3) and Aurora Kinase C (Aurkc) within the MI-upregulated ATF6-KO downregulated overlap. In addition, comparison with the 290 transcription factor gene list recovered Creb5 and Gata5 within the MI-up ATF6-KO downregulated intersection. These findings identify a focused set of candidate regulatory genes linking the CD4^+/TIP population to ATF6- and PERK-associated stress-response architecture.

The UPR-linked CD4^+/TIP program also showed a strong immune-signaling structure (Fig. 6C). Across plasma membrane, innate immunity, immune response, cytokine, chemokine, and CD marker gene lists, the MI-upregulated ATF6-KO downregulated (ATF6 promoted) intersection recovered a coherent set of receptors, complement-associated genes, inflammatory recruiters, and immune regulatory molecules. These included Colony Stimulating Factor 1 Receptor (Csf1r), Platelet Activating Factor Receptor (Ptafr), CD300A (encoding a cell surface immune receptor, Cd300a), Toll Like Receptor 2 (Tlr13), Triggering Receptor Expressed On Myeloid Cells 1 (Trem1), Triggering Receptor Expressed On Myeloid Cells Like 4 (Treml4), Complement C5a Receptor 1 (C5ar1), Complement C3d Receptor 2 (Cr2), Complement Factor Properdin (Cfp), CD84 (encoding a SLAM family self-ligand receptor and membrane glycoprotein), C-X-C Motif Chemokine Ligand 9 (Cxcl9), and C-X-C Motif Chemokine Ligand 14 (Cxcl14), among additional immune-associated genes (Fig. 6C). In contrast, the PERK-KO and IRE1α-KO downregulated intersections contained fewer shared immune genes, indicating that the immune-signaling component of the CD4^+/TIP program was most strongly represented within the ATF6-aligned arm of the UPR intersection structure.

A parallel pattern was observed within the secretome intersections (Fig. 6D). The MI-upregulated ATF6-KO downregulated secretome overlap included Apolipoprotein E (Apoe), Collagen Type IV Alpha 2 Chain (Col4a2), Laminin Subunit Alpha 4 (Lama4), Thrombospondin 2 (Thbs2), Csf1r, Trem1, Lcn2, Cxcl9, Cxcl14, and Cfp, together with additional matrix-associated, extracellular communication, and immune-recruitment genes (Fig. 6D). These data suggest that the CD4^+/TIP population contains a coordinated extracellular communication program that integrates matrix remodeling, immune signaling, and tissue-interaction features. In contrast, the PERK-KO and IRE1α-KO downregulated intersections showed more limited secretome overlap, further supporting predominant ATF6 alignment within the MI-upregulated CD4^+/TIP secretory and immune-interaction program [8,64,65].

To validate the MI-upregulated immune and secretome-associated CD4^+/TIP program, we assessed selected candidate genes by quantitative polymerase chain reaction (qPCR) in flow cytometry-sorted CD4^+/TIP cells isolated from sham and 1-week post-MI hearts (Fig. 6E). Consistent with the intersection analysis, MI-derived CD4^+/TIP cells showed significant increases in the repair-linked and matrix-associated genes Col4a2, Lama4, and Thbs2 (Fig. 6E1, E2, E5). Immune-signaling and innate-adjacent genes were also significantly elevated after MI, including Cd300a, Csf1r, Tlr13, Trem1, Treml4, C5ar1, Cr2, Cfp, Ptafr, and Cxcl9 (Fig. 6E3, E4, E6-E8, E10-E14). Lipocalin 2 (Lcn2) and Cxcl14 showed increased expression trends in MI-derived CD4^+/TIP cells, although these comparisons did not reach statistical significance (Fig. 6E9, E15). These qPCR data validate that the MI-upregulated CD4^+/TIP population is enriched for immune-signaling, complement-associated, chemokine-associated, and tissue-interaction genes.

In contrast to the MI-up immune and secretome program, the MI-downregulated CD4^+/TIP intersections identified a constrained ER proteostasis and stress-resolution module. This module included genes associated with ER protein handling, redox-linked proteostasis, and UPR signaling, including Heat Shock Protein Family A (Hsp70) Member 5 (Hspa5), SEC63 Protein Translocation Regulator (Sec63), Thioredoxin Domain Containing 11 (Txndc11), Endoplasmic Reticulum To Nucleus Signaling 1 (Ern1), Atf4, MAF BZIP Transcription Factor G (Mafg), and NFE2 Like BZIP Transcription Factor 1 (Nfe2l1) (Supp. Fig. 3A–C) [17,18,55,56,66,67]. To further validate this MI-down stress-resolution module, we assessed these genes by qPCR in flow-sorted CD4^+/TIP cells from sham and 1-week post-MI hearts (Supp. Fig. 3D). Consistent with the predicted suppression of ER proteostasis and redox-regulatory machinery, Sec63, Txndc11, Mafg, and Nfe2l1 were significantly decreased in MI-derived CD4^+/TIP cells compared with sham CD4^+/TIP cells (Supp. Fig. 3D2-D5). Hspa5, Ern1, and Atf4 also showed reduced expression trends in MI-derived CD4^+/TIP cells, although these comparisons did not reach statistical significance (Supp. Fig. 3D1, D6-D7). These findings provide qPCR support for the computationally identified MI-down CD4^+/TIP stress-resolution arm and reinforce the conclusion that CD4^+/TIP cells exhibit restrained ER protein-handling, proteostasis, and redox-amplifying regulatory programs after MI.

Collectively, Fig. 6 identifies a UPR-linked framework, in which the MI-specific CD4^+/TIP population couples an ATF6-aligned immune and secretome-associated program with suppression of ER proteostasis and stress-resolution genes. The qPCR validation in Fig. 6E confirms that flow cytometry-sorted CD4^+/TIP cells from 1-week post-MI hearts are enriched for repair-linked matrix genes, innate-adjacent receptors, complement-associated genes, and chemokine-associated outputs. In parallel, Supplementary Fig. 3D supports coordinated suppression of select ER proteostasis and redox-regulatory genes. Together, these data support a model in which CD4^+/TIP cells integrate injury-sensing and extracellular communication programs with restrained UPR-linked proteostasis, defining a stress-adaptive CD4⁺ T cell state within the early post-MI repair environment.

2.7. Candidate transcription factor and epi(genetic)factor genes define a focused regulatory profile of the CD4^+/TIP population at 1-week post-MI

To further refine the transcriptional identity of the CD4^+/TIP population identified in the preceding analyses, we prioritized a focused set of transcription factors and epi(genetic)factor candidates emerging from the CD4^+/TIP-associated Fig. 5 intersections (Fig. 7A). These candidates were selected to capture distinct regulatory features of the injury-polarized state, including stress-responsive transcriptional control, cardiogenic and vascular-associated developmental regulation, calcium- and stress-linked transcriptional regulation, nuclear receptor-associated plasticity, chromatin remodeling, cell-cycle or genome-integrity-associated signaling, and RNA fate control.

Fig. 7.

Myocardial infarction enriches a distinct regulatory profile within cardiac CD4^+/TIP cells.

(A) Functional summary of eight transcription factor and epifactor candidates prioritized from the CD4^+/TIP intersection analyses based on their associations with stress-responsive transcription, developmental regulation, nuclear receptor-linked plasticity, chromatin remodeling, genome maintenance, and RNA fate control.

(B.1–B.8) Candidate gene expression was measured by qPCR in flow cytometry-sorted CD4^+/TIP cells isolated from sham and 1-week post-MI hearts. Compared with sham, MI-derived CD4^+/TIP cells showed significantly increased expression of Creb5 (B.1), Gata5 (B.2), Mef2b (B.3), Nr2e1 (B.5), Smarcd3 (B.6), Aurkc (B.7), and Rbm24 (B.8). Rorb expression showed an increased trend that did not reach statistical significance (B4). Expression was normalized to S18, and groups were compared using Welch's t-test. Together, these findings define an MI-enriched CD4^+/TIP regulatory profile involving stress-responsive transcriptional control, nuclear receptor signaling, chromatin remodeling, and post-transcriptional regulation.

Within this candidate set, Creb5 was classified as a stress-responsive transcriptional regulator linked to adaptive and repair-associated gene programs within the CD4^+/TIP state (Fig. 7A). Gata5 was prioritized as a cardiac and vascular-associated developmental regulator, while Mef2b was included as a MEF2-family calcium- and stress-responsive transcriptional regulator. RAR Related Orphan Receptor B (Rorb) and Nuclear Receptor Subfamily 2 Group E Member 1 (Nr2e1) were identified as nuclear receptor-associated candidates linked to patterning, plasticity, and survival-associated transcriptional control. Smarcd3, also known as BAF60c, was prioritized as a SWI/SNF chromatin-remodeling subunit, Aurkc as a cell-cycle and genome-integrity-associated kinase, and Rbm24 as an RNA-binding and splicing regulator linked to post-transcriptional control of MI-associated stress-adaptive programs (Fig. 7A) [[46], [47], [48],[68], [69], [70], [71], [72]].

To validate whether these candidates were enriched within the CD4^+/TIP population after MI, CD4^+/TIP cells were isolated from sham and 1-week post-MI hearts using the flow cytometric strategy described above, and candidate gene expression was assessed by qPCR. Creb5 expression was significantly increased in MI-derived CD4^+/TIP cells compared with sham CD4^+/TIP cells (Fig. 7B1). Gata5 was also markedly increased following MI, supporting enrichment of a cardiogenic and repair-associated transcriptional cue within the CD4^+/TIP population (Fig. 7B2). Mef2b expression was significantly elevated in MI CD4^+/TIP cells, consistent with activation of a calcium- and stress-responsive regulatory program (Fig. 7B3). Rorb showed an increased trend in MI-derived CD4^+/TIP cells, although this comparison did not reach statistical significance (Fig. 7B4).

Additional CD4^+/TIP-associated regulatory candidates were also increased after MI. Nr2e1 expression was significantly elevated in MI CD4^+/TIP cells, supporting the presence of a nuclear receptor-linked plasticity and survival-associated regulatory program (Fig. 7B5). Smarcd3 was significantly increased, indicating enrichment of chromatin remodeling-associated regulatory structure within the post-MI CD4^+/TIP population (Fig. 7B6). Aurora Kinase C (Aurkc) was significantly upregulated following MI, consistent with engagement of cell-state, proliferative, or genome-maintenance-associated programs during injury-associated CD4⁺ T cell remodeling (Fig. 7B7). Rbm24 was also significantly increased in MI-derived CD4^+/TIP cells, linking this population to RNA-binding and post-transcriptional regulatory mechanisms (Fig. 7B8).

Collectively, these data identify a focused set of transcription factors and epi(genetic)factor candidates enriched within the CD4^+/TIP population at 1-week post-MI. These findings support the interpretation that the CD4^+/TIP population is not only defined by exclusion of canonical Th and Treg lineage markers, but also carries a positive regulatory profile associated with stress-responsive transcriptional control, chromatin remodeling, nuclear receptor-linked plasticity, and RNA fate regulation. The enrichment of these candidate genes provides a practical regulatory framework for profiling the injury-polarized CD4⁺ T cell state. It establishes a focused set of targets for downstream validation of CD4^+/TIP identity and functions after MI.

2.8. Thapsigargin (TG)-induced ER stress amplifies CD4^+/TIP candidate transcriptional regulators in post-MI cardiac CD4⁺ T cells

To determine whether the CD4⁺ T cell compartment containing the tissue injury-polarized (TIP) population exhibits altered transcriptional responsiveness to ER stress, CD4⁺ T cells were isolated from sham and 1-week post-MI hearts and cultured under control conditions or treated with thapsigargin (TG, 1.5 μM) for 24 or 48 h before qPCR analysis (Fig. 8A). Candidate transcriptional regulators associated with the CD4^+/TIP program were then assessed to determine whether injury-conditioned CD4⁺ T cells display a distinct response to TG-induced ER stress.

Fig. 8.

Thapsigargin-induced ER stress selectively amplifies CD4^+/TIP associated regulatory genes in post-MI cardiac CD4⁺ T cells.

(A) Experimental workflow. CD4⁺ T cells were isolated from sham and 1-week post-MI hearts and cultured under untreated control conditions or treated with thapsigargin (TG; 1.5 μM) for 24 or 48 h before qPCR analysis. Created in BioRender. Knight, C. (2026) https://BioRender.com/80xtl7c.

(B.1–B.8) Expression of the CD4^+/TIP-associated candidate regulators Aurkc (B.1), Mef2b (B.2), Rorb (B.3), Gata5 (B.4), Creb5 (B.5), Smarcd3 (B.6), Nr2e1 (B.7), and Rbm24 (B.8). MI-derived CD4⁺ T cells exhibited higher basal expression of these candidates than sham-derived cells and a more pronounced transcriptional response to TG. TG further increased Aurkc at 24 and 48 h; Mef2b, Rorb, Creb5, and Nr2e1 at 48 h; and Gata5 at 24 h. Smarcd3 increased selectively at 24 h, then attenuated at 48 h, while Rbm24 also exhibited an ER stress-responsive increase in MI-derived cells. Sham-derived CD4⁺ T cells maintained comparatively low expression and showed limited TG responsiveness. Expression was normalized to Gapdh. These findings indicate that prior myocardial injury establishes a CD4⁺ T-cell regulatory state that is selectively amplified by subsequent ER stress. Circle indicates the control group, square indicates the 24-h TG-treated group, and triangle indicates the 48-h TG-treated group.

Under basal control conditions, MI-derived cardiac CD4⁺ T cells showed increased expression of multiple CD^4+/TIP-associated candidate regulators compared with sham-derived CD4⁺ T cells. This pattern was observed across Aurkc, Mef2b, Rorb, Gata5, Creb5, Smarcd3, Nr2e1, and Rbm24, supporting the presence of an injury-associated transcriptional state within the post-MI CD4⁺ T cell compartment (Fig. 8B1-B8). In contrast, sham-derived CD4⁺ T cells showed low expression of these candidate regulators and did not exhibit a comparable TG-induced transcriptional response.

TG treatment further amplified expression of several candidate CD4^+/TIP regulators in MI-derived CD4⁺ T cells. Aurkc was elevated in post-MI CD4⁺ T cells and increased further following TG exposure at both 24 and 48 h (Fig. 8B1). Mef2b, Rorb, Creb5, Nr2e1, and Rbm24 showed a strong ER stress-responsive pattern, with expression increasing in MI-derived CD4⁺ T cells following TG treatment, most prominently at 48 h for Mef2b, Rorb, Creb5, and Nr2e1 (Fig. 8B2, B3, B5, B7, B8). Gata5 was robustly induced in MI-derived CD4⁺ T cells under basal conditions and remained highly responsive to TG treatment, with the strongest induction observed at 24 h (Fig. 8B4). Smarcd3 showed a more selective response, with increased expression following 24-h TG treatment and attenuation by 48 h (Fig. 8B6).

Collectively, these data demonstrate that post-MI cardiac CD4⁺ T cells exhibit a distinct ER stress-responsive transcriptional profile involving candidate CD4^+/TIP regulatory genes. The selective amplification of these regulators by TG in MI-derived CD4⁺ T cells indicates that the injury-conditioned CD4⁺ T cell compartment is not only transcriptionally distinct at baseline, but also differentially responsive to ER stress challenge. These findings provide functional qPCR support for the model that the CD4^+/TIP program is linked to ER stress adaptation during the early post-MI repair phase.

2.9. Injury-imprinted lineage programs define a previously missed MI injury polarization-specific CD4⁺ T cell subset

To highlight lineage-defining regulators of the injury-specific CD4^+/TIP compartment, we next prioritized a short list of standout transcriptional and epigenetic factors from Fig. 9 (Fig. 7). Rather than recapitulating canonical CD4⁺ helper or Treg lineage drivers, the CD4^+/TIP transcription factor genes converge on regulators linked to chromatin accessibility control, RNA fate decisions, and cardiac-associated regulatory programs, consistent with an MI-imprinted state distinct from canonical Th and Treg axes. Within this set, Gata5, Smarcd3, CREB5, Nuclear Receptor Subfamily 2 Group E Member 1 (Nr2e1), Rorb, and RNA Binding Motif Protein 24 (Rbm24) emerged as prominent candidates. Collectively, Fig. 7 defines candidate genes through which the MI-specific CD4^+/TIP population may be specified as an injury-adaptive state program involving cardiogenic transcriptional cues, chromatin remodeling, nuclear receptor-linked competency, and RNA fate control. These transcription factors provide a focused set of markers for characterizing the CD4^+/TIP population (Fig. 9).

Fig. 9.

Noncanonical lineage TFs appear in the MI-associated CD4^+/TIPprogram.

These highlighted genes are lineage-associated transcription factors in other cell types and developmental contexts, and they are not established CD4⁺ T cell lineage regulators. Created in BioRender. Knight, C. (2026) https://BioRender.com/c8a2bqa.

3. Discussion

Myocardial infarction (MI) triggers a sequential wound-healing process characterized by early inflammation, reparative extracellular matrix remodeling, neovascularization, and scar stabilization. While innate immune cells dominate the earliest inflammatory response, CD4⁺ T cells also influence infarct organization, immune resolution, and long-term remodeling after injury [[1], [2], [3], [4],73]. The current study identifies a prominent CD4⁺ T cell transcriptional program at 1-week post-MI that extends beyond canonical Th1, Th2, Th17, nTreg, and iTreg gene signatures and organizes into a distinct tissue injury-polarized CD4⁺ T cell state, termed CD4^+/TIP. This population is associated with repair-linked tissue interaction, innate-adjacent immune signaling, redox remodeling, and UPR-linked stress adaptation.

A central finding is that a substantial fraction of the 1-week post-MI CD4⁺ transcriptional response falls outside established lineage-enriched gene lists. Rather than reflecting only proportional shifts among known helper and regulatory lineages, the CD4^+/TIP fraction contains a coherent pathway structure centered on extracellular matrix organization, adhesion, vascular programs, developmental pathways, and tissue interaction. The corresponding downregulated CD4^+/TIP program indicates constrained engagement of RNA-handling, chromatin-regulatory, and protein catabolic pathways, suggesting that this injury-associated state combines acquisition of repair-linked programs with restraint of stress-amplifying infrastructure [22,26].

The addition of flow cytometric validation strengthens this interpretation. Using a negative-selection strategy within viable CD45⁺CD3⁺CD4⁺ cells, the CD4^+/TIP population was defined by lack of Tbet, Gata3, RORγt, and Foxp3 expression, excluding major Th1-, Th2-, Th17-, and Treg-associated lineage programs. The expansion of this flow-defined CD4^+/TIP compartment at 1-week post-MI demonstrates that the noncanonical population identified by transcriptomic analysis is not only represented as a gene module, but is also a responsive CD4⁺ T cell population after injury. This supports the conclusion that CD4^+/TIP cells represent an injury-associated CD4⁺ T cell state rather than a residual unassigned transcriptional fraction.

The redox and regulatory analyses further position the CD4^+/TIP population as a stress-adapted repair-associated state. nTreg and iTreg programs showed subset-specific stress and redox attenuation features consistent with preservation of regulatory function under post-MI conditions. The CD4^+/TIP program extended this logic beyond canonical regulatory lineages, integrating cardiogenic, matrix-remodeling, metabolic, developmental, and inflammatory-restraint features. In parallel, downregulation of genes linked to proteostasis recovery, antioxidant regulation, and mitochondrial metabolic entry supports a model in which CD4^+/TIP cells limit broader ER stress-linked and redox-amplifying pathways that could promote effector-like escalation [27,55,56,60].

The updated candidate-regulator validation provides additional support for a positive CD4^+/TIP identity. Flow cytometry-sorted CD4^+/TIP cells from 1-week post-MI hearts showed increased expression of candidate transcription factor and epifactor genes, including Creb5, Gata5, Mef2b, Nr2e1, Smarcd3, Aurkc, and Rbm24, with Rorb showing a trend toward increase. These findings indicate that CD4^+/TIP cells are not defined solely by exclusion of canonical lineage markers, but also carry a regulatory profile associated with stress-responsive transcriptional control, cardiogenic and repair-associated cues, chromatin remodeling, nuclear receptor-linked plasticity, genome maintenance, and RNA fate regulation [[46], [47], [48],[68], [69], [70], [71], [72]]. This focused candidate set provides a practical framework for profiling CD4^+/TIP identity and prioritizing future perturbation studies.

The UPR intersection analyses mechanistically sharpen this model. The CD4^+/TIP population showed dominant ATF6-aligned UPR structure, while its downregulated arm contained broad RNA- and protein-regulatory modules. qPCR validation of sorted CD4^+/TIP cells further supported this restrained proteostasis profile, with significant decreases in Sec63, Txndc11, Mafg, and Nfe2l1 and downward trends in Hspa5, Ern1, and Atf4. Together, these data suggest that CD4^+/TIP cells are not characterized by broad activation of ER stress machinery, but instead by selective restriction of ER protein-handling, proteostasis, and redox-regulatory outputs [27,28,[60], [61], [62],66,67,74].

In parallel, the MI-upregulated CD4^+/TIP program contained a strong innate-adjacent immune and secretome-associated structure. qPCR validation confirmed increased expression of matrix-associated genes, including Col4a2, Lama4, and Thbs2, as well as immune-signaling and innate-adjacent genes, including Cd300a, Csf1r, Tlr13, Trem1, Treml4, C5ar1, Cr2, Cfp, Ptafr, and Cxcl9. These findings support the interpretation that CD4^+/TIP cells are equipped for injury sensing, complement-responsive signaling, inflammatory cell recruitment, and extracellular communication during early repair [8,64,65,[75], [76], [77], [78], [79]]. The recovery of C5ar1 and Cr2 further suggests complement-responsive signaling within this population [[80], [81], [82]]. Thus, “innate-adjacent” in this study refers to a CD4⁺ T cell state enriched in innate-sensing, complement-associated, chemokine-associated, and tissue-interaction features within a tissue injury-polarized (TIP) niche.

The thapsigargin response data provide functional support for the stress-adaptive framework. MI-derived cardiac CD4⁺ T cells showed elevated baseline expression of multiple CD4^+/TIP-associated candidate regulators compared with sham-derived CD4⁺ T cells. They displayed selective amplification of these regulators following TG-induced ER stress. This response suggests that the injury-conditioned CD4⁺ T cell compartment is not only transcriptionally distinct at baseline, but also differentially responsive to ER stress challenge. Importantly, these data support selective stress-responsive regulatory capacity rather than uniform activation of all ER stress programs [27,28,67,74].

Several limitations define the scope of interpretation. Permanent LAD ligation provided a sustained and reproducible ischemic injury model suitable for examining CD4⁺ T-cell responses during infarct healing. However, this model does not incorporate the restoration of blood flow or the associated reperfusion response. Future studies using complementary ischemia-reperfusion models will be valuable for determining whether reperfusion-associated signals similarly induce the CD4^+/TIP population or alter its transcriptional and functional characteristics. In addition, transcriptomic profiling was restricted to a single time point, 1 week after MI, to examine CD4⁺ T-cell reprogramming during the transition from inflammation to active tissue repair. Consequently, the present study does not capture the temporal evolution of the post-infarct tissue environment, its changing wound-healing processes, or the corresponding development and persistence of CD4^+/TIP and other CD4⁺ T-cell states. Bulk RNA sequencing cannot fully exclude technical or biological mixture effects, particularly for transcripts frequently associated with innate immune lineages. The flow cytometric expansion of CD4^+/TIP cells and qPCR validation in sorted CD4^+/TIP populations strengthen CD4⁺ T-cell-intrinsic attribution, but single-cell resolution and protein-level validation of candidate receptors and regulatory markers would further refine this population. Establishing causality will require perturbing UPR pathways and nominated regulatory drivers in CD4⁺ T cells, followed by assessment of CD4^+/TIP identity, innate-adjacent signaling, and repair-linked outcomes after MI.

In summary, these findings support a novel model in which the 1-week post-MI environment fosters a prominent CD4^+/TIP state that excludes major Th1-, Th2-, Th17-, and Treg-associated lineage programs and is characterized by repair-linked tissue interaction, innate-adjacent immune and secretome programs, selective redox remodeling, restrained UPR-linked proteostasis, and stress-responsive regulatory capacity. By integrating transcriptomic discovery, flow cytometric population validation, flow cytometry-sorted-cell qPCR, and Thapsigargin (TG)-induced ER stress challenge, this study provides a framework for understanding how tissue injury-polarized CD4⁺ T cells may support cardiac repair outside traditional helper and regulatory lineage classifications.

4. Materials and methods

4.1. Animal MI models

All animal procedures were approved by the Institutional Animal Care and Use Committee of the Lewis Katz School of Medicine at Temple University (protocol #5060) and conducted in accordance with AAALAC guidelines. Mice were housed under controlled environmental conditions (22–24 °C, 12 h light/dark cycle) with unrestricted access to food and water. C57BL/6J mice were obtained from The Jackson Laboratory (000664, Bar Harbor, ME, USA). Permanent myocardial infarction (MI) was induced by ligation of the left anterior descending (LAD) coronary artery as previously described [83,84]. Takedowns of animals were performed at 1-week post-MI for cell sorting. Adult male mice were used to minimize potential variability related to sex differences.

4.2. Single-cell suspension Collection

Hearts were collected 1-week post-MI for CD4⁺ T cell isolation under 3% isoflurane. Tissue was diced using surgical scissors and digested in a buffer containing 60 U/mL Hyaluronidase (#LS005474, Worthington), 450 U/mL Collagenase Type I (#LS004197, Worthington), 125 U/mL Collagenase Type XI (#C7657, Sigma-Aldrich), 60 U/mL Deoxyribonuclease I (#LS002060, Worthington), and 20 mM HEPES at 37 °C for 30 min. The digested suspension was filtered through a 40 μm mesh and then spun at 50g for 5 min to remove the cardiomyocyte fraction. The supernatant was transferred and treated with red blood cell lysis buffer (#00433357, Invitrogen) for 10 min. Cells were resuspended in sorting buffer (#20144, STEMCELL Technologies) for flow cytometry staining.

4.3. CD4⁺ T-cell sorting

Flow cytometry was performed to sort CD4⁺ T cells for RNA sequencing. The collected single-cell suspensions were stained with Live/Dead Fixable Green for 30 min (#L23101, Thermo Fisher Scientific). Cells were then stained with APC-eFluor780-conjugated anti-CD45 (#47045180, eBioscience), PE-conjugated anti-CD3 (#100205, BioLegend), and PerCP-conjugated anti-CD4 (#100431, BioLegend) for 20 min. Gating was performed to sort living CD45/CD3/CD4⁺ cells using a FACSAria II (BD Biosciences).

4.4. CD4⁺ T cell subset sorting

Hearts and spleens were collected from sham and myocardial infarction (MI) mice at 1-week post-injury for CD4⁺ T cell subset isolation. Single-cell suspensions were generated from cardiac and splenic tissues and prepared for flow cytometric staining. Cells were first stained with a fixable live/dead viability dye to exclude nonviable cells from downstream analysis. Following viability staining, cells were surface stained to identify leukocyte and T cell populations, including CD45, CD3, and CD4. After surface staining, cells were fixed and permeabilized for intracellular transcription factor staining. Intracellular staining was performed for canonical CD4⁺ T cell lineage-associated markers, including Tbet, Gata3, RORγt, and Foxp3, corresponding to Th1-, Th2-, Th17-, and Treg-associated programs, respectively. Flow cytometric sorting was performed using a BD FACSDiscover S8. A negative-selection gating strategy was used to isolate the CD4^+/TIP population. Briefly, viable CD45⁺CD3⁺CD4⁺ cells were first identified, and canonical CD4⁺ T cell subsets were separated based on positive expression of Tbet, Gata3, RORγt, or Foxp3. The CD4^+/TIP population was defined as the viable CD45⁺CD3⁺CD4⁺ fraction lacking expression of Tbet, Gata3, RORγt, and Foxp3. This strategy enabled isolation of a noncanonical CD4⁺ T cell population outside the major Th1, Th2, Th17, and Treg-associated lineage markers. Sorted CD4⁺ T cell subsets, including the CD4^+/TIP population, were collected and frozen for downstream validation experiments.

4.5. RNA sequencing analysis

RNA was extracted from the sorted CD4⁺ T cells using an RNeasy Kit (#74106, Qiagen). RNA library preparation and sequencing were performed by Azenta US Inc. Ultra-low-input RNA sequencing libraries were generated using SMART-Seq HT (#634455, Takara) for full-length cDNA synthesis and amplification. Sequencing libraries were prepared using the Nextera XT DNA Library Preparation kit (#FC1311024, Illumina) and quantified by quantitative PCR (KAPA Biosystems). Libraries were multiplexed and sequenced using a 2 × 150 bp paired-end configuration on an Illumina HiSeq 4000 platform (Illumina). Raw base call files were converted to FASTQ format and demultiplexed using bcl2fastq version 2.17. Because cardiac T cells are virtually absent under steady-state conditions, healthy CD4⁺ control groups (sham) were represented using a naïve CD4⁺ T cell dataset [26]. Transcript quantification was performed against the mouse reference transcriptome (GRCm39). K-means clustering (top 4000 genes) and differential gene expression analysis were conducted using DESeq2, using the integrated Differential Expression & Pathway (iDEP) analysis from South Dakota State University [85]. Genes with <0.5 counts per million were excluded. Differentially expressed genes were defined as those with an absolute Log2(FC) > 1.5 and a false discovery rate <0.05. Metascape was used to perform pathway enrichment analysis of the upregulated and downregulated differentially expressed genes to identify significantly overrepresented biological processes and pathways [86]. Volcano plots were generated from unfiltered differentially expressed gene sets to highlight genes with Log2(FC) > 1.5 using GraphPad Prism version 11.0.0 for Windows (GraphPad Software, San Diego, California, USA).

4.6. CD4⁺ subset gene list preparation and analysis

Publicly available feature-count data were used to define gene expression lists specific to each subset (Th1, Th2, Th17, nTreg, and iTreg) relative to naïve CD4⁺ T cells [26]. Differentially expressed gene lists (p-adj. <05, Log2(FC) > 1.5) were generated as described above. These gene lists for each subset were intersected using InteractiVenn to capture genes uniquely upregulated in each subset and repressed in that subset [87]. These gene lists provided genes distinct to each subset. Cross-referencing these distinct DEGs for each subset against the CD4⁺ MI DEG list identified genes up- and downregulated specifically in each subset at 1-week post-MI. Distinct subset genes post-MI were presented as pie charts, showing the total number of genes and their percentage of the total gene population relative to the original CD4⁺ T cell MI differentially expressed gene dataset. Pathway enrichment was performed as described above. All pie charts and population analyses were performed using GraphPad Prism version 11.0.0 for Windows (GraphPad Software, San Diego, California, USA).

4.7. Gene list architecture and analysis

Distinct Sunset post-MI gene lists, generated as described above, were intersected with a variety of publicly available datasets from GEOR2 (NIH-NCBI-Geo Datasets, https://www.ncbi.nlm.nih.gov/gds), the Human Protein Atlas (HPA, https://www.proteinatlas.org/), Gene Set Enrichment Analysis (GSEA, https://www.gsea-msigdb.org/gsea/index.jsp), and InnateDB (https://www.innatedb.com/) using InteractiVenn [[88], [89], [90], [91]]. Cross-sectional gene lists underwent pathway enrichment analysis as described above and were prepared in tables corresponding to their CD4⁺ subset alignment. ER stress KO datasets were generated using GEOR2 for comparison from (ATF4-KO [GSE10470], ATF6-KO [NCBI-Geo Datasets ID: GSE49646], PERK-KO [GSE29929], IRE1α-KO [GSE130952], and thapsigargin treatment [GSE200626]). A list of 290 lineage-defining transcription factors (290 TFs) across cell types was used to interrogate expression across CD4⁺ T Cell subsets [92]. Transcription Factors obtained from the HPA were also used for comparison. A curated database of epigenetic factors and complexes (Epifactors, https://epifactors.autosome.org/) was also used for comparison [90]. Reactive oxygen species (ROS) gene lists were used to compare genes that are pro-mitochondrial (mito) ROS, Anti-mito ROS, Pro-Cellular ROS, and Anti-Cellular ROS [93]. A separate list of ROS upregulated genes was also used for comparison (GSEA). Additionally, the ER Stress KO and distinct CD4⁺ subset post-MI overlapping gene lists were cross-referenced with the transcription and epifactor gene lists described above to identify ER-stressed genes aligned within these lists during MI. The same ER Stress KO and distinct CD4⁺ subset post-MI overlapping gene lists were also used to explore immune signaling genes associated with the Plasma Membrane, Cytokines, Chemokines, and clusters of differentiation (CD) Markers on the HPA, as well as innate immunity genes from InnateDB, and immune response genes from Gene Ontology terms (https://geneontology.org/docs/ontology-documentation/, GO:20240314_152727). Following the same logic, comparisons were made to explore a canonical secretome gene from the HPA and exosome-related secretome genes from Exocarta [94], as well as non-canonical secretome gene lists under caspase-1 and caspase-4 activation [95,96].

4.8. Quantitative real-time PCR

Total RNA was extracted using the RNeasy Kit (#74106, Qiagen, Hilden, Germany). Reverse transcription was performed using the High-Capacity cDNA Reverse Transcription Kit (#4368814, Applied Biosystems, Foster City, CA, USA). Quantitative PCR was conducted using a CFX96 Real-Time PCR System (Bio-Rad, Hercules, CA, USA). Fixed sorted cell pellets were processed for RNA recovery using a modified reverse-crosslinking approach adapted from previously described methods for transcriptomic profiling of fixed and permeabilized sorted cells [97,98]. Briefly, sorted cells were pelleted and resuspended in reverse-crosslinking lysis buffer containing 100 mM NaCl, 10 mM Tris-HCl pH 8.0, 1 mM EDTA, 0.5% SDS, and 0.5 mg/mL Proteinase K. Samples were incubated at 50 °C for 60 min to promote protein digestion and reversal of fixation-associated RNA-protein crosslinking. Following digestion, QIAzol was added directly to the lysate, and RNA was isolated using chloroform phase separation followed by silica-column purification according to the manufacturer's RNA cleanup workflow. Relative gene expression was calculated using the ΔΔCt method. Primer sequences are listed in Table S1.

4.9. Thapsigargin-induced ER stress assay

To assess ER stress responsiveness in cardiac CD4⁺ T cells, CD4⁺ T cells were isolated from sham heart controls and 1-week post-myocardial infarction (MI) hearts as described above. Following isolation, cells were cultured in complete culture medium (#30-2001, ATCC) and assigned to one of six experimental conditions: sham control without thapsigargin (TG), sham treated with TG for 24 h, sham treated with TG for 48 h, MI control without TG, MI treated with TG for 24 h, and MI treated with TG for 48 h. TG was used at a final concentration of 1.5 μM to induce ER stress. Following treatment, cells were collected for RNA isolation and quantitative real-time PCR analysis. Expression of candidate CD4^+/TIP-associated regulatory genes was assessed, including Aurkc, Mef2b, Rorb, Gata5, Creb5, Smarcd3, Nr2e1, and Rbm24. Gene expression was normalized to Gapdh, and relative expression was calculated using the ΔΔCt method [99]. Expression patterns were compared across sham and MI-derived CD4⁺ T cells under control, 24-h TG, and 48-h TG conditions to determine whether the injury-conditioned CD4⁺ T cell compartment exhibited altered transcriptional responsiveness to TG-induced ER stress.

CRediT authorship contribution statement

Thomas WC. Knight: Writing – review & editing, Writing – original draft, Visualization, Methodology, Formal analysis, Conceptualization. Fatma Saaoud: Data curation. Ngefor Asangwe: Writing – review & editing. Ying Shao: Data curation. Iman Khan: Formal analysis. Hajime Kubo: Writing – review & editing. Mohsin Khan: Writing – review & editing. Hong Wang: Writing – review & editing. Raj Kishore: Writing – review & editing, Supervision, Resources. Xiaofeng Yang: Writing – review & editing, Resources, Conceptualization. Sadia Mohsin: Writing – review & editing, Resources, Conceptualization, Funding acquisition, Supervision, Project administration, Validation.

Ethics declaration

This study was conducted in accordance with the following guidelines for animal welfare and/or reporting: AAALAC. This study was approved by the Animal Care and Use Committee of the Lewis Katz School of Medicine at Temple University. (Approval No. Protocol #5060)

Declaration of generative AI

During the preparation of this work, the author(s) used Grammarly to check text for grammar, spelling, punctuation errors, and to improve text clarity. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article.

Declaration of competing interest

Xiaofeng Yang, MD, PhD, is a member of the Editorial Board of Redox Biology, and Hong Wang, MD, PhD, EMBA, serves as an editor for Redox Biology. The other authors declare no competing interests.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.