Fibroblast cells are known as key players in the repair of damaged heart structures. New findings show that injury also induces fibroblasts to become endothelial cells, helping to mend damaged blood vessels. See Article p.585
Heart attacks caused by a blockage in the coronary artery induce severe injury to cardiac muscle cells, leading to cell dysfunction and death. The damage elicits repair and regenerative responses that provoke the removal of dying cells and cell debris, recruit immune cells and initiate the formation of new blood vessels to recover blood supply. Fibroblast cells play a central part in this repair response. In this issue, Ubil et al.1 (page 585) describe how the plasticity of cardiac fibroblasts contributes to this process, by showing that fibroblasts can, in response to the activity of the transcription factor p53, convert into the endothelial cells that line the interior surface of blood vessels.
The heart consists of myocytes (muscle cells) and non-myocytes, which include cardiac fibroblasts and endothelial cells. The fibroblasts produce growth factors and extracellular matrix (ECM) proteins to maintain proper cardiac architecture, contraction and function2. They also interact with endothelial cells and myocytes to aid angiogenesis (blood-vessel formation) and maintain physiological homeostasis3. Following heart damage, cardiac fibroblasts are activated to produce ECM proteins and soluble factors to compensate for structural defects, contain the spread of damage, reinforce cardiac stiffness and prevent cardiac rupture. These activities, collectively referred to as fibrosis, aid in remodelling the heart musculature. Controlled fibrosis is crucial for restoring cardiac function after injury. However, excessive fibrosis is considered a pathological process that can lead to adverse effects, including reduced cardiac stiffness (diastolic dysfunction) and irregular electrical connectivity (arrhythmia).
Although endothelial cells in blood vessels are typically thought of as terminally differentiated cells, they can take on the characteristics of mesenchymal cells4,5, which are generally mobile cells surrounded by interstitial ECM proteins. During this endothelial-to-mesenchymal transition (EndMT), the endothelial cells lose the tight junctions that hold neighbouring cells together, and gain the ability to move, produce ECM proteins and contribute to excessive fibrosis, while also depleting functional capillaries and the endocardium tissue layer. EndMT is induced in cardiac endothelial cells by signalling pathways that depend on the growth factor TGF-β1 and that can be reversed by the activity of the protein BMP7 (ref. 4).
Now, Ubil and colleagues show that, following acute cardiac injury, fibroblasts (which belong to the mesenchymal cell lineage) can undergo a reverse conversion — from mesenchymal to endothelial cells (MEndT) — and become components of blood vessels (Fig. 1). To study this plasticity, they used mice in which cells that gain or lose expression of cell-type-specific markers can be tracked by fluorescence, a technique called genetic fate mapping.
The authors induced ischaemia-reperfusion injury by blocking the coronary artery and then restoring blood flow in the hearts of these mice. Three days later, they found that 35% of fibroblasts in the injury zone expressed the endothelial marker VECAD and were located in the interior of the vessel. Of these fibroblast-derived endothelial cells, 41% took up acetylated low-density lipoprotein, which is suggestive of endothelial-cell functionality. Most of the cells undergoing this transition expressed the fibroblast markers Col1α2 or FSP1, whereas very few expressed αSMA, a marker shared by a subset of cardiac fibroblasts (myofibroblasts) and the mesenchymal cells generated through EndMT (Fig. 1). This finding highlights a functional heterogeneity of recruited fibroblasts in injured cardiac tissue with respect to their plasticity. Understanding this heterogeneity will require future studies using fate mapping of myofibroblasts6.
Ubil and co-workers also found that the fibroblast-derived endothelial cells in the mice express increased levels of p53, a transcription factor known for its multiple functions, including regulation of the cell cycle, apoptotic cell death and DNA repair. To investigate the involvement of the p53 signalling pathway in the MEndT program, the authors studied cardiac fibroblasts cultured in vitro under serum deprivation — a stress condition that induces upregulation of p53. The cells formed tubular structures reminiscent of endothelial-cell architecture and expressed endothelial markers, including VECAD and transcription factors such as HoxA9 and HoxD3. However, the fibroblasts did not generate these tubules without serum deprivation, even when p53 was artificially overexpressed, suggesting that p53 expression alone is not sufficient to induce MEndT and that other signalling pathways are involved in launching this program.
The authors went on to show that treating mice with the small molecule RITA, which enhances p53 signalling, for three days after cardiac injury increased the number of fibroblast-derived endothelial cells. The treatment also accelerated angiogenesis and decreased cardiac fibrosis, leading to an improvement in cardiac function. These in vitro and in vivo findings suggest that p53 expression in fibroblast-derived endothelial cells has a key role in the recovery of cardiac function following injury.
Administration of other p53-activator molecules has previously been shown to impair angiogenesis, increase apoptotic cell death and cause dysfunctional muscle contraction in a mouse model of a condition called cardiac hypertrophy7. Conversely, Ubil et al. show no increase in myocyte p53 expression or the number of apoptotic cells after RITA treatment compared with control mice. These contrasting results could be explained by the different cardiac-injury models and drugs used, which will need to be considered in investigations of the potential clinical application of p53-inducing drugs for targeting cardiac fibrosis.
Nevertheless, Ubil and colleagues' study provides insight into the cardiac-repair process and highlights potential new therapeutic strategies. It also adds to the debate of whether the term 'terminally differentiated' in adult tissue might be too confining when cellular plasticity is rampant and seemingly functional in situations of disease or injury, just as in embryonic development.
Ubil, E. et al. Nature 514, 585–590 (2014).
Souders, C. A., Bowers, S. L. & Baudino, T. A. Circ. Res. 105, 1164–1176 (2009).
Kakkar, R. & Lee, R. T. Circ. Res. 106, 47–57 (2010).
Zeisberg, E. M. et al. Nature Med. 13, 952–961 (2007).
Von Gise, A. & Pu, W. T. Circ. Res. 110, 1628–1645 (2012).
LeBleu, V. S. et al. Nature Med. 19, 1047–1053 (2013).
Sano, M. et al. Nature 446, 444–448 (2007).
About this article
Journal of Crohn's and Colitis (2018)
Nature Reviews Cardiology (2018)
Journal of Clinical Investigation (2017)
Nature Reviews Cancer (2016)