The Infarcted Myocardium Calls for T cell help to Reg-ulate Repair

Ischemic heart disease affects over 120 million people worldwide. Cell death at the injury site triggers an immune and fibrotic response that work in tandem to form an indispensable scar for cardiac repair. Immune cell mobilization to the site of injury is necessary for repair, but the cardiac immune landscape changes after the scar formation and these alterations end up negatively impacting cardiac contractility in chronic ischemic heart disease¹. Understanding novel immunological mechanisms that interplay during the coordinated cardiac repair response is of utmost importance to discover new immunomodulatory therapeutics for ischemic heart disease.

In response to ischemia, innate immune cells capture and present antigens to CD4+ T cells via the major histocompatibility complex II (MHC-II). T cell antigen recognition through the T cell receptor (TCR) results in clonal expansion of antigen specific T cell subsets with defined effector functions dictated by the cytokine environment. The establishment of T helper (Th) and T regulatory (T reg) lineages is characterized by the expression of the signature transcription factors Tbx21 (Th1 cells), Gata3 (Th2 cells), Rorc (Th17 cells) and Foxp3 (Treg). Treg cells comprise two main subsets – thymus-derived (tTreg) and peripherally induced (iTreg) cells. Compared to tTreg, iTreg and Th17 cells exhibit more lineage plasticity in that they can interconvert under certain conditions^{2, 3}. In mouse models of myocardial infarction (MI), CD4+ T cells gradually infiltrate the heart during the first week after MI. Studies using CD4+ T-cell-deficient mice concluded that CD4+ T cells are necessary for cardiac repair⁴. More specifically, Treg depletion prior or at the time of ischemia was detrimental for survival to MI⁵, and Treg were found to accelerate the resolution of inflammation and scar formation post MI⁶. However, studies depleting Treg later post MI showed improved systolic function, demonstrating for the first time that Treg acquired a pro-inflammatory phenotype and that Treg plasticity took place in chronic ischemic heart failure⁷. Myocardial T cells, including Treg have also been detected in autopsy samples of patients post MI⁸, positioning CD4+ T cell subsets at the epicenter of cardiac repair.

In this issue of Circulation Research, Delgobo et al⁹ hypothesize that the local milieu of the infarcted myocardium influences the differentiation of antigen-specific T-helper cells, which in turn contribute to myocardial repair. The authors use scRNA-seq to characterize the phenotypic landscape of adoptively transferred myosin heavy chain alpha (MYHCA)-specific T-cell receptor (TCR) transgenic cells (TCR-M cells) in DO11.10 recipient mice prior to the induction of MI. This model system helped the authors identify in previous work that MYHCA is a dominant antigen triggering post MI CD4+ T cell activation¹⁰. The authors find that transferred TCR-M cells exhibit a predominantly iTreg phenotype in both the mediastinal lymph nodes (MedLN) and the infarcted heart. They find that TCR-M Treg in the infarcted host consist of two main subsets – a pro-fibrotic subset featuring Tgfb1 expression and a subset with characteristics of highly suppressive effector Treg. To complement and overcome the limitation of using one single transgenic TCR-M model, the authors generated a novel tool, MYHCA- MHC-II tetramers, to purify naturally-occurring endogenous MYHCA specific CD4+ T cells and performed additional scRNA seq. They compare endogenous tetramer+, and TCR-M T cells present in the ischemic heart, and the phenotype of endogenous cells in the presence and absence of the exogenous (TCR-M) T cells. With this elegant approach, the authors first demonstrate that the majority of naturally occurring MYHCA-specific, tetramer+ cells is in the MedLN/s and the heart of MI but not control mice. These share similarities with TCR-M cells, such as the enrichment within the Icos+ Treg cluster, suggestive of activated regulatory T-cells, increased expression of the iTreg score, Tgfb1 and the immune check point inhibitors Pdcd1, Ctla4 and Tigit. These data support that naturally occurring MYHCA specific T cells acquire a regulatory phenotype in the heart, similar to TCR-M cells. It is important to note that some endogenous cardiac tetramer+ cells additionally showed transcripts associated to pro-inflammatory responses such as Cxcr6 and Il18r1, that were not observed in TCR-M cells. The expression of Cxcr6 and Il18r1 marks terminally differentiated pathogenic Th17 cells, raising the possibility that endogenous MYHCA-specific T-cells contain a pathogenic Th17 subset refractory to Treg induction.

Using Flow cytometry, the authors show that adoptively transferred cardiac TCR-M cells preferentially express latency-associated peptide (LAP), a surrogate marker for TGF-β1 production, suppress endogenous Th17 cell responses and increase the Treg frequency in the heart. Interestingly, when TCR-M Th1, Th17, or Treg cells were pre-differentiated in vitro and then transferred into DO11.10 mice before MI induction, a substantial fraction of Th17 cells converted to Treg. Because endogenous cells in the DO11.10 recipients are restricted to chicken OVA and cannot respond to cardiac antigens, this further supports the notion that the infarcted myocardial milieu favors TCR-M Treg differentiation. The authors further showed that the MI mice which received pre-differentiated Th1 and Th17 TCR-M cells had exacerbated myocardial inflammation, whereas those that received Treg suppressed inflammation. The observation that some TCR-M Th17 cells converted to Treg, and the inverse correlation with cardiac inflammation support Th17 towards Treg plasticity in the infarcted heart associated with decreased myocardial inflammation. While antigen restricted mice are widely used tools to track antigen specific T cell immune responses, one caveat is that they may not fully represent the endogenous response to MI. The T cells restriction to OVA_323–339 peptide in the DO11.10 recipient mice may reduce the potential clonal competition between endogenous MYHCA-specific and transferred TCR-M cells, given that DO11.10 mice lack endogenous MYHCA-specific T cells and thus, a potentially pro-inflammatory milieu that results in a more favorable setting for Treg induction and Th17 to Treg conversion. Studies using mice with a normal polyclonal T cell compartment, such as the clever ones used by the authors to isolate naturally occurring endogenous MYHCA-specific CD4+ cells purified with tetramers could further assess this T cell subset conversion. In this line of thought, a recent study demonstrated in C57/BL6 mice that T conventional T cells make a small contribution to the ischemic heart Treg pool¹¹. While mouse strain differences may be taken into account, one cannot discard a potential skewed response in the DO.10.11 mice.

The specific signals that drive the differentiation of MYHCA-specific T cells into Treg in the infarcted myocardium remain to be uncovered and are an exciting area for future investigation. Environmental signals such as TGF-β, retinoic acid, and short-chain fatty acids induce Foxp3 expression in iTreg and contribute to immune tolerance to commensal bacteria, dietary antigens and the fetus^{12, 13}. Whether these, or novel cardiac specific signals mediate MYHCA-specific iTreg differentiation in the infarcted heart remains to be determined. The long term lineage stability of the iTreg described in this study is also an open question for future consideration. Stable Foxp3 expression is required for Treg lineage stability and is maintained by the methylation status of the conserved non-coding sequence 2 (CNS2) in the Foxp3 gene. CNS2 demethylation during iTreg differentiation is less efficient than in thymic Treg, resulting in lineage instability^{14, 15}. It is possible that the iTreg-promoting signals in the healing myocardium are transient and that the “repair iTreg” generated early after MI may lose their Treg identity and become dysfunctional, as demonstrated in experimental chronic ischemic heart failure using C57/BL6 mice⁷. Future studies that examine the CNS2 demethylation status of TCR-M iTreg, employ adoptive transfer of TCR-M iTreg into mice at later time points post-MI, or perform genetic fate-mapping of endogenous MYHCA-specific CD4+ T cells will likely provide novel insights to modify Treg stability. Perhaps through demethylation, MHCA-iTreg stability is maintained and prevents the pathogenic pro-inflammatory actions of T cells in the myocardium after cardiac repair has taken place. Lastly, as the majority of acute infarctions receive reperfusion, it will be interesting to tease out these exciting mechanism reported by the authors in ischemia/reperfusion.

Delgobo and colleagues convincing show that T cell plasticity is modulated by the ischemic myocardial environment. They remind us of the importance of investigating the active cardio-immune crosstalk necessary for antigen-specific T cell responses in the remodeling heart, and set the stage for new advances in immunomodulation of ischemic heart disease.