Mechanical strain promotes skin fibrosis through LRG-1 induction mediated by ELK1 and ERK signalling

Biomechanical force and pathological angiogenesis are dominant features in fibro-proliferative disorders. Understanding the role and regulation of the mechanical microenvironment in which pathological angiogenesis occurs is an important challenge when investigating numerous angiogenesis-related diseases. In skin fibrosis, dermal fibroblasts and vascular endothelial cells are integral to hypertrophic scar formation. However, few studies have been conducted to closely investigate their relationship. Here we show, that leucine-rich-alpha-2-glycoprotein 1 (LRG-1) a regulator of pathological angiogenesis, links biomechanical force to angiogenesis in skin fibrosis. We discover that LRG-1 is overexpressed in hypertrophic scar tissues, and that depletion of Lrg-1 in mouse skin causes mild neovascularization and skin fibrosis formation in a hypertrophic scarring model. Inhibition of FAK or ERK attenuates LRG-1 expression through the ELK1 transcription factor, which binds to the LRG-1 promoter region after transcription initiation by mechanical force. Using LRG-1 to uncouple mechanical force from angiogenesis may prove clinically successful in treating fibro-proliferative disorders.

There are a number of spelling errors and some of the wording is difficult to follow. I appreciate that English may not be the authors' first language but it would be helpful to improve this before publication.
Reviewer #2 (Remarks to the Author): The authors address a relevant question in the field of fibrosis, and identifies transcription factors of relevance.
The roles of FAK and ERK in fibrosis has been previously established; however, the present study identifies a new link between FAK-ERK and LRG-1 activation with subsequent promotion of angiogenesis.
The methodologies employed in this study are generally sound, with conclusions of some novelty.
However, there are several problematic areas the authors should address, potentially without the need for additional experiments.
(1) There is a lack of clarity on what cell types are over-expressing LRG-1 in Figure 1.
Overexpression of human LRG1 in endothelial cells has previously been shown to increase cell proliferation (Wang, et al., Nature, 2013). Is the LRG-1 over-expression presented in this study also found in endothelial cells? If so, why is LRG-1 protein added externally to HUVECs in Figure 2? Or is LRG-1 predominantly expressed by the fibroblasts, not endothelial cells?
The logic behind these experimental designs needs to be clarified.
(2) A minor point about this statement: "mechano-sensitive elements in the cell membrane are activated like integrin-focal adhesion kinase (FAK) complex, stretch-activated ion channels and G-Protein-coupled receptors".
-These are transmembrane receptors or ion channels that are mechano-sensitive. This is a minor issue where rewording would be recommended.
(3) The authors state: "This study providing a new sight that target LRG-1 to uncouple mechanical force from angiogenesis may prove clinically successful across diverse skin fibrosis or other fibroproliferative disorders." This major conclusions relies on some in vitro studies with HUVECS and the findings presented in Figure 3D. However, the quantification method for Figure 3D is not clear. Given the strength of the study relies heavily on the functional consequence of LRG-1 in relation to angiogenesis, the quantification methods need to be much more robust and transparent.
(4) There are typographical and grammatical errors throughout the manuscript that the authors should correct, otherwise the manuscript cannot be correctly interpreted by its readers. The manuscript focuses on the molecular mechanisms underlying neovascularization associated with hypertrophic scar formation. A strength of the research is the utilization of in vitro and animal models and the diverse tools used to address this issue. The authors demonstrate a pathway that includes focal adhesion kinase (FAK), ELK and LRG-1 in the response of fibroblasts to mechanical stimulation. The authors also demonstrate that LRG-1 acts on endothelial cells to promote neovascularization. In general, the experiments are well planned and the data are quite extensive. Overall the manuscript is well-written; however, substantial editing is needed to enhance the communication of the data and ideas in the paper.

Several suggestions include:
-Immunoblots should include size markers on the figures or in the figure legends.
-Rationale should be provided in the Methods section for the doses of LRG-1 used in the studies.
-Further explanation should be provided regarding the in vivo mechanical stimulation model.
What was the frequency, magnitude and duration of the stimulation? Also, further discussion regarding the relevance of this model to hypertrophic scar formation would be helpful.
-The Y-axis of PCR data should indicate "Relative mRNA Expression" -The X-axis of the graph in Figure 3E should state the units -is this hours?
Responses to Referees' comments: Response to Reviewer 1: "Overall, this is a very nice manuscript that, with some modifications, would make an important contribution to the field."

Comment 1
The photomicrographs in Fig   There is a lack of clarity on what cell types are over-expressing LRG-1 in Figure 1.

Overexpression of human LRG1 in endothelial cells has previously been shown to increase
cell proliferation (Wang, et al., Nature, 2013). Is the LRG-1 over-expression presented in this study also found in endothelial cells? If so, why is LRG-1 protein added externally to HUVECs in Figure 2? Or is LRG-1 predominantly expressed by the fibroblasts, not endothelial cells?
The logic behind these experimental designs needs to be clarified.

Comment 3
The authors state: "This study providing a new sight that target LRG-1 to uncouple mechanical force from angiogenesis may prove clinically successful across diverse skin fibrosis or other fibroproliferative disorders." This major conclusions relies on some in vitro studies with HUVECS and the findings presented in Figure 3D. However, the quantification method for Figure 3D is not clear. A major weakness of immunohistochemistry is its limited quantitative use. How representative are the images shown? Have non-consecutive tissue slides been used for the quantification? Most importantly, how exactly was the quantification performed?

The authors have left this very vague and this is not acceptable in its current form.
Given the strength of the study relies heavily on the functional consequence of LRG-1 in relation to angiogenesis, the quantification methods need to be much more robust and transparent.

Comment 4
The Y-axis of PCR data should indicate "Relative mRNA Expression" R: We thank the reviewer for the comment and sorry for the unclarity. According to this suggestion, the Y-axis of PCR data have been revised.