The chicken, the egg, and the elephant: eNOS and NRG1 in fibrosis

Endothelial cells produce many autocrine and paracrine factors that modulate tissue homeostasis. Endothelial cell-produced nitric oxide (NO) generated by endothelial nitric oxide synthase (eNOS) and neuregulin-1 (NRG1) are critical cardiovascular and renal homeostatic paracrine molecules (1, 2). Endothelial cell dysfunction, hallmarked by reduced NO production, contributes to heart failure and cardiovascular disease (1). In a recent issue of the American Journal of Physiology-Heart and Circulatory Physiology, Shakeri et al. (3) are the first to specifically address the relationship between endothelial cell dysfunction (NOS/NO) and NRG1/ErbB pathways.

NRG1 is a member of the epidermal growth factor superfamily that is secreted by endothelial cells (2). Activation of the erythroblastic leukemia viral oncogene homolog (ErbB4/ErbB2) receptors found on various cells, including cardiomyocytes, fibroblasts, endothelial cells, and myeloid cells, mediates the effects of NRG1 (2, 4). NRG1/ErbB signaling conveys cardioprotective, antifibrotic, and antiremodeling effects in the heart (2), impacting cardiac fibroblasts directly (4).

Patients with heart failure have reduced eNOS capacity (the chicken) (5), establishing the connection between endothelial cell dysfunction and heart failure. Patients with heart failure with preserved ejection fraction (HFpEF) also have impaired endothelial function (1). Therefore, understanding the implications of endothelial cell dysfunction is critical to developing novel therapeutic approaches to treat heart failure (1). The association between NRG1-β (the egg) and clinical outcomes in heart failure is complex. In the presence of microvascular dysfunction (nonischemic HFpEF), higher NRG1 levels are associated with improved outcomes (6). In contrast, in the presence of ischemia (HFpEF or HFrEF), elevated NRG1 levels are associated with worse clinical outcomes (6). Together, these clinical data in heart failure provide the rationale to examine the association between eNOS/NO and NRG1 pathways in nonischemic models of heart stress/failure.

Prior work by the University of Antwerp team has demonstrated that NRG1 can activate eNOS activity in cardiomyocytes (7). An essential current finding is that in eNOS-deficient mice, circulating NRG-1 is significantly increased, revealing a possible feedback mechanism by which these critical endothelial cell paracrine mechanisms (eNOS/NO and NRG1/ErbB) regulate one another. A complete dissection of the factors regulating the association between eNOS/NO and NRG1/ErbB pathways is needed. Equally important was the observation that the elevated circulating levels of NRG1 were inadequate to prevent the pathophysiological phenotype in eNOS-deficient mice. As discussed earlier, in humans, the level of NRG1 is associated with clinical outcomes (6); therefore, understanding the optimal levels of NRG1 that convey beneficial effects is essential. The critical finding in the current study that supplemental NRG1 treatment generates a cardiac and renal protective phenotype in the face of eNOS deficiency demonstrates that NRG1 treatment is a viable option for preventing progressive fibrosis in the setting of endothelial cell dysfunction. However, prior work by the authors examining the cell-specific expression of NRG1 receptors appears at odds with the current findings. The authors have demonstrated that in mice with endothelial cell-specific knockout of the NRG1 receptor (ErbB4), the fibrotic response induced by angiotensin II (ANG II) treatment is attenuated (8). Using this cell-specific knockout approach, the authors have also demonstrated that myeloid cell deletion of ErbB4 expression exacerbates myocardial fibrosis in response to ANG II treatment consistent with the findings reported in the current data (9). Furthermore, NRG1 treatment decreased inflammatory markers and macrophage accumulation in the heart in response to ANG II treatment (9). These findings with cell-specific approaches stress the complexity of neuregulin-1 biology that requires further investigation.

An alternative interpretation of the current data is that the protective phenotype conveyed by NRG1 treatment may represent a secondary effect. The findings that atrial natriuretic peptide (Nppa) mRNA levels are increased in response to NRG1 treatment support the supposition that the antifibrotic effects may not be a direct effect of NRG1/ErbB signaling. Additional studies are required to clarify this scenario. Atrial natriuretic peptide (ANP) is antihypertrophic and antifibrotic (10). Specific profibrotic and hypertrophic signaling pathways were analyzed in the current study. However, a nonbiased analysis using RNAseq could have revealed neuregulin-1 regulated genes in the heart and kidney, uncovering novel pathways associated with the protective phenotype observed in the presence of eNOS deficiency.

Although the use of a subpressor dose of ANG II negated the confounder of blood pressure effects from both ANG II and NRG1, it potentially limits the clinical relevance of the model. Hypertension is a critical risk factor for developing heart failure and both renal and cardiac fibrosis (10). The ability of NRG1 to compensate for increased mortality in eNOS-KO mice treated with pressor dose ANG II would provide clinically meaningful information given that in heart failure, the renin-aldosterone-angiotensin system (RAAS) is activated (10). Evaluation of the impact of the antihypertrophic and antifibrotic of the NRG1/ErbB pathway in an endothelial cell dysfunction model with hypertension is warranted.

A final intriguing finding is the downregulation of miR134 in eNOS-deficient endothelial cells. The provocative demonstration that miR-134 inhibitor treatment increases NRG1 mRNA levels in endothelial cells treated with ANG II suggests a novel approach to modulate NRG1 levels and thereby impact hypertrophy and fibrosis. Further evaluation of this finding with in vivo validation is warranted.

In summary, the current study has demonstrated a novel interplay between eNOS/NO and NRG1/ErbB pathways to regulate cardiac hypertrophy and fibrosis in the heart and kidneys. As with any good study, many new questions arise. A critical question is whether the cardiorenal protective effects a direct or secondary NRG1 effect? Understanding the association of eNOS/NO and NRG1/ErbB pathways in greater detail may reveal novel mechanisms to prevent fibrosis in the setting of endothelial cell dysfunction. Notably, the finding that NRG1 (the egg) conveys a protective phenotype in the absence of eNOS (the chicken) provides optimism that ongoing clinical studies examining NRG1 treatment in patients with heart failure (the elephant) may prove efficacious.

GRANTS

This work supported by National Heart, Lung, and Blood Institute Grant R01-HL127442-01A1.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.N. and R.J.G. drafted manuscript; edited and revised manuscript; and approved final version of manuscript.