Mapping the regenerative capacity of pericytes from embryonic development to ageing

Lead Research Organisation: University of Manchester
Department Name: School of Biological Sciences


Pericytes are small cells that line blood vessels and regulate blood flow, formation of new blood vessels and also facilitate immune reactions. Growing body of evidence suggests that these cells can give rise to various other cell types. In example studies have demonstrated that shortly after birth pericytes can give rise to skeletal muscle cells. For yet unknown reasons their capacity to develop into muscle cells diminishes in ageing. The tentative stemness properties of pericytes make them an attractive target to treat muscular dystrophies as well as other muscle diseases and injuries. About 1 in 3500 boys is born with muscular dystrophy in Europe. The disease leads to gradual death of skeletal muscle cells, progressive muscle weakness, and is associated with severely shortened life expectancy. There is no cure available to treat this disease. Previous studies with dystrophic mice and dogs have shown that pericyte-derived cells can successfully repair skeletal muscles upon injection into blood circulation. The cells leave blood vessels and target the inflamed muscle, where they will give rise to muscle cells. However, the process remains relatively ineffective. Considering that pericytes are abundant along most blood vessels it might be possible to instruct them to differentiate into beneficial skeletal muscle cells directly, without the need for cell isolation and injection. Yet, for this aim it is important to gain a detailed understanding of how pericyte fate is regulated. This is because in addition to potentially beneficial cell types pericytes may also give rise to myofibroblasts. In muscular dystrophies myofibroblasts accumulate between skeletal muscle cells and secrete various proteins that increase muscle stiffness and thereby prevent the healing process. Thus, in order to fully harness the regenerative potential of pericytes we would need to limit their negative contribution in muscular dystrophy.
I will study mouse embryonic development and adult skeletal muscles to show cells that can arise from pericytes. I will thereafter characterise the molecules that can guide pericytes to give rise to one or another cell type with an emphasis on finding ways how to influence pericytes to develop into skeletal muscle cells. This part of the project will rely on expert knowledge on cell microenvironment in Manchester and will make use of automated cell analysis techniques that can indicate possible pharmaceutical targets to manipulate pericyte fate. Finally, I will demonstrate why skeletal muscle pericytes lose their beneficial potential to contribute to skeletal muscle cells in ageing and whether it can be reversed. This study will for the first time address the stemness properties of pericytes throughout embryonic development and in tissue context in adults and will provide extensive novel information that can be used in regenerative medicine.

Technical Summary

Pericytes have well defines and specialised functions around blood vessels, including guiding blood vessel formation (angiogenesis) and regulating blood flow (vasoconstriction). Accumulating evidence shows that pericytes can leave their blood vessel associated niche and differentiate into other cell types. The pericyte progeny is poorly characterised due to the lack of truly specific markers. I will here take an innovative approach that makes use of a novel transgenic mouse line in which pericytes expressing two marker genes simultaneously will be labelled with red fluorescence and their progeny will be permanently marked by green fluorescent protein expression. Existing mouse models that allow specific tracking of adult skeletal muscle pericytes will complement the study. Using these methods I will for the first time map the fate of pericytes throughout embryonic development and in adult skeletal muscles. I will identify cell microenvironment conditions and signalling molecules that regulate pericyte differentiation into skeletal muscle cells. I will combine the fate mapping experiments and signalling pathway studies with direct imaging of pericyte differentiation in order to visualise their commitment into skeletal muscle cells and establish the possibilities of manipulating with pericyte plasticity. I will address the loss of pericyte myogenic potential in ageing and dissect the relative contribution of the aged tissue microenvironment to this change. I will show whether by exposing adult skeletal muscle pericytes to specific signals their regenerative potential can be reversed. The novel data on pericyte plasticity and lineage determination will be important in understanding the possible use of these cells in tissue regeneration.

Planned Impact

Academic Impact. The proposed research will have a direct and immediate impact on biomedical research. The novel mouse model established in this work will be valuable in a wide number of research areas related to vascular biology, including cancer research, neurobiology and studies of cardiac diseases. Understanding the role of pericytes in regulating blood vessel development and function in these areas is of immense medical significance, but lack of suitable mouse models has hindered progress. In addition to pericytes, several other cell types lack known specific marker genes. The novel approach to visualise pericytes by a double-transgenic approach will pave the way to implementing a similar method to study other cell types that suffer from this limitation. The databases on signalling pathways and cell microenvironment conditions that enhance pericyte differentiation will benefit researchers focusing on cell differentiation. Knowledge of possibly simple combinations of cell microenvironment components that guide cell fate will make it possible to apply the results in stem cell studies. I will present the results at high-level international conferences to facilitate collaboration and aim to published my work in leading journals.

Societal impact. The outcome of the work will benefit muscular dystrophy patients. Various forms of dystrophy are very common (i.e. 1 in 3500 boys is born with Duchenne dystrophy), yet there is no cure available. Current cell and gene therapy has shown only limited efficacy in improving patient's life. Pericytes are capable of differentiating into skeletal muscle cells, but unfortunately in dystrophic muscle may also give rise to myofibroblasts and enhance tissue fibrosis. By identifying signals that regulate pericyte fate these cells can be either directly targeted in the patient (if signalling pathways are highly specific) or isolated and activated in vitro before transferring them back to the patient. The potential benefit of this research is not limited to congenital dystrophies. There are significant socio-economical implications for the ageing of Western European population. Increased muscle weakness (age related atrophy) causes reduced mobility and leads to lower quality of life. My project targets the diminished myogenic potential of perivascular cells in ageing and will demonstrate whether this can be reversed. If we knew how to activate and drive pericytes towards skeletal muscle fate we could use this to target age related muscle weakness, in particular in patients suffering from various diseases (i.e. cancer), but also in case of acute skeletal muscle injuries in young adults.

Economic impact. I will identify pathways and molecules that regulate pericyte differentiation towards beneficial myogenic regenerative fate. These compounds can be combined to pharmaceuticals to target skeletal muscle diseases. The power of high throughput screening of chemical modulators by pharmaceutical industries can further narrow down the molecular pathways to achieve a cell-type specific effect. It is possible though that suitable safe drugs already exist but have not been tested for skeletal muscle repair, in which case there will be an immediate medical impact on patient care. The wider economic impact is based on increasing the mobility of patients suffering from muscle diseases but also from age related muscle weakness, thereby reducing their dependence on care services.

Educational/Public impact. In this study I will address one of the fundamental questions in developmental biology on the differentiation plasticity of cells. Pericytes have clearly well defined functional properties around blood vessels, but can give rise to cell types with very different roles, like the contractile skeletal muscle cells. Such an intriguing phenomenon is of interest to students of biology and to wider audience.
Description Perivascular cells cover blood vessels and regulate blood flow. Pericytes are wrapped around capillaries whereas vascular smooth muscle cells (VSMCs) surround larger blood vessels. Pericytes can contribute to fibrosis and may have a role in tissue regeneration. VSMCs are important in the context of cardiovascular diseases. Excessive VSMC proliferation can lead to vessel occlusion, is a frequent unwanted side-effect of vascular surgeries and is triggered in atherosclerosis. Analysis of the mechanisms of perivascular cell turnover is essential for the development of novel therapeutics for cardiovascular diseases and to enhance tissue regeneration, but has been hindered by the lack of specific animal models.
I established a novel triple-transgenic mouse model that relies on sequential activation of two recombinases controlled by the activity of two perivascular expressed genes, NG2 and CD146. By combining the specificity of the expression pattern of two genes this mouse models permits for the first time to characterize the fate of perivascular cells across mouse life.
I found that the origin of VSMCs in the adult aorta can be traced back to progenitor cells that colonize this blood vessel in early embryonic development. In parallel, already in embryonic development, a unique VSMC population emerges at aortic branching sites and is maintained in immature state to adulthood. It is well known that arterial branching sites are predisposed to atherosclerosis, however the reason for this has remained unknown. This study proves that a specialized immature cell population exists at arterial branching sites, explaining why they are more susceptible to cardiovascular diseases.
We revealed that arterial response to surgery is defined by the extent of injury. In vascular surgery that mainly affects the internal cell layers of the artery the surrounding VSMCs are activated and contribute to vascular wall thickening. In contrast, in more extensive surgeries local VSMCs die and VSMCs further away lack the migratory capacity to mount a regenerative response. In this case the vascular wall is colonized by adventitial cells that surround the arteries and can give rise to VSMCs. These results suggest that therapeutic options to enhance the recovery from arterial surgeries must take into account distinct cellular mechanisms in different types of surgeries.
I have demonstrated the significance of CD146 in regulating cell differentiation. CD146 is highly expressed in immature VSMCs and in pericytes, but its expression is lost in VSMC maturation. Using CRISPR-Cas9 genome editing I found that CD146 inhibits VSMC differentiation. This data explains how VSMCs are kept in immature state at arterial branching sites in the adult as they maintain high levels of CD146. I showed that CD146 regulates the differentiation of chondrocytes and skeletal muscle cells. My studies have shown that CD146 is induced in elastic cartilage injury response where it controls chondrocyte differentiation.
I did not find evidence of pericyte differentiation into skeletal muscle cells in development, postnatal skeletal muscle maintenance and in regeneration following injury. Pericytes give rise to a fraction of bone cells. Future transgenic mouse models are needed to shed light on the fate of adventitial cells.
Exploitation Route The published results demonstrate for the first time that a unique immature cell population exists at arterial branching sites that are susceptible to atherosclerosis. My work has also revealed the molecular signature of these cells and has shown the significance of CD146 in controlling vascular smooth muscle differentiation. This lays the foundation for screening for drugs that specifically affect vascular branching sites and CD146 in the context of atherosclerosis.
I showed how distinct cellular response is triggered in different types of common arterial surgeries. These results suggest that anti-proliferative drugs are effective in preventing vascular wall hyperplasia and occlusion only in certain types of surgeries whereas likely have detrimental effects in others. New clinical trials with surgery-specific drugs are needed to assess the results.
The results of the fellowship have wide academic impact as they have characterized fundamental principles in vascular biology. The developed transgenic strategy to map cell fate is applicable to study other cell types that lack unique marker genes. Comprehensive review article that has recently been published (Developmental Biology, 2018) provides an overview of arterial biology and response to injuries and helps to achieve educational impact for both medical and life science students.
Sectors Education,Healthcare,Pharmaceuticals and Medical Biotechnology

Description The main results of the fellowship have been circulated in press release from the University of Manchester ( and distributed in various popular science websites, helping to achieve wide public engagement and popularize biomedical research. My work showed how the type of arterial surgery determines the cellular injury response. The clinical implications of these results are evaluated in collaboration with reconstructive vascular surgeon Mr Jason KF Wong (University of Manchester NHS Foundation Trust, Wythenshawe Hospital). The results from this fellowship also identified a a unique cell population at arterial regions that are susceptible to atherosclerosis. It is foreseeable that this new knowledge will benefit pharmaceutical industry in designing drugs specifically targeting these cells.
First Year Of Impact 2017
Sector Healthcare
Impact Types Societal,Policy & public services

Title CD146 mutant cell lines 
Description CD146 (MCAM) is a membrane protein that is expressed in several embryonic tissues and upregulated in metastatic tumours. I found that its high expression level marks immature perivascular cells, whereas it is downregulated in blood vessel maturation. To characterize the molecular mechanisms by which this gene regulates cell differentiation I used CRISPR-Cas9 genome editing to establish three cell lines that lack CD146 or its endocytosis motif. The mutated cells are primitive embryonic cells (10T1/2) that represent widely used models for cell differentiation as they can be induced to differentiate into smooth muscle, chondrocyte and skeletal muscle cells. This new model enabled me to reveal the function of CD146 in the context of diverse differentiation scenarios. The model has been published in two papers (Moreno-Fortuny et al., 2017; Roostalu et al., 2018) and is available to the wider scientific community. 
Type Of Material Cell line 
Year Produced 2017 
Provided To Others? Yes  
Impact The mechanisms that control cell shape have remained poorly characterized. Among the most distinctive cell shapes is the elongated morphology of a multinucleated skeletal muscle cell. How this shape is achieved and why cell fusion leads to cell elongation instead of growth in multiple dimensions and rounding of the cell has remained unknown. We found that in early stages of skeletal muscle cell elongation CD146 is asymmetrically localized at distal ends of the cell. This localization pattern leads to polarization of cytoskeleton and cell elongation. Asymmetric distribution of CD146 is required for polar localization of several other cell signalling related proteins that coordinate cytoskeletal organization. In the absence of CD146 skeletal muscle cells initiate differentiation, fuse into multinucleated cells but fail to elongate. Our studies have shown that CD146 inhibits smooth muscle differentiation. Elimination of CD146 from embryonic fibroblasts leads to their spontaneous and rapid smooth muscle differentiation. These results are particularly significant in the context of the in vivo experiments that I carried out in parallel. The animal experiments revealed that high CD146 expression is maintained in a unique immature smooth muscle cell population at aortic branching sites in the adult mouse. These sites are known to be prone for atherosclerotic lesions, indicating CD146 as a potential drug target for cardiovascular diseases. 
Title Mouse model to label perivascular cells and study their fate 
Description Perivascular cells line all blood vessels and are crucial in vascular physiology. Pericytes control blood flow in capillaries, regulate leukocyte migration to underlying tissues and can contribute to tissue fibrosis and repair. Vascular smooth muscle cells strengthen the walls of large blood vessels and regulate blood flow. The analysis of these cells in animal models has been complicated due to the lack of cell type specific genes. Contractile proteins that are often used to mark smooth muscle cells are expressed by numerous other cell types, whereas only few markers have been identified for pericytes and even these are widely expressed in diverse mesodermal and neural tissues. I found that in most tissues CD146 shares expression with NG2 only in pericytes and not in other cell types where either one or the other gene is expressed separately. Both of these genes are also expressed in developing aortic smooth muscle cells and in developing as well as mature smooth muscle of smaller arteries. Here we established a novel triple-transgenic mouse model that combines the specificity of the expression patterns of CD146 and NG2 to label perivascular cells and accurately follow their fate. In a previously established mouse model NG2 gene regulatory elements drive the expression of CRE recombinase after tamoxifen administration (NG2-CRE-ERTM) (Zhu et al., Development 2011). I generated a new mouse model in which CD146 gene regulatory elements drive the expression of flippase and red fluorescent protein (Roostalu et al., Circ Res 2018), but only in cells that simultaneously express CRE recombinase. When this model is crossed to NG2-CRE-ERTM model CD146+NG2+ perivascular cells are marked by red fluorescence and start to express flippase. Cells that express either CD146 or NG2 alone remain unlabelled. When the double transgenic model is crossed to a third established line in which ubiquitous promoter drives the expression of green fluorescent protein only in cells that express flippase (Sousa et al., Cereb Cortex 2009) the CD146+NG2+ cells are also marked by green fluorescence. When CD146+NG2+ cells differentiate (in smooth muscle maturation or by pericyte differentiation into other cell types) they cease expressing red fluorescent marker protein but maintain green fluorescent marker, which is carried over to the following cell generations through mouse lifespan. This model provides a unique opportunity to specifically identify when and how perivascular cells are replaced in blood vessel walls in healthy physiology and in tissue injury response and whether pericytes can give rise to other cell types. 
Type Of Material Model of mechanisms or symptoms - mammalian in vivo 
Year Produced 2018 
Provided To Others? Yes  
Impact Using this transgenic mouse model we found that CD146+NG2+ cells emerge at first in early embryonic development in the wall of the primitive aorta. These cells share many characteristics with microvascular pericytes. They proliferate extensively to give rise to all the smooth muscle cell lineages that are maintained in the aortic wall to adulthood. These results suggest that there is very limited contribution from cells outside the aorta to its composition during postnatal growth period and in adult vascular wall homeostasis. The vast majority if not all the cells in aortic wall are descendants of embryonic CD146+NG2+ immature smooth muscle cells. We showed that while CD146 is rapidly downregulated in aortic smooth muscle already in the fetus it is maintained at aortic branching sites. We demonstrated that these cells represent a unique self-renewing immature smooth muscle cell population with distinct molecular and functional characteristics. It is widely recognized that aortic branching sites are prone to atherosclerosis, but the reasons have remained largely unknown. Our discovery of a specialized cell population at these locations provides new opportunities to specifically target these cells in mouse models of cardiovascular diseases and to develop drugs that precisely inhibit the activation of these cells and formation of atherosclerotic lesions. Smooth muscle cell activation and excessive proliferation following arterial surgery can lead to vascular wall thickening and occlusion. Anti-proliferative drugs are frequently applied with stents to prevent the process and have shown success in limiting in-stent restenosis. However this is not possible in most other types of surgeries and consequently the failure rates have remained high (30-50% failure in coronary bypass surgeries over 5 year period). Characterization of arterial cellular response to surgery is of paramount importance to find ways to improve the efficacy. This mouse model enabled to trace the fate of green fluorescent femoral artery smooth muscle cells following different types of surgery. Remarkably, our study showed that the cellular response is very different depending on whether surgical trauma is limited to a small area and vascular lumen or transects all arterial wall cell layers. In the first case smooth muscle cells respond by activation and cell division. In the latter case however the local smooth muscle cells die and are replaced by migratory adventitial cells. These results suggest that anti-proliferative drugs may impair the repair of arterial walls following anastomosis. The established mouse model has enabled me to follow the fate of pericytes across mouse organs from early embryonic development to adulthood. Contrary to previous studies we did not detect skeletal muscle differentiation of CD146+NG2+ pericytes in embryonic development, postnatal muscle growth and in injury repair. We found evidence for limited osteogenic differentiation of pericytes, but no epithelial conversion in the intestine or lung. Comparative immunohistochemical analysis suggests great diversity among pericytes and adventitial cells in the microvasculature, highlighting the possibility that distinct cells may contribute differently to tissue regeneration. 
Description Characterisation of pericyte fate in the central nervous system 
Organisation Ludwig Maximilian University of Munich (LMU Munich)
Country Germany 
Sector Academic/University 
PI Contribution We have carried out extensive experiments to identify the fate of mouse perivascular cells across diverse tissues and organs from early embryonic development to adulthood. In these experiments I have exposed the triple-transgenic mice, generated in this fellowship, for different periods to tamoxifen to initiate the expression of fluorescent marker proteins and have isolated tissue samples for histological and molecular analysis. The brain and spinal cord samples have been shipped to Dr. Marisa Karow in Munich to determine the identity of pericyte daughter cells in the nervous system.
Collaborator Contribution Dr. Marisa Karow is expert in central nervous system pericytes and has carried out histological analyses to characterize whether pericytes can in vivo give rise to other cell types in the brain and the spinal cord.
Impact The samples have been analyzed in Munich. Future research in this direction will depend on the acquisition of additional funding.
Start Year 2015
Description Role of CD146/MCAM in cartilage repair 
Organisation National Institutes of Health (NIH)
Department National Institute of Dental and Craniofacial Research (NIDCR)
Country United States 
Sector Public 
PI Contribution While analyzing the expression dynamics of pericyte associated gene CD146/Mcam I found that it is highly induced in chondrocytes following cartilage injury. I have characterized its activation pattern in the triple-transgenic fate-mapping mouse model by studying different types of articular and elastic cartilage by confocal and light-sheet microscopy. I established in vitro assays in which I used siRNAs to knock down CD146 in primary chondrocytes.
Collaborator Contribution I have established a collaboration with Prof. Pamela Gehron Robey whose lab has generated CD146 knockout mouse model and where the function of the gene has been evaluated in the context of bone and cartilage injuries.
Impact We are currently preparing a manuscript based on this collaboration that we aim to submit in April, 2018. The research has revealed an important role for CD146/MCAM in regulating elastic cartilage regeneration and has defined the downstream molecular signalling pathways. This data may eventually help to develop novel drugs to enhance cartilage repair.
Start Year 2017
Description Press release 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact I published a press release in collaboration with the University of Manchester media office ( This has been picked up by popular science websites (i.e. and distributed in social media.
Year(s) Of Engagement Activity 2017
Description Video introduction 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Postgraduate students
Results and Impact Recorded a video that was published at the MHS Fellowship Academy website to introduce how BBSRC funding helps me to advance biomedical research.
Year(s) Of Engagement Activity 2015