S100A8/A9 as a therapeutic target in myocardial infarction: cellular mechanisms, molecular interactions, and translational challenges

This editorial refers to ‘Inhibition of pro-inflammatory myeloid cell responses by short-term S100A9 blockade improves cardiac function after myocardial infarction’^†, by G. Marinković et al., on page 2713.

Following myocardial infarction, massive necrosis of myocardial cells releases damage-associated molecular patterns (DAMPs) that stimulate systemic inflammation, and promote recruitment of inflammatory leucocytes in the infarct zone. Infiltrating leucocytes clear the infarct of dead cells and matrix debris, and stimulate repair of the infarcted myocardium, secreting cytokines and growth factors that activate fibroblasts and vascular cells, and contribute to formation of a scar. Over the last 30 years, a large body of experimental evidence suggested that in addition to their reparative functions, pro-inflammatory cascades may also exert detrimental actions in the infarcted heart, extending ischaemic injury and accentuating adverse remodelling.¹ In experimental animal models of myocardial infarction, targeted inhibition of innate immune cascades, chemokines, cytokines, and adhesion molecules was protective, decreasing the size of the infarct, and reducing adverse dilative remodelling. Unfortunately, despite highly promising animal model evidence, clinical studies testing the effectiveness of anti-inflammatory strategies in human patients with myocardial infarction have produced disappointing results.²

The S100 calcium-binding proteins are abundantly expressed in sites of inflammation and serve as biomarkers reflecting disease activity in many inflammatory conditions. Some members of the family, including S100A7, S100A8, S100A9, S100A12, and S100A15, may act as DAMPs triggering or amplifying innate immune pathways following tissue injury.³ Induction and release of S100 family members has been documented in the infarcted myocardium, and has been suggested to mediate pro-inflammatory cascades.^4,5 S100A8 and S100A9 typically exist as a S100A8/S100A9 heterodimer, are predominantly produced by myeloid cells (both neutrophils and monocytes), and are known to activate innate immune pathways through stimulation of Toll-like receptor 4 (TLR4)- and receptor for advanced glycation end-products (RAGE)-mediated cascades.³ Although increased expression of S100A8 and S100A9 has been demonstrated both in animal models of myocardial ischaemia and in human myocardial infarction patients,⁶ the potential role of S100A8/A9 as an alarmin involved in ischaemic injury, post-infarction inflammation, and cardiac remodelling has not been investigated.

In the current issue of the European Heart Journal, Marinkovic and co-workers⁷ examined the role of S100A8/A9 in myocardial infarction in both human patients and mouse models. In patients with acute coronary syndromes, high plasma levels of S100A8/A9 during the acute event were associated with late development of systolic dysfunction and with a higher incidence of heart failure hospitalizations 1 year after the coronary event. In mouse models of reperfused and non-reperfused myocardial infarction, early short-term treatment with a small molecule inhibitor that blocks interactions between S100A9 and its receptors TLR4 and RAGE had lasting protective effects, attenuating systolic dysfunction, and increasing cardiac output for at least 3 weeks after infarction. The beneficial actions of the S100A9 inhibitor were attributed to attenuated recruitment of neutrophils and inflammatory monocytes, and to activation of a reparative phenotype in infiltrating macrophages. The findings provide important new information on the pathophysiological role of S100 proteins in mediating inflammatory injury and adverse remodelling following myocardial infarction, and suggest that S100A9 may be a novel and promising therapeutic target in patients with myocardial infarction. After three decades of frustrating translational failures in targeting inflammation following myocardial infarction, could S100A8/A9 represent the optimal inflammatory target that may allow effective and selective inhibition of injurious immune signalling in the infarcted heart without perturbing repair? If so, which cellular mechanisms may explain the protective effects of S100A8/A9 inhibition?

S100A8/A9 may act both as an intracellular mediator and as a secreted signal

Predominantly expressed by myeloid cells, S100A8 and S100A9 are known to act both as intracellular effectors and as secreted alarmins. As intracellular signals, S100A8/A9 control microtubule organization, mediating migratory responses of neutrophils and mononuclear phagocytes.⁸ Following inflammatory injury, increased S100A8/A9 levels in the extracellular space appear to reflect active secretion, rather than passive release from necrotic cells. S100A8 and S100A9 proteins lack a leader signal and are not secreted through the classical Golgi-associated pathway, but are actively secreted in response to cytokine stimulation via tubulin-dependent mechanisms.⁹ As a secreted alarmin, the S100A8/A9 heterodimer may modulate the phenotype and function of all cells involved in cardiac repair, including leucocytes, vascular endothelial cells, cardiomyocytes, and reparative fibroblasts (Figure 1A).

Figure 1.

The cellular and molecular targets of S100A8/A9 in the infarcted heart. (A) Following myocardial infarction, S100A8 and S100A9 are secreted primarily from activated myeloid cells (neutrophils and monocytes). The pro-inflammatory effects of the S100A8/A9 heterodimer in the infarcted myocardium may involve: (i) activation of oxidative responses in neutrophils,; (ii) induction of pro-inflammatory cytokines and chemokines in monocytes and macrophages; (iii) chemokine induction in endothelial cells; and (iv) increased synthesis of cytokines, proteases, and extracellular matrix (ECM) proteins in fibroblasts. Moreover, S100A8/A9 may promote contractile dysfunction in cardiomyocytes. (B) The pro-inflammatory actions of secreted extracellular S100A8/A9 in various cell types may be mediated through activation of Toll-like receptor 4 (TLR4) and receptor for advanced glycation end-products (RAGE) signalling, leading to downstream activation of nuclear factor (NF)-κB and mitogen-activated protein kinase (MAPK). Moreover, in myeloid cells, S100A8/A9 may transfer arachidonic acid (AA) to the NADPH oxidase complex, activating the oxidative burst.

Effects of S100A8/A9 on myeloid cells

Inflammatory activation of neutrophils and macrophages is a well-documented action of S100A8/A9 in vitro and in vivo, and may explain, at least in part, the protective actions of S100A9 blockade in the infarcted myocardium. The pro-inflammatory actions of S100A8/A9 may involve activation of several distinct mechanisms (Figure 1B); their relative contribution in vivo remains unclear. First, S100A8/A9 stimulates a pro-inflammatory programme through activation of TLR4 signalling and downstream stimulation of nuclear factor (NF)-κB signalling.¹⁰ Secondly, effects of S100A8/A9 may also involve activation of RAGE-dependent signalling.¹¹ Thirdly, S100A8/A9 may activate oxidative burst in neutrophils by transferring arachidonic acid to NADPH oxidase.¹²

Effects of S100A8/A9 on endothelial cells, fibroblasts, and cardiomyocytes

In addition to its effects on immune cells, secreted S100A8/A9 may also exert important actions on vascular endothelial cells, interstitial fibroblasts, and cardiomyocytes. In healing infarcts, endothelial cells are important sources of chemokines and, through their adhesive interactions with circulating leucocytes, play a critical role in recruitment of inflammatory cells in the area of injury. S100A8/A9 can bind to proteoglycans on the endothelial surface, triggering chemokine induction,¹³ and promoting leucocyte infiltration. Fibroblasts, the main matrix-producing cells in healing infarcts, also respond to S100A8/A9 stimulation, secreting cytokines, proteases, and extracellular matrix proteins.¹⁴ Surviving cardiomyocytes in the infarct border zone may also be directly targeted by S100A8/A9 released by myeloid cells. Experiments in endotoxin-stimulated cardiomyocytes demonstrated that S100A8/A9 transduces RAGE-dependent signals that mediate contractile dysfunction.¹⁵ In the healing infarct, the protective effects of S100A9 inhibition on systolic function may involve direct protective effects on cytokine-stimulated cardiomyocytes that may attenuate contractile dysfunction.

S100A8/A9 as a therapeutic target in myocardial infarction

Despite the promising experimental evidence provided by Marinkovic et al.,⁷ therapeutic targeting of S100A8/A9 in patients with myocardial infarction may pose major challenges. First, the broad cellular effects of S100A8/A9 on all cell types involved in myocardial injury and repair complicate therapeutic implementation. Inhibition of activity and function of professional phagocytes may delay clearance of apoptotic cells, leading to secondary necrosis and promoting sustained inflammation and injury. Moreover, abrogation of S100A8/A9-mediated actions on fibroblast activity may perturb reparative functions. Although no functional perturbations were noted in the current study upon S100A9 blockade, effects on the reparative response were not systematically studied. Secondly, the translational value of animal models of myocardial infarction is limited by the absence of directly translatable outcome endpoints, and by the pathophysiological heterogeneity of human post-infarction remodelling that cannot be recapitulated in animal investigations. In mouse models of myocardial infarction, injury is caused by surgical occlusion of the coronary artery; thus, any effects of the intervention on thrombotic events cannot be evaluated. On the other hand, mortality endpoints are difficult to interpret, as arrhythmic and heart failure-related deaths (the most common causes of mortality in human infarction) are relatively uncommon. Thus, conclusions regarding outcome are derived primarily through extrapolation of effects on systolic function, or geometric remodelling. Moreover, the mouse model is designed to dissect cellular and molecular mechanisms, by minimizing pathophysiological variability. For this purpose, genetically identical, sex- and age-matched young mice are studied in a well-standardized surgical model of infarctive injury. In contrast, human patient populations exhibit marked variations in the intensity and profile of inflammatory and remodelling responses following infarction. This variability is dependent not only on the size of the initial injury, but also on a wide range of factors that may affect the inflammatory response (such as age, sex, genetic differences, other concomitant conditions, the pattern of coronary disease, the use of medications, etc.). The pathophysiological heterogeneity of human myocardial infarction cannot be simulated by an animal model. It is likely that targeting inflammation may be effective only in patient subpopulations with excessive, sustained, or unrestrained inflammatory responses. Thus, successful translation of S100A8/A9 targeting in human patients may require identification of subsets of patients with exaggerated alarmin-driven immune reactions. Acute coronary syndrome patients with very high circulating levels of S100A8/A9 were identified in the current study, and may represent a promising target group for such an intervention.

Funding

Dr Frangogiannis’ laboratory is supported by NIH grants R01 HL76246 and R01 HL85440, and by Department of Defense grants PR151134 and PR151029.

Conflict of interest: none declared.

The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.