Contracting muscle for regeneration : myogenic dedifferentiation in limb progenitor cell formation

Abstract

The urodele amphibians, such as newts and axolotls, are unique among vertebrates since they regenerate their limbs, tail, jaws and part of their heart and CNS as adults. After amputation of a newt limb, regeneration proceeds by the reversal of differentiation of tissues in the limb stump. In this process, termed dedifferentiation, tissue characteristics are lost and cells with a more stem cell-like morphology appear. Such cells migrate distally and proliferate to form the blastema, a structure that gives rise to the tissues of the new limb. In this thesis, I have studied cellular and molecular mechanisms of how progenitor cells are formed from skeletal muscle during newt limb regeneration.

It is not established to what extent skeletal muscle tissue dedifferentiation, seen after amputation, involves cellularization of myofibres, and to what extent it involves activation of resident reserve cells. In paper I we identified a Pax7+ satellite cell population in newt skeletal muscle, which is mitotically activated after limb amputation. We established a method to isolate and cultivate newt single muscle fibres, in which activation of satellite cells as well as cellularization of the myofibre can be studied. Newt single fibres produced proliferating progeny that we showed to be satellite cell derived. When implanted into a newt limb before its amputation, the satellite cell progeny appeared to contribute to the blastema and to cartilage and epidermis of the regenerate.

In paper II we developed a way of expressing genes exclusively in terminally differentiated muscle cells by cytoplasmic injection of expression vectors. This method enables the study of the effect of various molecules on dedifferentiation without interfering with the differentiation process. In addition, injecting plasmids expressing fluorescent markers provides a tool for tracing of individual cells derived by cellular dedifferentiation, a method used in paper IV.

Newt myotubes reenter S-phase in response to a thrombin-activated serum component. Thrombin is activated after both amputation and lentectomy in the newt, and might provide a link between tissue injury and dedifferentiation. Mouse myotubes do not respond to serum by S-phase reentry, a feature that might be an aspect of the relatively poor regenerative ability in mammals. In paper III we used quantitative RT-PCR to study the responsiveness of mouse myotubes to various serum components in comparison to their newt counterparts. We show that mouse myotubes respond to the thrombin-activated factor by increased expression of Id1, c-fos and c-jun, indicating that mouse myotubes retain responsiveness to this factor as well as the ability to leave G0 and reenter the cell cycle. Furthermore, we found that in newt myotubes global methylation of H3K9 is reversed after serum stimulation, while this classical repressor mark is retained in serum stimulated mouse myotubes, which might reflect their block to S-phase reentry.

The molecular pathways inducing cellularization of myofibres are not known. However, in addition to major tissue damage, myofibre fragmentation requires direct injury to the fibre. This fact led us to hypothesize that inducing a programmed cell death (PCD) response in the fibre can lead to its dedifferentiation. In paper IV we tested this hypothesis, by lineage tracing of individual myotubes, and showed that a previously identified inducer of cellularization evokes a PCD response in myotubes. Furthermore, we showed that the pro-apoptotic drugs, STS and TSA, induce myotube cellularization, an effect that can be blocked by the anti-apoptotic drug DIDS.

In summary, the major findings of this thesis are the following. Similarly to mammals, newt skeletal muscle harbours a Pax7+ satellite cell population, which is activated during newt limb regeneration. Furthermore, cellular dedifferentiation and PCD share molecular pathways in both newt and mammalian myotubes. In addition, mammalian myotubes retain responsiveness to the thrombin activated serum factor. Our data pinpoint several features that characterize both newt and mammalian skeletal muscle and indicate that further understanding of skeletal muscle plasticity after injury may reveal ways to promote blastema formation in mammals.