Project 6

Background and preliminary work: For a long time, FB have simply been regarded as the cells producing extracellular matrix (ECM) and, thus, the structural framework for organs and tissues. The ECM is not only a physical scaffold, but also provides crucial biochemical and biomechanical cues required for tissue morphogenesis, differentiation and homeostasis. Although its principal components are water, proteins and polysaccharides, each tissue has an ECM with a unique composition ¹. In pathological settings, the ECM is remodeled by increased turnover of existing and deposition of new ECM. Therefore, FB are at center stage in coordinating both normal tissue homeostasis and response to injury. Upon injury, tissue-resident FB proliferate, migrate and differentiate into so-called myofibroblasts, a cell type combining features of smooth muscle cells, like formation of actin-fibers that can withstand contractile forces, with a spindle-like morphology typical for FB ². Myofibroblasts efficiently contribute to repair after injury by remodeling the ECM to restore the mechanical properties of the affected tissue. However, adverse remodeling, possibly driven by persistently activated myofibroblasts, promotes fibrosis in many diseases, including cardiovascular, pulmonary and liver pathologies like atherosclerosis ³, heart failure and myocardial infarction ⁴, chronic obstructive pulmonary disease (COPD) ⁵, and non-alcoholic fatty liver disease (NAFLD) including non-alcoholic steatohepatitis (NASH) ⁶. Thus, interference with this prolonged myofibroblast activation could be of therapeutic interest and identification of the pathways involved could provide the required mechanistic clues.

The conversion of resident FB into myofibroblasts can be induced by mechanical stretch and cytokines like transforming growth factor β1 (TGFβ1). Interestingly, several lipids including sphingosin-1-phosphate and oxidized low-density lipoprotein have a similar effect ^{7, 8}. In addition, the differentiation depends on respiratory chain function, assigning a critical role for mitochondria in this process. A connection between myofibroblast differentiation and mitochondrial respiration has long been demonstrated by an increase in mitochondrial oxygen consumption after treatment of murine FB with TGFβ1 ⁹. We have later shown that the mitochondrial electron transport chain activity is not only enhanced upon TGFβ1 treatment, but that it is rather required for this process, as complex I inhibition with rotenone prevented myofibroblast differentiation of cardiac FB ¹⁰.

In line with the notion of tissue-specific ECM composition, resident FB are very heterogeneous between organs, with a “core” subtype present in all tissues and more divergent, organ-specific subtypes ^{11, 12}. Importantly, isolated murine FB retain their organ-specific transcriptomic signatures even in culture [189]. This offers the opportunity to study myofibroblast differentiation also ex vivo in pure FB populations, which can reveal kinetics and potential cell-autonomous programs. Moreover, these cells are easily accessible for pharmacological and genetic manipulation for validation of central regulators. 

In our joint program we will analyze myofibroblast differentiation and fibrosis ex vivo and in vivo in different tissues and the role of mitochondria therein. The Altschmied group has well-documented experience in the isolation of FB from different organs, myofibroblast differentiation ^{10, 13, 14}, characterization of and interference with mitochondrial functions ^{10, 13-17} and techniques for the genetic manipulation of cells, including retro- and lentiviral gene transfer ^18-22. Moreover, we have created unique mouse models with decreased or increased mitochondrial electron transport chain function ¹⁰. The Leitinger group has a long-standing track record with respect to characterization of oxidized lipids and their impact on cardiovascular and other diseases, which include fibrotic changes, in vivo ^23-30. They have established a diet-induced progressive model for NAFLD defined by distinct stages of hepatic steatosis and subsequent inflammation and fibrosis. In this model the Leitinger laboratory has shown that liver-specific expression of a single chain IgM-fragment recognizing oxidized phosphatidylcholines (scFv-E06) with an adeno-associated virus (AAV) in mice with established steatosis prevents the progression to hepatic fibrosis ³⁰.

Hypothesis:  We hypothesize that the organ-specific micro-environment and metabolic conditions are critical determinants of FB differentiation. In particular, we want to test the hypothesis that myofibroblast differentiation follows organ-specific pathways that are modulated by extrinsic (hyperlipidaemia) and intrinsic (mitochondrial function) conditions. 

Work program: The German laboratory will focus on ex vivo myofibroblast differentiation using FB from heart and lung. Therefore, FB isolated from these tissues will be treated with human TGFβ1 to induce differentiation. In a complementary in vivo approach in the liver, the American partner laboratory will use the NAFLD-model described above and block fibrosis development by infection with AAV-scFv-E06. Mice infected with a control virus or not infected at all, will be used as controls, as they develop hepatic fibrosis. Furthermore, they will examine gene expression profiles in liver, heart, and kidneys in a mouse model of high fat high sucrose (HFHS) diet-induced cardiometabolic disease, and the effect of oxidized phospholipid lowering by AAV-scFv-E06 on disease progression will be investigated. FB from various organs from these animals will be isolated from fibrotic and non-fibrotic regions.

All FB will then be subjected to single-cell RNA sequencing (scRNA seq); the required facilities are available at the HHU Düsseldorf (BMFZ) and in Charlottesville (Genome Analysis and Technology Core, School of Medicine). Corresponding in vivo data for heart fibroblasts are generated in P7 (Bottermann/Fischer/Harris) and P9 (Gödecke/Wolf). In-depth bioinformatic analysis will be performed within the IRTG (P5/P10) and with the help of the Bioinformatics Core units at HHU (CUBI) and at the American partner location. FB clusters including myofibroblasts will be identified by their marker signatures. Pseudotemporal ordering together with pathway analysis will allow us to uncover early changes and pathways, which might be critical in driving myofibroblast differentiation. Of particular interest for the functional validation are central regulators of the differentiation process like e.g. transcription factors, which can be inferred from the scRNAseq data ³¹ Finally, the bioinformatic analyses will provide information, which FB subtypes have undergone myofibroblast conversion and which might have been unaffected. Thereby, we will reveal if this differentiation follows a conserved program or is specific for the different tissues.

After having identified candidates initiating general and/or tissue-specific myofibroblast differentiation, we will analyze their function in this process to establish causal relationships. Therefore, selected targets will be overexpressed or downregulated in FB isolated from heart, lung and liver using lentiviral vector systems, before the cells are treated with TGFβ1. Differentiation into myofibroblasts will be analyzed qualitatively and quantitatively by immunoblot and immunofluorescence staining for α smooth muscle actin; the latter analyses will also allow for morphological evaluation. If specific signaling pathways are involved, for which small molecule modulators exist, we will also use pharmacological inhibition or activation. We would expect that molecules affecting a common program executed by core fibroblasts, will affect differentiation of fibroblasts from all three sources. The more interesting question will be, if myofibroblast differentiation of cells from one tissue is improved or impaired by regulators affecting fibroblasts from another organ or not affected at all, which would indicate a strict organ specificity.

For the in vivo validation we will again use the NAFLD and the HFHS-diet mouse models and stain liver, heart, lung, and kidney sections for the modulators of myofibroblast differentiation identified in fibroblasts from either of the three or all organs. Using organs from different stages of the disease we will be able to define a sequence of events and the contribution of common and organ-specific differentiation processes in an intact tissue.

The contribution of mitochondrial functionality will be addressed by two different approaches. Firstly, we will analyze if inhibition of the mitochondrial respiratory chain affects not only the differentiation of cardiac, but also of lung and liver FB. Secondly, we will isolate FB from all three tissues from our unique mice containing Telomerase Reverse Transcriptase (TERT) exclusively in the mitochondria of all organs ¹⁰. We have previously shown that myofibroblast differentiation capacity of cardiac FB from these animals is enhanced ¹⁰. These studies will now be extended to lung and liver fibroblasts. Furthermore, we will address the question if there are additive or synergistic effects between improved ETC complex I activity and positive regulators of myofibroblast differentiation and if these regulators themselves affect complex I using the levels of specific subunits as surrogate markers for its activity ¹⁰.

Added value of the collaboration: Both sides, especially the graduate students, will profit from the complementary expertise and experimental models available in the partner laboratories. Therefore, the German graduate student will work on the ex vivo aspects and will be introduced into the NAFLD and HFHS diet models during the exchange phase in the American laboratory and participate in the respective analyses. Conversely, the American student will be made familiar with the genetic manipulation and ex vivo differentiation of isolated fibroblasts during his stay in the German partner laboratory. 

In summary, this joint project will identify how the differentiation of FB from different tissues into myofibroblasts is regulated and if these pathways are common to different organs or tissue-specific.