Progressive cardiac fibrosis is a major complication in heart disease and new efficacious anti-fibrotic drugs are urgently needed. Drug repurposing offers a possibility to accomplish this task faster. One example is PFD, which has been used for the treatment of IPF in Europe since 2011 and is now being tested for its anti-fibrotic action in patients with HFpEF. First published results with 39 patients in the treatment group showed a 1.2% reduction in extracellular volume after 52 days of treatment (Lewis et al., 2021).
Despite this significant result, considering anti-fibrotic drugs for general use with every fibrotic disease might be critical, because it ignores tissue- and cell type-specific characteristics. Therefore, more basic research regarding the effects of PFD on human cardiac cells is necessary, as the specific mechanisms of action of PFD are still unclear. One prominent example, which underlines the need for more research on human cells, is the reported consistent and inconsistent regulation of the myofibroblast marker SMA by PFD in human lung and cardiac fibroblasts, respectively (Molina-Molina et al., 2018, Jin et al., 2019, Palano et al., 2020, Widjaja et al., 2021, Bracco Gartner et al., 2022).
Initially, we were also interested in the PFD-dependent regulation of fibrosis-associated gene expression in human CF, however, in preliminary experiments we could not detect substantial changes in SMA and pro-collagen expression (data not shown). Therefore, we focused on the not yet investigated anti-mitogenic effect of PFD in human CF. By a concentration-response analysis, we determined an anti-proliferative IC50 of 0.43 mg/ml for PFD in 2D-cultured human CF, which is in a similar range as demonstrated for rat CF and other fibroblast types (Lin et al., 2009, Shi et al., 2011, Tao et al., 2020). Moreover, we found that 1 mg/ml PFD resulted in cell number stagnation and 3 mg/ml eliminated all seeded human CF within two days. For comparison, we used highly proliferative, immortalized tsA201 cells and found that these cells were less sensitive to PFD, as indicated by the higher IC50 and lower cytotoxicity. This data indicates that not all cell types display the same sensitivity to PFD, which could explain, e.g., the discrepancy between published cytotoxic PFD concentrations (Shi et al., 2011, Mediavilla-Varela et al., 2016).
We further investigated the effect of PFD on the TGF-β1-induced SMAD signaling, as well as on the central mitogenic MEK1/2-ERK1/2 cascade. We found that PFD was without effect on basal SMAD phosphorylation and inhibited moderately SMAD2, but not SMAD3 phosphorylation in the presence of TGF-β1. In contrast, MEK1/2 and ERK1/2 phosphorylation was prominently reduced under basal conditions, indicating that the mitogenic signaling is more sensitive to PFD than the basal pro-fibrotic signaling in the 2D cultured myofibroblasts. Of note, we could not detect any significant influence on the phosphorylation of both kinases by TGF-β1 as demonstrated by others for ERK1/2. We believe that this is due to the different experimental conditions as Widjaja and co-workers starved the cells overnight, which we did not. (Widjaja et al., 2021). We further show that PFD inhibits the phosphorylation of rpS6, which is located down-stream of the MEK1/2-ERK1/2 pathway and phosphorylated by p90 ribosomal S6 kinases (Carriere et al., 2008). Although the role of rpS6 phosphorylation is still not fully clear, it is discussed to play a role in mRNA translation, determination of cell size, and glucose homeostasis (Ruvinsky et al., 2005, Bohlen et al., 2021). Further studies on PFD might investigate if one of these processes is impaired by PFD. Moreover, it would be interesting to see if canonical up-stream mediators of MEK1/2, like Raf kinase, are also influenced by PFD.
In addition to our 2D proliferation analysis, we studied the effect of PFD on the cell number, viability, and cell cycle activity of human CF in 3D cultures. Importantly, we were interested if the PFD effect is dependent on the CF phenotype. Therefore, we generated ECT with different geometries, because we have recently shown that this is an effective way to influence the cellular phenotype in 3D. In brief, the uniform model allows for the cells to be exposed to a higher mechanical constraint as the ECT is in full contact with a central rod, whereas the ECT is restricted to the two poles in the non-uniform model. Consequently, the cells adapt with a less pronounced myofibroblast phenotype in the non-uniform ECT than in the uniform ECT. One sign of this differential adaptation process was the better survival of the human CF in the uniform model, which we could reproduce in this study (Santos et al., 2022).
Importantly, we found that PFD treatment affected the cell number, viability, and cell cycle activity of human CF in both models to a similar extent. We show that 1 mg/ml PFD further reduced the cell number by 40 to 50% in both models and cell viability as well as cell cycle activity were similarly impaired, suggesting that PFD does not act differentially on different CF phenotypes. This also indicates that the cytotoxic effect of PFD appears to be more prominent in cases where the rate of cell death surpasses cell proliferation as in 3D, but not in 2D culture (Santos et al., 2022). We further believe that the sudden stagnation in non-uniform ECT contraction after 2 days of treatment with PFD is not a sign of an impaired contractile ability of the embedded cells, but a consequence of the enhanced cell loss. We have shown before that inhibitors acting directly on the cell’s contractile machinery, like Latrunculin A, reduce ECT contraction immediately and continuously and not delayed and abruptly as found for PFD (Santos et al., 2021). Similarly, the PFD-dependent inhibition of ECT compaction and stiffening might be a consequence of the enhanced loss of cells. Thus, our data indicate that the anti-fibrotic effects of PFD in our ECT model merely resulted from its anti-proliferative and/or cytotoxic activity. Moreover, PFD effects were relatively similar in both of our ECT models, and together with the demonstrated anti-proliferative activity in an immortalized cell line, its action on proliferating cells can be considered highly non-cell type-specific. Thus, it is not surprising that PFD is nowadays also discussed as a potential anti-cancer therapeutic (Paliogiannis et al., 2021).