Biochemical and biomechanical regulation of the myofibroblast phenotype
Piersma, Bram
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2017
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Piersma, B. (2017). Biochemical and biomechanical regulation of the myofibroblast phenotype: focus on Hippo and TGFβ signaling. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
CHAPTER
7
7
The extracellular matrix (ECM) is a dynamic collection of proteins and carbohydrates, which are constantly synthesized, modified, and degraded by cells. Under normal conditions, the dynamic ECM turnover is in balance and the structure and composition of the matrix remain relatively constant. However, in chronic disease, matrix synthesis and degradation is out of balance and infiltrating or activated cells alter the composition of the matrix, and matrix cross-linking changes. The replacement of normal functional ECM with collagenous scar tissue causes distortion of tissue architecture by contraction of the connective tissue. This also changes the mechanical properties of the tissue, but that is an area for future work, in part because the basic properties of matrix are now beginning to be understood in simple systems.
The functional importance of the ECM becomes apparent when looking at the wide range of disorders that are caused by mutations in genes that encode ECM components1.
Although ECM related disorders are widespread in the population, the general public is often not familiar with their existence. This is in part due to the fact that ECM related disorders often present as secondary ailments or do not present themselves at all, only to become apparent at the time an organ or organ system fails to fulfil its function. The relatively slow and silent development of ECM related disorders—often termed fibrotic disorders or fibrosis—is what make them so deadly. Worldwide, the prevalence and incidence of fibrotic disorders are rising at an alarming rate: it has been estimated that about 45% of all deaths in developed countries are the result of fibrotic disorders2.
This coincides with an estimated annual economic burden of over US$300 billion in the United States alone. Despite the significant contribution to worldwide mortality, the mechanisms underlying fibrosis remain poorly understood.
Fibrotic disorders can affect many organ systems, including the lungs, liver, kidneys, heart, skin, fascia, gut, and eyes, ultimately resulting in organ failure3. Because each
organ or organ system has a different purpose, is constructed by a unique set of ECM components, and consists of different cell types, the fibrotic disorders in each organ all have different characteristic causes and consequences. There are also, however, universal mechanisms shared by fibrotic pathologies among multiple organs. It is therefore of paramount importance to unravel both the general and organ or cell type specific mechanisms that drive the fibrotic response. In this thesis we set out to understand part of the biochemical and biomechanical signaling that regulate the key cell phenotype responsible for the development of fibrosis: the myofibroblast. We provide new insights into the biology of myofibroblasts and present potential therapeutic avenues for the treatment of fibrosis and other ECM related disorders.
Understanding the myofibroblast phenotype
The main culprit in the development of fibrotic disorders is the myofibroblast. Myofibroblasts cannot be identified as a single cell type, but rather represent a phenotype or state that can be adopted by a variety of cells, including fibroblasts, pericytes, stellate cells, and according to some even epithelial and endothelial cells2,4. The promiscuous nature of the myofibroblast makes it difficult to generate
the myofibroblast are well known can be readily translated from in vitro studies to the in vivo situation. The formation of a well-developed cytoskeleton is one of the key characteristics that describe a myofibroblast, which allows them to contract the surrounding ECM via integrin adhesions. Especially the expression of the smooth muscle isoform of actin, αSMA, is required for myofibroblast contractility5,6. However,
because of the crucial role actin filaments play in a wide variety of cellular processes, therapeutic targeting of actins is not possible.
The second hallmark of myofibroblasts is their capability to synthesize enormous amounts of ECM components, in response to growth factors, cytokines, and increased tissue stiffness4. TGFβ signaling has been put forward as a key signaling pathway
in driving the myofibroblast phenotype switch. Canonical TGFβ signaling involves activation of the receptor pair, recruitment and phosphorylation of Smad transcription factors, and subsequent Smad nuclear accumulation and transcriptional activation of genes containing a so called Smad binding element (SBE). Although a large body of knowledge exists on these signaling cascades, therapeutic targeting of myofibroblasts and fibrosis remains challenging. As of to date only a handful of anti-fibrotic therapies exists, many of which cast a shadow of doubt with respect to efficacy and specificity7,8.
In this thesis we focused on the mechanisms underlying the TGFβ-mediated myofibroblast phenotype switch. We started out to investigate a signaling cascade that is crucial for the regulation of cell proliferation, organ size and apoptosis: the Hippo signaling pathway. We found that one of the members of the hippo signaling cascade, YAP, a mechanosensitive transcriptional coactivator, which is activated by stiffness-induced ECM signaling, can also be activated by TGFβ1 (Chapter 2). Moreover, we demonstrated that YAP amplifies the induction of a TGFβ-induced myofibroblast phenotype. This coincides with other reports that have studied YAP and its role in fibrosis or other ECM related disorders9–12.Moreover, we could corroborate our findings
in normal dermal fibroblasts to the fibroproliferative disorder termed Dupuytren disease. We found elevated YAP levels and increased YAP nuclear localization in clinical biopsies, suggesting that YAP could be a potential target for the treatment of Dupuytren disease. Dupuytren disease is a common disorder in people from Caucasian descent on the northern hemisphere and observations from several studies indicate that Dupuytren disease has a strong genetic component13. Especially the WNT signaling
pathway has been implicated in the susceptibility and development of Dupuytren disease14,15. Interestingly, none of the genetic studies hints at an involvement of hippo/
YAP signaling, although Smad3 has been found to be enriched in diseased samples14.
This could be explained by the fact that elevated levels of YAP per se do not necessarily contribute to increased YAP-dependent signaling16. However, caution should be taken
when using YAP nuclear localization or phosphorylation status as only measure of YAP activity, because these features do not necessarily imply YAP activation and thus physiological importance17. Therefore, when studying YAP mediated transcription, one
should always include transcriptional activity assays or chromatin precipitation assays. Moreover, our investigation only focused on resected tissues of patients receiving surgery at the end-stage of the disease, not taking into account the onset of Dupuytren
7
disease. Thus, whether YAP activation and subsequent mechanosignaling is a cause, or consequence of Dupuyten disease remains to be elucidated. Hence, I suggest further and more detailed investigation of YAP in the pathogenesis of Dupuytren disease, including, if possible, the early stages of the disease.
Recently, many investigators have taken interest in the study of the hippo signaling cascade in a variety of disorders, including cancer and fibrosis16. It is of interest that
both YAP, and its paralog, transcriptional coactivator with PDZ-binding motif (TAZ), communicate either directly or indirectly with other signaling cascades (Chapter
3). An example of these molecular interactions is YAP/TAZ sequestration within the
β-catenin destruction complex, and the regulation by WNT signaling18. This seminal
finding bridges the genetic studies with our own in Dupuytren disease fibroblasts and clinical biopsies. We hypothesize that deregulated WNT signaling in Dupuytren myofibroblasts destabilizes the β-catenin destruction complex, rendering YAP and TAZ free to translocate to the nucleus. This coincides with results from WNT signaling in Duputren’s fibroblasts, where β-catenin was found to be localized primarily in the nucleus19. Similar findings were reported for the interaction of YAP/TAZ with
TGFβ signaling in epithelial cells20,21. We could corroborate these findings in dermal
fibroblasts when we furthered our investigation by delineating the mechanism by which TGFβ1 induces nuclear accumulation of YAP (Chapter 4). We found that TGFβ1-induced nuclear accumulation of YAP is dependent on actin polymerization and myosin II-mediated contractility. Moreover, we demonstrated that YAP complexes with Smad2 and Smad3 and that Smad3 is a limiting factor for nuclear accumulation of YAP upon TGFβ1 exposure. Vice versa, YAP is not necessary for the nuclear accumulation of Smads. Because mechanically-induced YAP activation also relies on polymerization of the actin cytoskeleton, we think that TGFβ1 induces both actin polymerization as well as Smad-mediated translocation. These data uncover a complex mechanism of YAP regulation and refine the current models that describe the TGFβ1-induced myofibroblast phenotype. Additionally, we found that the benzoporphyrin derivate verteporfin is not a YAP specific inhibitor, but also targets the nuclear accumulation of Smad2 and Smad3. These features cannot be attributed to the YAP-inhibiting actions of verteporfin, because Smad nuclear accumulation is YAP-independent. Interestingly, the number of YAP/Smad interactions were not affected by verteporfin treatment, suggesting that verteporfin acts on specific residues in Smad2 and Smad3 which are necessary for nuclear anchorage, or on other proteins that regulate nuclear anchorage of Smads . Because total Smad protein levels were also affected by verteporfin treatment, it is conceivable to think that verteporfin changes the phosphorylation status of either the C-terminal MH2 domain or the central linker-region. Phosphorylation of serine and threonine residues in the Smad linker-region mediates the interaction with prolyl-isomerases and ubiquitin ligases22. Changes in phosphorylation status of Smads could
thereby lead to increased susceptibly to cytoplasmic retention, polyubiquination and proteasomal degradation. Thus, we identified a novel function of the clinically approved agent verteporfin, which may prove an interesting candidate for anti-fibrotic therapy.
In Chapter 5, we investigated another part of the cytoskeleton in relation to the myofibroblast phenotype. Spectrins are proteins that form tetrameric heterodimers that connect the plasma membrane to the actin cytoskeletal apparatus. Tang et al. demonstrated that βII-spectrin deficiency results in disruption of normal Smad3 and Smad4 localization and subsequent TGFβ signaling. Moreover, it was found that spectrins associate with, and regulate YAP function23,24, suggesting a link between
spectrins, TGFβ1, and YAP in myofibroblast biology. Against our expectations, we found that spectrin deficiency neither altered cell spreading on soft and stiff ECM substrates, nuclear accumulation of YAP, nor the capability of fibroblasts to adopt a myofibroblast phenotype in response to TGFβ1 exposure. These findings perplexed us, as several labs demonstrated how both the cytoskeleton and YAP can be regulated by spectrins. It seems, that in fibroblasts spectrins play a different role, when compared to epithelial cells, possibly because epithelial cells rely heavily on the maintenance of cell polarity, a process that involves spectrins. Furthermore, TGFβ1 is known for its different functions in a variety of cell types. In epithelial cells, it attenuates proliferation, whereas it can promote proliferation of fibroblasts25. These cell type specific functions of both
spectrins and TGFβ1 may explain our conflicting results. Additionally, we found that TGFβ1 exposure actually decreased expression of both αII- and βII spectrins, suggesting that myofibroblasts no not require high spectrin levels on maturation. Future studies should therefore focus on overexpression of spectrins to fully comprehend their functions in myofibroblasts.
Collagen synthesis is one of the hallmarks of myofibroblasts and fibrosis, and relies on a set of complex post-translational modifications, including hydroxylation of lysyl residues of the nascent collagen α-chains. Ascorbic acid, normally termed vitamin C, acts as co-factor for several of the enzymatic reactions that are required to form a mature collagen triple helix26. Humans have lost the ability to synthesize
ascorbic acid, due to mutations in the enzyme L-gulono-γ-lactone oxidase and therefore rely on ascorbic acid uptake from the diet27. For this reason, it is of uttermost
importance that scientists include ascorbic acid in their cell culture regimens when investigating collagen biosynthesis or fibrosis in human cells, something that is often neglected. In Chapter 6 we focused on the role of ascorbic acid in the myofibroblast phenotypical switch, because it has been reported that ascorbic acid is required in several other enzymatic reactions besides collagen hydroxylation28. We found that
ascorbic acid amplifies gene and protein expression of multiple genes involved in the pathogenesis of fibrosis, including ACTA2, CCN2, COL1A1, and COL4A1. Moreover, we could show that ascorbic acid is necessary for proper contraction of a collagen gel. In contrast to previous findings we could not demonstrate altered Smad2/3 signaling in the presence of ascorbic acid compared to the control situation. This could be explained by cells originating from different organ systems28. We think that
some of the effects that we observed may be attributed to the epigenetic changes induced by the addition of ascorbic acid to our culture conditions. Emerging evidence demonstrated that ascorbic acid influences several epigenetic processes, including demethylation of cytosine in the DNA and lysine residues in histones. Both DNA methylation and methylation of histone tails are thought to be extremely important
7
in the organization of chromatin, thereby modulating the accessibility of the DNA for gene regulation29,30. The two enzyme families implicated in the ascorbic acid-mediated
epigenetic modulation are the Ten-eleven translocation dioxygensases (TET) and the JmjC-domain containing histone demethylases (JmjCs). TETs catalyze the oxidation of 5-methyl cytosine (5mC) to 5-hydroxymethyl cytosine (5hmC), thereby paving the way for further modification into 5-formylcytosine and subsequent 5-carboxylcytosine, eventually resulting in the demethylation of a cytosine31. JmjC histone demethylases
function in the demethylation of mono- di- and tri-methylated lysine residues. When ascorbic acid is bound to the JmjC domain, it is thought to hydroxylate the histone lysine-methyl group, which is then spontaneously reduced to unmethylated lysine by removal of formaldehyde32. Future investigations should therefore focus on the
epigenetic changes that occur with regards to the modulation of pro-fibrotic genes.
Making cultural changes
One of the questions to which an answer has eluded scientists and clinicians for decennia is the following: why are results from cell culture experiments often so hard to replicate in the clinic? In other words, why do drugs work in a dish but not in a patient? One of the answers can be found in the way we set up our cell culture experiments. In the last decade, it has become evident that ECM and cell mechanics play an important role in the regulation of cell function and fate16,33. Tissue mechanics
therefore have a profound effect on the results obtained in daily cell culture routine. By keeping our cells in a 2D environment and an extreme stiff substrate to grow on, data that we obtain may not give correct answers to the questions we are asking. In the fields of developmental biology, stem cell research and cancer research, multiple investigators have shown that non-physiological culture conditions give significant different results than, for example using specific ECM components as culture substrate, 3D organoid culture systems, and cultures mimicking the stiffness of human organs in health and disease. In Chapter 2 and Chapter 4 we described how the mechanosensitive transcriptional co-activator YAP, is activated by substrate stiffness in a 2D setting. These data corroborate the notion that stiffness alone, induces significant changes in cell function and differentiation. However, 3D cultures or ex vivo culture systems should provide additional insight and confirmation of our findings. Hence, the factors described above should therefore be taken into account for all cell culture experiments. This shift in paradigm is a formidable challenge with respect to general laboratory practice and could pose financial burden of scientific research. However, in the long run, such changes will result in increased reliability of scientific data and a reduced drug attrition rate in the pharmaceutical pipeline45.
Another matter is the addition of ascorbic acid to our cell culture media. We demonstrated in Chapter 6 that ascorbic acid changed the myofibroblast phenotype, by amplifying expression of multiple pro-fibrotic genes. We hypothesize that this is in part due to the epigenetic changes that occur through the actions of TET and Jmj domain containing enzymes. By using culture medium deficient of ascorbic acid, many scientists may be analyzing data that would have been dramatically different if they had included the vitamin. Examples of the impact of ascorbic acid supplementation
can be found in the fields of bone research, neurosciences, and cancer research34–36. In
the past, addition of ascorbic acid was a cumbersome substance to work with, because the instable compound is readily metabolized into the biological inactive form 2,3- diketo-L-gulonic acid37. More recently, a more stable variant of ascorbic acid has been
introduced, which, in basic or neutral solutions remains stable for up to three days37.
Thus, we strongly suggest that ascorbic acid, in the form of 2-phosphate ascorbic acid, becomes a standard ingredient in all culture media formulations for human cells.
Therapeutic targeting of the myofibroblast
Many therapeutics against fibrosis target the end-product of these disorders: the ECM itself7,38. Although successful in pre-clinical models39, a phase II clinical trial using the
monoclonal antibody against the extracellular copper-dependent amine oxidase LOXL2, failed to show clinical efficacy in the setting of liver fibrosis and pancreatic
adenocarcinoma, a tumor type characterized by a well-developed fibrotic stroma40
(Gilead press release). So, why do many of the drugs that target the ECM fail to prove clinical efficacy? Answers to this question lie in the temporal regulation of ECM cross linking. The bulk of both enzymatic and non-enzymatic cross-linking occur over months or years after the ECM components have been deposited41. However, if successful, this
strategy only relieves the patient of the ECM buildup in tissues, but does not take into account the culprits of the disease, the myofibroblasts that produce and remodel the vast amount of ECM components. Although the challenge in targeting myofibroblasts is formidable, patients would be better served with anti-myofibroblast therapies in combination with drugs that target the production and cross-linking of the ECM. Multiple compounds that target the myofibroblast phenotype have been tested in both pre-clinical studies and clinical trials, but here too, many fail to provide clinically relevant results7,8. More specifically, targeting of the pro-fibrotic TGFβ signaling
cascade has provided mixed results, many of which can be attributed to the pleiotropic effects of TGFβ42. However, recent findings show that targeting of circulating TGFβ
with fresulimumab seems promising for the treatment of systemic sclerosis43. Other
options for the targeting of the myofibroblast phenotype may lie at crosstalk between different signaling cascades, such as TGFβ, WNT, and Hippo. Specifically modulating the interaction between specific transcriptional modulators, could provide clinically efficacy, without shutting down a complete signaling cascade at the level of the ligand or receptor.
In Chapter 4 we provide data on the effect of verteporfin on the inhibition of YAP/ Smad signaling and the expression of signature myofibroblast genes. Similar findings were obtained in a study on renal fibrogenesis, where the authors infused verteporfin into mice subjected to unilateral ureteral obstruction12. This model uses ligation of
the ureter to create urine backflow and increased pelvic and tubular fluid pressure, resulting in tubular epithelial apoptosis and subsequent activation of myofibroblasts. Although a crude model for fibrosis, the authors elegantly demonstrated that verteporfin reduced the levels of αSMA-positive cells, the amount of total collagen, and collagen type IV levels. Of note, in rat fibroblasts, verteporfin induced dramatic reduction of YAP and TAZ levels, whereas we did not find changes in YAP, highlighting
7
the possible differences of verteporfin treatment in rodents and humans. The study from Szeto et al. emphasizes the potential of verteporfin as anti-fibrotic therapy, and together with our findings demand thorough pre-clinical and drug safety studies. Next to organ fibrosis, several cancers, such as pancreatic adenocarcinoma (PDAC) display aberrant activation of myofibroblasts and excessive deposition of ECM. PDAC is a cancer of the pancreas which has an abysmal prognosis at the time of diagnosis, due to the aggressiveness and high metastatic burden. Treatment of the solid tumor is hampered by the well-developed fibrotic stroma: stiff connective tissue consisting of fibroblasts, stellate cells and cross-linked ECM44. PDAC fibrosis is thought to promote
tumor growth and favor chemotherapy resistance through impeded circulation and drug penetration. Therefore, we propose the investigation of verteporfin as potential anti-fibrotic therapy in fibrotic tumors such as PDAC. However, care must be taken with the use of verteporfin, because inhibition of TGFβ signaling in tumor epithelial cells could lead to adverse effects42.
Concluding remarks
With this thesis we aimed to elucidate how biomechanical and biochemical cues regulate the myofibroblast phenotype and fibrosis. To this end, we focused on three main areas of research. First, unraveling the functions of YAP signaling in the myofibroblast phenotype and the fibroproliferative disorder Dupuytren disease. Second, delineate the functions of spectrin proteins in fibroblast spreading, wound healing, and the TGFβ1-induced myofibroblast phenotype. Third, investigate how ascorbic acid modulates ECM synthesis and myofibroblast contraction.
Understanding how myofibroblasts transduce biomechanical and biochemical cues to modulate the regulation of gene transcription is of paramount importance for the development of anti-fibrotic therapeutics. With this thesis we provide novel insights in how YAP regulates the myofibroblast phenotype, by interacting with Smad transcription factors. These findings change the classical dogma of TGFβ1 signaling in fibrosis. Moreover, the finding that the drug verteporfin antagonizes myofibroblast activation, provides ample support for the initiation of pre-clinical studies that target the myofibroblast.
In conclusion, our findings do not only provide a better understanding of the myofibroblast phenotype and fibrosis, but also provide support for changes in general laboratory practice and tools for the possible treatment of fibrosis and cancer.
REFERENCES
1. Bateman, J. F., Boot-Handford, R. P. & Lamandé, S. R. Genetic diseases of connective tissues: cellular and extracellular effects of ECM mutations. Nat. Rev. Genet. 10, 173–83 (2009).
2. Wynn. Cellular and molecular mechanisms of fibrosis. The Journal of Pathology 214, 199–210 (2008). 3. Rockey, D. C., Bell, P. D. & Hill, J. A. Fibrosis--a common pathway to organ injury and failure. N. Engl. J.
Med. 372, 1138–49 (2015).
4. Hinz, B. The myofibroblast: Paradigm for a mechanically active cell. J Biomech 43, 146–155 (2010). 5. Hinz, Celetta, Tomasek, Gabbiani & Chaponnier. Alpha-smooth muscle actin expression upregulates
fibroblast contractile activity. Mol Biol Cell 12, 2730–41 (2001).
6. Goffin, J. et al. Focal adhesion size controls tension-dependent recruitment of α-smooth muscle actin to stress fibers. The Journal of Cell Biology 172, 259–268 (2006).
7. Friedman, S., Sheppard, D., Duffield, J. & Violette, S. Therapy for Fibrotic Diseases: Nearing the Starting Line. Sci Transl Medicine 5, 167sr1–167sr1 (2013).
8. Gourdie, R. G., Dimmeler, S. & Kohl, P. Novel therapeutic strategies targeting fibroblasts and fibrosis in heart disease. Nat Rev Drug Discov 15, 620–38 (2016).
9. Fujii, M. et al. Convergent signaling in the regulation of connective tissue growth factor in malignant mesothelioma: TGFβ signaling and defects in the Hippo signaling cascade. Cell Cycle 11, 3373–9 (2012).
10. Liu, F. et al. Mechanosignaling through YAP and TAZ drives fibroblast activation and fibrosis. Am. J. Physiol. Lung Cell Mol. Physiol. 308, L344–57 (2015).
11. Mannaerts, I. et al. The Hippo pathway effector YAP controls mouse hepatic stellate cell activation. J. Hepatol. 63, 679–88 (2015).
12. Szeto, S. G. et al. YAP/TAZ Are Mechanoregulators of TGF-β-Smad Signaling and Renal Fibrogenesis. J. Am. Soc. Nephrol. (2016). doi:10.1681/ASN.2015050499
13. Shih, B. & Bayat, A. Scientific understanding and clinical management of Dupuytren disease. Nature Reviews Rheumatology 6, 715–26 (2010).
14. Becker, K. et al. Meta-Analysis of Genome-Wide Association Studies and Network Analysis-Based Integration with Gene Expression Data Identify New Suggestive Loci and Unravel a Wnt-Centric Network Associated with Dupuytren’s Disease. PLoS ONE 11, e0158101 (2016).
15. Dolmans, G. H. et al. Wnt signaling and Dupuytren’s disease. N. Engl. J. Med. 365, 307–17 (2011). 16. Piccolo, S., Dupont, S. & Cordenonsi, M. The biology of YAP/TAZ: hippo signaling and beyond. Physiol.
Rev. 94, 1287–312 (2014).
17. Chen, Q. et al. Homeostatic control of Hippo signaling activity revealed by an endogenous activating mutation in YAP. Genes Dev. 29, 1285–97 (2015).
18. Azzolin, L. et al. YAP/TAZ incorporation in the β-catenin destruction complex orchestrates the Wnt response. Cell 158, 157–70 (2014).
19. Van Beuge, M. M., Ten Dam, E.-J. P. M. J., Werker, P. M. & Bank, R. A. Wnt pathway in Dupuytren disease: connecting profibrotic signals. Transl Res 166, 762–771.e3 (2015).
20. Varelas, X. et al. The Crumbs Complex Couples Cell Density Sensing to Hippo-Dependent Control of the TGF-β-SMAD Pathway. Developmental Cell 19, 831–844 (2010).
21. Varelas, X. et al. TAZ controls Smad nucleocytoplasmic shuttling and regulates human embryonic stem-cell self-renewal. Nature Cell Biology 10, 837–848 (2008).
22. Burch, M. L., Zheng, W. & Little, P. J. Smad linker region phosphorylation in the regulation of extracellular matrix synthesis. Cell. Mol. Life Sci. 68, 97–107 (2011).
23. Fletcher, G. et al. The Spectrin cytoskeleton regulates the Hippo signalling pathway. The EMBO journal
34, 940–54 (2015).
24. Wong, K. K. et al. β-Spectrin regulates the hippo signaling pathway and modulates the basal actin network. J. Biol. Chem. 290, 6397–407 (2015).
7
25. Strutz et al. TGF-beta 1 induces proliferation in human renal fibroblasts via induction of basic fibroblastgrowth factor (FGF-2). Kidney international 59, 579–92 (2001).
26. Traber, M. G. & Stevens, J. F. Vitamins C and E: beneficial effects from a mechanistic perspective. Free Radic. Biol. Med. 51, 1000–13 (2011).
27. Drouin, G., Godin, J.-R. R. & Pagé, B. The genetics of vitamin C loss in vertebrates. Curr. Genomics 12, 371–8 (2011).
28. Park, J.-K. K. et al. Vitamin C deficiency attenuates liver fibrosis by way of up-regulated peroxisome proliferator-activated receptor-gamma expression in senescence marker protein 30 knockout mice. Hepatology 51, 1766–77 (2010).
29. Schübeler, D. Function and information content of DNA methylation. Nature 517, 321–6 (2015). 30. Greer, E. L. & Shi, Y. Histone methylation: a dynamic mark in health, disease and inheritance. Nat. Rev.
Genet. 13, 343–57 (2012).
31. Tahiliani, M. et al. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 324, 930–5 (2009).
32. Tsukada, Y. et al. Histone demethylation by a family of JmjC domain-containing proteins. Nature 439, 811–6 (2006).
33. Engler, A. J., Sen, S., Sweeney, H. L. & Discher, D. E. Matrix elasticity directs stem cell lineage specification. Cell 126, 677–89 (2006).
34. Aghajanian, P., Hall, S., Wongworawat, M. D. & Mohan, S. The Roles and Mechanisms of Actions of Vitamin C in Bone: New Developments. J. Bone Miner. Res. 30, 1945–55 (2015).
35. Covarrubias-Pinto, A., Acuña, A. I. I., Beltrán, F. A., Torres-Díaz, L. & Castro, M. A. Old Things New View: Ascorbic Acid Protects the Brain in Neurodegenerative Disorders. Int J Mol Sci 16, 28194–217 (2015). 36. Solary, E., Bernard, O. A., Tefferi, A., Fuks, F. & Vainchenker, W. The Ten-Eleven Translocation-2 (TET2)
gene in hematopoiesis and hematopoietic diseases. Leukemia 28, 485–96 (2014).
37. Austria, R., Semenzato, A. & Bettero, A. Stability of vitamin C derivatives in solution and topical formulations. J Pharm Biomed Anal 15, 795–801 (1997).
38. Bonnans, C., Chou, J. & Werb, Z. Remodelling the extracellular matrix in development and disease. Nat. Rev. Mol. Cell Biol. 15, 786–801 (2014).
39. Miller, B. W. et al. Targeting the LOX/hypoxia axis reverses many of the features that make pancreatic cancer deadly: inhibition of LOX abrogates metastasis and enhances drug efficacy. EMBO Mol Med 7, 1063–76 (2015).
40. Meissner, E. G. et al. Simtuzumab treatment of advanced liver fibrosis in HIV and HCV-infected adults: results of a 6-month open-label safety trial. Liver Int. 36, 1783–1792 (2016).
41. Van der Slot, A. J. et al. Increased formation of pyridinoline cross-links due to higher telopeptide lysyl hydroxylase levels is a general fibrotic phenomenon. Matrix Biol. 23, 251–7 (2004).
42. Akhurst, R. & Hata, A. Targeting the TGFβ signalling pathway in disease. Nature Reviews Drug Discovery
11, 790–811 (2012).
43. Rice, L. M. et al. Fresolimumab treatment decreases biomarkers and improves clinical symptoms in systemic sclerosis patients. J. Clin. Invest. 125, 2795–807 (2015).
44. Laklai, H. et al. Genotype tunes pancreatic ductal adenocarcinoma tissue tension to induce matricellular fibrosis and tumor progression. Nat. Med. 22, 497–505 (2016).
