Growth and differentiation potential of main- and side-population cells derived from murine skeletal muscle

Available online at www.sciencedirect.com R

Experimental Cell Research 291 (2003) 83–90

www.elsevier.com/locate/yexcr

Growth and differentiation potential of main- and side-population cells derived from murine skeletal muscle Tetsuro Tamaki,a,* Akira Akatsuka,b Yoshinori Okada,b Yumi Matsuzaki,d Hideyuki Okano,d and Minoru Kimurac a

Department of Physiology, Division of Human Structure and Function, Tokai University School of Medicine, Bohseidai, Isehara, Kanagawa 259-1193, Japan b Laboratory for Structure and Function Research, Tokai University School of Medicine, Bohseidai, Isehara, Kanagawa 259-1193, Japan c Department of Molecular Life Science, Tokai University School of Medicine, Bohseidai, Isehara, Kanagawa 259-1193, Japan d Department of Physiology, Keio University School of Medicine, 35 Shinanomachi, Shinjyuku-ku, Tokyo 160-8582, Japan Received 14 March 2003, revised version received 19 June 2003

Abstract Skeletal muscle-derived CD34⫹/45⫺ (Sk-34) cells were identified as a new candidate for stem cells. However, the relationship between Sk-34 cells and side-population (SP) cells is unknown. Here, we demonstrate that Sk-34 cells prepared from murine skeletal muscles consist wholly of main-population (MP) cells. The Sk-34 cells included only a few SP cells (1:1000, SP:MP). Colony-forming units of Sk-34 cells of both SP and MP possessed the same potential to differentiate into adipocytes, endothelial, and myogenic cells and showed the same colony-forming activity (1.6%). In addition, the colony-forming units of the CD34⫺/45⫺ (double negative: DN) population were found to begin CD34 expression and to possess the potential to differentiate into myogenic and endothelial cells. We also found that expression of CD34 antigen precedes MyoD expression during the myogenic process of DN cells. Furthermore, both Sk-34 and DN cell populations were mostly negative for CD73 (93–95%), whereas the CD45⫹ cell population was ⬎25% positive for CD73, and this trend was also seen in bone marrow-derived CD45⫹ cells. These results indicate that the MP cell population is about 99.9% responsible for the reported in vitro myogenic-endothelial responses of skeletal muscle-derived cells. © 2003 Elsevier Inc. All rights reserved. Keywords: CD34; CD45; Side-population cells; Main-population cells; Endothelial cells; MyoD; CD73; Myoblast; Bone marrow; Stromal cell

Introduction In order to open new pathways for tissue reconstitution therapy via cell transplantation, researchers continue to seek more convenient sources of stem cells. Several tissue-specific stem cells have been identified in adult brain [1,2], bone marrow [3,4], and skeletal muscle [5,6]. Skeletal muscle is the largest organ in the body, comprising about 40 – 50% of total body mass, and presumably, it allows donor cells to be obtained relatively easily and safely. Classically, myoblasts in postnatal muscle have been considered to be derived from satellite cells located between the plasma membrane and the basal lamina of muscle fibers. * Corresponding author. Fax: ⫹81-463-95-0961. E-mail address: [email protected] (T. Tamaki). 0014-4827/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0014-4827(03)00376-8

More recently, hematopoietic and myogenic stem cell populations in the muscle, called side population (SP) cells, have been purified based on the efflux of the fluorescent dye Hoechst 33342 [5,6]. SP cells were shown to be clearly distinct from satellite cells in mutant mice lacking Pax-7, which exhibited a complete absence of satellite cells, while a normal population of SP cells was present [7]. In a recent study, we also identified “myogenic-endothelial” progenitor cell populations residing in the interstitial spaces of skeletal muscle by immunohistochemistry and immunoelectron microscopy based on an expression of CD34 antigen [8]. We characterized these cells using fluorescence-activated cell sorting (FACS) on the basis of cell surface antigen expression, and sorted them as a CD34⫹ and CD45⫺ fraction from enzymatically isolated cells. Cells in the CD34⫹/CD45⫺ fraction (designated Sk-34

84

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

cells) were ⬃94% positive for Sca-1 and mostly negative (⬍3% positive) for CD14, 31, 49, 144, c-kit, and FLK-1. This demonstrates that Sk-34 cells are not committed endothelial progenitors because endothelial progenitor cells are positive for CD31, FLK-1, and CD144 (VE-cadherin) as well as for Sca-1 [9 –11]. However, Sk-34 cells formed colonies in clonal cell culture, and colony-forming units (CFU) displayed the potential to differentiate into adipocytes, endothelial, and myogenic cells. Furthermore, Sk-34 cells fully differentiated into vascular endothelial cells and skeletal muscle fibers in vivo after transplantation. However, immediately after sorting, Sk-34 cells expressed c-met mRNA, but did not express any other myogenic cell-related mRNA, such as MyoD, myf-5, myf-6, myogenin, M-cadherin, Pax-3, and Pax-7. After culturing for 3 days, however, these cells expressed mRNA for all myogenic markers [8]. Interestingly, both the Sk-34 and the CD34⫺/45⫺ (double negative: DN) cell fractions also expressed Bcrp1/ ABCG2 mRNA, which has been shown to be sufficient to confer the “SP phenotype” [12]. Thus, it was suggested that at least some of the muscle SP cell activity described by Gussoni et al. [5] and Jackson et al. [6] might be attributed to the behavior of Sk-34 cells. The aim of the current study was to clarify the relationship between SP cells and Sk-34 cells derived from the skeletal muscle. The growth and differentiation potentials of SP-Sk-34, MP-Sk-34, SP-DN, and MP-DN fractions were compared using clonal cell culture, and were characterized by immunocytochemistry and flow cytometry.

suspension was incubated for 90 min at 37°C. After Hoechst staining, cells were centrifuged, resuspended in cold HBSS, and then stained with fluorescein isothiocyanate (FITC)conjugated anti-mouse CD34 (RAM34) and phycoerythrin (PE)-conjugated anti-mouse CD45 (30-F11). Cell analysis and sorting were performed using a triple laser VantageSE (Becton Dickinson, CA). Hoechst 33342 was excited at 350 nm, and fluorescence emission was detected using 405/ BP30 and 585/BP20 optical filters against Hoechst blue and Hoechst red, respectively. A 555-nm-long pass dichroic mirror (Omega Optical Inc.) was used to separate emission wavelengths. Both Hoechst blue and red fluorescence were shown using a 488-nm Argon laser, and live cells were counted after cells positive for PI (propidium iodide) were excluded. After collecting 105 events, MP and SP populations were defined, as described previously [14]. The muscle SP cell population was confirmed by addition of 50 ␮M verapamil. Biotin-conjugated anti-mouse CD34 and streptavidin-APC and FITC- and PE-conjugated CD45 were used for the flow cytometric analysis of enzymatically extracted cells (EECs) and cultured Sk-34 and DN cells. All antibodies were purchased from Pharmingen (San Diego, CA). PE-conjugated Sca-1 (stem cell antigen) and CD73 (ecto5⬘-nucleotidase) were also used to compare the characteristics between fractionated skeletal muscle-derived cells and bone marrow-derived mononuclear cells. In this respect, we obtained nucleated whole bone marrow cells from femurs of the same mice that were used for muscle analysis. Nucleated bone marrow cells were prepared by suspension in Histopaque 1083 (Sigma, St. Louis, MO) and used as a positive control in CD73 staining.

Materials and methods Clonal cell culture Mouse strain C57BL/6 mice were used for all cell cultures, immunocytochemistry, and flow cytometric characterizations. Cell purification and characterization Interstitial cells were extracted from entire muscles of the thigh and lower legs (tibialis anterior, extensor digitorum longus, soleus, plantaris, gastrocnemius, and quadriceps femoris) of 3- to 8-week-old mice using an isolation method for intact, living individual muscle fibers associated with satellite cells, as described previously [13]. Entire muscles were treated with 0.1% collagenase type IA (Sigma-Aldrich, Tokyo, Japan) in Dullbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum (FCS) with gentle agitation for 2 h at 37°C. Extracted cells were filtered through a 70-, 40-, and then a 20-␮m nylon mesh, washed, and resuspended at 106 cells/ml in Hanks balanced salt solution (HBSS, GIBCO) containing 10% FCS and 10 mM Hepes. Hoechst 33342 (Sigma-Aldrich) was added to a final concentration of 5 ␮g/ml and the

Purified CD34⫹/45⫺ cells (Sk-34) were plated in Methocult GFH4434V (StemCell Tech., Vancouver, Canada) complete methylcellulose medium and incubated in humidified atmosphere at 37°C in 5% CO2. All cultures were performed in quadruplicate and scored at 14 days of culture using an inverted microscope. The growth potential of purified CD34⫺/45⫺ (Sk-DN) cells was also determined using a clonal cell culture system based on a human collagen gel (CollagenCult H4742, StemCell Tech.) with 10 ng/ml bFGF and 20 ng/ml EGF. Immunostaining The specific cellular composition of each colony was determined by staining with monoclonal anti-MyoD (5.8A, Dako), anti-CD34 (RAM34) antibody, and fluorescein-labeled Isolectin B4 (Griffonia Simplicifolia Lectin I, Vector Lab., Burlingame, CA), and evaluating uptake of acetylated low-density lipoprotein labeled with 1,1⬘-dioctadecyl3,3,3⬘,3⬘-tetramethylindocarbocyanine perchlorate (DiI-AcLDL, Biomedical Technologies Inc., Stoughton, MA).

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

When staining for putative myogenic colonies in methylcellulose, each colony was manually extracted using a finetip glass pipette and then fixed with 4% paraformaldehyde/ 0.05 M phosphate buffer (4% PFA/PB) for 10 min. Cells were subjected to cytospin (1000 rpm) and then stained with monoclonal anti-MyoD. For sphere-forming cells in collagen gel, cells were resuspended in 0.1% collagenase (10 min at 37°C) and fixed with 4% PFA/PB. Cells were then subjected to cytospin and stained. Immunoreactions were visualized with streptavidin-biotin complex and 3,3-diaminobenzidine (DAB; Wako Pure Chemical, Osaka, Japan) and 4-chloro-1-naphthol (Wako). For the remaining adhesive and spreading cells in methylcellulose medium, the medium containing methylcellulose was washed out by DMEM containing 5% FCS, and the cells were cultured in DMEM containing 10 ␮g/ml DiI-Ac-LDL with 5% FCS for 3 h. The medium was then changed to DMEM containing 5 ␮g/ml Isolectin B4 with 5% FCS, and cells were incubated for 15 min. After labeling with Isolectin B4, cultures were washed with PBS, fixed using 10% formaldehyde neutral buffer solution, and stained with oil red O, a marker of lipid deposition. Flow cytometric analysis was also performed for the total Sk-34 and DN cells (a floating and/or weakly attached cell population and an adhesive, spreading cell populations) at 10 days of culture. For this purpose, an adhesive, spreading cell populations were resuspended by 0.05% trypsin solution containing 0.53 mM EDTA.

Results MP and SP population in the enzymatically extracted cells To clarify the relationship between SP cells and Sk-34 cells, we sorted SP and MP cells using Hoechst dye efflux [14], and further divided the two populations based on expression of cell-surface antigens CD34 and CD45. The typical pattern of isolation and characterization of muscle MP and SP fractions from EECs is shown in Fig. 1. The fraction of SP cells was small (0.08% of total cells), whereas MP cells accounted for 93% of total cells (Figs. 1A, C, and D). Both MP and SP fractions were stained for CD34 and CD45 to divide them into four subpopulations: CD34⫺/45⫹, CD34⫹/45⫹, CD34⫹/45⫺, and CD34⫺/ 45⫺ (Figs. 1B and C). The SP cells were about 58% CD34⫹/45⫺ (Sk-34), 36% CD34⫺/45⫺ (DN), and the remaining 6% were in the CD45-positive fractions (CD34⫺/45⫹ and CD34⫹/45⫹)(Figs. 1C and D). In the case of MP cells, there were about 71% in Sk-34, 18% in DN, and the remaining 10% were in the CD45⫹ fractions. The number of MP-Sk-34 cells was about 1000-fold higher than the number of SP-Sk-34 cells (Fig. 1D). The number of MP-DN cells was also about 360 fold higher than SP-DN cells (Fig. 1D).

85

Fig. 1. Isolation and characterization of muscle MP and SP cells (A). MP and SP cells were further characterized by CD34 and CD45 (B, C). Both populations were divided into four subpopulations: CD34⫺/45⫹, 34⫹/ 45⫹, 34⫹/45⫺ (Sk-34), and 34⫺/45⫺ (DN). The percentage of total MP and SP in each fraction is shown in D. Values are calculated based on total events in A and expressed as mean ⫾ SE for five trials. In parentheses, the mean frequency (%) of the four subpopulations in the MP and SP gate (mean ⫾ SE) is shown (D).

Growth and differentiation characteristics, MP vs SP cells To determine the growth characteristics and multilineage differentiation of purified SP-Sk-34 and MP-Sk-34 cells in vitro, we performed clonal cell culture in a semisolid medium. Both cell types began to form colonies within 5–7 days when cultured in medium containing methylcellulose (Figs. 2A and B). Some of these colonies formed spherelike shapes (Figs. 2C and D). As the number of cells gradually increased, two distinct populations of cells were seen in colonies after 7–10 days: a floating and/or weakly attached cell population and an adhesive, spreading cell population. This is one of the typical characteristics of Sk-34 cells [8]. Both cell populations continuously increased their numbers and showed spontaneous, intermittent contractions. Numerous myotubes were also observed in the colony after 14 days of culture (Figs. 2E–H). To characterize the cells composing these colonies in MP-Sk-34 and SP-Sk-34 cells after 10 days of culture, floating and/or weakly attached cells in individual colonies were picked up by micropipette, subjected to cytospin preparation, and stained for MyoD. The majority of these cells were strongly or weakly positive for MyoD (Figs. 2I and J). Expression of MyoD was relatively low in larger cells. Interestingly, a few cells, typically smaller cells, still expressed CD34 while coexpressing MyoD (Figs. 2K and L). Most of adhesive cell populations of MP- and SP-Sk-34 demonstrated uptake of DiI-Ac-LDL and were positive for Isolectin B4 (Fig. 2M–O; MP and 2Q–S; SP). In addition, the total Sk-34 cells after 10 days of culture showed totally CD45 negative (Fig. 2U). Thus, it was considered that the adhesive cell populations demonstrating uptake of DiI-AcLDL and positive for Isolectin B4 were endothelial cells.

86

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

Fig. 2. Growth and differentiation of MP-Sk-34 and SP-Sk-34 cells in vitro. Colonies derived from each population after 7 days (A–D), 10 days (I–T), and 14 days (E–H) under the same culture conditions are shown. Typically, colonies were composed of floating and/or weakly attached large round cells and adherent spreading cells (A, B). Some colonies became roughly spherical (C, D). After 14 days of culture, numerous myotubes were formed in colonies of both populations (E, F). Panels G and H are higher magnification views of E and F. In cytospin preparations, the majority of weakly attached cell populations expressed MyoD after 10 days of culture (I, J, brown reaction), and note that a few small round cells coexpressed MyoD (brown) and CD34 (K, L, dark purple, arrow). Adherent cells remaining at the bottom of the plate, after washing the semisolid medium with the floating and/or weakly attached cells, are shown in (M–T). Most of these cells demonstrated DiI-Ac-LDL uptake and were positive for Isolectin B4 (M–O and Q–S), and several cells showed oil droplet-like staining typical of fat cells (P, T). The insets of P and T show higher magnification views of the region outlined in the panel. Arrows in M–P and Q–T show the same cell in the MP and SP colonies. Resuspended total cells in Sk-34 cell cultured at 10 days showed totally CD45 negative (U). Colony-forming activity (%) of MP-Sk-34 and SP-Sk-34 cells was 1.6 ⫾ 0.08 and 1.6 ⫾ 0.07. Bars ⫽ 10 ␮m.

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

87

Fig. 3. Growth and differentiation of MP-DN and SP-DN cells in vitro. Colonies derived from MP-DN (A, B, E–G), and SP-DN (C, D) in collagen culture containing bFGF and EGF are depicted. DN cells in both populations typically formed sphere-like colonies at 7 days of culture (A, C). These spheres gradually increased in diameter and finally formed numerous myotubes at 14 d (B, D). When the spheres in A and C were resuspended in collagenase and subjected to cytospin and MyoD staining, about 60% of cells were positive for MyoD (E, brown). Adherent cells were also observed in the colonies (F), as they were in the Sk-34 colonies, and some cells demonstrated uptake of DiI-Ac-LDL at 10 days of culture (G, arrows). Interestingly, when the cells in the 4-day cultures were resuspended and analyzed by flow cytometry using CD34 and CD45 antigens, a lot of cells in both MP-DN and SP-DN cells expressed CD34 (I, J), and the relative number of CD34⫹ cells decreased after 10 days of culture (K). Note that there were no CD34⫺/45⫹ cells, and very few cells were observed in CD34⫹/45⫹ fraction. For comparison, the characteristics of total DN cells immediately after sorting (before culture) were shown in H. Colony-forming activity (%) in MP-DN and SP-DN cells was 9.8 ⫾ 0.07 and 9.7 ⫾ 0.2, respectively. Bars in A–G ⫽ 10 ␮m.

We also observed several cells that had an oil droplet-like structure (typical of fat cells) in the adhesive cell populations of MP- and SP-Sk-34 (Figs. 2P and T). The colonyforming activity was the same (1.6%) in both MP-Sk-34 and SP-Sk-34 populations. Characteristics of DN cells To assess the growth and differentiation potential of DN cell populations, we performed collagen-based clonal cell culture with bFGF and EGF (see Materials and methods) on purified MP-DN and SP-DN cells (Fig. 3). Both MP-DN (Fig. 3A) and SP-DN (Fig. 3C) cells formed sphere-like colonies after 4 –7 days of culture. The diameter of these spheres gradually increased, and the colonies formed numerous myotubes after 14 days of culture (Figs 3B and D). The spheres were tested for MyoD expression by resuspension in collagenase and subjected to cytospin preparation and staining for MyoD. At 4 –5 days of culture, a few MyoD⫹ cells were observed (data not shown). After 7 days, a majority of cells expressed MyoD (Fig. 3E). Adherent cells were also observed in DN cell culture at 10 days, as in

Sk-34 culture, and these cells demonstrated DiI-Ac-LDL uptake (Figs. 3F and G, arrows). In addition, when total cells in DN cell culture was resuspended by collagenase and trypsin/EDTA and analyzed by flow cytometry using CD34 and CD45 at 10 days, there were totally CD45 negative (Fig. 3K). Interestingly, the colony-forming activity of DN cells were about 6-fold higher than that of Sk-34 cells (about 10% both in MP-DN and SP-DN cells, while 1.6% in Sk-34). After 4 days of culture, when the cells were resuspended and stained for CD34 and CD45, over 86% of DN cells from both the MP and SP fractions expressed CD34 (Figs. 3I and J). The proportion of cells expressing CD34 was lower at 10 days of culture (Fig. 3K), while an increase in the MyoD-positive population was observed around this time (Fig. 3E). Expression of CD73 in the bone marrow cells, Sk-34, DN, and CD45⫹ fractions To test the possibility that the Sk-34 and DN cells include bone marrow-derived stromal cells, these cells were stained by CD73 as well as Sca-1 (Fig. 4). In this respect,

88

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

Fig. 4. Isolation and characterization of Sk-34, DN, and CD45⫹ cell populations. Shown are the basic sorting patterns of muscle-derived cells (A), and bone marrow-derived mononuclear cells (F). Fractions, enclosed by squares in A and F were further characterized by Sca-1 (B, C) and CD73 (D–H). Both Sk-34 and DN cells were ⬎94% (C) and ⬎26% (B) positive for Sca-1 and ⬎95% (E) and ⬎93% (D) negative for CD73. The skeletal muscle-derived CD45⫹ fraction contained ⬎25% CD73⫹ cells (H). Similarly, bone morrow-derived CD45⫹ cells were ⬎24% positive for CD73 (G).

we used nucleated whole bone marrow cells for positive control of CD73 analysis. Basic sorting patterns of EECs and bone marrow cells are shown in Figs. 4A and F. Analyzed gates of Sk-34, DN and CD45⫹ cell fractions in the muscle and CD45⫹ fraction in the bone marrow cells were enclosed by squares and further characterized by Sca-1 and CD73. Both Sk-34 and DN cells were ⬎94% (C) and ⬎26% (B) positive for Sca-1 and ⬎95% (E) and ⬎93% (D) negative for CD73, respectively. In contrast, skeletal musclederived CD45⫹ fraction contained ⬎25% CD73⫹ cells (H). Similarly, bone morrow-derived CD45⫹ cells were ⬎24% positive for CD73 (G).

Discussion In the present study, we found that EECs include a very small SP cell population (0.08%), but are composed mostly of MP cells. Both in MP and in SP, cell populations were divided into four subpopulations: CD34⫺/45⫹, CD34⫹/ 45⫹, CD34⫹/45⫺, and CD34⫺/45⫺, and the patterns wholly corresponded to the total EEC sorting pattern that we recently reported [8]. About 58% of SP cells were present in the CD34⫹/45⫺ (Sk-34) fraction, 36% in the CD34⫺/45⫺ (DN) fraction, and the remaining 6% were CD45⫹ fractions. This result concurs with our previous report that the expression of Bcrp1/ABCG2 mRNA was detectable both in Sk-34 and in DN cell fractions [8]. SPSk-34 and MP-Sk-34 cells showed the same growth char-

acteristics and multilineage differentiation to myogenic, endothelial, and fat cells. All of these characteristics in MPand SP-Sk-34 cells correspond well with the characteristics of total Sk-34 cells, as reported previously under the same culture conditions [8]. However, the number of MP-Sk-34 cells was about 1000-fold higher than that of SP-Sk-34 and the colony-forming activity was the same in both populations. Thus, we conclude that the reported characteristics of Sk-34 cells [8] are primarily (99.9%) attributable to MPSk-34 cells. Additionally, the majority of floating and/or weakly attached cells after 10 days of culture were strongly or weakly positive for MyoD (Figs. 2I and J). Expression of MyoD was relatively low in larger cells. However, a few cells, typically smaller cells, still expressed CD34 while coexpressing MyoD (Figs. 2K and L). This clearly indicated that CD34⫹ cells differentiated into myogenic cells following an increase in their size and the disappearance of CD34 antigen. As we have previously reported, the DN cell populations did not grow well under the culture conditions used for Sk-34 cells [8]. However, the DN cell populations also contained 0.06% of SP cells (Figs. 1C and D). To assess the growth and differentiation potential of MP-DN and SP-DN cell populations, we performed collagen-based clonal cell culture with bFGF and EGF (see Materials and methods). Importantly, when the DN cells were cultured without bFGF and EGF, the CFU with multilineage did not appear. The growth and differentiation characteristics of MP-DN and SP-DN cells were also identical in the collagen-based

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

culture medium. Both MP-DN and SP-DN cells formed sphere-like colonies after 4 –7 days of culture (Figs. 3A and C). Spheres at Day 7 included MyoD-positive cells (Fig. 3E) and gradually increased their diameter and formed numerous myotubes after 14 days of culture (Figs. 3B and D). After 10 days of culture, adherent cells demonstrated DiIAc-LDL uptake (Figs. 3F and G), and when the total cells were resuspended from DN cell cultures, there were totally CD45 negative (Fig. 3K). These results indicate that the DN cells were also able to differentiate into myogenic and endothelial cells. Interestingly, the colony-forming activity of DN cells was about 6-fold higher than that of Sk-34 cells (about 10% in DN and 1.6% in Sk-34). Additionally, over 86% of DN cells in both the MP and the SP fractions expressed CD34 after 4 days of culture (Figs. 3I and J). The fraction of cells expressing CD34 decreased at 10 days of culture (Fig. 3K), while the fraction of MyoD-positive cells increased around this time (Fig. 3E). Taken together, these results suggest that DN cells differentiate into myogenic and endothelial cells through the expression of CD34 antigen. However, the functional relationship between Sk-34 and DN cells in vivo is still not clear. More importantly, these results clearly indicate that expression of CD34 antigen precedes MyoD expression during the myogenic process. It is thought that the myogenic lineage-positive cells in Sk-34 and DN populations are derived from bone marrow stromal cells and remained in the interstitial spaces of skeletal muscle because muscle regeneration by bone marrowderived myogenic progenitors has been suggested by Ferrari et al. [3]. Recently, it was also suggested that CD73 (ecto5⬘-nucleotidase) may play a role in bone marrow stromal interactions and in the differentiation of mesenchymal stem cells, and that it is a possible marker for mesenchymal stem cells [15]. Based on this report, we performed CD73 staining to Sk-34, DN, and CD45⫹ cell populations. The results showed that both Sk-34 and DN cell populations were over 93–95% negative for CD73 (Figs. 4D and E). Thus, the possibility that Sk-34 and DN cell populations are derived from bone marrow stromal cells was low. In addition, CD73 was also found on endothelial cells in the thymus and peripheral lymphoid organs [16]. In this respect, both Sk-34 and DN cell populations include very few endothelial cells immediately after sorting, as we have reported previously [8]. Furthermore, over 94% of Sk-34 and 26% of DN cells were Sca-1 positive (Figs. 4B and C). These results indicate that over 80% of EECs were Sca-1 positive. Given the results shown in Figs. 1D, 4B, and 4C, it is likely that the muscle SP cell population that we sorted in the present study was mostly CD34⫹/⫺, CD45⫺, and Sca-1⫹. These results correspond to those of Gussoni et al. [5], who suggested that muscle SP cells are CD34⫹/⫺, Sca-1⫹, and CD45⫺. Interestingly, CD45⫹ fractions were ⬎25% positive for CD73 (Fig. 4H). A recent bone marrow transplantation study has been interpreted to suggest that hematopoietic progenitors in the skeletal muscle have a bone marrow origin [17]. Similarly, two groups demonstrated that

89

CD45⫹ cells in the skeletal muscle had hematopoietic stem cell activity [18,19]. Thus, it is reasonable to speculate that the CD45⫹ fractions in the skeletal muscle include hematopoietic progenitors and these cells were bone marrow in origin. In fact, we have also detected several hematopoietic colonies in the CD45⫹ cell population in muscle using clonal cell culture appropriately stimulated by several factors (data not shown). Therefore, cells in the CD45⫹ fractions in the skeletal muscle may have a unique multilineage differentiation potential when compared to Sk-34 and DN cells. However, it seems consistent that the skeletal musclederived CD45⫺ cells were highly myogenic, whereas CD45⫹ cells displayed limited myogenic activity [19]. The relationship between satellite cells, Sk-34, and DN cells still remain unclear, and we are currently investigating these questions further. Based on these observations, we conclude that the growth characteristics and multilineage differentiation were identical in skeletal muscle-derived MP and SP cells in clonal cell culture. However, the Sk-34 cells include only a few SP cells (⬍0.1%), and are wholly composed (99.9%) of an MP cell population. The CFU of DN cells can differentiate into myogenic and endothelial cells as well as Sk-34 cells, and these cells express CD34 antigen preceding MyoD expression during the myogenic process. These findings may provide new strategy for muscle tissue reconstitution therapy by defining a new source of stem cells that can be obtained more easily, rapidly, and effectively from skeletal muscle.

Acknowledgments The authors thank Miss Hiroko Kouike (Keio University School of Medicine) for technical assistance with cell sorting. This work was supported by a Grant-in-Aid-for Scientific Research (14658234) from the Ministry of Education, Science, and Culture of Japan, and by Tokai University School of Medicine Research Aid.

References [1] C.B. Johansson, S. Momma, D.L. Clarke, M. Risling, U. Lendahl, J. Frisen, Identification of a neural stem cell in the adult mammalian central nervous system, Cell 96 (1999) 25–34. [2] B.A. Reynolds, S. Weiss, Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system, Science 255 (1992) 1707–1710. [3] G. Ferrari, G. Cusella-De Angelis, M. Coletta, E. Paolucci, A. Stornaiuolo, G. Cossu, F. Mavilio, Muscle regeneration by bone marrowderived myogenic progenitors, Science 279 (1998) 1528 –1530. [4] D.J. Prockop, Marrow stromal cells as stem cells for nonhematopoietic tissues, Science 276 (1997) 71–74. [5] E. Gussoni, Y. Soneoka, C.D. Strickland, E.A. Buzney, M.K. Khan, A.F. Flint, L.M. Kunkel, R.C. Mulligan, Dystrophin expression in the mdx mouse restored by stem cell transplantation, Nature 401 (1999) 390 –394.

90

T. Tamaki et al. / Experimental Cell Research 291 (2003) 83–90

[6] K.A. Jackson, T. Mi, M.A. Goodell, Hematopoietic potential of stem cells isolated from murine skeletal muscle, Proc. Natl. Acad. Sci. USA. 96 (1999) 14482–14486. [7] P. Seale, L.A. Sabourin, A. Girgis-Gabardo, A. Mansouri, P. Gruss, M.A. Rudnicki, Pax7 is required for the specification of myogenic satellite cells, Cell 102 (2000) 777–786. [8] T. Tamaki, A. Akatsuka, K. Ando, Y. Nakamura, H. Matsuzawa, T. Hotta, R.R. Roy, V.R. Edgerton, Identification of myogenic-endothelial progenitor cells in the interstitial spaces of skeletal muscle, J. Cell Biol. 157 (2002) 571–577. [9] T. Asahara, T. Murohara, A. Sullivan, M. Silver, R. van der Zee, T. Li, B. Witzenbichler, G. Schatteman, J.M. Isner, Isolation of putative progenitor endothelial cells for angiogenesis, Science 275 (1997) 964 –967. [10] T. Asahara, T. Takahashi, H. Masuda, C. Kalka, D. Chen, H. Iwaguro, Y. Inai, M. Silver, J.M. Isner, VEGF contributes to postnatal neovascularization by mobilizing bone marrow-derived endothelial progenitor cells, EMBO J. 18 (1999) 3964 –3972. [11] J. Yamashita, H. Itoh, M. Hirashima, M. Ogawa, S. Nishikawa, T. Yurugi, M. Naito, K. Nakao, Flk1-positive cells derived from embryonic stem cells serve as vascular progenitors, Nature 408 (2000) 92–96. [12] S. Zhou, J.D. Schuetz, K.D. Bunting, A.M. Colapietro, J. Sampath, J.J. Morris, I. Lagutina, G.C. Grosveld, M. Osawa, H. Nakauchi, B.P. Sorrentino, The ABC transporter Bcrp1/ABCG2 is expressed in a

[13] [14]

[15]

[16]

[17]

[18]

[19]

wide variety of stem cells and is a molecular determinant of the side-population phenotype, Nature Med. 7 (2001) 1028 –1034. R. Bischoff, Proliferation of muscle satellite cells on intact myofibers in culture, Dev. Biol. 115 (1986) 129 –139. M.A. Goodell, K. Brose, G. Paradis, A.S. Conner, R.C. Mulligan, Isolation and functional properties of murine hematopoietic stem cells that are replicting in vivo, J. Exp. Med. 183 (1996) 1797–1806. F. Barry, R. Boynton, M. Murphy, J. Zaia, The SH-3 and SH-4 antibodies recognize distinct epitopes on CD73 from human mesenchymal stem cells, Biochem. Biophys. Res. Commun. 289 (2001) 519 –524. Y. Yamashita, S.W. Hooker, H. Jiang, A.B. Laurent, R. Rest, K. Khare, A. Coe, P.W. Kincade, L.F. Thompson, CD73 expression and fyn-dependent signaling on murine lymphocytes, Eur. J. Immunol. 28 (1998) 2981–2990. H. Kawada, M. Ogawa, Bone marrow origin of hematopoietic progenitors and stem cells in murine muscle, Blood 98 (2001) 2008 – 2013. A. Asakura, P. Seale, A. Girgis-Gabardo, M.A. Rudnicki, Myogenic specification of side population cells in skeletal muscle, J. Cell Biol. 159 (2002) 123–134. S.L. Mckinney-Freeman, K.A. Jackson, F.C. Camargo, G. Ferrari, F. Mavilio, M.A. Goodell, Muscle-derived hematopoietic stem cells are hematopoietic in origin, Proc. Natl. Acad. Sci. USA 99 (2002) 1341– 1346.