Identification and Analysis of Myogenic Progenitors In Vivo During Acute Skeletal Muscle Injury by High-Dimensional Single-Cell Mass Cytometry

The protocol presented here enables the identification and high-dimensional analysis of muscle stem and progenitor cells by single-cell mass cytometry and their purification by FACS for in-depth studies of their function. This approach can be applied to study regeneration dynamics in disease models and test the efficacy of pharmacological interventions.

Skeletal muscle regeneration is a dynamic process driven by adult muscle stem cells and their progeny. Mostly quiescent at a steady state, adult muscle stem cells become activated upon muscle injury. Following activation, they proliferate, and most of their progeny differentiate to generate fusion-competent muscle cells while the remaining self-renews to replenish the stem cell pool. While the identity of muscle stem cells was defined more than a decade ago, based on the co-expression of cell surface markers, myogenic progenitors were identified only recently using high-dimensional single-cell approaches. Here, we present a single-cell mass cytometry (cytometry by time of flight [CyTOF]) method to analyze stem cells and progenitor cells in acute muscle injury to resolve the cellular and molecular dynamics that unfold during muscle regeneration. This approach is based on the simultaneous detection of novel cell surface markers and key myogenic transcription factors whose dynamic expression enables the identification of activated stem cells and progenitor cell populations that represent landmarks of myogenesis. Importantly, a sorting strategy based on detecting cell surface markers CD9 and CD104 is described, enabling prospective isolation of muscle stem and progenitor cells using fluorescence-activated cell sorting (FACS) for in-depth studies of their function. Muscle progenitor cells provide a critical missing link to study the control of muscle stem cell fate, identify novel therapeutic targets for muscle diseases, and develop cell therapy applications for regenerative medicine. The approach presented here can be applied to study muscle stem and progenitor cells in vivo in response to perturbations, such as pharmacological interventions targeting specific signaling pathways. It can also be used to investigate the dynamics of muscle stem and progenitor cells in animal models of muscle diseases, advancing our understanding of stem cell diseases and accelerating the development of therapies.

Skeletal muscle constitutes the largest tissue by mass in the body and regulates multiple functions, from eyesight to respiration, from posture to movement, as well as metabolism¹. Therefore, maintaining skeletal muscle integrity and function is critical to health. Skeletal muscle tissue, which consists of tightly packed bundles of multinucleated myofibers surrounded by a complex network of nerves and blood vessels, exhibits remarkable regenerative potential^1,2.

The main drivers of skeletal muscle regeneration are adult muscle stem cells (MuSCs). Also known as satellite cells, due to their unique anatomical location adjacent to the plasma membrane of the myofiber and beneath the basal lamina, they were first identified in 1961³. MuSCs express a unique molecular marker, the transcription factor paired box 7 (Pax7)⁴. Mostly quiescent in healthy adults, they become activated upon muscle injury and proliferate to give rise to progeny that will (i) differentiate into fusion-competent muscle cells that will form new myofibers to repair muscle damage or (ii) self-renew to replenish the stem cell pool⁵.

At the cellular and molecular level, the process of regeneration is quite dynamic and involves cell-state transitions, characterized by the coordinated expression of key myogenic transcription factors, also known as myogenic regulatory factors (MRFs)^6,7. Prior in vivo developmental studies, lineage tracing experiments, and cell culture work using myoblasts have shown that sequential expression of these transcription factors drives myogenesis, with myogenic factor 5 (Myf5) being expressed upon activation, myogenic differentiation 1 (MyoD1) expression marking commitment to the myogenic program, and myogenin (MyoG) expression marking differentiation^{8,9,10,11,12,13,14}. Despite this knowledge and the discovery of cell surface markers to purify MuSCs, strategies and tools to identify and isolate discrete populations along the myogenic differentiation path and resolve a myogenic progression in vivo have been lacking^15,16,17,18.

Here, we present a novel method, based on recently published research, which enables the identification of stem and progenitor cells in skeletal muscle and the analysis of their cellular, molecular, and proliferation dynamics in the context of acute muscle injury¹⁹. This approach relies on single-cell mass cytometry (also known as Cytometry by Time of Flight [CyTOF]) to simultaneously detect key cell surface markers (α₇ integrin, CD9, CD44, CD98, and CD104), intracellular myogenic transcription factors (Pax7, Myf5, MyoD, and MyoG) and a nucleoside analog (5-Iodo-2′-deoxyuridine, IdU), to monitor cells in S phase^{19,20,21,22,23}. Moreover, the protocol presents a strategy based on the detection of two cell surface markers, CD9 and CD104, to purify these cell populations by fluorescence-activated cell sorting (FACS), therefore enabling future in-depth studies of their function in the context of injury and muscle diseases. While primary myoblasts have been extensively used in the past to study the late stages of myogenic differentiation in vitro, it is not known whether they recapitulate the molecular state of muscle progenitor cells found in vivo^{24,25,26,27,28,29,30}. The production of myoblasts is laborious and time-consuming, and the molecular state of this primary culture changes rapidly upon passaging³¹. Hence, freshly isolated myogenic progenitors purified with this method will provide a more physiological system to study myogenesis and the effect of genetic or pharmacological manipulations ex-vivo.

The protocol presented here can be applied to address a variety of research questions, for example, to study the dynamics of the myogenic compartment in vivo in animal models of muscle diseases, in response to acute genetic manipulations or upon pharmacological interventions, therefore deepening our understanding of muscle stem cell dysfunction in different biological contexts and facilitating the development of novel therapeutic interventions.

Animal procedures were approved by the Danish animal experiments inspectorate (protocol # 2022-15-0201-01293), and experiments were performed in compliance with the institutional guidelines of Aarhus University. Analgesia (buprenorphine) is provided in drinking water 24 h prior to injury for the mice to adapt to the taste. Supplying buprenorphine in drinking water is continued for 24 h post-injury. Together with a subcutaneous (s.c.) injection of buprenorphine at the time of acute muscle injury, buprenorphine in the drinking water after notexin injection will alleviate the pain associated with the injury. While it is recommended to administer a s.c. injection of buprenorphine at the time of acute muscle injury, followed by buprenorphine in the drinking water, buprenorphine in the drinking water prior to injury is optional. However, researchers must follow the animal welfare standards and guidelines established by the appropriate regulatory agency.

NOTE: For single-cell mass cytometry (CyTOF) experiments of injured hindlimb muscles, start at section 1: Analgesia in water 24 h prior to muscle injury until 24 h post-injury. For sorting of muscle stem and progenitor cells from uninjured mice, perform sections 5 and 6: Euthanasia + Skeletal muscle dissection and dissociation, and continue to section 11: Staining with fluorophore-conjugated antibodies for FACS. An overview of the experimental setup and the protocol is shown in Figure 1.

1. Analgesia in water 24 h prior to muscle injury until 24 h post-injury

1. To a dark or aluminum foil wrapped drinking bottle, add 3 mL of buprenorphine (0.3 mg/mL) and fill to 100 mL with filtered water to reach a final concentration of 0.009 mg/mL 24 h pre-injury. Attach to the mouse cage.
2. Remove the drinking bottle 24 h post-injury, and reconnect the mouse cage to the drinking valve system.

2. Preparing for acute injury procedure

NOTE: Use 70% ethanol to disinfect the work bench, nose cone setup and induction box.

1. Prepare a 1.5 mL tube with diluted notexin in PBS (5 µg/mL). Dilute buprenorphine (0.3 mg/mL) in sterile 0.9% saline to 0.015 mg/mL in a light-safe 1.5 mL tube. Keep on ice. Prepare insulin syringes for buprenorphine (28 G, 0.5 mL) and notexin injections (29 G, 0.3 mL).
2. Configure isoflurane-based anesthetic units:
   1. For anesthesia in an induction box, use 3% isoflurane with a flow of 1.5 L/min (50% O₂, 50% atmospheric air).
   2. For maintenance utilizing a nose cone setup, use 1.5% isoflurane with a flow of 0.6 L/min (50% O₂, 50% atmospheric air).

3. Acute injury by notexin injection

CAUTION: Notexin has Phospholipase A₂ activity and is the principal component of venom from the Australian tiger snake (Notechis scutatus), with an intravenous LD₅₀ of 5–17 mg notexin/kg in mice^32,33. In the present protocol, the Tibialis Anterior (TA) muscle of each hindlimb is injected with 10 µL of 5 mg/mL notexin, and the Gastrocnemius (GA) muscle of each hindlimb is injected twice (once into each head of the muscle) with 15 µL of 5 mg/mL notexin. It is important to perform the intramuscular (i.m.) injections correctly to limit damage and frequently inspect the injected animals to ensure minimal pain.

1. Anesthetize the mouse in the induction box (3% isoflurane). When the mouse is unconscious, lower the isoflurane level to 1.5% in the induction box.
2. Weigh and individually mark the mice according to institutional approved procedures. Under the nose cone setup, shave the hindlimbs with a trimmer and transfer back to the induction box. 
3. Calculate buprenorphine volume for each mouse :
4. Mix the notexin solution by pipetting up and down. Prepare 2 insulin syringes (29 G, 0.3 mL) loaded with 10 µL of notexin solution (5 µg/mL) for TA injections, and 4 insulin syringes loaded with 15 µL of notexin solution (5 µg/mL) for GA injection.
   NOTE: The Tibialis Anterior (TA) muscle is located on the anterior side of the lower leg of the mouse, extending from the knee to the ankle. The Gastrocnemius (GA) muscle is a two-headed muscle that is located in the back of the lower leg, superficial to the soleus. It runs from its two heads (medial and lateral) just above the knee to the heel, extending across a total of three joints (knee, ankle, and subtalar joints).
5. Perform intramuscular injections
   1. TA injection (2 injections in total)
      1. Place the mouse in a supine position under the nose cone setup and disinfect the injection site with sterile alcohol wipes (70% ethanol or isopropanol). Insert the needle, bevel down, in the belly of the TA (distal to the mid-belly) at 30° angle. Advance the needle along the muscle, moving parallel to the tibia, to reach the mid-belly of the TA. Inject the notexin slowly and continuously, leaving the needle in place for 10 s before withdrawing it. Do not insert the needle too deep initially (to avoid injecting notexin below the TA into the extensor digitorum longus muscle), and not too far proximally when advancing the needle (to avoid injecting too close to the knee).
   2. GA injection (4 injections in total)
      1. Insert the needle, bevel down, at ~45° angle in the mid-belly of the lateral head of the GA. Inject the notexin slowly and continuously, leaving the needle in place for 10 s before withdrawing it. Rotate the hindlimb and inject notexin in the medial head of the GA muscle as above. Care should be taken not to insert the needle too deeply.
6. Turn the mouse around and inject buprenorphine s.c. using an insulin syringe (28 G, 0.5 mL). Transfer the mouse to an empty recovery cage on a heating plate. Only half of the cage bottom should be placed over the heating plate to allow the mouse to thermoregulate during recovery³⁴. Repeat steps 3.3–3.6 for the remaining mice.
7. When the mouse in the recovery cage is fully awake, move it back into the original cage. Return it to the stable and supplement it with wetted chow. Remove the buprenorphine drinking bottle 24 h after injection.
8. Monitor the mice for 6 h after notexin injection and then every 12 h for 2 days for signs of pain, impaired mobility, and decreased food consumption^35,36.  

4. 5-Iodo-2’-deoxyuridine injection

CAUTION: 5-Iodo-2’-deoxyuridine (IdU) is suspected of causing genetic defects and damaging fertility or the unborn child. Read the safety data sheet (SDS) before handling. Personal protective equipment should be worn during handling. Use a fume hood when weighing the IdU powder. Materials that have been in contact with IdU should be discarded according to local safety regulations.

NOTE: IdU labeling in vivo is used to monitor cell division during the injury time course because IdU, an iodinated thymidine analog, gets incorporated into the DNA of cells in S phase. IdU is injected intraperitoneally (i.p.) at 20 mg/kg body weight 8 h prior to sacrificing the mouse.

1. Use a sonicator at 37 °C to prepare a 2 mg/mL stock solution of IdU in sterile PBS. IdU is light-sensitive; use within the same day or freeze at -20 °C for up to 3 months. If using frozen IdU: thaw, vortex, centrifuge at 10,000 x g for 30 s, and use the supernatant for injections below.
2. Anesthetize the mouse in the induction box (3% isoflurane). Note earmarking, weigh the mice, and calculate the volume of the IdU solution:
3. Perform i.p. injection using an insulin syringe (28 G, 0.5 mL) and transfer the mouse to an empty recovery cage on a heating plate. When awake, move the mouse back to its original cage. Repeat steps 4.2–4.3 for the remaining mice and return them to the stable.

5. Euthanasia

NOTE: See Table 1 for buffer recipes. Prepare wash media (Nutrient mixture F-10 (Ham's), 10% horse serum, 1x Pen/Strep) and filter through a polyethersulfone (PES) membrane into a polystyrene container. Prepare dissociation buffer (wash media supplemented with 650 U/mL Collagenase, Type II) and keep on ice. CyTOF mass cytometry measurements are very sensitive to contaminants. For this reason, it is essential to use reagents of the highest analytical grade for sample processing. To prevent metal contamination, it is highly recommended to use sterile plasticware and new glassware that has never been washed with detergent because many laboratory soaps contain high levels of barium. It is recommended to use double-filtered, distilled, deionized water for reagent preparation. Phosphate-buffered saline (PBS) is prepared in-house. Dilute the 10x stocks to 1x and filter the 1x PBS with 0.2 μm filters. Filter the 1x PBS again at the start of each experiment. Dissection tools must not be cleaned with detergent due to the presence of barium.

1. Prepare the weighing scale, container, and isoflurane-based anesthetic unit with the induction box. Disinfect appropriate dissection tools.
2. Prepare and mark 50 mL tubes for each mouse with 5 mL of dissociation buffer and a Petri dish. Keep dissociation buffer on ice.
3. Anesthetize the mouse in the induction box (3% isoflurane, increase to 5% when unconscious) and perform cervical dislocation.

6. Skeletal muscle dissection and dissociation

1. Dissect TA and GA muscles from both hindlimbs and transfer into the lid of a Petri dish. Use scissors to cut the tissue into a minced slurry (~1 mm³ pieces). Transfer to a 50 mL tube containing 5 mL of ice-cold dissociation buffer. Keep on ice.
2. Repeat the euthanasia procedure (step 5.3) and skeletal muscle dissection and dissociation procedure (step 6.1) until all mice have been processed.
3. Pre-heat all sample tubes in a 37 °C water bath (2 min), and incubate for 45 min on rotation in an incubator (37 °C).
4. Vortex and wash by adding 10 mL of wash media. Centrifuge (380 x g, 10 min, room temperature [RT]), and aspirate to 4 mL.
5. Add 0.5 mL of collagenase (1000 U/mL) and 0.5 mL of dispase (11 U/mL). Vortex and incubate for 20 min on rotation at 37 °C (e.g., in an incubator). Centrifuge  (380 x g, 1 min, RT) and resuspend with a 5 mL pipette.
6. Add 40 µm cell strainers to 50 mL tubes and pre-wet with 5 mL of wash media. Aspirate and eject cell suspension 10 times using a 5 mL syringe with a 20 G needle. Strain through the pre-wetted 40 µm cell strainer. Wash the 50 mL tube with 10 mL wash media, transfer it to the 40 µm cell strainer, and centrifuge (380 x g, 10 min, 4 °C) the cell suspension.

7. Live/dead staining with cisplatin and paraformaldehyde fixation

CAUTION: Cisplatin and paraformaldehyde (PFA) are carcinogenic. Read the SDS before handling. Paraformaldehyde (PFA; 16%) is a skin, eye, and respiratory irritant. Wear personal protective equipment and handle these substances under a fume hood. During fixation of cells, the final concentration of PFA will be 1.6%. Correct protective measures should be taken, and waste should be handled according to local regulations.

NOTE: Prepare cold (4 °C) and warm (37 °C) serum-free DMEM. Prepare DMEM supplemented with 10% FBS, filter through a PES membrane into a polystyrene container, and keep on ice. Prepare PBS and cell stain media (CSM; PBS, 0.5% BSA, 0.02% sodium azide) in a CyTOF-dedicated glass bottle and filter through a PES membrane. CSM can be stored at 4 °C for up to 6 months.

1. Aspirate the supernatant, flick the pellet, and resuspend in 1 mL of cold serum-free DMEM. Count the cells and adjust the cell density to 1 x 10⁶ cells/mL in serum-free DMEM.
2. Add cisplatin stock (25 mM) to a final concentration of 25 µM. Vortex 10 s and incubate for exactly 1 min (the reaction is very time-sensitive). Quench the reaction with ice-cold DMEM+10% FBS (3 times the sample volume) and keep it on ice. Centrifuge (380 x g, 10 min, 4 °C), aspirate the supernatant and resuspend the pellet thoroughly (up to 10 x 10⁶ cells/mL in CSM). Filter the suspension through a 35 µm cell strainer.
3. Fix the cell suspension with filtered PFA stock solution (16%) and pipette up and down (perform under a fume hood) to reach the final PFA concentration (1.6%) (e.g., add 100 µL of 16% PFA to 900 µL of cell suspension). Vortex for 30 s, incubate for 10 min on ice, and wash twice with 2 mL of CSM by centrifugation (800 x g, 5 min, RT).
   NOTE: At this point, the samples can either be snap-frozen on dry ice and stored at -80 °C or used directly for antibody staining. If cells are to be frozen, transfer cells into 5 mL polypropylene tubes, as polystyrene can crack at low temperatures. If freshly fixed cells are used for staining, proceed to step 8.2.

8. Staining with metal-conjugated antibodies

CAUTION: Methanol (MeOH) is highly flammable and corrosive to the respiratory tract. Read the SDS before handling. Wear personal protective equipment and handle this substance under a fume hood. Handle waste in accordance with local regulations.

NOTE: List of antibodies (Ab) targeting surface markers and intracellular markers can be found in Table 2.

Antibody conjugation: Most of the antibodies used in this protocol were conjugated in-house because they were not commercially available. Protocols for metal conjugation of antibodies have been previously published, and conjugation kits are now commercially available^37,38. Immunoglobulin type G (IgG) is compatible with the available conjugation protocols. It is of high importance that the antibody formulation used for metal conjugation is free of cysteine-containing carrier proteins (e.g., bovine serum albumin (BSA)), which can affect conjugation efficiency by competing for the free maleimide groups of the polymer, and can interfere with quantification of the metal-conjugated antibody. The cysteine content of gelatin is much lower than that of BSA. However, it is recommended that if the antibody formulation contains carrier proteins, such proteins are removed prior to conjugation. It is now possible to request BSA- and gelatin-free antibodies from the manufacturer. Small molecule preservatives (e.g., sodium azide, glycerol, and trehalose) are compatible with metal conjugation protocols^37,38.

Antibody titration: After each metal conjugation, antibodies should be titrated to determine the optimal antibody concentration that provides the maximal signal-to-noise ratio. For antibody titration, perform a 6-step two-fold serial dilution and stain both samples known to express (e.g., muscle cells, positive controls) and lack (negative controls) the protein of interest^19,21,37,38.

Prepare fresh Cell-ID Intercalator-Ir (stock = 500 μM; intercalator-ir solution) working solution by diluting the stock to 0.1 μM in PBS/1.6% PFA.

1. Thaw frozen samples for 5 min at RT and centrifuge (800 x g, 1 min, RT). Wash with 2 mL of CSM by centrifugation (800 x g, 5 min, RT). 
2. Prepare a 2.5x surface antibody (Ab) staining mix in CSM. Aspirate the supernatant to ~60 µL and resuspend the pellet thoroughly. Stain the cells by adding 40 µL of surface Ab staining mix and incubating it for 1 h at RT. 
3. Flick the samples every 20 min to mix. Wash twice with 1 mL of CSM by centrifugation (800 x g, 5 min, RT). Aspirate the supernatant and flick the pellet. 
   NOTE: If no intracellular staining is needed, methanol permeabilization is not required. However, if only surface staining is performed, the intercalator-ir solution has to be diluted in a buffer containing a permeabilizing agent (e.g., Maxpar fix and perm buffer) to penetrate the nuclear membrane. Proceed to step 8.5 and dilute the Cell-ID Intercalator-Ir in a buffer containing a permeabilizing agent and 1.6% PFA. Otherwise, continue with the intracellular staining below.
   CAUTION: Iridium is hazardous, and precautions for safe handling must be taken. Iridium is flammable and is an irritant to the eyes/skin. Avoid creating and breathing dust or fumes. Avoid contact with skin and eyes. However, the intercalator-ir solution mixture is provided at a concentration of <1% in water, which is considered non-hazardous according to the globally harmonized system (GHS) of classification and labeling of chemicals. Read the SDS from Standard Biotools before handling. Personal protective equipment should be worn during handling. Materials that have been in contact with the intercalator-ir solution should be discarded according to local safety regulations.
4. To permeabilize the cells, add 0.5 mL of ice-cold MeOH dropwise while vortexing. Incubate for 15 min on ice under a fume hood. Centrifuge (800 x g, 5 min, RT) and then wash twice with 1 mL of CSM by centrifugation (800 x g, 5 min, RT). After the last wash, aspirate the supernatant to ~60 µL and resuspend the pellet thoroughly.
5. Prepare a 2.5x intracellular Ab staining mix in CSM. Stain the cells by adding 40 µL of intracellular Ab staining mix and incubating for 1 h at RT. Flick the samples every 20 min to mix.
6. Wash twice with 1 mL of CSM by centrifugation (800 x g, 5 min, RT). Aspirate the supernatant and flick the pellet. Resuspend the samples in 0.5 mL of intercalator-iridium solution (Table 1) and vortex. Incubate the samples for 1 h at RT or overnight (O/N) at 4 °C (see note below).
   NOTE: Samples can be stored in intercalator-iridium solution at 4 °C for up to 48 h. The staining protocol used in this study has been developed based on pioneering work by the Nolan lab{C}23. It differs from the Standard Biotools protocols, as (i) in-house prepared buffers are used for washing and staining, (ii) the cells are fixed prior to surface staining, and (iii) cells are permeabilized with methanol for intracellular staining with antibodies to transcription factors or signaling molecules. Researchers developing a new protocol must thoroughly test the compatibility of their antibody panel with methanol as a permeabilization agent. When adding new antibodies to a panel, it is recommended to test antibody specificity by flow cytometry using positive and negative control samples prior to metal conjugation and titration.

9. Sample preparation for loading into mass cytometer

NOTE: Cell pellets are very loose when in CAS buffer (Table of Materials). During washes with CAS buffer, do not aspirate to dryness. Instead, keep a residual volume as described below.

1. Vortex and centrifuge (800 x g, 10 min, RT) the samples. Pour off the supernatant (handle as PFA waste) and vortex.
2. Wash with 1 mL of CSM by centrifugation (800 x g, 10 min, RT). Aspirate the supernatant and flick the pellet.
3. Wash with 1 mL of CAS buffer by centrifugation (800 x g, 10 min, RT). Aspirate to ~200 µL. Vortex and add 1 mL of CAS buffer. Take out 5 µL aliquot for cell count. Centrifuge (800 x g, 10 min, RT) and carefully aspirate to ≤50 µL. Do not disturb the pellet.
4. Resuspend the pellet to 1–2 x 10⁶ cells/mL in CAS buffer and add calibration beads (1x; Table of Materials) to a final concentration of 0.1x (e.g., add 100 µL of 1x calibration beads to 900 µL of cell suspension). 
5. Load the sample into the mass cytometer and collect data using a flow rate of 400–500 cells/s.
6. Perform data normalization of FCS files using the CyTOF software or previously developed normalization tools³⁹.
   NOTE: Operation of CyTOF mass cytometers is instrument-specific^40,41. It is recommended to consult the CyTOF user manual prior to operation. The calibration beads are metal-embedded polystyrene normalization bead standards containing known concentrations of the metal isotopes cerium (140/142Ce), europium (151/153Eu), holmium (165Ho), and lutetium (175/176Lu). The calibration beads allow for control for machine sensitivity, which can vary over time, primarily due to the buildup of biological material and variations in plasma ionization over time.

10. CyTOF data analysis

NOTE: For downstream analysis, normalized FCS files can be analyzed locally or uploaded to cloud-based software solutions such as Cytobank, Cell Engine, OMIQ, or FCS Express⁴².

1. Concatenate individual FCS files for each sample into a single file, if necessary.
2. Identify single cells by gating on Iridium-intercalator positive events, which enables discrimination of single nucleated cells from debris or doublets.
3. Gate for live cells by selecting cisplatin negative events. Cisplatin binds covalently to cellular proteins and labels dying and dead cells with compromised membranes to a greater extent than live cells⁴³.
4. Gate on the population of interest, for example, the myogenic compartment (Live/CD45^-/CD31^-/Sca1^-/α₇integrin⁺/CD9⁺) (Figure 2A) and quantify the relative proportion of stem and progenitor cells. This approach requires prior knowledge of cell surface or intracellular marker expression to define individual populations.
5. To perform high-dimensional analysis that will enable the identification of previously unrecognized rare cell subsets within a complex population, export the population of interest, and use clustering algorithms that have been developed specifically for CyTOF data analysis⁴⁴.
   NOTE: Previous work employed the X-shift algorithm, which uses weighted k-nearest neighbor density estimation (kNN-DE) in high-dimensional space to perform unsupervised clustering based on defined parameters¹⁹. X-shift has been shown to be highly effective in identifying rare cell populations⁴⁵.
6. For X-shift analysis, download the vortex software package (from the Nolan lab GitHub page [https://github.com/nolanlab/vortex]) and Java 64-bit⁴⁶. Instructions can be found here: https://github.com/nolanlab/vortex/wiki/Getting-Started.
7. Upload the exported cell populations to a local database and define the clustering parameters. Among the user-defined parameters are (i) the markers used for clustering, (ii) the range of k values (e.g., 5 to 150), and (iii) the number of clustering steps.
   NOTE: Previous work used, as clustering markers, a combination of surface markers known to be expressed in muscle stem cells, novel surface markers identified in a high throughput flow cytometric screen of muscle cells and myoblasts, and myogenic transcription factors (TFs) known to define different stages of myogenesis. This approach enabled the identification of two cell surface markers, CD9 and CD104, whose co-expression pattern distinguishes previously unrecognized progenitor cell populations¹⁹.  The careful choice of clustering markers will allow the researcher to answer specific research questions. For example, if identification and analysis over time of dividing cell subsets is desired, including IdU as one of the clustering markers is recommended. X-shift can perform clustering at multiple k values and automatically identify, by calculating the “elbow point”, the k value that results in optimal cluster number, therefore avoiding under-clustering or over-fragmentation. It is recommended to calculate the elbow point and use the clustering analysis defined by the optimal k value for data visualization and downstream applications.
8. To visualize the spatial relationships between cell populations within the X-shift clusters, perform a force-directed layout, which will generate a 2D map where the distance between clusters indicates their similarity in marker expression in high dimensional phenotypic space. By coloring the map using one marker at a time, it is possible to discover novel cell populations and follow their dynamics. Follow-up quantification by manual gating in software, such as Cytobank and Cell Engine, is recommended.
9. Perform heatmap analysis to quantify the dynamic expression of multiple surface and intracellular proteins throughout the time course and reveal new trends.
   NOTE: When performing an injury time course, it is possible to cluster data from all time points together to generate a map of regeneration. In addition, it is possible to dissect such map into the individual time points to follow the cellular and molecular dynamics of regeneration over time¹⁹.

11. Staining with fluorophore-conjugated antibodies for FACS

NOTE: Cells used for unstained, single-color controls and fluorescence minus one (FMO) controls can originate from the TA and GA set from an extra mouse if available. Alternatively, the quadriceps (upper anterior thigh muscle) can be dissected and digested into a single cell suspension, following the same procedure as for the TA+GA set above and used for controls. Prepare FACS buffer (PBS, 2.5% Goat serum, 2 mM EDTA), filter through a PES membrane into a polystyrene container, and keep on ice. FACS buffer can be stored at 4 °C for up to 1 month. A list of antibodies used for FACS can be found in Table 3.

1. Prepare a lineage mix. Stain each control in 50 µL (containing approx. 3–5 x 10⁵ cells). Calculate the amount of lineage mix needed. Make 20%–30% volume excess. The stock concentration of antibodies used for the lineage mix is 0.2 mg/mL.
   1. In a 0.5 mL tube, add anti-CD45 APC-Cy7, anti-CD31 APC-Cy7, anti-Sca1 APC-Cy7, and anti-CD11b APC-Cy7 to reach final concentrations of 1 µg/mL, 2.5 µg/mL, 2.5 µg/mL, and 0.63 µg/mL, respectively.
2. Prepare all-stain mix. Stain each all-stained sample in 500 µL (containing approx. 3–5 x 10⁶ cells). Make 10% volume excess of all-stain-mix.
   1. In a 1.5 mL tube, add lineage-mix antibodies as above + anti-α₇ integrin PE, anti-CD9 APC, and anti-CD104 FITC to reach final concentrations of 2 µg/mL, 1.2 µg/mL, and 3 µg/mL, respectively.
3. Prepare FACS buffer + DAPI: For 10 mL of FACS buffer, add 1 µL of DAPI (stock = 1 mg/mL) to reach a 100 ng/mL final DAPI concentration.
   CAUTION: DAPI (4',6-diamidino-2-phenylindole dihydrochloride) is classified as a possible skin and respiratory irritant. However, DAPI solution at a concentration of <1% is considered non-hazardous, according to GHS. Read the SDS before handling. Personal protective equipment should be worn during handling. Materials that have been in contact with DAPI should be discarded according to local safety regulations.
4. Resuspend cells from section 6: Skeletal muscle dissection and dissociation:
   1. For all-stained samples, resuspend TA and GA sets from a single mouse in 500 µL of FACS buffer and transfer it to 15 mL tubes for staining.
   2. For controls, use either TA and GA sets or quadricep sets. If using 1 mouse for controls, resuspend cells in 750 µL of FACS buffer. If multiple mice, resuspend each TA and GA or quadricep set in 500 µL of FACS buffer and combine the sets into a single 1 mL sample by taking a fraction from each sample. Add 50 µL of control cells (3–5 x 10⁵ cells) to 5 mL polypropylene tubes for staining.
5. Stain as described in Table 4 for 45 min at 4 °C in the dark.
6. Wash all-stained samples by adding 5 mL of FACS buffer. Wash the controls by adding 1 mL of FACS buffer and centrifuge (380 x g, 10 min, 4 °C).
7. Resuspend all-stained samples in 1 mL of FACS buffer + DAPI. Resuspend the controls in 300 µL of FACS buffer with or without DAPI.
8. Keep cells at 4 °C in the dark until sorting on a flow cytometer with 4 lasers (405 nm, 488 nm, 561 nm, 633 nm).
9. Run the controls and samples on the sorter and create sorting gates using the relevant software. For unstained/single color controls and FMO controls, record 1 x 10⁴ and 0.3–1 x 10⁵ events, respectively. For all-stained samples, record up to 1 x 10⁶ events. Sort all the all-stained samples, or as much as needed, depending on the downstream assay requirements. For the flow cytometer used here (FACSAria III), sort in Purity Mode using a 70 mm nozzle.
10. See Figure 4 for the gating strategy.
    1. Use the unstained and single-color controls to set the voltage for all detectors.
    2. Use the single-color controls to set up the compensation matrix.
    3. Use the FMO controls to establish sorting gates.
11. Distinct muscle stem and progenitor cell populations are sorted into ice-cold FACS buffer.
12. Centrifuge (380 x g, 10 min, 4 °C) the cell populations. Resuspend in the appropriate buffer, count, and continue with downstream analysis.

Here we present an overview of the experimental setup for using this combined approach which includes (i) high-dimensional CyTOF analysis of an acute injury time course by notexin injection to study the cellular and molecular dynamics of stem and progenitor cells in skeletal muscle (Figure 1, top scheme); and (ii) FACS of stem and progenitor cells using two cell surface markers, CD9 and CD104, to isolate these populations and perform in-depth studies of their function (Figure 1, bottom scheme). 8-10-week-old female C57BL/6 mice were used in the experiments described below.

CyTOF analysis of skeletal muscle for identifying stem and progenitor cells and studying their cellular and molecular dynamics during the course of acute injury
We have reanalyzed publicly available, previously published CyTOF datasets from an acute injury time course by notexin injection, where skeletal muscle samples were stained with a panel of 21 metal-conjugated CyTOF-compatible antibodies¹⁹ (Figure 2 and Figure 3). Here we present the hierarchical gating strategy for the identification of stem and progenitor cells in vivo in skeletal muscle (Figure 2A). This approach identifies a sequence of 4 populations, stem cells (SC), and progenitor populations 1-3 (P1, P2, P3). The progenitor populations P1 and P2 correspond to increasingly more mature progenitor cells. P3 was previously identified but not characterized due to its low abundance.

We first identify cells by gating on Ir191/Ir193 double positive events, which enables us to discriminate single-nucleated cells from debris or doublets due to the nucleic acid intercalating properties of this reagent⁴⁷. We then identify live cells by gating on cisplatin negative events, given that cisplatin is a chemical that binds covalently to cellular proteins and labels cells with compromised cell membranes (dying and dead cells) to a greater extent than live cells⁴³. We further enrich for myogenic cells by excluding immune cells (CD45⁺, CD11b⁺), endothelial cells (CD31⁺), and mesenchymal cells (Sca1⁺), and gating on α₇ integrin⁺/CD9⁺ myogenic cells^16,19. Stem and progenitor cells are then identified by visualizing the expression of CD9 and CD104 within the α₇ integrin⁺/CD9⁺ myogenic cell gate using a CD9 (y-axis) by CD104 (x-axis) biaxial dot plot. Stem cells (SC) express intermediate levels of CD9 and lack expression of CD104 (lower left quadrant); P1 progenitor cells express high levels of CD9 and lack expression of CD104 (upper left quadrant); P2 progenitor cells express high levels of both CD9 and CD104 (upper right quadrant), P3 progenitor cells express intermediate levels of CD9 and intermediate to high levels of CD104 (lower right quadrant) (Figure 2A). The CyTOF panel used in the experiment presented here included antibodies to cell surface markers, intracellular myogenic transcription factors and signaling molecules (Figure 2B).

To visualize the expression of the myogenic transcription factors within the different stem and progenitor cell populations, we show a representative CD9 by CD104 biaxial dot plot from an uninjured muscle sample, colored in third dimension by expression of Pax7, Myf5, MyoD, and MyoG, as well as CD9 and CD104 to highlight their differential expression within the different populations (Figure 2C).

To visualize the cellular, molecular and proliferation dynamics of stem and progenitor cell populations during the time course of acute injury, we show representative CD9 by CD104 biaxial dot plots at different injury time points (Day 0, Day 3, Day 6), colored in third dimension by incorporation of IdU (left), which enable the identification of cells in S phase, or expression of MyoD (right), a transcription factor that is upregulated in activated stem cells that have entered the cell cycle (Figure 3A). As previously described, the stem cell population (SC) proportionally increases at day 3 post-injury (red arrow), suggesting expansion, which is corroborated by a large increase in IdU incorporation, indicating cell proliferation (Figure 3A, left). This increase in IdU incorporation is accompanied by increased MyoD expression (Figure 3A, right).

Given the well-established molecular definition of activated muscle stem cells (MuSCs; highly proliferative and MyoD^high), high dimensional CyTOF analysis of muscle cells investigating the expression of known and novel cell surface markers, myogenic transcription factors, and markers of S phase, enabled the identification of an activated stem cell signature based solely on cell surface markers^19,48,49,50. Such signature, which is based on the co-expression of cell surface markers CD98 and CD44, now enables the identification and prospective isolation of this transient MuSC subset at day 3 post-injury, opening the door to studies of stem cell self-renewal that were previously unfeasible¹⁹. To visualize the subset of activated stem cells, we show representative CD98 (y-axis) by CD44 (x-axis) biaxial dot plots of the SC population at different injury time points (Day 0, Day 3, Day 6), colored in the third dimension by IdU incorporation (left) or expression of MyoD (right). Activated stem cells, defined by co-expression of high levels of cell surface markers CD98 and CD44, as well as high MyoD expression and high IdU incorporation, are indicated by a red arrow (Figure 3B). These data highlight the high proliferative state and the activated phenotype of MuSCs at day 3 post-injury.

Purification of stem and progenitor cells by FACS
The gating strategy used to isolate muscle stem and progenitor cell populations by FACS is shown in Figure 4. The main gate is used to exclude debris. The singlets gate enables the exclusion of doublets⁵¹. The live cells gate excludes dying and dead cells that stain with DAPI, and the lineage-negative gate excludes immune cells (CD45⁺/CD11b⁺), endothelial cells (CD31⁺), and mesenchymal cells (Sca1⁺), which are stained with individual antibodies labeled with the same fluorophore APC-Cy7. Cells within the myogenic compartment are gated based on expression of α₇ integrin and intermediate to high expression levels of CD9. Muscle stem and progenitor cell populations are identified based on differential expression of CD9 and CD104 (Figure 4A), and their relative proportion is quantified (Figure 4B). Each gate is established using appropriate FMO controls, which take into account the background signal generated by the combination of the fluorophore-conjugated antibodies used in that control (Figure 4C). SC population is defined based on the expression of intermediate levels of CD9 and lack of CD104 expression (CD9^int/CD104^-). While P1 and P2 express high levels of CD9, CD104 is not expressed in P1 progenitor cells (CD9^high/CD104^-) but is highly expressed in P2 progenitor cells (CD9^high/CD104^high).

Figure 1: Combined approach to analyze and purify muscle stem and progenitor cells in the context of acute muscle injury. Protocol scheme. (Upper scheme) Investigation of stem and progenitor cell dynamics by high-dimensional CyTOF analysis of skeletal muscle in the context of acute injury. The injury time course experimental setup requires intramuscular (i.m.) injection of notexin in the Tibialis Anterior (TA) and Gastrocnemius (GA) muscle of mice 6 (Day -6) or 3 (Day -3) days prior to tissue harvest. IdU injection is performed intraperitoneally (i.p.) 8 h prior (-8 h) to tissue harvest to detect proliferating cells. The TA and GA muscles are dissected, and tissue is dissociated by a combination of mechanical dissociation and enzymatic digestion. Cells are then incubated with cisplatin to detect dying and dead cells, fixed with paraformaldehyde to preserve marker expression, stained with metal-conjugated antibodies, and permeabilized with methanol for intracellular staining. Cells are finally analyzed on the CyTOF instrument. After CyTOF analysis and event acquisition, data are normalized, concatenated, and stored using servers like Cytobank or Cell Engine⁴². Data can be analyzed using (i) traditional biaxial plots, and (ii) high-dimensional clustering algorithms, such as X-shift, a k-nearest neighbor-based algorithm⁴⁷. Cell clusters can be visualized by single-cell force-directed layout visualization as previously described¹⁹. (Lower scheme) FACS of stem and progenitor cells. TA and GA muscles from uninjured mice (homeostasis) are dissected and dissociated as above. Cells are stained with antibodies to lineage markers (CD45, CD11b, CD31, and Sca1), cell surface markers used to define muscle stem cell and progenitor cell populations (α₇ integrin, CD9, and CD104), and DAPI to determine viability. Cells are analyzed on the FACS instrument to establish the sorting strategy and sorted into distinct muscle stem and progenitor cell populations based on differential expression of cell surface markers CD9 and CD104. Please click here to view a larger version of this figure.

Figure 2: Identifying stem and progenitor cells using single-cell mass cytometry (CyTOF). (A) CyTOF gating strategy to identify muscle stem and progenitor cells. Cells are distinguished from debris and doublets by staining with a cationic nucleic acid intercalator (Cell-ID Intercalator-Ir) and gating on Ir191/Ir193 double positive events. Live cells are then identified based on lack of cisplatin staining, which labels cells with compromised cell membranes (dying and dead cells) to a greater extent than live cells. Immune cells (CD45⁺ and CD11b⁺), endothelial cells (CD31⁺), and mesenchymal cells (Sca1⁺) are excluded based on expression of the respective markers. The myogenic compartment is identified by co-expression of α₇ integrin and CD9. Stem and progenitor cells are distinguished by differential expression of CD9 and CD104 in a CD9 (y-axis) by CD104 (x-axis) biaxial dot plot of α₇ integrin⁺/CD9⁺ cells. (B) Summary of CyTOF marker panel displaying antibodies to cell surface markers, intracellular transcription factors, phosphorylated signaling molecules, and reagents for DNA staining (Cell-ID Intercalator-Ir), viability staining (cisplatin), and cell proliferation (IdU). (C) Representative CD9 (y-axis) by CD104 (x-axis) biaxial dot plots of the myogenic compartment (α₇ integrin⁺/CD9⁺ cells) from uninjured mice colored by expression of the transcription factors Pax7, Myf5, MyoD, and MyoG and the cell surface markers CD9 and CD104. Expression of the individual markers in the stem and progenitor populations reveals a myogenic progression. Please click here to view a larger version of this figure.

Figure 3: Resolving the cellular and molecular dynamics of muscle stem and progenitor cells during an acute injury time course. (A) Representative CD9 (y-axis) by CD104 (x-axis) biaxial dot plots of the myogenic compartment (α₇ integrin⁺/CD9⁺ cells) during the injury time course (Day 0, Day 3, Day 6), colored by IdU incorporation (left panels) and MyoD expression (right panels). Expression of the individual markers in the stem and progenitor populations reveals a myogenic progression. The red arrow indicates the expansion of the stem cell (SC) population at day 3 post-injury. (B) (Left) The muscle stem cell (SC) population is highlighted in red in a scheme depicting a CD9 (y-axis) by CD104 (x-axis) biaxial plot. (Right) Representative CD98 (y-axis) by CD44 (x-axis) biaxial dot plots of the stem cell (SC) population during the injury time course (Day 0, Day 3, Day 6), colored by IdU incorporation (left panels) and MyoD expression (right panels). Co-expression of cell surface markers CD98 and CD44 identifies a subset of activated stem cells at day 3 post-injury (red arrow), which is marked by high IdU incorporation (left panels) and increased MyoD expression (right panels). Please click here to view a larger version of this figure.

Figure 4: Sorting stem and progenitor cells for downstream assays using a unique strategy defined by cell surface markers CD9 and CD104. (A) Hierarchical gating strategy used to identify stem and progenitor cells during a sort. The main gate, based on FSC and SSC, excludes debris. The singlets gate based on FSC-Area and FSC-Height is used to exclude doublets. The lineage gate is used to exclude immune cells (CD45⁺ and CD11b⁺), endothelial cells (CD31⁺), and mesenchymal cells (Sca1⁺), stained with antibodies conjugated to the same fluorophore, APC-Cy7. The α₇ integrin⁺/CD9⁺ gate is used to define cells within the myogenic compartment. Stem and progenitor cells within the myogenic compartment are distinguished based on differential expression of cell surface markers CD9 and CD104. (B) Quantification of stem and progenitor cells as a fraction of α₇ integrin⁺/CD9⁺ cells. (C) Fluorescence minus one (FMO) controls are used to establish positive gates for the different populations in the hierarchical gating strategy (A). Please click here to view a larger version of this figure.

Table 1: Table of buffer recipes. Buffer type, content, and description of the preparation. Please click here to download this Table.

Table 2: Antibodies and reagents used for CyTOF analysis. See also Figure 2B. Most antibodies used for CyTOF were conjugated in-house with the indicated metals. Please click here to download this Table.

Table 3: Antibodies used for FACS of stem and progenitor cells. Please click here to download this Table.

Table 4: Staining scheme used for FACS. Samples are categorized into unstained, single-stained color controls (single), FMO controls (FMO), and all-stained samples. The antibody concentration used in the current experiment is given. It is important to note that the addition of DAPI is performed during the final cell resuspension. Please click here to download this Table.

Skeletal muscle regeneration is a dynamic process that relies on the function of adult stem cells. While prior studies have focused on the role of muscle stem cells during regeneration, their progeny in vivo has been understudied, primarily due to a lack of tools to identify and isolate these cell populations^15,16,17,18. Here, we present a method to simultaneously identify and isolate muscle stem and progenitor cells in vivo in skeletal muscle and to resolve their cellular and molecular dynamics in the context of acute muscle injury¹⁹.

The approach presented here, which combines high-dimensional CyTOF analysis with cell sorting, provides tools to gain mechanistic insights into how cell-state transitions drive effective tissue regeneration, and reveal alterations in stem cell function during aging and muscle diseases⁵². Changes in the relative proportion of the different populations in genetic mouse models or upon pharmacological intervention can provide mechanistic insights into the process of myogenesis and elucidate the role of individual molecules and downstream signaling pathways in stem cell self-renewal or differentiation. Such changes need to be carefully evaluated because they can stem from alterations in quiescence, cell survival, progression time from stem to progenitor cells, or accumulation of one subset over another due to a differentiation block, which is characteristic of cancer cells^53,54,55,56. The power of CyTOF analysis is the ability to detect up to 40 different cell surface and intracellular proteins simultaneously on a single-cell basis, allowing the study of cell cycle status, DNA damage, and post-translational modifications, to name a few²¹. Importantly, CyTOF analysis enables easy detection of IdU without the need for intracellular staining, which might affect the detection of surface antigens^19,20. This method, therefore, offers the opportunity to distinguish different mechanisms by which stem cell function is modulated by injury or disease because it enables investigation of cellular function and molecular alterations in freshly isolated muscle tissue, such as cell survival, cell cycle status, intracellular signaling, and epigenetic modifications, by simultaneous antibody-based detection of multiple proteins⁵⁷. Molecular and functional analysis of muscle stem cells and their progeny can therefore be performed in a variety of biological contexts or pharmacological treatments. Importantly, the ability to simultaneously purify stem and progenitor cells allows in-depth functional studies of their properties ex-vivo.

This protocol contains critical steps that should be performed with careful planning to ensure high-quality data and reproducibility of the results. Notexin injection has to be performed with high precision in order to obtain reproducible results. Moreover, for reproducibility purposes, it is important to use the same lot of notexin and test the toxin potency prior to each experiment⁵⁸. In addition to the injection of a myotoxin (notexin or cardiotoxin), acute skeletal muscle damage can be induced by mechanical (eccentric contraction), chemical (e.g., Barium Chloride (BaCl₂)), and freeze injuries^{58,59,60,61,62,63,64,65}. It should be noted that BaCl₂ injection is not an ideal injury method for CyTOF experiments because it can introduce high levels of Barium in some of the samples, which can (i) lead to signal spillover in adjacent channels, (ii) generate oxides in M+16 channels, and (iii) damage the CyTOF detector in the long-term⁶⁶. Myotoxic injury is usually preferred because it is minimally invasive, does not require open surgery (e.g., freeze injury), and can be applied to multiple muscles. Skeletal muscle dissociation, which employs both mechanical and enzymatic digestion to a single-cell suspension, is also a critical step, which, if not performed according to protocol, can lead to under- or over-digestion of the muscle tissue, therefore impacting yield. Key factors in the dissociation process are the purity and activity of the enzymes, the extent of tissue cutting, and the length of incubation^18,67.

Importantly, for CyTOF analysis, careful sample preparation and processing is critical to obtain high-quality data. The staining procedure to determine viability, which employs cisplatin staining, is extremely time-sensitive and needs to be completed within 1 min⁴³. A high-quality single-cell suspension is key to preventing aggregate formation during fixation and downstream processing. The permeabilization is also a critical step, and methanol, the permeabilization reagent used here, might not be compatible with the detection of intracellular proteins not described in this protocol^{19,23,68,69,70,71,72}. For this reason, it is important to test several permeabilization reagents when adding to the panel new antibodies to detect intracellular proteins.

Limitations of CyTOF analysis include (i) the inability to simultaneously analyze and prospectively isolate cell populations; (ii) the need to develop and test panels of in-house metal-conjugated antibodies, as some of the antibodies used in this protocol are not commercially available in a metal-conjugated format (Table 2); (iii) the high cost of instrumentation and reagents^19,66,73. Among its several advantages are (i) the ability to simultaneously analyze up to 40 markers, which enables in-depth studies of rare biological samples; (ii) the low background due to the use of rare earth elements for antibody conjugation; (iii) the minimal overlap between channels, due to the high precision of mass resolution; and (iv) the ease of investigating cell proliferation by IdU incorporation without the need for intracellular staining⁷⁴.

Therefore, this technology offers the opportunity to build on the antibody panel presented here (e.g., by including antibodies to additional intracellular markers) in order to study the molecular properties of stem and progenitor cells^20,23,75. When developing an updated CyTOF panel to detect additional surface and/or intracellular proteins, it is important, in order to obtain high-quality data, to determine antibody specificity and investigate cross-reactivity, which could be tissue-specific. Hence, the addition of novel antibodies to the panel requires extensive testing by flow cytometry using positive (e.g., overexpression) and negative (e.g., knock out) muscle tissue samples^19,76. Moreover, to maximize signal and minimize signal spillover, it is important to choose metal isotopes in the high sensitivity range (153-176 Da) for the detection of low abundance targets, while taking into account the potential for oxidation of lighter isotopes that should be chosen for detection of high abundance targets. Finally, once the antibodies have been tested and metal-conjugated, it is essential to perform a titration to identify the antibody concentration that provides the optimal signal/noise ratio^21,66. While building a CyTOF panel can be time-consuming, once established, a new CyTOF panel can provide unprecedented insights.

An alternative to CyTOF analysis, to reduce cost and circumvent the need for the CyTOF instrument and in-house antibody conjugation, will involve the design of multiple smaller panels of commercially available fluorophore-conjugated antibodies to perform traditional flow cytometry analysis at a smaller scale. However, such an approach would not be suitable for analyzing precious samples with limited availability (e.g., rare knock-out or transgenic animals).

When performing multiparametric flow cytometry or cell sorting, it is key to include both single-color controls for establishing a compensation matrix, and FMO controls for establishing the background for individual populations and/or subsets^77,78,79 (Figure 4C). FMO controls, which exhibit, in addition to autofluorescence, the background signal generated from all the other fluorophore-conjugated antibodies, are critical for establishing positive gates that define cell populations in a hierarchical gating strategy. Sort experiments aimed at purifying multiple populations should include non-overlapping gates, and purity checks should be performed to ensure enrichment of the populations of interest⁸⁰. Cell sorting of stem and progenitor cell populations using cell surface markers CD9 and CD104 can be used to perform in-depth studies of these subsets ex-vivo, such as (i) RNA-seq or protein-based analysis in response to perturbations (e.g., genetic manipulations, pharmacological interventions); (ii) drug screenings to identify targets that preserve stemness (e.g., to improve regeneration - stem cell therapy), or induce differentiation (e.g., to suppress cancer cell growth); and (iii) functional assays (e.g., cell transplantation upon ex-vivo genetic manipulations or pharmacological interventions).

In summary, the approach presented here combines CyTOF analysis and cell sorting to enable in vivo studies of the cellular and molecular dynamics of muscle stem and progenitor cells in response to genetic manipulations or pharmacological interventions, as well as studies aimed at investigating the response of prospectively isolated cell populations in culture. This approach can be used to develop drug screening strategies aimed at identifying compounds that push cells toward differentiation to target cancer cells, or enhance stemness for stem cell therapy applications. Such an ex-vivo platform would be a powerful tool to gain a deep understanding of stem cell function and develop targeted therapeutic interventions. Finally, the ability to identify and prospectively isolate the activated stem cell population in response to injury, based solely on cell surface marker expression (CD98^high/CD44^high), can open the door to studies that were previously not feasible, such as studies aimed at understanding stem cell return to quiescence or uncovering the molecular defects that impair muscle stem cell regenerative function in aging, and muscle diseases.

The authors declare no conflict of interest.

We thank the members of the FACS Core Facility in the Department of Biomedicine at Aarhus University for technical support. We thank Alexander Schmitz, the manager of the Mass Cytometry Unit at the Department of Biomedicine, for discussion and support. Scientific Illustrations were created using Biorender.com. This work was funded by an Aarhus Universitets Forskningsfond (AUFF) Starting Grant and a Start Package grant (0071113) from Novo Nordisk Foundation to E.P.