Three studies reveal that augmentation of a signalling pathway involving the growth factor neuregulin 1 and its receptor protein ERBB2 can promote the generation of muscle cells in zebrafish, mice and infant heart tissue.
Whether mature heart-muscle cells called cardiomyocytes can proliferate to make new muscle has been an area of contention for many years. The debate has implications for the treatment of cardiovascular disease, because the loss of cardiomyocytes can lead to heart failure and death. Mammalian cardiomyocytes were thought to stop proliferating in the first few days after birth1. However, evidence has emerged to suggest that adult mouse and human cardiomyocytes do proliferate, albeit in very small numbers2,3. Now, three studies (published in eLIFE4, Nature Cell Biology5 and Science Translational Medicine6) report that the growth factor neuregulin 1 (Nrg1) promotes heart regeneration in zebrafish, and cardiomyocyte proliferation in mammalian hearts.
Cardiomyocytes in adult mice and humans are thought to renew at a rate of less than 1% per year2,3. By contrast, adult zebrafish can regenerate up to one-third of their heart muscle from existing cardiomyocytes in a few months after injury7. What factors mediate this regeneration? Gemberling et al.4 found that Nrg1 induces cardiomyocyte proliferation and therefore heart regeneration in zebrafish by signalling through its co-receptor protein, Erbb2.
Turning to mammalian hearts, D'Uva et al.5 examined whether Erbb2 is required for heart-muscle growth in mice just after birth. Pups older than one week had lower expression of Erbb2 than younger mice, indicating that expression of the receptor is downregulated within a week of birth, at the same time as cardiomyocytes stop proliferating. The authors found that transient induction of an activated form of Erbb2 for 10–21 days in juvenile or adult cardiomyocytes prolonged the regenerative capacity of the hearts into adulthood or restored regenerative capacity, respectively. Expression of activated Erbb2 in mature cardiomyocytes enabled the cells to partially dedifferentiate to a less-specialized cell type after injury, resulting in disassembly of the muscle contractile apparatus and leading to proliferation in the hearts of juvenile and adult mice (Fig. 1).
In the third study, Polizzotti et al.6 provide evidence to support D'Uva and colleagues' data. The authors injured the hearts of newborn mice through localized freezing. Control hearts showed scarring and had reduced function following injury, but heart function was preserved and cardiomyocyte proliferation was maintained in newborn mice treated with Nrg1. However, treatment started later than four days after birth did not improve heart function, possibly owing to the postnatal loss of Nrg1-receptor expression reported by D'Uva and colleagues. Together, then, these three studies support activation of the Nrg1–Erbb2 pathway as a way to prolong the ability of the newborn mouse heart to regenerate in the first few days after birth.
Whether there is a regenerative window in infants comparable to that in mice is unclear. Polizzotti and colleagues addressed this issue by adding NRG1 to diseased human heart biopsies that were cultured in vitro. The treatment promoted cardiomyocyte proliferation in biopsies taken from newborns, but the proliferative effect was significantly reduced by six months of age. Thus, there may be a narrow therapeutic window in which NRG1 treatment could improve the success of reconstructive surgery for infants born with severe heart defects.
Could NRG1 treatment also be a good way to treat heart disease in adults? D'Uva et al. demonstrated that transient induction of activated Erbb2 following a heart attack in adult mice led to improved heart function, reduced scarring and increased cardiomyocyte proliferation compared with mice that did not receive activated Erbb2. But the current studies4,5also demonstrate that, in both fish and mice, the unrestrained cardiomyocyte proliferation induced by Nrg1–Erbb2 signalling leads to enlargement and eventual failure of the heart. The correct level of signalling is therefore imperative — it may be beneficial to induce cardiomyocytes to proliferate with Nrg1, but it is equally important to shut off the pathway to maintain the appropriate number of cells for a healthy heart.
Activated ERBB2 was first identified in tumour cells, and expression of this mutated form of the receptor is an indicator of poor prognosis in breast cancer8. Therapies targeting activated ERBB2 are used to treat breast cancer, but heart-muscle damage is a known side effect and must be monitored in these patients8. Thus, another concern with using NRG1 therapy for heart disease is the possibility that this treatment might increase the risk of cancer. To minimize this risk, if activation of the NRG1–ERBB2 pathway is used to treat heart disease in the future, therapies should ideally specifically target cardiomyocytes, and the duration of treatments should be limited.
Other studies have tested the efficacy of Nrg1 treatment in adult mice following a heart attack, and have produced conflicting results. Although one study9 reported that Nrg1 signalling stimulates proliferation of differentiated cardiomyocytes following injury, another10 found no evidence of DNA synthesis, an indicator of proliferation, in Nrg1-treated hearts. The authors of the latter study speculated that Nrg1 might also protect the mammalian heart from injury, possibly by stimulating the growth of new blood vessels, which could explain the positive results of some studies in adult animals.
NRG1 is currently being tested in humans as a treatment for heart failure (https://clinicaltrials.gov), but these trials are ongoing and it is too soon to tell whether the treatment is effective. The current studies suggest that NRG1 treatment might help the injured heart not only by supporting existing muscle, but also by promoting the production of new muscle. Both facets of NRG1 function should be considered in the development of treatments for cardiovascular disease based on regenerative pathways.Footnote 1
Soonpaa, M. H. & Field, L. J. Circ. Res. 83, 15–26 (1998).
Bergmann, O. et al. Science 324, 98–102 (2009).
Senyo, S. E. et al. Nature 493, 433–436 (2013).
Gemberling, M., Karra, R., Dickson, A. L. & Poss, K. D. eLIFE 4, e05871 (2015).
D'Uva, G. et al. Nature Cell Biol. http://dx.doi.org/10.1038/ncb3149 (2015).
Polizzotti, B. D. et al. Sci. Transl. Med. http://dx.doi.org/10.1126/scitranslmed.aaa5171 (2015).
Poss, K. D., Wilson, L. G. & Keating, M. T. Science 298, 2188–2190 (2002).
Cote, G. M., Sawyer, D. B. & Chabner, B. A. N. Engl. J. Med. 367, 2150–2153 (2012).
Bersell, K., Arab, S., Haring, B. & Kühn, B. Cell 138, 257–270 (2009).
Reuter, S., Soonpaa, M. H., Firulli, A. B., Chang, A. N. & Field, L. J. PLoS ONE 9, e115871 (2014).