Established cell surface markers efficiently isolate highly overlapping populations of skeletal muscle satellite cells by fluorescence-activated cell sorting

Abstract

Background: Fluorescent-activated cell sorting (FACS) has enabled the direct isolation of highly enriched skeletal muscle stem cell, or satellite cell, populations from postnatal tissue. Several distinct surface marker panels containing different positively selecting surface antigens have been used to distinguish muscle satellite cells from other non-myogenic cell types. Because functional and transcriptional heterogeneity is known to exist within the satellite cell population, a direct comparison of results obtained in different laboratories has been complicated by a lack of clarity as to whether commonly utilized surface marker combinations select for distinct or overlapping subsets of the satellite cell pool. This study therefore sought to evaluate phenotypic and functional overlap among popular satellite cell sorting paradigms.

Methods: Utilizing a transgenic Pax7-zsGreen reporter mouse, we compared the overlap between the fluorescent signal of canonical paired homeobox protein 7 (Pax7) expressing satellite cells to cells identified by combinations of surface markers previously published for satellite cells isolation. We designed two panels for mouse skeletal muscle analysis, each composed of markers that exclude hematopoietic and stromal cells (CD45, CD11b, Ter119, CD31, and Sca1), combined with previously published antibody clones recognizing surface markers present on satellite cells (β1-integrin/CXCR4, α7-integrin/CD34, and Vcam1). Cell populations were comparatively analyzed by flow cytometry and FACS sorted for functional assessment of myogenic activity.

Results: Consistent with prior reports, each of the commonly used surface marker schemes evaluated here identified a highly enriched satellite cell population, with 89-90 % positivity for Pax7 expression based on zsGreen fluorescence. Distinct surface marker panels were also equivalent in their ability to identify the majority of the satellite cell pool, with 90-93 % of all Pax7-zsGreen positive cells marked by each of the surface marker schemes. The direct comparison among surface marker schemes validated their selection for highly overlapping subsets of cells. Functional analysis in vitro showed no differences in the abilities of cells sorted by these different methods to grow in culture and differentiate.

Conclusions: This study demonstrates the equivalency of several previously published and widely utilized surface marker schemes for isolating a highly purified and myogenically active population of satellite cells from the mouse skeletal muscle, which should facilitate cross-comparison of data across laboratories.

Keywords: Fluorescence-activated cell sorting; Satellite cells; Surface marker.

Fig. 1

Experimental design and applied gating strategy for flow cytometry. a Skeletal muscles were harvested from Pax7-zsGreen animals and mononuclear and myofiber-associated cells were purified using a two-step enzymatic and mechanical digestion procedure. All cells were stained with the viability markers PI and calcein, as well as negatively selecting markers. Due to fluorophore overlap, positive selecting antibodies were split between two staining conditions including all hematopoietic and stromal factors (Sca1, CD31, CD45, Mac1, and Ter119) and either β1-integrin, CXCR4, and VCam1 or β1-integrin, CXCR4, α7-integrin, and CD34. Cells were analyzed and sorted by flow cytometry. b Gating conditions applied to all downstream analysis included initial physical parameter gate to exclude cells larger than our target population, viability selection of cells that were PI⁻ and calcein⁺, and exclusion of hematopoietic and stromal cell types. All comparative flow cytometric analyses (see Figs. 2, 3, and 4) began with this Sca1, CD31, CD45, Mac1, and Ter119 negative cell population (red)

Fig. 2

The majority of skeletal muscle satellite cells isolated by distinct surface marker combinations express Pax7. All gating conditions were set using non-transgenic C57BL/6J mouse myofiber-associated cells to distinguish cells with positive signal for Pax7-zsGreen (see “FMO” control, a–c, top row). a–c The populations marked by β1-integrin and CXCR4, VCam1, or α7-integrin and CD34 are highly enriched for cells expressing Pax7-zsGreen. d Comparative analysis across surface marker strategies shows no significant differences in Pax7-zsGreen expression between groups (n = 12)

Fig. 3

Canonical Pax7-expressing satellite cells are positive for distinct surface marker combinations. All gating conditions were set using Pax7-zsGreen mouse cells stained as FMO controls to distinguish cells with positive signal of the marker population (a–c, top row). a–c Satellite cells expressing Pax7-zsGreen are highly marked by β1-integrin and CXCR4, VCam1, or α7-integrin and CD34 markers. d Comparative analysis shows no significant differences in surface marker composition among Pax7⁺ satellite cells (n = 12)

Fig. 4

High overlap between unique surface marker combinations that enrich for skeletal muscle satellite cells. All gating conditions were set using cells stained as FMO controls to distinguish cells with positive signal of the marker population (a–d, top row). a Cells expressing β1-integrin and CXCR4 markers also express α7-integrin and CD34. b Cells expressing α7-integrin and CD34 also express β1-integrin and CXCR4. c Cells expressing VCam1 also express β1-integrin and CXCR4. d Cells expressing β1-integrin and CXCR4 markers also express VCam1. e Comparative analysis of surface marker presence within surface marker defined populations indicates expression of multiple surface markers on the same satellite cells (n = 12)

Fig. 5

Functional characterization of skeletal muscle progenitor cells isolated by distinct surface marker combinations. a Similar levels of colony formation were seen across cells isolated by β1-integrin and CXCR4, VCam1, and α7-integrin and CD34 sorting paradigms. Data were collected for cells harvested independently from n = 5 mice (two female and three male). Each dot represents colony-forming efficiency of an individual mouse, calculated from analysis of at least 95 wells. Overlay represents mean ± SD. b No differences in myogenic differentiation indices (see “Methods” section) among β1-integrin and CXCR4, VCam1, and α7-integrin and CD34 sorted populations. Data were collected for cells harvested independently from n = 3 female mice. Each dot represents one mouse, with two technical replicates per biological replicate. Overlay indicates mean ± SD. c Representative ×10 images of cultures quantified in (b), derived from sorted β1-integrin and CXCR4 (left), VCam1 (middle), and α7-integrin and CD34 (right) cell populations after 72 h in differentiation media. d Quantification of the percentage of Pax7, MyoD, or MyoG protein expressing cells within myogenic cultures expanded in growth media for 5 days. Error bars represent standard deviations. N = 3, with two technical replicates per biological replicate