Heart muscle cells die en masse after injury, yet the adult mammalian heart retains little capacity to regenerate them. Regulatory microRNA sequences may stimulate self-renewal of these muscle cells. See Article p.376
Muscle cells of the mammalian heart exit the cell-division cycle shortly after birth and rarely divide again. Not surprisingly, therefore, after a heart attack — which can kill upwards of a billion of these cardiomyocytes — heart function is compromised, and this can lead to chronic heart failure or even sudden death. This dismal picture has prompted a long-standing search for ways to mend the damaged heart, with particular emphasis on understanding the reasons why cardiomyocytes do not re-enter the cell cycle. Eulalio et al.1 report on page 376 of this issue that certain microRNA sequences can stimulate division of adult cardiomyocytes. Their observations hint at the possibility of identifying the molecular regulators that maintain cardiomyocytes in the quiescent stateFootnote 1.
Over the past 30 years, the prevailing view that heart muscle cells are not replaced has gradually given way to one that there is a modest capacity for cardiomyocyte renewal, albeit outmatched by the loss of these cells after cardiac ischaemia (limited blood supply, and thus oxygen, to the heart) or other types of injury. Perhaps the most compelling evidence for cardiomyocyte regeneration comes from analysis of the incorporation of radioactive carbon-14 into human DNA that occurred worldwide during the era of above-ground testing of atomic weapons. These data2 showed that some 1% of cardiomyocytes in young adults are replaced annually.
Although the incidence of cardiomyocyte turnover declines with age, it might nonetheless contribute to repair after injury. What's more, in mice3,4 and humans5, a heart attack (myocardial infarction) seems to increase cardiomyocyte turnover within the preserved area of the heart bordering the region of damage, supporting the view that regeneration is an adaptive response to injury.
New muscle cells can come from two sources — from de novo differentiation of stem cells and from replication of pre-existing cardiomyocytes. Studies that tracked the fate of existing cardiomyocytes after injury in genetically engineered mice did not detect their replication, attributing renewal to de novo differentiation instead3. However, also published on Nature's website today, a study based on a sensitive, mass-spectroscopy technique confirms6 that a low level of replication occurs. Moreover, a number of soluble factors (including neuregulin7, periostin8 and fibroblast growth factor plus p38 inhibitors9) stimulate replication, thereby preserving cardiac function after ischaemic injury. This leads to the question of what normally prevents cardiomyocyte division.
To address this, Eulalio et al. conducted an unbiased search for microRNAs (miRNAs) that can induce re-entry of postnatal cardiomyocytes into the cell cycle. MicroRNAs are natural, single-stranded RNAs that are around 22 nucleotides long and bind directly to and suppress messenger RNA targets through imprecise base-pairing. Active miRNAs thus suppress expression of cellular proteins from mRNAs, and so their identification can provide insight into the protein regulators that keep cells in a quiescent state.
The authors' screen yielded 204 miRNAs that more than doubled the rate of division of postnatal rat cardiomyocytes (and more than 300 miRNAs that suppressed division). Of the 204, a minority (40) also functioned in mice, and Eulalio et al. selected two of these (dubbed hsa-miR-590-3p and hsa-miR-199a-3p) for further analysis. In the follow-up studies, they gathered impressive microscopy images to document mitotic cell division and even its final step, cytokinesis.
The researchers also tested the effect of the two miRNAs in vivo. When delivered directly to the rat heart — by means of a viral vector that is used in gene therapy — these miRNAs increased cardiomyocyte proliferation. Moreover, the two miRNAs boosted the normally ineffective process of heart repair after infarction in mice. Specifically, several parameters of cardiac function and structure improved over time, despite the initial loss of cardiac muscle due to the infarction.
These findings are notable because they not only add to evidence for the replicative capacity of cardiomyocytes, but also show that miRNAs can modulate cell regeneration in the heart. Furthermore, by overriding the cellular processes that maintain the quiescent state, they open the door to a broader, systems-level understanding of what prevents cardiomyocytes from re-entering the cell cycle and dividing. This information might enable the design of drugs to target some of the proteins whose expression is affected by these miRNAs, enhancing regeneration.
The miRNAs that Eulalio et al. tested did not affect the division of another class of cell in the heart, cardiac fibroblasts, suggesting that these regulatory sequences tap into fundamental mechanism(s) of cardiomyocyte differentiation and quiescence. It remains to be discovered whether the same miRNAs also mediate the regrowth of blood vessels in ischaemic tissue and/or blunt adverse changes that occur upon prolonged stress, such as a detrimental shift in energy utilization or a decline in contractility of the heart muscle. Moreover, it should be investigated whether boosting cardiomyocyte proliferation using miRNAs has undesirable consequences that might compromise heart function in the long term.
“The main strength of this study lies in its unbiased search of the miRNAs, and thus of the total cellular pool of proteins that they could have affected.”
In my opinion, the main strength of this study lies in its unbiased search of the miRNAs, and thus of the total cellular pool of proteins (the proteome) that they could have affected, in cardiomyocytes. Functional screening using synthetic miRNAs has also been performed previously to study heart development10 and disease-related increase in cardiomyocyte size (hypertrophy)11. Given that the 1,500 or so known human miRNAs control an estimated 60% of the proteome12, this approach is emerging as an effective means of gaining comprehensive insight into complex biology. Eulalio et al. used a library of about 900 synthetic miRNAs, thus probing a large swathe of the proteome. The miRNA hits that they got might not be the natural regulators of cardiomyocyte quiescence (using miRNAs with an antisense sequence in a similar screening is more likely to lead to miRNAs that sustain quiescence), but are nevertheless valuable because of the proteins they downregulate.
Unfortunately, identifying the target proteins of miRNAs is not easy. The binding of an miRNA to its cognate mRNA is not just a matter of base-pair recognition: sequences surrounding the recognition site and RNA-binding cofactors also influence site accessibility12. This makes computational approaches too error-prone to use as the sole means of target prediction. Eulalio et al. therefore profiled changes in the levels of mRNA gene transcripts in cardiomyocytes treated with hsa-miR-590-3p and hsa-miR-199a-3p. They found about 700 gene transcripts that showed decreased expression and 1,000 transcripts with enhanced expression; there was little overlap between the mRNAs affected by the two miRNAs.
The authors attempted to determine the importance of particular transcripts by reducing their expression using specific short interfering RNAs (siRNAs). Individual siRNAs only partially reproduced the miRNA effects, in agreement with the expectation that the miRNAs act on multiple cellular mRNA targets and, thus, that multiple proteins cooperate to maintain quiescence. Clearly, systems-level analysis of changes to the proteome is needed to further the understanding of cardiomyocyte quiescence.
Knowledge of the mechanisms that keep cardiomyocytes in a differentiated, non-dividing state will greatly aid the development of regenerative drugs. Ideally, such a drug would target an intracellular signalling pathway coordinating several processes that enable a cardiomyocyte to divide successfully, including cell-cycle entry, faithful DNA synthesis to avoid triggering apoptotic cell death, and disassembly and rebuilding of the contractile apparatus. This is a tall order. The present study is important because it takes a step towards achieving a systems-level understanding of what cardiomyocyte quiescence means, and addressing whether it can be overcome therapeutically.
*This article and the paper under discussion1 were published online on 5 December 2012.
Eulalio, A. et al. Nature 492, 376–381 (2012).
Bergmann, O. et al. Science 324, 98–102 (2009).
Hsieh, P. C. H. et al. Nature Med. 13, 970–974 (2007).
Olivetti, G. et al. Am. J. Pathol. 141, 227–239 (1992).
Quaini, F. et al. Circ. Res. 75, 1050–1063 (1994).
Senyo, S. E. et al. Nature http://dx.doi.org/nature11682 (2012).
Bersell, K., Arab, S., Haring, B. & Kühn, B. Cell 138, 257–270 (2009).
Kühn, B. et al. Nature Med. 13, 962–969 (2007).
Engel, F. B., Hsieh, P. C. H., Lee, R. T. & Keating, M. T. Proc. Natl Acad. Sci. USA 103, 15546–15551 (2006).
Colas, A. et al. Genes Dev. 26, http://dx.doi.org/10.1101/gad.200758.112 (2012).
Jentzsch, C. et al. J. Mol. Cell. Cardiol. 52, 13–20 (2012).
Bartel, D. P. Cell 136, 215–233 (2009).
About this article
European Journal of Heart Failure (2016)
Circulation Research (2013)
The Lancet (2013)