Skeletal muscle retains the ability to regenerate following injury by the expansion and differentiation of resident myogenic precursor cells. Our studies demonstrate that Sca-1 deletion in vivo
results in a hyperproliferative state that resists differentiation, consistent with previous in vitro
observations made by us and subsequently by others [19
]. This disturbance in the balance between myoblast proliferation and differentiation contributes to a subtle delay in muscle regeneration after myonecrotic injury. The current studies constitute the first report of hyperproliferation and delayed differentiation in Sca-1−/−
myoblasts in vivo
, and support the hypothesis that Sca-1 functions as a regulator of muscle regeneration in vivo
through its direct effect on myoblast proliferation and secondary effect on differentiation. We therefore propose a model whereby Sca-1, through its expression on myogenic progenitors and proliferating myoblasts, and downregulated coincident with differentiation into myocytes, controls the balance between proliferation and differentiation during myogenic repair (). This model explains why the absence of Sca-1 results in an imbalance between these two processes, with increased cellular proliferation and delayed myogenic differentiation ultimately resulting in the delayed regeneration observed in Sca-1−/−
myofibers following injury.
Model for Sca-1 function during muscle regeneration
Interestingly, Sca-1 appears to have little impact during primary myogenesis. Sca-1−/−
mice are viable and breed normally [11
]. They appear to have normal levels of physical activity, and histological analysis suggests that Sca-1 is not necessary for myogenic determination and myofiber formation during embryonic development (Fig. S2
). We observed after weaning that body and muscle weight gain, tibial length, and myofiber CSA between Sca-1+/+
mice were comparable (Fig. S2
). We followed growth curves of Sca-1−/−
mice until 35 weeks of age and found no difference when controlled for gender. In addition, we have not observed increased fibrosis or histologic abnormalities in Sca-1−/−
muscle with age (data not shown). Furthermore, the numbers of α7-integrin+
resident myogenic cells were unaffected by Sca-1 deletion () which suggests that Sca-1 is not required for the initial specification and establishment of myogenic progenitors.
The proposed role of Sca-1 in regulating precursor cell proliferation versus differentiation, as illustrated by the response to muscle injury in Sca-1−/−
animals, is supported by parallel observations obtained after serial bone marrow transplantation [6
], where Sca-1−/−
bone marrow failed to repopulate irradiated mice. These authors suggested that the absence of Sca-1 affected the balance between hematopoietic stem cell proliferation and self-renewal. Similarly, Sca-1−/−
mice develop age-related osteoporosis [10
], indicating that Sca-1 may regulate proliferation versus differentiation of mesenchymal stem cells. These findings in muscle, bone marrow, and bone support a role for Sca-1 in regulating tissue progenitor cell function rather than in initial cell and tissue development [38
]. They also are consistent with an overall model where Sca-1 expression is pivotal for determining the balance between proliferation, differentiation, and self-renewal in a variety of tissues [6
]. While a focal injury model (i.e.,
cardiotoxin injection) is unlikely to elicit exhaustion of the myogenic progenitor compartment through diminished self-renewal, studies combining the Sca-1−/−
background with advanced age or crossing with a mouse strain that undergoes chronic muscle injury, e.g.
, the mdx
], should be able to test this hypothesis.
Recently published studies suggested age-dependent alterations in primary myogenesis in Sca-1-null mice [20
], however, we observed no difference in body and muscle weight, tibial length, or myofiber CSA between Sca-1−/−
animals evaluated from 4 through 35 weeks of age (Fig. S2
). It is possible that those observations reported by others were made in animals that were significantly older, or of mixed gender compared to the cohort we have studied. Gender itself can contribute significantly to skeletal muscle regenerative capacity [40
]. In addition, those mice were from a different genetic background (C57BL/6), and it is well-established that Sca-1 haplotype expression varies by strain [41
]. More important, however, is the consistent conclusion drawn from both studies regarding a regulatory role for Sca-1 during myoblast proliferation. Specifically, in vitro
data previously reported by us and others [19
] overlap well with our current in vivo
observations, lending significant validity to the model of Sca-1 as a regulator of myoblast proliferation during myogenic repair.
It is worth noting that while we observed an initial cell cycle acceleration of proliferating α7+ myoblasts, this was a transient phenomenon and eventually lead to appropriate cell cycle withdrawal and differentiation, although this was delayed. The regulatory mechanisms that eventually slow cell division and allow for myoblast differentiation in the absence of Sca-1 are not known. The observation that these cells continue to proliferate ex vivo, where they are not subject to other signals from the tissue milieu, only serves to reinforce a hypothesis that while the absence of Sca-1 disinhibits myoblast proliferation in response to injury in vivo, this most likely represents a higher threshold for initiating cell cycle withdrawal and differentiation.
For this study, we did not distinguish between myogenic precursors on the basis of anatomical location traditionally used to identify satellite cells. Instead, our α7-integrin-based strategy pooled a heterogeneous population of myogenic precursors that become M-cadherin+
with activation (), and should include both satellite cells as well as interstitial cells with myogenic potential, while excluding hematogenous cells present in adult muscle () [2
]. While the myogenic potential of the α7+
compartment has been well-documented [24
], the extrapolation of these results to other myogenic progenitor pools characterized by more restrictive criteria is cautioned.
While we observed equivalent size distributions among regenerating fibers in Sca-1+/+
animals by day 25, we also noticed an increase in the frequency of larger myofibers (≥2000 μm2
) in the area immediately adjacent to the previously injured region in Sca-1−/−
muscle (Fig. S3
). This could indicate either an exaggerated compensatory hypertrophy adjacent to the site of injury due to a qualitative difference in regenerating Sca-1−/−
muscle, but also might be explained by extension of the expanded Sca-1−/−
myoblast pool into the surrounding muscle. The latter seems more likely, since no qualitative defect in muscle function (e.g.
, ambulation, food seeking behavior, exercise) was observed in the recovered Sca-1−/−
The molecular footprints of progenitor cell activation, myoblast proliferation, and differentiation into myocytes have previously been characterized [49
]. We followed these transcriptional changes in the myogenic population over time. Our results suggest that Sca-1−/−
myoblasts experience an early differentiation delay, specifically affecting MyoD, myogenin, desmin and myosin expression. Interestingly, the expression of Myf5, one of the earliest markers of myogenic commitment, persisted. This is consistent with our previous in vitro
observation that skeletal myoblasts treated with Sca-1 antisense or blocking antibodies undergo differentiation arrest after the expression of Myf5 [19
]. With the delay in expression of desmin, a mature myoblast marker, and in myosin, which heralds differentiation into myocytes, we also uncovered a subtle delay in early fusion. These observations not only provide insight into the molecular events occurring in the absence of Sca-1, but strongly support that Sca-1−/−
myoblasts proliferate at the expense of progression along the differentiation pathway. Previously, we ascribed to Sca-1 a specific role in myoblast fusion in vitro
]. However, with these new insights from in vivo
study, we now conclude that the effect of Sca-1 interference on myoblast differentiation and fusion we had observed in vitro
was likely the consequence of its direct effect on myoblast proliferation. This explanation similarly is supported by studies from other investigators [20
Thus, based on previous studies in vitro
] and our current in vivo
observations reported here, we revise our model to propose that Sca-1 controls the tempo of myogenic repair in vivo
, and that its appropriately regulated expression controls myoblast cell cycle withdrawal during differentiation in response to injury. More generally, these observations emphasize the importance of balancing progenitor proliferation and differentiation during tissue repair. A through understanding of the mechanisms guiding these cellular decisions, as well as those governing progenitor activation and self-renewal, remain critical to developing strategies for targeted tissue regeneration.