Using experimental mouse models, hematopoietic potential has been shown to exist within skeletal muscle. In humans, the clinical utility of using muscle-derived hematopoietic progenitors remains uncertain. Here, we evaluate the hematopoietic potential of human skeletal muscle. De novo adult muscle contained markedly reduced levels of hematopoietic colony-forming units (hCFU) and negligible responsiveness to hematopoietic ex vivo culture conditions that augment hematopoietic activity of fetal muscle. Neither fetal nor adult muscle yielded significant engraftment in transplanted immune-deficient mice. Although adult muscle possessed 1.5±0.9 hCFU/g, similar hematopoietic activity (2.3±0.17 hCFU) could also be demonstrated from as little as 3–10 μl of contaminating peripheral blood. We suggest that the clinical utility of adult skeletal muscle as an alternative source of hematopoietic cells in humans appears limited due to the low yield of blood-forming precursors and their lack of responsiveness to ex vivo expansion.
The presence of cells with hematopoietic potential residing in adult skeletal muscle in the mouse has generated much excitement in the search for alternative sources of blood-forming stem and progenitor cells. Adult murine skeletal muscle has been shown to contain significant in vitro hematopoietic colony-forming capacity and is able to reconstitute all blood-cell lineages following transplantation into irradiated allogeneic recipients,1, 2, 3, 4 but seem to originate from the bone marrow and subsequently migrate to skeletal muscle.5, 6 The hematopoietic potential of human skeletal muscle remains less certain, and requires quantitative assessment of blood-forming capacity to determine clinical potential. Here, we sought to directly compare the hematopoietic capacity of adult to fetal human skeletal muscle recently shown to possess hematopoietic activity in response to EPO and BMP-4,7 in order to characterize the clinical hematopoietic utility of adult muscle.
Materials and methods
Adult human skeletal muscle tissue (n=6) and 16–22-week gestational age human muscle tissue (n=10) was obtained in accordance with local ethical and biohazard authorities of the University of Western Ontario and London Health Sciences Centre. Adult skeletal muscle was obtained just distal to the operative site of lower limb amputations. Adult and fetal skeletal muscle dissected from the surrounding tissues was rinsed three times with 10–25 ml Hank's Balanced Salt Solution (HBSS; Gibco, Burlington, Ontario) or perfused with isotonic PBS solution as indicated, and minced using dissection scissors for incubation with collagenase type II (400 units/ml) at 37°C for 30 min. Adult human skeletal muscle tissue used was not from diseased, or necrotic or fibrous tissue. Single cell suspensions were obtained by subsequent filtration through a 70 μm and then 40 μm cell strainers. Cells were then seeded into tissue culture plates for 1 h, and nonadherent cells were removed, counted, and replated in fibronectin-coated plastic dishes (Biocoat, Becton Dickinson, Bedford MA, USA). Peripheral blood samples were processed in the identical manner.
In vitro hematopoietic culture conditions
Serum-free essential media for in vitro cultures contained 1% bovine serum albumin, 10 μg/ml bovine pancreatic insulin, and 0.2 mg/ml human transferrin in Iscove's modified Dulbecco medium (IMDM; Stem Cell Technologies, Vancouver, BC, Canada) and 300 ng/ml stem cell factor (Stemgen; Amgen Inc., Thousand Oaks, CA, USA) and FLT-3 ligand, 50 ng/ml granulocyte colony-stimulating factor (Neupogen; Amgen), and 10 ng/ml interleukin-3 (IL-3) and interleukin-6 (IL-6) termed HGFs. These conditions were used for all cell populations cultured throughout the study, and are referred to as ‘hematopoietic culture conditions’ herein. In some of the hematopoietic cultures, 25 ng/ml BMP-4 (R&D Systems) and 3 U/ml EPO (Amgen) were added.
Flow cytometric analysis of human muscle cells
De novo muscle cells or cells cultured under hematopoietic conditions were resuspended in PBS at 1–10 × 106 cells/ml for analysis using human CD45-fluorescein isothiocyanate (FITC or APC; Becton Dickinson Immunocytometry Systems (BDIS), clone #H130 or 2D1 respectively, San Jose, CA, USA), human CD34-allophycocyanin (APC, BDIS, clone #581) and mouse immunoglobulin G1 (IgG1) conjugated to each fluorochrome as controls (BDIS). The exclusion dye 7-aminoactinomycin D was also added to all samples assuring analysis of viable cells.
In vitro clonogenic colony assays
Human clonogenic progenitor assays were performed by plating 2.5–20 × 106 de novo or cultured muscle cells into serum-free Methocult H4236 (Stem Cell Technologies) containing 50 ng/ml rhu-SCF, 10 ng/ml each rhu-granulocyte macrophage colony-stimulating factor (GM-CSF) and rhu-IL-3, and 3 U/ml rhu-EPO with or without 25 ng/ml BMP-4. Both adherent and nonadherent (cells in suspension) cellular fractions of fetal and adult muscle cells were harvested from hematopoietic cultures for plating into hCFU assays. Differential hematopoietic colony counts were assessed by morphology following incubation for 10–14 days at 37°C and 5% CO2 in a humidified atmosphere.
Transplantation of human muscle cells and analysis of NOD/SCID mice
Cells were injected intravenously into the tail vein of sublethally irradiated (325 rad total dose) NOD/SCID mice as previously described,7 and stained with monoclonal antibodies CD45-FITC for detection of human hematopoietic cells, along with analysis of genomic DNA from bone marrow of transplanted animals using PCR for human-specific CART-1.7
Results and discussion
Freshly isolated fetal muscle contains significantly greater hematopoietic potential than adult muscle
Freshly isolated mononuclear cells from adult muscle and fetal muscle were analyzed by flow cytometry for the presence of hematopoietic cell surface markers CD45 and CD34. Although there was no significant difference in the yield of CD45+ cells between fetal and adult muscle (2.0±0.85 vs 1.2±0.40%), there was an eight-fold greater yield of more primitive CD45+CD34+ cells in unmanipulated fetal muscle compared with adult muscle (1.2±0.33 vs 0.15±0.14%, P<0.01) (Table 1). Hematopoietic colony-forming capacity was 360-fold greater in de novo fetal muscle relative to de novo adult muscle (47±16 vs 0.13±0.04 total hCFU/106 cells, Table 1). Transplantation of fetal muscle cells into NOD/SCID mice failed to yield evidence of detectable human engraftment (two muscle samples into six mice at 1.2 × 106 cells/recipient, data not shown). Transplantation of freshly isolated adult muscle into NOD/SCID mice yielded similar results with no detectable human engraftment, consistent with a previous report indicating that freshly isolated adult muscle cells also lacked NOD/SCID repopulating capacity.8
Responsiveness of adult muscle cells to hematopoietic cytokines, EPO and BMP-4 is reduced compared with fetal muscle cells
Fetal and adult-derived muscle mononuclear cells were cultured under hematopoietic conditions as previously shown.7 There was no significant difference in the recovery of fetal compared with adult-muscle-derived cells (14.9±9.9 vs 6.6±0.96%) (Table 1). hCFU increased from 47 to 550/106 fetal cells (17±6-fold increase over de novo) after 5 days of hematopoietic culture, similar to the increase from 0.13 to 2.2/106 adult cells plated in methylcellulose (14±12-fold increase) (Table 1). The percentage of CD45+ and CD45+CD34+ cells present after 5 days of culture diminished to low levels in both fetal and adult muscle samples (Table 1).
Factors such as erythropoietin (EPO) and bone morphogenic protein 4 (BMP-4) have been shown to induce hematopoietic cell fate during embryogenesis and may act as regulators of hematopoietic initiation in nonhematopoietic tissues.9, 10 Addition of BMP-4 and EPO to the hematopoietic cultures of fetal muscle enhanced hematopoeitc progenitors (1970 total hCFU/106 cells or 24±8-fold increase over de novo fetal muscle), but did not augment the hCFU of cultured adult muscle cells (1.3 total hCFU/106 cells or 4.3±3.2-fold increase over de novo adult muscle) (Table 1). Signaling responsiveness to EPO and BMP-4 in fetal muscle as compared to adult tissue may account for these distinctions. Hematopoietic potential of adult muscle appears to be distinct from fetal muscle, characterized by limited responsiveness to hematopoietic cytokines and by the negligible effects of BMP-4 and EPO.
Capacity of fetal and adult muscle to repopulate NOD/SCID mice is negligible
Although not amenable to kinetic analysis and lethal irradiation of recipients for competitive repopulating assays used for mouse hematopoietic stem cells, NOD/SCID recipients provide the best xenotransplantation model to experimentally assess candidate human hematopoietic stem cells. A total of 11 NOD/SCID mice were transplanted with adult muscle cells cultured under hematopoietic conditions (935 000±135 000 cells) and 10 mice with fetal muscle cells cultured under hematopoietic conditions (723 000±200 000 cells). None of the mice yielded detectable human engraftment upon flow cytometric or PCR analysis at 8 weeks post transplantation in the bone marrow compartment of recipient mice. A single mouse transplanted with freshly isolated fetal cells yielded molecular evidence of human chimerism by PCR. More recently in a unique set of experiments, upwards of 100 million freshly isolated fetal muscle cells or cells cultured in hematopoietic conditions were transplanted into NOD/SCID mice without detection of hematopoietic engraftment (data not shown). The combined results of our data and the one previous report8 suggest that very large numbers of muscle-derived cells are required for any detectable human engraftment in transplanted NOD/SCID mice.
Large samples of human adult muscle are required to detect any in vitro hematopoietic activity and may be due to contaminating peripheral blood
Human adult muscle specimens of 4–10 g (average weight of 8.4±1.4 g, Table 2) were obtained prior to amputation procedures and represent 40–100-fold larger sample sizes than most typical muscle biopsies. Only 1.5±0.9 hCFU/g of adult muscle tissue were present in our samples (Table 2); however, the total hCFU content of peripheral blood from normal volunteers (n=5) treated in a similar manner as skeletal muscle samples yielded 2.3±0.17 total hCFU/10 μl of blood, similar to previous reports.11, 12 We suggest that this small volume of peripheral blood present in the adult muscle sample (3–10 μl/g of adult human skeletal muscle13, 14) could account for the entire yield of hCFU observed in de novo adult skeletal muscle (Table 2). To address this concern, adult skeletal muscle samples were perfused with isotonic solution (PBS) to remove circulating blood, and then assayed for hCFU potential (n=3, Table 2). No hCFU were detected in perfused muscle, supporting the notion that the limited hCFU potential from de novo isolated skeletal muscle most likely arises from peripheral blood. In contrast, the number of hCFU observed in fetal blood is insufficient to account for hCFU yields of fetal muscle,7 suggesting that significant numbers of hematopoietic progenitors must reside within human fetal skeletal muscle.
Adult skeletal muscle contains limited hematopoietic potential
Aside from the arguments of the cellular plasticity of muscle giving rise to hematopoietic cells via transdifferentiation, more recent papers continue to suggest that human skeletal muscle may provide a source of hematopoietic cells for transplantation. In contrast to fetal muscle, our study was unsuccessful in demonstrating that adult skeletal muscle contains hematopoietic potential beyond that in contaminating peripheral blood cells. We, furthermore, found no significant responsiveness to hematopoietic cytokines and BMP-4. Although these experiments were performed separately, concurrent comparisons of adult and fetal muscle cells cultured under hematopoietic conditions were not possible, given the impracticality of obtaining donor adult amputation and fetal abortus muscle tissues on the same day. Differences in hemogenic capacity of adult vs fetal tissue may be due to distinct microenvironments between these stages of ontogeny, such as reduced levels of specific growth factors. Alternatively, circulating adult hematopoietic progenitors may have a reduced ability to migrate into the skeletal muscle in the adult. Although hematopoietic culture conditions used in our study may require optimization in order to detect greater blood-forming capacity in human adult skeletal muscle, we conclude that the clinical utility of human adult muscle appears negligible due to their low frequency and limited capacity for in vitro expansion.
Jackson KA, Mi T, Goodell MA . Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc Natl Acad Sci USA 1999; 96: 14482–14486.
McKinney-Freeman SL, Jackson KA, Camargo FD et al. Muscle-derived hematopoietic stem cells are hematopoietic in origin. Proc Natl Acad Sci USA 2002; 99: 1341–1346.
Kawada H, Ogawa M . Bone marrow origin of hematopoietic progenitors and stem cells in murine muscle. Blood 2001; 98: 2008–2013.
Geiger H, True JM, Grimes B et al. Analysis of the hematopoietic potential of muscle-derived cells in mice. Blood 2002; 100: 721–723.
Issarachai S, Priestley GV, Nakamoto B, Papayannopoulou T . Cells with hemopoietic potential residing in muscle are itinerant bone marrow-derived cells. Exp Hematol 2002; 30: 366–373.
Majka SM, Jackson KA, Kienstra KA et al. Distinct progenitor populations in skeletal muscle are bone-marrow derived and exhibit different cell fates during vascular regeneration. J Clin Invest 2003; 111: 71–79.
Jay KE, Gallacher L, Bhatia M . Emergence of muscle and neural hematopoiesis in humans. Blood 2002; 100: 3193–3202.
Dell'Agnola C, Rabascio C, Mancuso P et al. In vitro and in vivo hematopoietic potential of human stem cells residing in muscle tissue. Exp Hematol 2002; 30: 905–914.
Maeno M, Mead PE, Kelley C et al. The role of BMP-4 and GATA-2 in the induction and differentiation of hematopoietic mesoderm in Xenopus laevis. Blood 1996; 88: 1965–1972.
Kelley C, Yee K, Harland R, Zon LI . Ventral expression of GATA-1 and GATA-2 in the Xenopus embryo defines induction of hematopoietic mesoderm. Dev Biol 1994; 165: 193–205.
Dubé ID, Kalousek DK, Coulombel L et al. Cytogenetic studies of early myeloid progenitor compartments in Ph1-positive chronic myeloid leukemia. II. Long-term culture reveals the persistence of Ph1-negative progenitors in treated as well as newly diagnosed patients. Blood 1984; 63: 1172–1177.
Eaves CJ, Eaves AC . Erythroid progenitor cell numbers in human marrow – implication for regulation. Exp Hematol 1979; 7 (Suppl. 5): 54.
Lillioja S, Young AA, Culter CL et al. Skeletal muscle capillary density and fiber type are possible determinants of in vivo insulin resistance in man. J Clin Invest 1987; 80: 415–424.
McGuire BJ, Secomb TW . A theoretical model for oxygen transport in skeletal muscle under conditions of high oxygen demand. J Appl Physiol 2001; 91: 2255–2265.
Funding for this research project was provided by a research grant from the Canadian Institutes of Health Research (CIHR) and Krembil Foundation, and Canada Research Chair in Stem Cell Biology and Regenerative Medicine to MB and a research fellowship award from the Canadian Leukemia Foundation to DSA and postgraduate scholarship from StemNet National Centres of Excellence in Canada to KEJ. We thank T Forbes and Ian H Chin-Yee for their assistance in obtaining tissue samples for this study.
About this article
Cite this article
Allan, D., Jay, K. & Bhatia, M. Hematopoietic capacity of adult human skeletal muscle is negligible. Bone Marrow Transplant 35, 663–666 (2005). https://doi.org/10.1038/sj.bmt.1704866
- stem cells
Muscle-derived stem cells: isolation, characterization, differentiation, and application in cell and gene therapy
Cell and Tissue Research (2010)
In vitro and in vivo evaluation of the haematopoietic potential of skeletal muscle in a non-human primate model
Bone Marrow Transplantation (2008)