*2.13. In Vivo Studies*

Two-month-old male αSGKO/SCIDbg mice (n = 5) were anesthetized with an intramuscular injection of physiologic saline (10 mL/kg) containing ketamine (5 mg/mL) and xylazine (1 mg/mL) and then 5 <sup>×</sup> 105 PC-derived iPSCs were injected into the Tibialis Anterior muscle (TA), according to standardized procedures [5]. Mice were sacrificed 20 days after implantation for morphological analysis. Experiments on animals were conducted according to the rules of good animal experimentation I.A.C.U.C. no 432 of 12 March 2006 and under Italian Health Ministry approval no. 228/2015-PR. In vivo experiments were conducted in accordance with the principles of the 3Rs (replacement, reduction and refinement).

#### *2.14. Engrafted Human Muscular Cell Identification*

Human differentiated iPSCs were identified by immunofluorescence for anti-human lamin A/C (1:100, SIGMA). Anti-laminin (1:100, SIGMA) was used to identify the fibers. The images were obtained by confocal laser scanning microscope.

#### *2.15. Statistical Analysis*

Statistical significance of the differences between means was assessed by one-way analysis of variance (ANOVA), followed by the Student-Newman-Keuls test, to determine which groups were significantly different from the others. When only two groups had to be compared, we used the unpaired Student's t-test. *p* < 0.05 was considered significant. Values are expressed as means ± standard deviation (SD). All the analyses were performed using Graph-Pad PRISM 7 and 8.

#### **3. Results**

#### *3.1. Pericyte and Fibroblast Isolation and Characterization*

Pericytes were isolated from three healthy human skeletal muscle biopsies and characterized by flow cytometry using a specific panel of markers according to previous studies [6]. The harvested cells highly expressed well-known pericyte markers, such as ALP (alkaline phosphatase), PDGFRβ (platelet derived growth factor receptor-beta), CD146 (MCAM, melanoma cell adhesion molecule), NG2 (Neuron/glial antigen 2), and CD44 (HCAM, homing cell adhesion molecule) (Figure 1A) as well as CD56 (NCAM, neural-cell adhesion molecule), a glycoprotein specifically expressed in muscle by human satellite cells [6,34].

Skeletal muscle resident PCs, expressing ALP, represent a myogenic cell compartment, distinct from satellite cells, capable of promoting myofiber regenerating [6]. Therefore, we selected pericytes with myogenic potential by fluorescence-activated cell sorting (FACS), combining the cell surface markers ALP and CD56. We enriched the pericyte population selecting the fraction ALP+CD56−, which represented the 28% of the total population (Figure 1B).

After expansion and before reprogramming, ALP+CD56<sup>−</sup> subpopulation was analyzed for the expression of the canonic muscular pericyte markers—almost the totality of the tested cells expressed ALP, PDGFRβ, CD146, NG2, and CD44, while the expression of CD56 was dramatically reduced (Figure 1C), suggesting that ALP+CD56<sup>−</sup> fraction retains pericyte features.

These results are in line with previous studies in which pericyte identification was performed through the combination of NG2, PDGFβ, and CD146 markers [35,36]. PC phenotype was further confirmed by the ALP colorimetric assay (Figure 1D) and immunofluorescence positivity for NG2, PDGFRβ, and αSMA (Figure 1E).

We further examined the myogenic potential of ALP+CD56<sup>−</sup> cells by measuring myosin heavy chain (MHC) expression, upon skeletal muscle differentiation, induced by cellular confluence and serum depletion [37]. After two weeks, ALP+CD56<sup>−</sup> cells spontaneously differentiated into myosin positive multinucleated myotubes, as confirmed by the expression of MHC (Figure 1F). These results indicate that pericytes possess myogenic potential, along with supporting vessel formation and angiogenesis.

PCs are crucial in several phases during angiogenesis and vascular homeostasis, regulating the germination of the capillaries and the stabilization of the vessels. We therefore evaluated the ability of ALP+CD56<sup>−</sup> cells to generate networks and to cooperate with endothelial cells (HUVECs) to form capillary-like structures. For this purpose, we transduced cells with a lentivirus expressing GFP, co-cultured with HUVECs (GFP<sup>+</sup> ALP+CD56<sup>−</sup> cells/HUVECs in a 1:4 ratio), and assembled on Matrigel for 6 h. We found that the ALP+CD56<sup>−</sup> cells significantly enhanced capillary-like structure formation of HUVECs. Indeed, PCs co-cultivated with HUVECs displayed higher segment total length, total mesh area and total branch length compared to HUVECs cultured alone (Figure 1G). These results demonstrate that pericytes isolated from skeletal muscle maintain their ability to support vessel formation and myogenic potential after isolation, sorting and expansion procedures.

**Figure 1.** Pericyte characterization. (**A**) Representative histograms indicating the percentage of alkaline phosphatase (ALP+), platelet derived growth factor receptor-beta (PDGFRβ+), MCAM, melanoma cell adhesion molecule (CD146+), (αSMA+), Neuron/glial antigen 2 (NG2+), HCAM, homing cell adhesion molecule (CD44+) and NCAM, neural-cell adhesion molecule (CD56+) positivity (black peaks) determined by flow cytometry in pre-sorted cells isolated from muscular biopsy (n = 4). Matched isotypes were used as negative controls (grey peaks). (**B**) Representative gating strategy for ALP<sup>+</sup> and CD56<sup>−</sup> cell sorting (n = 3). Cells were first gated for cell size (side light scatter SSC-A vs. forward light scatter FSC-H) and vitality (Live Qdot-525-A). The muscular cell gate was further analyzed for singlets (SSC-A vs. SSC-H) and their expression for ALP and CD56. Pericytes, ALP<sup>+</sup> and CD56− were then sorted from this gated population. The lower set of four plots confirmed the efficiency of the sorting. (**C**) Representative post-sorting histograms for key pericyte markers after two passages in vitro, indicating an enhanced expression of ALP, NG2, PDGFRβ, CD146, and CD44. (**D**) Sorted pericytes stained for ALP showing fibroblast colony-forming units (CFU-F) when seeded at low confluence. Scale bar represents 300 μm. (**E**) Immunofluorescence labeling for NG2 (red) and the co-staining for PDGFRβ (green) and αSMA (magenta) on sorted ALP+CD56<sup>−</sup> cells. Nuclei were stained with DAPI. Scale bar represents 50 μm. (**F**) Representative fluorescence image for myosin heavy chain (MHC) (red), validating the differentiation of sorted pericytes toward skeletal muscle phenotype. Scale bar represents 100 μm. (**G**) Illustrative images of human umbilical vein endothelial cells (HUVEC) in co-culture with pericytes displaying the formation of capillary-like networks with HUVEC labeled for von Willebrand factor (vWF; magenta), and GFP<sup>+</sup> pericytes. Nuclei were identified by DAPI (blue). Scale bar represents 100 μm. Tubular structures were photographed at 5× magnification and quantified by the angiogenesis analyzer ImageJ tool. Total segment length, total mesh area and total branching length exhibited significant differences between HUVEC alone and in co-culture with pericytes, as shown in the graphs.

Epigenetic memory inherited from their original tissue have been demonstrated to influence the iPSC differentiation potential [29], suggesting that pericyte myogenic and angiogenic potential could be advantageous in tissue regeneration. Hence, we isolated human adult skin fibroblasts from the same donor of pericytes in order to compare the capability of iPSCs derived from pericytes (PC-derived iPSCs, ALP+CD56<sup>−</sup> subpopulation) and fibroblasts (FB-derived iPSCs) to re-differentiate into muscle cells. Skin fibroblasts, isolated by enzymatic digestion, were characterized by immunofluorescence (Figure S1) and FACS analysis (Figure S1) for the expression of vimentin and CD90.
