Exploring pericyte and cardiac stem cell secretome unveils new tactics for drug discovery

ABSTRACT Ischaemic diseases remain a major cause of morbidity and mortality despite continuous advancements in medical and interventional treatments. Moreover, available drugs reduce symptoms associated with tissue ischaemia, without providing a definitive repair. Cardiovascular regenerative medicine is an expanding field of research that aims to improve the treatment of ischaemic disorders through restorative methods, such as gene therapy, stem cell therapy, and tissue engineering. Stem cell transplantation has salutary effects through direct and indirect actions, the latter being attributable to growth factors and cytokines released by stem cells and influencing the endogenous mechanisms of repair. Autologous stem cell therapies offer less scope for intellectual property coverage and have limited scalability. On the other hand, off‐the‐shelf cell products and derivatives from the stem cell secretome have a greater potential for large‐scale distribution, thus enticing commercial investors and reciprocally producing more significant medical and social benefits. This review focuses on the paracrine properties of cardiac stem cells and pericytes, two stem cell populations that are increasingly attracting the attention of regenerative medicine operators. It is likely that new cardiovascular drugs are introduced in the next future by applying different approaches based on the refinement of the stem cell secretome.


Introduction
Coronary heart disease (CHD) caused by the narrowing of arteries that feed the heart is the UK's single biggest killer, being responsible for ~73,000 deaths each year, an average of 200 people each day. Acute myocardial infarction (MI) represents the most harmful form of CHD. Over the last decade, mortality due to CHD has declined in the UK, but more people live with secondary consequences. In fact, most of the current treatments are palliative, i.e. they reduce symptoms associated with heart dysfunction, without providing a definitive repair. Consequently, CHD patients undergo a progressive decline in the pumping function of the heart that ultimately leads to heart failure (HF). Today, post-infarct HF is the leading cause of invalidity, hospitalization and mortality in patients over 65. In 2012-13, the UK National Health System (NHS) expenditure for cardiovascular disease was £7.02billion, 63% of which devoted to secondary care (Bhatnagar, Wickramasinghe et al. 2015) The NHS analysts have predicted a mismatch between total budget and patient needs of nearly £30 billion by 2020/21. Therefore, efficiency actions to increase quality and reduce expenditure growth are essential for all services, including those for treatment and care of CHD patients.
However, efficiency alone may not suffice without the introduction of new technologies having a transformative impact on this unmet clinical field.

The urgent need for new therapies
Current care of CHD comprises pharmacotherapy and revascularisation. However, medical treatment can be ineffective as in the case of refractory angina (which has an estimated prevalence of 1.8 million in the USA and an incidence of 30-50 000/year in Europe).
Additionally, a steadily increasing number of patients fall into the category in which revascularization cannot be applied or fails because of restenosis. This is especially true of patients with occlusive pathology extending to the microcirculation and diabetic or elderly patients who have had multiple bypasses and stenting operations. Also, the most important A C C E P T E D M A N U S C R I P T

ACCEPTED MANUSCRIPT
6 limitation of current treatments is that they do not replace cells irreversibly damaged by ischaemia.
Cardiovascular regenerative medicine is a fast-growing field of research that aims to improve the treatment of CHD through innovative restorative methods, such as gene therapy, stem cell therapy and tissue engineering (Assmus, Schachinger et al. 2002, Wollert, Meyer et al. 2004. Clinical studies with skeletal myoblasts, bone marrow-derived cells, mesenchymal stem cells (MSCs) and cardiac stem cells (CSCs) have shown feasibility and initial evidence of efficacy (Assmus, Schachinger et al. 2002, Menasche, Alfieri et al. 2008, Hare, Traverse et al. 2009, Sant'anna, Kalil et al. 2010, de Jong, Houtgraaf et al. 2014. After multiple systematic reviews and meta-analyses, the consensus is that transplantation of adult bone marrow cells modestly improves ventricular function, infarct size, and remodeling in patients with CHD compared with standard therapy, and these benefits persist during long-term follow-up (Martin-Rendon 2016). Bone marrow cell transplantation also reduces the incidence of death, recurrent MI, and stent thrombosis in patients with CHD (Jeevanantham, Butler et al. 2012). Moreover, Steven Chamuleau, Andreas Zieher and collegaues have recently utilized interaction models in a multivariable fashion to identify subgroups of patients that are defined as potential treatment responders, while simultaneously correcting for relevant factors that affect general disease outcome.
This kind of approach could be the next step towards optimal cell therapy in clinical care (Zwetsloot, Gremmels et al. 2016).
The SCIPIO clinical trial, the first in man to investigate c-kit+ CSCs, reported that 16 patients with ischemic cardiomyopathy received intracoronary infusions of 0.5-1x10 6 c-kit+, autologous CSCs and compared to controls these patients benefited from an 8 and 12 unit increase in left ventricular ejection fraction, 4 and 12 months after infusion, respectively (Bolli et al. 2011). A subset of 7 patients was subject to cMRI analysis, which A C C E P T E D M A N U S C R I P T

ACCEPTED MANUSCRIPT
7 showed that the infarct region had significantly decreased in size by ~10g up to 12 months following c-kit+ CSC transplantation (Bolli et al. 2011).
However, there is a persisting dispute regarding the mechanisms underpinning the benefit of cell therapy. The direct contribution of transplanted cells in vascular and cardiac reconstitution has been questioned (Balsam, Wagers et al. 2004, Murry, Soonpaa et al. 2004, and presently the concept of paracrine promotion of spontaneous healing processes prevails (Gnecchi, He et al. 2005. Indeed, the general consensus is that cell therapy and resultant improvements in cardiac function and decreased infarct size in human trials is due to a 'paracrine' effect . However, the lack of cardiomyocyte differentiation capability of bone marrow cells or CSCs could be due to lack of characterisation of the transplanted cell type, poor cell survival and retention, hostile host environment and subsequent restriction of cell proliferation, integration and differentiation in this damage-regeneration infarct model. Despite the adult mammalian heart being composed of terminally differentiated cardiomyocytes that are permanently withdrawn from the cell cycle (Nadal-Ginard 1978, Chien andOlson 2002), it is now apparent that the adult heart has the capacity, albeit low, to self-renew cardiomyocytes over the human lifespan (Bergmann, Zdunek et al. 2012, Bergmann, Zdunek et al. 2015. This is supported by the detection of small, newly-formed, immature cardiomyocytes, which incorporate BrdU/EdU and/or stain positive for Ki67, Aurora B, and embryonic/neonatal myosin heavy chain, as well as cardiomyocytes undergoing mitosis, under normal conditions and in response to diverse pathological and physiological stimuli (Urbanek, Quaini et al. 2003, Urbanek, Torella et al. 2005, Bostrom, Mann et al. 2010, Ellison, Vicinanza et al. 2013, Waring, Vicinanza et al. 2014, Bergmann, Zdunek et al. 2015. The source of these newly formed cardiomyocytes is a matter of debate (Laflamme and Murry 2011). Three main sources of origin of the new A C C E P T E D M A N U S C R I P T

ACCEPTED MANUSCRIPT
8 cardiomyocytes have been claimed: (a) circulating progenitors, which through the bloodstream home to the myocardium and differentiate into cardiomyocytes (Quaini, Urbanek et al. 2002); (b) mitotic division of the pre-existing cardiomyocytes (Engel, Hsieh et al. 2006, Bersell, Arab et al. 2009, Bostrom, Mann et al. 2010, Senyo, Steinhauser et al. 2013); and (c) a small population of resident multipotent stem cells able to differentiate into the main cell types of the heart (i.e., cardiomyocytes, smooth and endothelial vascular and connective tissue cells) (Torella, Ellison et al. 2007, Rasmussen, Frobert et al. 2014).
Blood-borne precursors, although well documented for having a role in inflammation and healing, and when adult mouse bone marrow cells were injected into the chick embryo they converted to a myocardial phenotype (Eisenberg, Burch et al. 2006), their cardiomyogenic potential in the damaged adult heart is very limited, if any (Loffredo, Steinhauser et al. 2011, Ellison, Vicinanza et al. 2013. The evidence so far presented in support of re-entry of terminally differentiated cardiomyocytes into the cell cycle has been limited to show division of cells that express proteins of the contractile apparatus in their cytoplasm (Kuhn, del Monte et al. 2007, Bersell, Arab et al. 2009, Bostrom, Mann et al. 2010, Senyo, Steinhauser et al. 2013). This evidence is equally compatible with new myocyte formation from the pool of multipotent cardiac stem/progenitor cells, which as precursor cells express contractile proteins and because newly born myocytes are not yet terminally differentiated they are capable of a few rounds of division before irreversibly withdrawing from the cell cycle (Nadal-Ginard 1978. However, the mechanisms underlying a strict postmitotic state in the heart during pathological remodelling have yet to be fully elucidated (Zebrowski, Becker et al. 2016).
The best documented source of the small, immature, newly formed cardiomyocytes in the adult mammalian heart, including the human (Torella, Ellison et al. 2006), is a small population of endogenous cardiac stem and progenitor cells (eCSCs) distributed throughout A C C E P T E D M A N U S C R I P T

ACCEPTED MANUSCRIPT
9 the atria and ventricles, which are clonogenic, self-renewing and can give rise to functional cardiomyocytes and vasculature in vitro and in vivo. Importantly, owing to genetic labelling and transitional tracking it is now documented that newly formed cardiomyocytes observed in the adult mammalian heart are the product of eCSC differentiation (Hsieh, Segers et al. 2007, Ellison, Vicinanza et al. 2013, van Berlo and Molkentin 2014.
Here, we provide an overview of current knowledge regarding the therapeutic potential of using the stem cell-derived secretome instead of source stem cell therapy to repair and regenerate the damaged heart. Furthermore, we illustrate methodological aspects of secretome-based cardiovascular regenerative medicine, with particular reference to functional transcriptomics and proteomics as a combinatory strategy to cherry pick the most beneficial components of the stem cell secretome. Finally, we report current evidence regarding the salutary aspects of the secretome from two stem cell populations, namely CSCs and pericytes.

Generation of a therapeutic product from stem cell secretome
Stem cells secrete potent combinations of cytokines, growth factors, enzymes, microvesicles/exosomes and genetic material, which help cardiac repair and regeneration at multiple points. The stem cell secretome supports cardiomyocyte survival and proliferation, differentiation of resident stem cells, and neovascularization, while limiting inflammatory and pro-fibrotic processes (Gnecchi, He et al. 2005, Baraniak and McDevitt 2010, Zhou, Honor et al. 2011, Rao, Aronshtam et al. 2015. Some paracrine factors have pleiotropic actions. For instance, one of the key pathways in stem cell-based cardiac repair is the stromal cell-derived factor-1 (SDF-1)/CXCR4 axis. It was proposed that substituting SDF-1 gene therapy for source stem cells might represent a sensible therapeutic approach (Penn 2009). The blinded placebo-controlled STOP-HF trial demonstrated a single endocardial administration of plasmid SDF-1 is safe, attenuates left ventricle remodeling and improves A C C E P T E D M A N U S C R I P T

ACCEPTED MANUSCRIPT
10 ejection fraction in ischaemic cardiomyopathy (Chung, Miller et al. 2015). However, there are also paracrine factors that boost specific responses, thereby providing a more defined approach for selective treatments inspired by the stem cell secretome. For instance, mesenchymal stem cell-derived Insulin Growth Factor-1 (IGF-1) and ABI Family Member 3 Binding Protein (Abi3bp) reportedly induce resident CSC mobilization (Mourkioti and Rosenthal 2005) and differentiation to the cardiac lineage, (Engels, Rajarajan et al. 2014, Hodgkinson, Gomez et al. 2014) while CSC-derived basic Fibroblast Growth Factor (bFGF), Vascular Growth Factor-A (VEGF-A) and Hepatocyte Growth Factor (HGF) exert potent pro-angiogenic effects (Zhou, Honor et al. 2011, Rao, Aronshtam et al. 2015.
Intracoronary administration of IGF-1 and HGF, in doses ranging from 0.5 to 2µg HGF and 2 to 8µg IGF-1, just below the site of left anterior descendent occlusion, 30 minutes after MI during coronary reperfusion in the pig, triggered a regenerative response from the CSCs, which is potent and able to produce physiologically significant regeneration of the damaged myocardium (Ellison, Torella et al. 2011). IGF-1 and HGF induced CSC migration, proliferation and functional cardiomyogenic and microvasculature differentiation. Furthermore, IGF-1/HGF, in a dose-dependent manner, improved cardiomyocyte survival, and reduced fibrosis and cardiomyocyte reactive hypertrophy.
Interestingly, the effects of a single administration of IGF-1/HGF is still measurable 2 months after its application, suggesting the existence of a feedback loop triggered by the external stimuli that activate the production of growth and survival factors by the targeted cells, which explains the persistence and long duration of the regenerative myocardial response. These histological changes were correlated with a reduced infarct size and an improved ventricular segmental contractility and ejection fraction at the end of the followup assessed by cMRI (Ellison, Torella et al. 2011).
Neuregulin-1 (NRG-1) is another key factor implicated in stimulating cardiac repair and regeneration (Wadugu andKuhn 2012, Waring, Vicinanza et al. 2014). An Ig-domain containing form of NRG-1β, also known as glial growth factor 2 (GG2) has been shown to improve left ventricular ejection fraction and remodelling in pigs post-MI, compared to controls (Galindo, Kasasbeh et al. 2014). It is thought that NRG-1 imparts functional benefits by activating and increasing eCSC proliferation (Waring, Vicinanza et al. 2014), inducing cardiomyocyte replacement (Bersell, Arab et al. 2009, Cohen, Purcell et al. 2014, Polizzotti, Ganapathy et al. 2015. A bioengineered hydrogel system enables targeted and sustained intramyocardial delivery of NRG-1, activating the cardiomyocyte cell cycle and enhancing ventricular function in a murine model of ischemic cardiomyopathy (Cohen, Purcell et al. 2014), protecting cardiomyocytes from apoptosis and improving mitochondrial function (Galindo, Kasasbeh et al. 2014). However, the role of NRG-1 in inducing cardiomyocyte proliferation in the adult heart has been challenged, with NRG1β1 treatment not increasing cardiomyocyte DNA synthesis and consequent cardiomyocyte renewal in normal or infarcted adult mouse hearts (Reuter, Soonpaa et al. 2014). Therefore, the role of NRG-1 administration in inducing cardiomyocyte proliferation and replacement in the adult failing heart remains controversial.
The stem cell secretome is collected in a form of cell culture conditioned medium (CM) or supernatant. The use of stem cell-derived secretome has several advantages compared to the use of stem cells, as CM can be manufactured, sterilized, freeze-dried, packaged, stored and transported more easily. Therefore, the stem cell-derived secretome has a promising prospect to become a successful pharmaceutical/medicinal product for regenerative medicine. On the other hand, variability represents a major limitation. The first level of inconsistency is represented by inter-individual variability, due to the patient's characteristics. Culture conditions, growth media and passages also affect the composition of CM, as recently reviewed (Pawitan 2014). In some occasions, changes in culture conditions are intentionally introduced to enhance the production of certain biological components. For instance, stem cells could be exposed to hypoxic conditions, with the aim of stimulating the production of growth factors (Di Santo, Yang et al. 2009, Ranganath, Levy et al. 2012). The three-dimensional spheroid culture of human adipose-derived stem cells with clinically relevant medium composed of amino acids, vitamins, glucose, and human serum leads to 23-to 27-fold higher production of angiogenic factors than that by conventional monolayer culture (Bhang, Lee et al. 2014). On the other hand, the purity of preparations may become an issue when complete growth medium is used, or collection of CM is performed shortly after removal of growth medium supplements. In this case, contamination may derive from the presence of left-over serum carrier proteins, due to dynamic recycling of extracellular proteins by cellular vesicles (Iso, Rao et al. 2014). The presence of serum or culture media in the CM-or supernatant-derived product might elicit immune responses and/or side effects. In this respect, alternative methods to culture cells (such as the adoption of serum-free culture, etc) and the refinement of the CM-derived secretome products should be considered. However, a recent study indicates that the formation of complexes of serum-derived immunoglobulins with paracrine factors could enhance the stability and biological activity of the CM therapeutic component (Rao, Aronshtam et al. 2015). A challenging objective of future research is to define the whole composition of crude CM preparations, measure activity of the main components and determine therapeutic doses. In this respect, manufacturing and quality control protocols need to be further refined and standardized to accomplish the classical regulatory path for drugs. The goal is to generate synthetic versions, inspired by the stem cell secretome, containing consistent dosages of therapeutic factors for the treatment of thousand patients.

ACCEPTED MANUSCRIPT
13 Different methods have been used for studying the secretome composition and for distinguishing the essential components that are advantageous for therapeutic applications.
Multiplex antibody-based techniques, such as antibody arrays, have high sensitivity (in the range of 1-10 pg/ml), specificity, and reproducibility across a broad range of concentrations. High-throughput analysis of the human MSC secretome using a human cytokine antibody array has identified about 40 proteins with high expression levels (Parekkadan, van Poll et al. 2007). Studies using multiplex antibody-based arrays showed the contribution of MSC-derived paracrine factors in determining cardiac improvement in a swine MI models (Nguyen, Maltais et al. 2010)  RNA interference, by siRNA or shRNA, represents a powerful method to evaluate the function of candidate genes (Lemons, Maurya et al. 2013). Throughput interference screening of mRNAs and microRNAs in cell-based assays can also help to decipher the functional importance of single components of the stem cells secretome. It is necessary to validate the outcome of silencing by rescue experiments where the CM from silenced cells is supplemented with a candidate recombinant protein to determine if it restores the original phenotype. Figure 1 illustrates an application of siRNA technology to the discovery of CM components implicated in the activation of stem cell migration.
A more systematic approach to tackling the secretome complexity combines LC-MS/MS detection, antibody arrays, microarrays, and bioinformatics. This approach identified 201 unique proteins (132 using LC-MS/MS and 72 using antibody arrays) typical of human MSC secretome (Sze, de Kleijn et al. 2007). Also, Sze exploited a computational analysis of data to predict the roles of the secretome components in metabolism, immune response, and development.

Pre-clinical cardiovascular studies using stem cell-derived secretome
Some preclinical studies have demonstrated that delivery of stem cell-derived CM has salutary effects in models of limb and myocardial ischaemia.  (Mirabella, Cilli et al. 2012), cord blood-derived EPCs (Kim, Song et al. 2010) and foetal aorta CD133 vascular cells (Barcelos, Duplaa et al. 2009) in limb ischaemia and wound healing models and with CM from EPCs (Hynes, Kumar et al. 2013) and MSCs in MI (Timmers, Lim et al. 2007, See, Seki et al. 2011, Timmers, Lim et al. 2011. Furthermore, the CM of human embryonic stem cell- derived ECs reportedly restored the healing activity of circulating proangiogenic cells from diabetic patients upon combined injection in ischaemic limbs of severe combined immunodeficient mice (Ho, Lai et al. 2012 (Khan, Nickoloff et al. 2015). A recent study showed the pathogenic communication between vascular ECs and pericytes in the diabetic microvasculature is mediated by the shedding of endothelial microparticles carrying miR-503, which transfer miR-503 from ECs to vascular pericytes (Caporali, Meloni et al. 2015).
To date, no clinical trial employing stem cell-derived CM for the treatment of cardiovascular disease has been reported. Two pilot studies described potential therapeutic activity of MSC CM for hair follicle regeneration and resurfacing wound healing in humans (Zhou, Xu et al. 2013, Fukuoka and Suga 2015, Shin, Ryu et al. 2015. These initial trials are expected to encourage clinical studies for the treatment of cardiovascular patients.

Commercialization of the stem cell secretome
The stem cell secretome is receiving increasing attention not only from researchers but also and Cosmetic Act (http://www.ipscell.com/2016/08/3-more-fda-warning-letters-to-stemcell-cosmetics-makers/). Therefore, marketing of these products with claims evidencing the use in the diagnosis, cure, mitigation, treatment, or prevention of disease violates the Act.
If this applies to skin creams, even more stringent regulation should be installed for the therapeutic use of CM in human cardiovascular diseases.

Exploiting the secretome of CSCs and pericytes for cardiovascular repair
One important caveat of secretome-based therapy is represented by the specificity of the source cell about the target tissue. It is, in fact, more likely that a therapeutic effect is achieved by exploiting the paracrine activity of tissue-specific cells rather than using cells

ACCEPTED MANUSCRIPT
18 isolated from a different tissue. In this respect, CSCs and pericytes may be uniquely suited to produce paracrine factors instrumental to cardiac and vascular repair and regeneration.
The first report of CSCs able to repopulate the adult heart was in 2003 (Beltrami, Barlucchi et al. 2003). Following this initial report, several cardiac stem/progenitor cell types have been identified based on the specification of membrane markers and transcription factors and resident CSCs exhibit spontaneous regenerative capacity when tested in an appropriate injury model (Ellison, Vicinanza et al. 2013). Table 2 provides a summary of different CSC subtypes based on antigenic characteristics.
Pericytes are mural cells that encircle capillaries in different organs. Pericyte-like cells that express stem cell markers and possess clonogenic expansion capacity have been identified in the adventitia of large vessels from human fetuses (Invernici, Emanueli et al. 2007) and adult individuals (Campagnolo, Cesselli et al. 2010). They share several markers that are typical of microvascular pericytes but do not express the surface marker CD146 (Table 3) cardiomyocytes, or more likely from the resident eCSC population (Loffredo, Steinhauser et al. 2011), has yet to be determined. Moreover, there are a number of technical limitations to prove myogenesis in these models, and the reader is directed to comprehensive reviews on this subject (Zebrowski, Becker et al. 2016).

Evidence for paracrine healing properties of pericytes
In 2007 Barcelos' study (Chen, Saparov et al. 2014) highlights this was the first to demonstrate the efficacy and associated healing mechanisms of topical therapy with human progenitor cellderived CM in a preclinical model of diabetic ischemic foot ulcer. Adding a level of M A N U S C R I P T

ACCEPTED MANUSCRIPT
21 complexity to this system, miR-15a and miR-16-1 can antagonize Wnt signaling (Bonci, Coppola et al. 2008). Importantly, the two microRNAs impair the biological functions of proangiogenic cells, and their expression is increased in the proangiogenic cells and serum of patients with critical limb ischaemia (Spinetti, Fortunato et al. 2013).
Pericyte-like cells isolated and expanded from the adult saphenous vein produce large amounts of angiogenic factors, in particular, VEGF-A, VEGF-B, Ang-1, and miR-132, which are delivered to neighboring ECs through the establishment of integrin-mediated interactions (Campagnolo, Cesselli et al. 2010, Katare, Riu et al. 2011). Secretion of VEGF-A, Ang-1 and miR-132 is further augmented by hypoxia/starvation, which mimics in vitro the environment encountered by cells upon transplantation into ischaemic tissues (Katare, Riu et al. 2011). Importantly, anti-miR-132-transfected pericytes were inferior to naïve pericytes or scrambled-transfected pericytes in improving reparative angiogenesis in the infarcted mouse heart. Nonetheless, the improvement afforded by pericytes was not completely abrogated by the miR-132 silencing, thus suggesting the participation of miR-132-dependent and independent mechanisms (Katare, Riu et al. 2011). of Ang-2 in both the cell types. Also, CSCs exposed to pericyte CM showed reduced SDF-1 mRNA levels as compared with CSCs exposed to the unconditioned medium. These data suggest a transcriptional interference on Ang-2 and SDF-1. However, while Ang-2 expression was reduced at mRNA and protein level, the increase in SDF-1 content in coculture media was not attributed to the induction of gene transcription, but rather to an increase in secretion rate. Exploring a possible involvement of Dipeptidyl peptidase-4 (DPP-4) we found higher expression levels by CSCs. Furthermore, DPP-4 is downregulated at

ACCEPTED MANUSCRIPT
23 mRNA and protein level in CSCs exposed to pericytes CM. These findings may have important implication for association cell therapy, where the outcome may depend on the balance between cooperative and competitive interactions at the level of the secretome. For instance, in our study, the combined transplantation of pericytes and CSCs additively reduced the infarct size and promoted vascular proliferation and arteriogenesis, but did not surpass single therapies concerning contractility indexes .
Skeletal muscle-derived stem cells have been among the first cell therapy models for regenerative treatment of the infarcted heart (Menasche 2008). A recent study investigated the therapeutic potential of human skeletal muscle pericytes for treating ischaemic heart disease and mediating associated repair mechanisms in mice (Chen, Okada et al. 2013). The authors found that pericyte transplantation attenuates left ventricular dilatation and myocardial fibrosis and improves cardiac contractility in infarcted mouse hearts. In line with findings in saphenous vein-derived pericytes, hypoxia-induced the expression of VEGF-A as well as PDGF-β, Transforming Growth Factor beta1 (TGF-β1), and corresponding receptors while expression of bFGF, HGF, and Ang-1 was repressed.

Resident cardiac pericytes: are they better suited for the heart?
New knowledge on microvessel-associated regenerative precursor cells in cardiac muscle opens prospectives for organ-specific treatments of patients with congenital and acquired heart defects. Peault's team showed that microvascular pericytes within the human myocardium exhibit phenotypes and multipotency similar to their anatomically and developmentally distinct counterparts (Chen, Baily et al. 2015). However, they have no ability for skeletal myogenesis, diverging in this respect from pericytes of all other origins.
In contrast, a cardiomyogenic potential was evidenced both in vitro and after intramyocardial transplantation in vivo.

24
Using the same standard operating protocol employed for the derivation of pericytes from adult saphenous veins, we have isolated and expanded pericyte-like progenitors from small biopsies of congenitally defective hearts (antigenic features illustrated in Table 3) . The long-term objective of research on neonatal cardiac pericytes is their perusal for definitive correction of congenital heart defects . A spectrum of prostheses in the form of conduits, patches, and valves is employed in congenital cardiac surgery, but none of them is perfect.
In particular, currently, available grafts do not possess growing capacity and become incompetent with time, thus requiring surgical replacement. The basic concept of tissue engineering is to create living material made by cellularized grafts that, once implanted into the heart, grows and remodels in parallel with the recipient organ. Neonatal pericytes produce a powerful secretome uniquely fit for the purpose. This includes growth factors that promote vasculogenesis and cardiomyogenesis and chemokines able to attract ECs and endothelial progenitor cells instrumental to graft endothelialisation. Comparison with saphenous vein-derived pericytes indicates similarity of the 2 cell populations, though cardiac pericytes secrete more HGF (6-fold), Ang-2 (8-fold), bFGF (4-fold), and VEGF-A (6-fold) than saphenous vein-derived pericytes. Furthermore, neonatal cardiac pericytes release procollagen type 1, a major constituent of the cardiac extracellular matrix, which is fundamental to the maintenance of graft integrity.
The umbilical cord represents a valuable source of stem cells immediately available at birth for corrective strategies. In a recent study, human cord-derived pericytes or cord blood-derived MSCs were delivered before or after alveolar injury into the airways of newborn rats exposed to hyperoxia, a model of bronchopulmonary dysplasia which complicates extreme prematurity and currently lacks efficient treatment (Pierro, Ionescu et al. 2013). Rat pups exposed to hyperoxia showed typical alteration in alveolar growth with

Evidence for paracrine healing properties of cardiac stem/progenitor cells
There is a growing consensus that the beneficial effects of any myocardial autologous or allogeneic cell therapy so far tested is, at least in part, mediated by a paracrine effect on the patient's cells at risk and activation of the host's CSCs by growth factors secreted by the transplanted cells (Lorkeers, Eding et al. 2014); (Braunwald 2015). Major contributors to this cardioprotective and CSC stimulatory effect are HGF and IGF-1 acting through their relative receptors present both on the myocytes, vascular and CSCs of the recipient's heart (Ellison, Torella et al. 2011).
MSCs have a broad repertoire of secreted trophic and immunomodulatory cytokines; however, they also secrete factors that negatively modulate cardiomyocyte apoptosis, inflammation, scar formation, and pathological remodeling (Ranganath, Levy et al. 2012).
Moreover, it is questionable whether they are the most optimal cell to use regarding survival and homing to and engraftment in the myocardium. Furthermore, cells can become entrapped in the microvasculature and impede cell entry into the myocardium.
The safety and efficacy of transplanting 2 million allogeneic, mismatched Cardiosphere-derived cells (CDCs) was tested in infarcted rats. Three weeks post-MI, animals that received allogeneic CDCs exhibited smaller scar size, increased infarcted wall thickness, and attenuation of ventricular remodeling. Allogeneic CDC transplantation
We have previously shown that CSCs that express high levels of the transcription factor GATA-4 exert a paracrine survival effect on cardiomyocytes through increased IGF-1 secretion and induction of the IGF-1R signaling pathway (Kawaguchi, Smith et al. 2010).
Furthermore, unlike bone marrow derived cells (Hofmann, Wollert et al. 2005), CSCs have a very high tropism for the myocardium (Ellison, Vicinanza et al. 2013). This cardiac tropism is governed by the SDF1-CXCR4 signaling axis, and when CSCs are injected either intracoronary or systemically, they home to and nest into the damaged heart with a high efficiency and significantly restore the myocardium, anatomically and functionally (Ellison, Vicinanza et al. 2013).
Cloned male eGFP-transduced heterologous HLA not-matched porcine CSCs, were administered intracoronary at differential doses (5x10 6 , 5x10 7 and 1x10 8 ) in 3 groups of pigs, 30 minutes after coronary reperfusion. Pig serum was injected to 6 control pigs after MI. BrdU was administered via osmotic pumps to track myocardial regeneration. Pigs were sacrificed at 30 min, 1 and 21 days. We found that heterologous CSC administration was well tolerated and without adverse effects. CSCs nested into the damaged myocardium with an efficiency of >95%, at 30 minutes through to 1 day after MI. Minimal spill over of CSCs was detected in the coronary sinus, spleen, lung or live and all injected CSCs had disappeared from the myocardium at 21 days. CSC-treated infarcted pig hearts showed a significant increase in the number of endogenous c-kit pos (GFP neg ) CSCs in the border and infarct regions, compared to CTRL. CSC-treated hearts exhibited an increase in the number

ACCEPTED MANUSCRIPT
27 of small, newly formed BrdU pos myocytes and capillaries. CSC injection significantly preserved myocardial wall structure and attenuated remodeling by reducing hypertrophy, apoptosis and fibrosis (Figure 3). Moreover, cardiac function was significantly preserved/improved by heterologous CSC-treatment. To summarize, intracoronary injection of heterologous CSCs after MI in pigs, which is a clinically relevant MI model, activates the endogenous CSCs through a paracrine mechanism resulting in improved myocardial cell survival and physiologically meaningful regeneration (Ellison, Torella et al. 2009).
Allogeneic CSC therapy is conceptually and practically different from any presently in clinical use. The proposed cell therapy is only a different form of growth factor therapy, where the cells naturally home to the damaged myocardium, deliver a more complex mixture of growth factors, elicit a 'paracrine' effect and activate the endogenous target cells.
Then the allogeneic CSCs are eliminated, and the regeneration triggered by activated endogenous CSCs is completely autologous. Therefore, the allogeneic CSCs survived long enough in the allogeneic host to produce their paracrine effect before being eliminated by the host immune system. Once more information is available, the allogeneic cells could be used either alone or in combination with the available factor therapy to improve the activation of the CSCs and the maturation of their progeny.

Conclusions
The market analysis predicts protein and peptide-based drugs will compose >50% of novel drugs within the next 10 years. Pharmacological companies have been using unbiased discovery methods to generate new druggable compounds since many years. The approach is now increasingly used in academic research. It is likely that new cardiovascular drugs will be introduced in the next future by applying these approaches to the study of stem cell paracrine function. Due to the variability of the product and high costs associated with clinical grade production, stem cell-based therapy is not amenable to the majority of A C C E P T E D M A N U S C R I P T

ACCEPTED MANUSCRIPT
28 ischaemic patients. The use of drugs inspired by the stem cell secretome will instead offer unprecedented therapeutic opportunities, resulting in a fundamental shift in the initial concept of regenerative medicine.