CFs are the main contributors of cardiac fibrosis development3. The availability of human CFs is limited hampering the field to move forward. To date, CFs can be generated from iPSCs, which could provide an unlimited source of human CFs13,14,15. However, the behaviour of iPSC-CFs in relation to mechanical stimulation had not been investigated yet.

In this study we demonstrated that iPSC-CFs are comparable to primary CFs with regard to the expression of key CF markers at gene and protein levels. Expression of the cardiac markers GATA4 and TCF21 indicate the cardiac lineage of the cells. Furthermore, expression of the mesenchymal markers VIM and PDGFRA as well as the ECM component COL1A1 and the collagen binding receptor DDR2 support their fibroblast phenotype. In addition, we showed that iPSC-CFs respond to pro-fibrotic and mechanical stimulation. TGF- induces CFs transdifferentiation into myofibroblasts and promotes ECM remodelling. Mechanical stimulation in the form of cyclic stretch at physiological levels reduces collagen expression in iPSC-CFs. Interestingly, cyclic stretch also protects against TGF- stimulation, preventing the cells from transdifferentiating into myofibroblasts.

One can only use iPSC-derived cells when they accurately represent their primary counterparts. Key characteristics of CFs are a defined mRNA profile, responsiveness to pro-fibrotic cytokines, interaction with the ECM and mechanical sensitivity. iPSC-CFs generated using the protocol developed by Zhang et al. showed a comparable RNA sequencing profile in iPSC-CFs and primary CFs13. Using our iPSCs lines, following the same protocol we generated iPSC-CFs with an mRNA profile comparable to primary CFs. Furthermore, at a functional level we demonstrated that iPSC-CFs interact with their environment in a similar way as primary CFs, and respond to pro-fibrotic stimulation. These results indicate that iPSC-CFs possess several key characteristics of primary CFs and may be suitable to investigate the behaviour of CFs and develop disease models of cardiac fibrosis.

In order to investigate the behaviour of CFs in their native environment, we next investigated the behaviour of iPSC-CFs under physiologically relevant conditions. In an effort to mimic the dynamic environment of the continuously beating heart, we investigated the effects of cyclic mechanical stretch on iPSC-CFs. The importance of mechanical stimulation has been acknowledged, but the effects of mechanical stimulation on CFs remain controversial in in vitro studies23. On one hand, it has been reported that cyclic stretch may induce transdifferentiation of CFs into myofibroblasts. On the other hand, it has been shown that cyclic stretch may have a protective effect instead. One of the main factors influencing this controversy is the usage of cell sources from different species.

As primary human CFs are limited in availability, iPSC-CFs could provide a representative and stable source of cells to move forward. In order to study how iPSC-CFs and primary CFs behave in a mechanically dynamic environment similar to the heart, cells were exposed to 10% cyclic stretch at 1Hz for 72h19. With this approach, we demonstrated that: Cyclic stretch alone inhibits expression of collagen 1 but does not affect iPSC-CFs transdifferentiation or expression of matrix remodelling genes. In addition, cyclic stretch is protective against TGF- mediated myofibroblast transdifferentiation in iPSC-CFs, resulting in normalised expression of collagen 1, -SMA and matrix remodelling genes such as TIMP1 and MMP1.

The cause of the aforementioned controversy in literature regarding either the pro-fibrotic or anti-fibrotic response of CFs to mechanical stimulation is hard to pin-point; experimental conditions vary widely between studies, such as cell origin, the duration of the experiment, the surface coating and the presence of serum. A common trend in all those studies is that there may be a time-dependent response of stretch. It was shown in primary mouse CFs that the response starts with an initial increase in phosphorylation of AKT, a downstream kinase involved in the transduction of mechanical stimuli24,25. At the gene level, it was shown in primary rat CFs that there is an initial increase in fibrotic markers (i.e. ACTA2, TGFB1, CTGF) after 4 h followed by a reduced increase after 24h26. Roche et al. observed a similar effect in primary rat CFs with an apparent reduced increase of COL1A1 gene expression after 48h compared to 24h27. 72h of cyclic stretch was instead shown to inhibit TGF- induced fibroblast activation in primary human CFs16,18. Furthermore, it has been demonstrated that 96h of cyclic stretch can promote or inhibit the response of primary mouse CFs to a broad spectrum of biochemical stimuli, including TGF-, angiotensin II, interleukin-1 and others17. Overall, it appears that longer stimulation results in a gradual decrease of an initial pro-fibrotic response with eventually cells balancing the fibrotic response to the mechanically active environment in order to reach homeostasis. We may hypothesize that the duration of this response curve is dependent on different factors, including the origin and age of the cells, their culture conditions (surface coating, substrate stiffness, or medium supplementation with serum) and the presence of other cell types23. A clear association between mechanosensing and a response of CFs is apparent, but there is a need for a reproducible cell type to better understand this phenomenon.

TGF- signalling is one of the main pathways involved in the activation of CFs and development of cardiac fibrosis28. Exposure of iPSC-CFs to TGF- promotes the expression of fibrotic and myofibroblast markers, such as -SMA. When stretched however, this effect is diminished. How mechanical changes communicate with the TGF- pathway is not well understood. On one hand, mechanical strain has been shown in tissue to release active TGF- from the ECM, which would promote fibroblast activation29. On the other hand, in this in vitro study mechanical strain appears to inhibit fibroblast activation, indicating that there may be other mechanisms at play in this model. It is unknown whether this anti-fibrotic effect is directly caused by interplay between mechanosensitive complexes and the TGF- pathway. Mechanosensitive receptors such as integrins or mechanoresponsive factors such as YAP/TAZ may communicate with the TGF- pathway30,31. Alternatively, cyclic stretch may have an indirect effect, for example through internalization of extracellular receptors, altering the response to ligand stimulation. Regardless, the field of mechanotransduction in CFs remains requires further investigation.

While iPSCs have started a new era of research, the usage of these cells comes with limitations. iPSC-CFs showed many similarities with primary CFs, but the maturity of iPSC-derived cell lineages remains an important topic of contention. Although maturation is clearly defined for some cell types, such as cardiac myocytes, a clear definition lacks for CFs. The heterogeneity and plasticity of this cellular population under physiological conditions makes it difficult to set well defined standards of mature CFs32. iPSC-CFs present with various characteristics of primary cells, but they differ in several aspects as well. For example, Zhang et al. noted an increased proliferation capacity in iPSC-CFs and foetal CFs compared to adult CFs, indicating the iPSC-CFs may be more foetal-like13. This increased proliferation capacity and ability to stay in an inactivated state while in culture increases the applicability of the iPSC-CFs in research, as it has been demonstrated that CFs which have transdifferentiated into myofibroblast will have an altered response to mechanical stimulation33. In addition, little is known about the electrophysiological characteristics of iPSC-CFs and their interaction with other conducting cells such as cardiomyocytes34. Further electrophysiological characterisation should be performed to better understand the behaviour of these cell in the electrical circuit of the heart.

To conclude, in this study we demonstrated that iPSC-derived CFs show similar gene and protein expression as primary CFs. In addition, pro-fibrotic stimulation promoted transdifferentiation of iPSC-CFs into a myofibroblast phenotype. When stimulated with cyclic stretch, this transdifferentiation is inhibited. Together, the mechano- and TGF--responsive characteristics support the use of iPSC-CFs for physiological relevant disease modelling. Future studies could further dive into the mechanisms driving cardiac fibroblast behaviour and cardiac fibrosis.

Read more from the original source:
Mechanical stimulation of induced pluripotent stem derived cardiac fibroblasts | Scientific Reports - Nature.com