Male and female mice (C57BL/6, Jackson Laboratory; Bar Harbor, ME) were bred and housed in animal care facilities of the University of Missouri. Mice (weight, ~ 30 g) were studied when ~ 4 months old. In reporter mice bred on a C57BL/6 background [Cdh5-CreERT2 [23] x ROSA26mTmG (#007676, Jackson Laboratory)], Cre recombination for expression of membrane-bound green fluorescent protein (GFP) in ECs was induced through intraperitoneal injection of 100 μL of tamoxifen (#T5648, Sigma-Aldrich; St. Louis, MO; 10 mg/mL + 5% ethanol in corn oil) on 3 consecutive days; at least 1 week elapsed after the first injection prior to study. All mice were maintained under a 12:12 h light/dark cycle at 22–24 ºC with fresh food and water ad libitum. To avoid any order effect, the collection of data at criterion timepoints was randomized. Prior to performing a muscle injury, intravital microscopy, or tissue collection, a mouse was anesthetized [ketamine (100 mg/kg) + xylazine (10 mg/kg) in sterile saline; intraperitoneal injection]. Mice were euthanized at the end of an experiment by anesthetic overdose and cervical dislocation.

An anesthetized mouse was positioned on an aluminum warming plate to maintain body temperature at 37 ºC. As needed, supplemental injections of anesthetic (~ 20% of initial) were given to maintain a stable plane of anesthesia as confirmed by lack of withdrawal to a toe pinch (monitored every 15 min). The skin overlying the left GM was shaved and sterilized by swabbing 3 × with betadine and 70% alcohol. While viewing through a stereomicroscope, the mouse was positioned on its abdomen and a ~ 1-cm incision was made through the skin to expose the GM, which was continuously irrigated with sterile saline. A hole was made through the center of the GM with a sterile 2-mm diameter biopsy punch (MediChoice #DP0200, Owens & Minor; Mechanicsville, VA) positioned perpendicular to the muscle surface; remaining adhesions around the circumference of the injury were carefully cut with fine-tipped Vannas scissors to free the biopsy from surrounding tissue. Larger vessels were avoided to minimize bleeding. This volume of muscle loss allows for regeneration [22] while creating a void into which the regeneration of microvessels and myofibers could be studied. Anatomical landmarks provided a reference for consistency in the biopsy site. The skin was closed with sterile 6–0 nylon sutures (UNIFY #S-N618R13, AD Surgical; Sunnyvale, CA). The entire procedure required ~ 20 min. For recovery, the mouse was placed on a heated platform, monitored until consciousness and ambulation were restored (~ 2–3 h), then returned to its original cage. Normal activity and behavior were routinely observed within 24 h. Regeneration of tissue components into the wound was evaluated at designated timepoints through 21 days post-injury (dpi) with uninjured mice (0 dpi) serving as controls.

To evaluate cellular damage at the biopsy site (Fig. S1), Evans Blue dye [EBD [24]; 1% solution in sterile saline; #E2129, Sigma] was injected intraperitoneally (10 μL/g body mass) following the surgical procedure. The GM was dissected (as described below) at 1 dpi, and images of EBD staining were acquired with a 4 × objective on an E800 microscope coupled to a DS-Fi3 camera using Elements software (Nikon; Tokyo, Japan).

A mouse was anesthetized, the surgical area was shaved, the mouse was placed on a warming plate, and the skin overlying the GM was removed with scissors. Exposed tissue was continuously superfused with a bicarbonate-buffered physiological salt solution (bbPSS; pH 7.4, 34–35 °C) containing (in mM) 131.9 NaCl2 (Fisher Scientific; Pittsburg, PA), 4.7 KCl (Fisher), 2 CaCl2 (Sigma), 1.17 MgSO4 (Sigma), and 18 NaHCO3 (Sigma) equilibrated with 5% CO2/95% N2. While viewing through a stereomicroscope, the GM was cut from its origin along the lumbar fascia, sacrum, and iliac crest and then reflected away from the body to view its vascular supply from the ventral surface [7, 25].

The exposed GM was spread onto a transparent rubber pedestal (Sylgard 184; Dow Corning; Midland, MI) and pinned at its edges to approximate in situ dimensions. Exposed tissue on the mouse was covered with plastic film (Glad Press n’ Seal) to prevent dehydration, and the preparation was transferred to the stage of a Nikon E600FN microscope, where the GM was equilibrated for 30 min while continuously superfused with bbPSS at 3 mL/min. Supplemental doses of anesthetic were given throughout the experimental protocol (duration, 2–3 h) to maintain a stable plane of anesthesia (as above). To assess vascular perfusion, 200 μL of fluorescein isothiocyanate (FITC) conjugated dextran (70 kDa; 10 mg/mL sterile saline) was injected into the systemic circulation via the retroorbital sinus and allowed to circulate for ~ 10 min [7, 26]. The GM was illuminated with a mercury lamp for fluorescence imaging using an appropriate filter cube. Images were acquired through Nikon Plan Fluor 4x/0.13 or Plan Fluor 10x/0.3 objectives coupled to a low light CMOS FP-Lucy camera [Stanford Photonics, Inc.; Palo Alto, CA (SPI)] and displayed on a digital monitor. Time-lapse images were recorded at 40 frames/s using Piper Control software (SPI).

The GM was removed from a Cdh5-mTmG mouse and placed in a custom imaging chamber with the ventral surface facing the objective to optimize the resolution of the microvasculature. A drop of PBS (~ 10 μL) was added to the chamber and the GM was flattened by placing a glass block (2 cm × 2.5 cm × 1 cm; mass, 7.8 g) on the dorsal surface. Images were acquired with a HC PL APO 10/0.40 CS2 objective on an inverted laser scanning confocal microscope (TCS SP8) using LASX software (all from Leica Microsystems; Buffalo Grove, IL). Images were also acquired with a Dragonfly 200 Confocal Microscope System [Andor Zyla camera coupled to a Leica DMi8 microscope with a Leica HC PL APO 10x/0.45 objective using Fusion software (Oxford Instruments; Abingdon, UK)]. To image the entire wound, tile scans (3 × 3 grid encompassing ~ 3.5 mm × 3.5 mm × 200 μm deep) were stitched. Using maximum projection z-stacks, 10 random microvessels were chosen within a region of interest (ROI; 493 × 493 μm). Diameters of regenerating microvessels were measured at the midpoint between two branch points and averaged for a given GM; 5–7 GM were analyzed at respective dpi timepoints.

An excised GM was immobilized by pinning the edges in the well of a 12-well plate coated with Sylgard 184. After washing with PBS, a region of muscle (~ 5 mm × 5 mm) containing the injury surrounded by undamaged tissue was trimmed and prepared for either whole mount immunostaining or frozen in Tissue Tek OCT compound (VWR International LLC; Radnor, PA) to obtain tissue cross sections for histology.

Whole mount preparations were fixed in 2% paraformaldehyde for 30 min, washed in PBS, and placed in blocking buffer (2% bovine serum albumin, 4% normal donkey serum, 0.5% triton X-100 in PBS) for 30 min. Preparations were incubated overnight at 4 °C with validated primary antibodies: rat monoclonal anti-CD31 [platelet endothelial cell adhesion molecule [PECAM-1; 1:400, clone MEC 13.3, #550,274, BD Pharmingen; San Diego, CA [27]] to identify ECs, rabbit polyclonal anti-CD45 [leukocyte common antigen; 1:200, #ab10558, Abcam; Cambridge, UK [28]] to identify inflammatory cells, and goat polyclonal anti-PDGFRα [1:200, #AF1062, R&D Systems; Minneapolis, MN [29]] to identify fibroadipogenic progenitor cells (FAPs) [30,31,32]. Preparations were washed in blocking buffer, incubated with secondary antibodies for 30 min, washed again, then rinsed in PBS before transfer to an imaging chamber for confocal image acquisition (as above).

For tissue cross-sections, trimmed GM samples were transferred to a cryomold containing OCT compound with 2–0 black silk suture (length, 2 mm) positioned adjacent to the injury for a visual reference. The cryomold was frozen in isopentane cooled in liquid nitrogen and stored at − 80 °C until sectioning. Cross-sections of frozen GM were cut (thickness, 10 μm) with a cryostat (HM 550 Cryostat, Thermo Scientific; Waltham, MA) at the center of the injury with reference to the silk suture.

For fluorescence imaging, sections were fixed in 4% paraformaldehyde for 10 min and stained with CD31 (1:500, BD Pharmingen), mouse monoclonal anti-myosin heavy chain [1:5, MF-20, Developmental Studies Hybridoma Bank, The University of Iowa Department of Biology; Iowa City, IA [33]], rabbit polyclonal anti-laminin [1:200, #PA1-16,730, Invitrogen; Waltham, MA [34]], and appropriate secondary antibodies (1:400, AlexaFluor, Fisher). Prolong Gold containing DAPI (Fisher) was added before slides were coverslipped. Sections were imaged using appropriate filters on an E800 microscope with a DS-Qi2 camera and Elements Software (Nikon).

The distance devoid of myofibers (i.e., a diameter of remaining visible wound) was measured along GM cross-sections accounting for curvatures within the specimen (Fig. 5A, B). Within this region, the total area of CD31+ staining was measured to evaluate microvascular density within the wound that preceded myofiber ingrowth.

To visualize collagen deposited within the provisional matrix in whole-mount preparations, the GM was removed, permeabilized in 0.5% Triton X-100 in PBS, and incubated for 1 h with PicroSirius Red [1% Direct Red in saturated picric acid; [35]]. Thereafter, GM was treated for 30 min with 0.5% acetic acid in ddH2O and rinsed in 100% EtOH. Images were acquired as described for EBD.

Summary data are displayed for individual GM preparations along with means ± s.e.m. Statistical analyses were performed using Prism 9 software (GraphPad Software, Inc.; La Jolla, CA). One-way ANOVA with Tukey’s post hoc tests was used to determine statistical significance between time points with P < 0.05 considered significant. Values for “n” refer to the number of GM analyzed; one GM was studied per mouse.