Immune cells and Immunotherapy for Cardiac Injury and Repair

Abstract

Cardiac injury remains a major cause of morbidity and mortality worldwide. Despite significant advances, a full understanding of why the heart fails to fully recover function after acute injury, and why progressive heart failure frequently ensues, remains elusive. No therapeutics, short of heart transplantation, have emerged to reliably halt or reverse the inexorable progression of heart failure in the majority of patients once it has become clinically evident. To date, most pharmacologic interventions have focused on modifying hemodynamics (reducing afterload, controlling blood pressure and blood volume) or on modifying cardiac myocyte function. However, important contributions of the immune system to normal cardiac function and the response to injury have recently emerged as exciting areas of investigation. Therapeutic interventions aimed at harnessing the power of immune cells hold promise for new treatment avenues for cardiac disease. Here, we review the immune response to heart injury, its contribution to cardiac fibrosis, and the potential of immune modifying therapies to affect cardiac repair.

Introduction

The body responds to various cardiac injuries via complex acute and chronic adaptive processes to maintain pump function. Central to this process is inflammation and immune cell signaling. For example, in the setting of acute ischemic injury, immune-mediated activation of cardiac fibrosis is necessary to avoid catastrophic myocardial rupture. Activation of resident interstitial fibroblasts and recruitment of those derived from the epicardium result in scar formation that maintains chamber integrity. However, in many cases, the initially beneficial fibroblast response becomes mal-adaptive, resulting in excess extracellular matrix (ECM). This over-accumulation of ECM both stiffens the myocardium and negatively alters the cardiomyocyte niche resulting in progressive deterioration of cardiac function. While fibroblast activation has been well studied, the multifaceted role of immune signaling and its effects on acute and chronic fibrosis have only recently gained widespread traction in the cardiovascular research community. The emerging understanding is poised to revolutionize the treatment of myocardial diseases through targeted immune modulators. Furthermore, advances in T cell engineering offer exciting future directions for development of novel therapeutics.

Immune response to cardiac injury

The immune response to cardiac injury is complex (see figure 1) and remains only partially understood. Differences in the types of injury, host and environmental factors all modulate the immune response. The potent sterile inflammation of the injured heart depends on a multitude of signals. Damage associated molecular patterns (DAMPs) released by dying cells, the cytosolic DNA sensing pathway (cGAS-STING), and cardiomyocyte NLRP3 inflammasome (activated through CaMKIIδ) are each responsible for initiating the pro-inflammatory environment during myocardial injury.^1–5 Furthermore, the many different cell types that make up the immune system interact with one another and with resident cardiac fibroblasts, endothelial and myocardial cells. Some examples of the beneficial and injurious immune cell sub-type responses to injury are summarized below.

Figure 1. Immune cell involvement after cardiac injury.

Numerous immune components are involved in the injured myocardium. The main components are represented here. Stressed cells initiate inflammatory secretory programs through the cytosolic DNA sensing cGAS-STING pathway. Dying cells release damage associated patterns (DAMPs) which act as potent signaling molecules. C-reactive protein (CRP), IgM, and other circulating factors permeate the injured heart. Tissue-resident macrophages proliferate and additional CCR2⁺ monocytes are recruited from circulation. Macrophages polarize between M1-like (inflammatory) and M2-like (reparative) phenotypes. Dendritic cells are potent activators of recruited T cells, which are major immune infiltrates. CD8⁺ cytotoxic and several effector T (T_eff) cells signal with M1 macrophages and many other pro-inflammatory pathways. In contrast some T_eff and regulatory T cells (T_reg) infiltrate the heart and stimulate M2-like macrophages, ECM remodeling and overall heart repair. B cells produce many chemokines, cytokines, and autoantibodies. Natural killer (NK) cells are recruited and play a mainly reparative role. Many neutrophils are recruited where they stimulate M1-like macrophages and secrete pro-inflammatory cytokines. Eosinophils also infiltrate the heart and help stimulate tissue recovery. One of the major downstream targets of this inflammatory milieu are cardiac fibroblasts. Tissue-resident fibroblasts proliferate and activate and produce detrimental fibrosis.

One of the first immune cell types to respond to injury is tissue-resident macrophages. These self-renewing cells are established early in the embryonic myocardium from the yolk sac and fetal liver.⁶ Macrophages are critical for myocardial development, including the proper formation of the cardiac lymphatic system and they facilitate proper electrical conduction.^7–9 Recently, a fascinating role for macrophages has been described in the maintenance of cardiac energy homeostasis, mediated by disposal of spent/defective mitochondria that are extruded by cardiac myocytes and engulfed by resident macrophages.¹⁰ This process utilizes components of the autophagy pathway and is enhanced in times of myocardial stress. Depletion of cardiac macrophages disrupts this process and results in activation of the inflammasome and cardiac dysfunction.¹⁰ This observation may help to explain the long-appreciated role of autophagy in metabolic dysfunction in heart failure.¹¹

In the setting of injury, activated fibroblasts (discussed below), cardiomyocytes, and various other immune cells release cytokines to polarize existing macrophages and chemokines to recruit additional monocytes from the circulation.^8,12,13 Tissue-resident CCR2⁻ macrophages are activated and proliferate.¹⁴ These cells seem to serve several key protective/reparative roles in the post-injury heart, including promoting neo-vascularization.^6,8,12–14 In contrast, the potent inflammatory signaling environment (including IL-1β) setup by numerous cell types recruits many infiltrating immune cells from the circulation.¹⁵ Infiltrating CCR2⁺ monocytes mature into macrophages and may outnumber those arising from tissue-resident cells.⁸ Pro-inflammatory Ly6c^high macrophages produce a potent cocktail of cytokines including (but not limited to) IL-1β, IL-6, and TNFα.^8,16 Historically, the polarization of macrophages has been referred to as M1-like (detrimental) and M2-like (reparative) phenotypes without consideration of their tissue source. Using this terminology, the incoming CCR2⁺ macrophages are thought to mainly polarize into the M1-like phenotype, especially in the early phases of injury-response. Once there, pro-inflammatory macrophages secrete many cytokines and play apoptotic, cytolytic and proteolytic roles.^8,13 The timing and ratio between these two phenotypes of macrophages is critical.¹⁷ Tipping the balance towards M2-like macrophages is sufficient to stimulate heart repair and improve function following injury in animal models.^8,18,19 Further breakdown of the individual contributions of subgroups of macrophages in cardiac disease is the subject of intense research.²⁰ Similarly, the mechanism underlying atherosclerotic disease is heavily dependent on infiltration of clonal monocytes.²¹ Clonal hematopoiesis of indeterminant potential (CHIP) has also been implicated in ischemic heart failure.²² Further refinement of our understanding of macrophage subtypes responding to various myocardial injuries, along with the development of interventions capable of targeting specific subtypes, may provide attractive therapeutic avenues for development.

Alongside the recruitment of monocytes, the inflammatory environment of the injured heart also attracts significant neutrophil infiltrates. Neutrophils participate in various aspects of myocardial injury response. These include amplifying damaging pro-inflammatory signals, production of reactive-oxygen species, and secretion of proteolytic enzymes to remodel the ECM.^23–25 However, experimental depletion of all neutrophils before cardiac injury negatively impacts the ability of the heart to recover.²⁶ Mechanistically, neutrophils promote cardiac recovery by polarizing macrophages towards a reparative phenotype.²⁶ Neutrophil recruitment and participation in the post-infarct heart have recently been comprehensively reviewed by Daseke, et al.²⁷ Targeting subsets of pro-inflammatory neutrophils might therefore offer opportunities for therapy.

Eosinophils are also recruited to the injured heart. Increased levels of circulating eosinophils have been associated with risk of coronary artery disease and, in rare examples, excessive myocardial infiltration can result in eosinophilic myocarditis.²⁸ However, recent research also suggests a protective role for eosinophils in the heart following acute cardiac injury. Depletion of eosinophils exacerbates cardiac dysfunction and fibrosis after infarction.^29,30 Eosinophil-specific IL-4 and mEar1 expression has been demonstrated to block H₂O₂ and hypoxia induced cardiomyocyte death and dampen fibroblast activation.³⁰ The full role of eosinophils in the injured myocardium remains to be elucidated.

Tissue-resident mast cells proliferate in response to numerous cardiac injuries. Degranulated mast cells release many pro-inflammatory signaling molecules (including TNF-α and histamines) and proteases (including chymase, renin, and tryptase) involved in fibroblast activation.^31,32 Mast cells are mainly pro-fibrotic in the injured heart and interestingly, this negative impact is somewhat abrogated in the presence of estrogen.³³ In contrast, natural killer cells appear to play multiple, complex roles to protect the injured myocardium. Natural killer cells limit damaging innate immune cell infiltration and activity through restraining chemokine production as well as secreting INF-γ, perforin and other anti-inflammatory chemokines.³⁴ Additionally, natural killer cells have been shown to directly inhibit activated fibroblasts from over-producing collagen, limit cardiomyocyte apoptosis and stimulate neovascularization.^35,36

The adaptive immune system is also integral to the pro-inflammatory environment of the injured heart. It plays multifaceted roles following acute myocardial infarction that have only recently begun to be fully elucidated.³⁷ In chronic pressure-overload, dendritic cells (professional antigen-presenting cells) accumulate potent γ-ketoaldyhydes, which activate a pro-inflammatory program of reactive oxygen species, IL-1β, IL-6, and IL-23 secretion.³⁸ Dendritic cells also increase expression of T cell co-stimulatory proteins. Ultimately, dendritic cells in the hypertensive heart serve to promote T cell proliferation (especially CD8+ subsets) and promote polarization into a pro-inflammatory phenotype.^38–40 Isolated dendritic cells from hypertensive mice transplanted into naïve mice prime the recipient towards the ill-effects of high blood pressure.³⁸ Post-myocardial infarction, cross-priming dendritic cells are partially responsible for sustained myocardial injury and decreased left ventricular function.³⁹ On the other hand, there is some evidence that injection of dendritic cells primed in injured hearts helps to coordinate a beneficial, pro-reparative myocardium through coordination of regulatory T cells and shifting macrophage polarization to the M2-like phenotype.⁴¹ More work is required to dissect and understand the potentially beneficial and deleterious functions of dendritic cells in the injured heart.

Both B and T lymphocytes also play important roles in cardiac homeostasis and response to injury. For example, mice lacking PD-1 (essential for B cell differentiation) spontaneously develop dilated cardiomyopathy with increased levels of cardiomyocyte-specific IgG autoantibodies.⁴² Interestingly, patients with end-stage heart failure can often harbor multiple anti-cardiomyocyte antibodies as recently reviewed in detail by García-Rivas, et al.⁴³ In the setting of acute decompensation and ischemia-reperfusion, B cells are responsible for production of potent apoptotic signals and autoantibody-activation of the complement cascade.^43,44 B cell depletion in hypertensive mice results in lower fibrotic burden.⁴⁵ Furthermore, B cells in the injured heart signal to CD4+ T cells through MHC-II and to monocytes.^43,46 This interplay is dependent on the IL-1 pathway and MyD88, with MyD88 knockout mice lacking the robust fibrosis seen in induced myocarditis in wild-type animals.⁴⁷ IL-10, produced primarily by B cells, has emerged as a critical anti-inflammatory (protective) cytokine that targets the potent NF-κB/STAT3 signaling pathway, which can be augmented through blockade of T cell costimulation.^48,49

Certain infiltrating T cells are protective in the injured myocardium. Subsets of CD4⁺ non-specific effector T (T_eff) cells have been shown to protect against post-inflammatory fibrosis in an experimental model of myocarditis.⁵⁰ The authors suggest multiple potential mechanisms for protective effects of non-specific T_eff cells. Essentially, if the balance shifts away from antigen-specific T_eff cells to heart non-specific T_eff cells, the levels of pro-inflammatory cytokines such as IL-17A are reduced and fibrosis is limited.⁵⁰ Separately, our lab previously elucidated an immune-based cardioprotective mechanism through which the epicardium recruits CD4⁺ T regulatory (T_reg) cells in a Hippo/INF-γ –dependent process.⁵¹ Furthermore, Foxp3⁺ T_reg cells have been shown to stimulate recovery of both injured skeletal and heart muscle.^52–54 Accumulating T_reg and certain MYHCA-specific T helper cells have multiple roles, driving pro-reparative phenotypes.⁵²

Many classes of pro-inflammatory T cells infiltrate the injured myocardium in response to dendritic cell activation and chemotactic signals. Activated fibroblasts, macrophages, endothelial cells, and cardiomyocytes produce additional chemotactic signals attracting T cells (through CXCR3 and endothelial ICAM-1), including the NLRP3 inflammasome, CCL2 ligand, GM-CSF, IL-6 and CXCL1/9/10, amongst others.^55–57 Once in the myocardium, activated macrophages also contribute to the stimulation of T cells.⁸ Pro-inflammatory CD8⁺ T cells, activated by dendritic cells, produce potent cytokines including IL-17, IFN-γ, and TNFα.^38,39,58 CD8⁺ T cells are required for M1-like macrophage infiltration and pro-inflammatory cytokine and chemokine secretion.⁵⁸ In addition to cytotoxic T cells, certain CD4⁺ T cells also exhibit damaging effects in the injured heart and are required for the progression of pressure-overload injury into heart failure.^56,59–61 Adoptive transfer of splenic T cells from heart failure mice was sufficient to induce heart failure in naïve recipient mice.⁵⁹ T cell infiltrates have pleotropic effects on virtually every cell type in the injured myocardium. One main role is the direct activation of cardiac fibroblasts and the induction of fibrosis.⁶² The rapid, pro-inflammatory response of CD4⁺ and CD8⁺ T cells may be beneficial in the acute phase of injury to enhance scar formation but may be maladaptive in the long term.

The development of myocardial fibrosis

The multi-faceted immune response to injury impacts many aspects of cardiac function. Among these, regulation of cardiac fibrosis is particularly important in both acute and chronic settings.^63,64 Tissue-resident fibroblasts differentiate from multiple sources early during cardiogenesis and are a self-sustaining population.⁶⁵ At homeostasis, these cardiac fibroblasts are primarily responsible for the secretion and maintenance of the extracellular matrix. After an injury, resident fibroblasts activate, alter gene expression programs (including pro-inflammatory cytokine production), proliferate and accelerate matrix production.^66–70 Additionally, in the setting of myocardial infarction, activated fibroblasts are recruited from the epicardium through a process of epithelial-mesenchymal transformation and migration into underlying regions of damaged myocardium.^51,71,72 In the setting of pressure-overload, fibroblasts are also activated in the vessel wall leading to peri-arteriolar fibrosis that can impair vascular function.⁷³ Fibroblast activation occurs in response to multiple cytokines such as transforming growth factor beta (TGFβ), which is produced by a variety of cells including immune cells and active when liberated from its latent form in the matrix.^74–77 TGFβ signaling is mainly propagated within fibroblasts through phosphorylation of SMAD2/3 which complexes with SMAD4 to directly regulate transcription.^75,78–81 Additional signaling cues are received from the local inflammatory environment and spread through multiple signaling pathways including WNT, MEK/ERK, and JNK.⁸² Recently, autocrine interleukin-11 has also been implicated in fibroblast activation.⁸³ The extracellular matrix (ECM) produced by activated fibroblasts maintains tissue integrity, limiting cardiac dilatation and rupture.^84,85 However, in chronic injury the resulting fibrosis contributes negatively to myocardial health.

Myocardial fibrosis has been associated with poor clinical outcomes and negatively impacts heart physiology both directly and indirectly.^86–89 First, the excess ECM directly alters the compliance of the heart muscle. The stiffness associated with myocardial fibrosis places increase workload on the cardiomyocytes, fueling their dysfunction and decreasing both diastolic and systolic function.^63,64,89,90 Furthermore, excess collagen networks have been linked to electrical conduction defects and arrhythmias.⁹¹ Perivascular fibrosis has been associated with impaired coronary blood flow and microvessel rarefication contributing to latent myocardial hypoxia.⁷³ Finally, both activated fibroblasts and the composition of the ECM signal to infiltrating immune cells.^87,92 For example, the extracellular matrix protein tenascin-c has been shown to accelerate macrophage recruitment and to activate a pro -inflammatory/-fibrotic (M1-like) transcriptome.⁹³ Interestingly, another neonatal extracellular matrix protein, agrin, can stimulate murine cardiac regeneration.⁹⁴

Fibroblast activation is less homogenous than was originally appreciated. Revolutions in single cell and single nuclei RNA sequencing have uncovered many seemingly unique fibroblast populations before and after injury with distinct transcriptome signatures.^20,95–98 The lineage relationship between these subsets and their physiologic importance remains to be fully elucidated. Further evidence for different subsets of fibroblasts contributing to disease pathology comes from recent research in rheumatoid arthritis. Croft et al., identified two subpopulations of fibroblasts in injured joints.⁹⁹ The first group of fibroblasts (marked by FAP⁺/Thy1⁻ expression) were responsible primarily for bone and cartilage-degradation, while a second population of FAP⁺/Thy1⁺ fibroblasts contributed to joint inflammation.⁹⁹ This example suggests that further refinement in our understanding of distinct sub-classes of activated fibroblasts in the heart may reveal distinct roles in cardiac pathology and inform targeted therapeutics.

Immune modulators as cardiac therapies

The potential for immune modulation to impact the heart has been appreciated for decades, and clinicians have long recognized that acute infection and associated immune activation can negatively impact cardiac function.^100–102 Recently, the role of immune modulation in atherosclerosis has been clearly demonstrated, as reviewed elsewhere.^21,103 Broad immunosuppression in acute ischemic heart failure through the administration of corticosteroids or nonsteroidal anti-inflammatory drugs (NSAIDS) is contraindicated and can lead to an increased incidence of cardiac rupture, and their use in chronic forms of heart failure has generally produced either no or a negative effect.^104–106 Similarly, broad immune suppression through colchicine administration has also failed to demonstrate any benefit in multiple heart diseases and is often contraindicated.^107–109 The challenge is to identify targeted immune-modulators that can specifically affect components of the immune response to favor recovery after injury and/or diminish adverse remodeling.

Evidence to support the potential for immune modulation in heart failure has emerged, surprisingly, from studies focused on the mechanism of potential benefit observed in some cardiac stem cell studies. Despite decades of work, cardiac stem cells remain elusive, yet improvement in cardiac function is sometimes observed after delivery of various cell types to the injured heart. The delivered cells generally do not survive, engraft, or differentiate into functional myocytes. Instead, they seem to stimulate an inflammatory response through activated macrophages that can result in enhanced cardiac function, at least temporarily.¹¹⁰ In seminal work, Vagnozzi, et al. showed that following ischemia-reperfusion injury, a myocardial injection of freeze-thaw killed cells resulted in equal improvement in function to injection of live putative stem cells.¹¹⁰ More importantly, a local injection of zymosan (a potent stimulator of the innate immune system) was also sufficient to produce functional improvements.¹¹⁰ This effect was due to recruitment of CCR2⁺ and CX3CR1⁺ (M2-like) macrophages which are beneficial following injury.¹¹⁰ The authors described endothelial cell, but not cardiomyocyte, proliferation and beneficial extracellular matrix remodeling. Thus, agents that augment M2-like macrophages may prove beneficial.

Targeted disruption of the complement cascade following myocardial infarction was hypothesized to tip the balance towards myocardial recovery. Early evidence of monoclonal antibodies blocking C5 cleavage were encouraging in animal models.^111,112 However, clinically similar antibodies (pexelizumab) have yet to demonstrate beneficial effect above standard percutaneous coronary reperfusion therapy following reperfusion injury associated with myocardial infarction.¹¹³ While clinical trials failed to show clear benefit of targeted interference of the complement cascade, the approach may still hold promise.¹¹⁴

One of the first cytokines targeted in heart failure was TNFα.¹¹⁵ This potent inflammatory cytokine was identified as a circulating biomarker, clinically correlated with heart failure severity.¹¹⁶ Research laboratories were quick to follow up this observation with animal model experimentation. It was found that administration of exogenous TNFα was sufficient to induce progressive left ventricular dysfunction.¹¹⁷ Further, animals co-infused with TNFα and its antagonist partially reversed the induced left ventricular dysfunction.¹¹⁷ Based on these observations and others, several clinical trials were conducted to test safety and efficacy in humans of anti-TNFα therapy. Unfortunately, while intravenous etanercept infusions were well tolerated, RENEWAL analysis of the RECOVER and RENAISSANCE clinical trials revealed that TNFα blockade did not yield beneficial reductions in morbidity and mortality.¹¹⁸ Despite these initial unsuccessful clinical trials, the original concept of therapeutically modulating the post-injury immune system remains attractive.¹¹⁹

Another cytokine that has been tested in heart failure is C-X-C motif chemokine ligand 12 (CXCL12; also known as stromal cell-derived factor 1 or SDF-1). Studies in animal models suggested potential benefit, perhaps via recruitment of monocytes or other cells through a CXCR4 dependent mechanism.¹²⁰ Unfortunately, a phase II clinical trial in patients with heart failure with reduced ejection fraction (EF<40%) determined that transcatheter endomyocardial delivery of CXCL12 provided no cardiac functional improvements detectable at four months.¹²¹ Further refinement in dosing, delivery and timing of cytokine-based therapeutics may beget clinical benefit.

Targeted disruption of the fibroblast activation induced by immune cells after injury has also been considered as a potential therapeutic approach. In this regard, inhibition of the TGFβ receptor kinase pathway has received particular attention.^122,123 Experiments in animals have yielded conflicting results, with some finding cardiac benefits and others worrisome increases to mortality and left ventricular dysfunction.^{122,124–127} In systemic sclerosis, disruption of the TGFβ pathway was not successful at controlling fibrosis and may have even been responsible for adverse events (although sample size was small).¹²⁸ In contrast, TGFβ blockade with perfenidone has demonstrated benefits in fibrosis-based interstitial lung diseases.^129,130 At least one clinical trial of anti-TGFβ blockade in heart failure patients has been initiated.¹³¹ While we await these results, developing more targeted approaches may yield beneficial results.⁶⁵ Partial evidence for this comes from studies in mice with cardiomyocyte-specific TGFβ receptor inactivation, which favors cardiac recovery through an immune-modulatory mechanism.¹³²

Targeted or engineered immune-based therapeutics

Recent efforts have begun to focus on the use of biomaterials, biologics and targeted cell and gene therapy to modify specific components of the immune response to cardiac injury (see Figure 2).¹³³ For example, targeted inhibition of the coagulation cascade (factor Xa, but not IIa) has demonstrated beneficial cardiac remodeling following injury by dampening harmful inflammation.¹³⁴ Clever chemical engineering has resulting in graphene oxide complexes functionalized with IL-4, that appear to coax cardiac macrophages from M1-like to M2-like pro-reparative phenotype and limit damaging inflammation when injected into post-myocardial infarction hearts.¹³⁵ Several other macrophage modulating molecules are in various stages of development in non-cardiovascular injuries that may be worth evaluating in heart injury models.¹³⁶ These include lipoxin A4 and resolvin D1, which were shown to improve resolution following implanted biomaterial scaffolds through an M2-like macrophage mechanism.¹³⁷ Many other naturally derived biomaterials have been shown to scavenge radical oxygen species and can be loaded with other anti-inflammatory molecules to further alter the microenvironment.¹³⁶ It may be possible to target lipid nanoparticles loaded with mRNA or other agents to specific immune cell types, taking advantage of therapeutic platforms developed and validated as part of the COVID-19 vaccination efforts.

Figure 2. Potential immune-modulating targets to promote cardiac repair.

Many immune modulatory targets exist within the complex signaling network in the injured myocardium. Chimeric antigen receptor (CAR) constructs include antigen recognition domains and intracellular signaling domains, which are engineered into cytotoxic T cells. CAR T cells designed against activated fibroblasts and senescent cells both offer exciting directions to improve cardiac repair in the failing heart. Shifting the balance of macrophage polarization to favor the M2-like, reparative phenotype, may produce beneficial effects. Limiting the cytotoxic CD8⁺ and heart-specific CD4⁺ T cell infiltrate in favor of boosting regulatory (T_reg) cells, beneficial subsets of effector T (T_eff) cells and natural killer (NK) cells have each been shown to benefit cardiac repair. Similarly, shifting the balance to eosinophils over neutrophils may benefit recovery. Disrupting B cell pathways offers potential benefit. Targeted interleukin blockade (IL-1, IL-11, IL-15, TGFβ, etc.) or reparative cytokine (IL-10) administration may limit damage of excessive (or improperly timed) inflammation. Lastly, the development of novel chemical materials and biological interventions (including miRNAs and exosomes) offer exciting targeted immune modulators.

Immune-modulating exosomes and miRNAs may hold potential for enhancing repair following cardiac injury. Exosomes are naturally-occurring small extracellular vesicles filled with a variety of biological molecules, which may offer avenues for immune modulation in cardiac injury.¹³⁸ Advances in engineered exosomes offers the potential for unique cardiac delivery vehicles of immune-modulating therapeutics such as miRNAs. Multiple miRNAs including miR-155, miR-146a, miR-181b, miR-208, miR-29b and others are under investigation for promoting heart repair through immune modulation, with several in current clinical trials.¹³⁹

Blocking specific cytokines offers a potential avenue for therapy. Interleukin 1β (IL-1β) is a pleotropic cytokine with various inflammatory functions in the injured heart, mainly augmenting injury. Systemic blockade of IL-1 signaling is typically achieved with either a competitive inhibitor of the receptor (canakinumab) or blocking antibodies against IL-1β (anakinra).¹⁰⁵ This approach has shown modest beneficial effect in animal models of myocardial infarction.^140,141 In clinical trials canakinumab has been well tolerated and resulted in avoidance of post-myocardial infarction heart failure and reduction in mortality in heart failure patients.^142,143 Furthermore, canakinumab has demonstrated beneficial effect in atherosclerotic disease.¹⁴⁴ The beneficial outcomes observed following IL-1 pathway disruption are exciting with the caveat that less targeted IL-1 interference (low-dose methotrexate) did not recapitulate the results obtained with canakinumab in atherosclerotic events and resulted in increased incidence of negative side effects.¹⁴⁵ This trial re-enforces that targeted immunotherapeutics are likely to offer more potential benefits than broad immune modulators. Interestingly, SGLT2 inhibitors, which have demonstrated benefits in heart failure patients, may be operating through IL-1β and/or NLRP3 inflammasome mechanisms, supporting the notion of careful inflammation modulation.^146–148 Similarly, both interleukin 15 and interleukin 11 blockade are under investigation as a potential cytokine targets for cardiac disease.^83,149 Interleukin 11 appears to be a central player in the immune-fibrosis axis in multiple organs, reinforcing its attractiveness as a potential therapeutic target.^83,150,151

Certain cytokines have been shown to stimulate repair and may be attractive therapeutics. In the injured heart, IL-10 is typically thought of as pro-reparative; however, in other contexts it can be pro-inflammatory.¹⁵² A recent structural study has elucidated a biased signaling mechanism for IL-10, analogous to GPCR/β-arrestin signaling.¹⁵³ Furthermore, the authors demonstrate how engineered variants of IL-10 can strongly skew the intracellular signaling response.¹⁵³ This paradigm of fine-tuning biased receptor signaling through ligand optimization, well accepted in the field of GPCR signaling, offers exciting opportunities for anti-inflammatory therapeutics.

A recent study identified cardioprotective T regulatory-like cells in the post-myocardial infarcted heart. These CD4⁺ T cells react with myosin heavy chain alpha and accumulate in the injured myocardium of both mice and human patients, and appear to be beneficial in mice when delivered exogenously before induction of myocardial infarction.⁵² Related Foxp3⁺ T_reg cells have been shown to be beneficial in the injured mouse and rat heart, either with an infusion of autologous T_reg or a CD28 antibody (which augments the natural recruitment of T_reg cells).^51,54,154 Boosting the recruitment and accumulation of T_reg in the injured heart as demonstrated in animal models, may be an attractive future direction for human therapeutics.

CAR T cells for heart disease

Genetically engineered cytotoxic T cells that have been redirected to recognize a specific antigen were developed in the late twentieth century following the success of adoptive transfer of T cells to control viral infection.^155–160 This was accomplished by expressing a chimeric antigen receptor (CAR) on activated CD8⁺ T cells. The chimeric antigen receptor is a combination of a single-chain variable fragment (scFv) against a specific antigen linked to intracellular T cell signaling and co-stimulatory molecules (see figure 2 inset). Typically, these include at least a CD3-zeta tyrosine phosphorylation activation motif, as well as 4–1BB and CD28 co-stimulatory signaling domains, though this design has many iterations. This all-in-one protein design results in one-step cytotoxicity, induced through antigen binding alone, as opposed to T cell receptor (TCR) mediated cytotoxicity which requires MHC class I antigen presentation, CD3 recruitment to the TCR, and costimulatory signaling.¹⁶¹ CAR T cells were quickly adapted to target cancerous cells and have revolutionized the treatment of refractory acute lymphoblastic leukemia and diffuse large B-cell lymphoma.^161,162

In the setting of heart failure, it is attractive to consider targeting activated cardiac fibroblasts with engineered T cells. This is made possible because activated cardiac fibroblasts are sufficiently different from quiescent, homeostatic fibroblasts such that they express unique cell surface markers and can therefore be specifically targeted. An early indication that this approach might be effective to reverse fibrosis and improve function after injury came from mouse studies in which activated fibroblasts were eliminated by genetic ablation using diphtheria toxin in animals engineered to express the diphtheria toxin receptor on activated fibroblasts.¹⁶³ In the setting of pressure-overload injury, heart function was improved and fibrotic burden was decreased by removal of activated fibroblasts.¹⁶³

In order to provide a mechanism to translate these findings to humans, we sought to target activated fibroblasts with an engineered T cell. To find a unique molecular marker of activated fibroblasts, we utilized transcriptomics to compare healthy human hearts to dilated and hypertrophic hearts. Consistent with the literature, we found that both fibroblast activation protein alpha (FAP) and periostin were highly upregulated in both disease states when compared to healthy hearts.¹⁶⁴ In the adult, FAP protein expression is largely restricted to activated fibroblasts in various disease states and after injury. It is also expressed by proliferative mesenchyme in many forms of cancer.¹⁶⁵ Although a secreted form of FAP can be identified in the circulation of healthy humans, it does not correlate with stromal cell expression or serve as a reliable biomarker.^166,167 Importantly, this protein is virtually absent in healthy adult tissues.¹⁶⁵ FAP CAR T cells have been demonstrated to effectively target and reduce FAP⁺ cancer associated fibroblasts.¹⁶⁸ Using a mouse model of cardiac fibrosis and heart failure produced by the constant infusion of angiotensin II and phenylephrine, we showed that FAP CAR T cells were effective at restoring cardiac function and significantly lowering the burden of fibrosis even when the CAR T cells were delivered after fibrosis has already accumulated.¹⁶⁴ Resorption of previously accumulated ECM is likely the result of macrophage, non-targeted fibroblasts and neutrophil activity which predominates once the cells producing the excessive matrix are destroyed by the CAR T therapy.

One of the advantages of FAP as an antigen target is the ability to visualize expression in vivo using imaging probes. Initial FAP-PET (Positron Emission Tomography) probes were developed using antibodies and tested extensively in mice.¹⁶⁹ Loktev, et al. extended these findings by attaching gallium-68 to a synthetic FAP inhibitor and showed that cancer-associated fibroblasts, which resemble activated cardiac fibroblasts, can be directly imaged in both mice and humans.¹⁷⁰ These powerful imaging tools have been used successfully to image cardiac fibrosis in humans and represent an exciting screening tool for clinical evaluation of heart failure patients prior to FAP CAR T intervention and for monitoring of therapeutic efficacy.^171,172

The use of CAR T cells to target senescent cells is also in development. Senescent cells are characterized by permanent cell cycle arrest and secretion of high levels of inflammatory molecules producing a senescence-associated secretory phenotype (SASP). Cellular senescence has been associated with many age-related disorders including cardiac pathologies and arrhythmias.^173,174 Clearance of senescent cells may ameliorate the decrease in cardiac performance associated with aging and improve recovery following injury.^175,176 Recently, Amor, et al. described an anti-senescence CAR T cell directed against urokinase-type plasminogen activator receptor, a marker of at least some senescent cells.¹⁷⁷ Following adoptive transfer of anti-senescence CAR T cells, mice with induced liver fibrosis or lung adenocarcinoma showed improvement in organ function.¹⁷⁷ To our knowledge, this approach has not yet been studied in animal models of cardiac disease.

CAR T cell therapy is associated with important acute toxicities. Infusion of CAR T products, including the FDA approved anti-CD19 CAR T cells (tisagenlecleucel and axicabtagene ciloleucel) can induce potentially lethal cytokine release syndrome.^178,179 Fortunately this syndrome is typically transient and can be managed with IL-6 blockade (tocilizumab) and supportive care.^179–181 Anti-CD19 CAR T therapy is also rarely associated with neurotoxicity, potentially through unintended, on-target elimination of CD19⁺ mural cells.¹⁸² Especially pertinent for this discussion are cases of cardiotoxicity following engineered T cell administration potentially due to unforeseen cross-reactivity.^183,184 The binding strength of the individual CAR to the target antigen is also important. In animal trials of preliminary FAP CAR constructs, some toxicity was observed, although it is mitigated by changing the antibody upon which the CAR was based.^168,185 These examples highlight the need for restricted target antigen expression limited to the pathologic cells of interest, and for further safety studies for potential cardiac CAR T products. In support of FAP as a potential target, a recent study found that genetic ablation of FAP does not impair cardiac function in healthy animals or result in cardiotoxicity following myocardial infarction.¹⁸⁶ Furthermore, target-cell specificity can be greatly increased through the use of dual-specific engineered T cells. In this approach, two antigens are chosen and the intracellular signaling domains are split between the two separate chimeric antigen receptors.¹⁸⁷ Therefore, the CAR T must engage both antigens before cytolysis occurs. An alternative approach is using a bispecific T cell engager (BiTE) to increase specificity and reduce unwanted off-target effects.^188,189 By refining the target antigen beyond FAP, anti-fibrotic CAR T cells hold great promise.

Unlike in cancer where curative therapy requires elimination of every cancerous cell, CAR T therapy for fibrosis may be effective at restoring function even if only a fraction of the disease-causing fibroblasts are destroyed. Dosing can therefore be titrated, and cells can be engineered to have only transient activity, eliminating potential long-term toxicities. It is critical to tune the quantity, potency, and longevity of the CAR to ensure that the beneficial, homeostatic fibroblasts remain within the myocardium to maintain the healthy ECM after the excessive fibrosis is reduced. Furthermore, subsequent acute myocardial ischemic events require robust fibroblast activation. A transient anti-fibrotic CAR T is therefore required. Various mechanisms to produce transient CAR T cells have been described, including the use of kill switches or required co-activator small molecules.¹⁹⁰ Alternatively, CARs can be engineered using mRNA rather than lentiviral vectors such that the CAR is no longer expressed after the mRNA decays (over the course of several days) or after dilution due to cell division.¹⁹¹ The ability to target specific subsets of pathologic fibroblasts in various disease states, in the heart and in other organs, may emerge as additional unique cell surface markers are identified. Cell therapy for cardiac diseases is an exciting field, still in its infancy.

Conclusion

The pathology of myocardial injury and repair is coordinated by a complex web of intersecting inflammatory pathways and immune cell types. Our understanding of the beneficial and harmful contributions of specific immune cells and cytokines is an exciting and emerging field of investigation. Targeted modulation of the immune system as a mechanism to boost myocardial recovery and repair is an important goal of cardio-immunology. M2-like macrophages, natural killer cells, and regulatory T cells represent especially promising targets for therapeutic intervention. Furthermore, the prospect of engineering the immune system to reduce both fibrosis and senescence in aging and failing hearts offers potential future therapeutic avenues. Other engineered immune cells, such as CAR macrophages may also be on the horizon.¹⁹² Altogether, targeted manipulation of the immune system to benefit the injured heart is an exciting and promising field of active study.

Nonstandard Abbreviations and Acronyms:

CAR

Chimeric antigen receptor

cGAS-STING

cyclic GMP-AMP synthase stimulator of interferon genes

CHIP

Clonal hematopoiesis of indeterminate potential

COVID-19

Coronavirus disease 2019

CXCL12

C-X-C motif chemokine ligand 12 (previously known as stromal cell-derived factor 1 or SDF-1)

DAMPs

Damage associated molecular patterns

ECM

Extracellular matrix

EF

Ejection fraction (%)

FAP

Fibroblast activation protein

IL-1

Interleukin 1

M1-like

Pro-inflammatory macrophage

M2-like

Pro-reparative macrophage

MHC-II

Major histocompatibility complex class II

NK

Natural killer cell

NLRP3

NLR family pyrin domain containing 3

PET

Positron emission tomography

scFv

Single-chain variable fragment

SGLT2

Sodium/glucose cotransporter 2

TCR

T cell receptor

T_eff

Effector T cell

TGFβ

Transforming growth factor β

TNFα

Tumor necrosis factor

T_reg

Regulatory T cell