This is a short post just to wish all of you a happy new year. I hope 2014 will be an exciting year for regeneration and it will bring us tens of papers reporting novel data on how our beloved friends are able to regrow injured or lost organs, structures and body parts. I will stay here blogging about basic research focussed on understanding this fascinating biological process.
Autophagy is a biological process through which cells self-degrade unnecessary or dysfunctional components. Autophagy may be important in certain contexts such as for example nutrient starvation where a minimum level of energy is required to maintain the basic functions that warranty the survival of the cells. Another context in which autophagy can be important is during the remodelling of the cytoplasm that takes place in the process of cell reprogramming. In animal models in which regeneration depends upon an initial dedifferentiation of differentiated cells into proliferative cells, an extensive cytoplasmic remodelling must occur. However, few data is available about the role that autophagy may play during regeneration. Now, a recent paper from the laboratories of DJ Klionsky and T Vellai reports on the role of autophagy in zebrafish fin regeneration (http://www.ncbi.nlm.nih.gov/pubmed/24317199).
The authors use first a transgenic line carrying a GFP reporter under the control of Lc3 promoter, a specific marker for autophagosomes and autolysosomes. They found that Lc3 was upregulated within the tail fin blastema at 2 days after amputation (dpa). From that time the levels of Lc3 gradually decreased until reaching basal levels around 6 dpa. Lc3 was expressed in newly differentiated cells within the blastema. As those cells come from dedifferentiation, Lc3 expression could be reflecting an autophagic activity during this dedifferentiation-redifferentiation process. This increase in autophagy during regeneration was further corroborated by observations of electron microscopy. Thus, an increase of autophagic features was detected in epidermal cells, osteocytes and pigment cells of 2-day blastemas.
In order to see whether this increase in autophagy had a role during regeneration the authors used different methods to block this process. They either injected an antisense morpholino oligonucleotide against atg5 or a drug that acts as an autophagy inhibitor. Silencing of atg5 at 2 dpa blocked regeneration and induced the degeneration of the existing blastema. Similar results were obtained after using the inhibitory drug. Because in other models the inhibition of autophagy induces apoptosis the authors checked next whether the defects in regeneration after blocking autophagy could be explained in terms of increased apoptosis. Indeed, an increase of apoptotic cells within the blastemas was observed. Not only apoptosis was missregulated but also they observed a significant reduction of proliferative cells as well as problems with cell differentiation. All these data lead the authors to conclude that autophagy would promote cell survival and proliferation and would mediate cell differentiation during regeneration.
Finally, the authors looked for upstream regulators of autophagy in this context. It is well known that the FGF signalling is required for fin regeneration and it silencing inhibits cell dedifferentiation. MAPK/ERK is one of the downstream effectors of FGF signalling and has been involved with the regulation of autophagy. Therefore, the authors checked whether the blocking of MAPK/ERK itself influenced fin regeneration. Using an inhibitory drug they found a complete impairment of fin regeneration. Remarkably, they also observed a decrease of autophagy in those treated animals, suggesting that MAPK/ERK activity was required for the upregulation of autophagy during regeneration. From all that data the authors propose a model in which Fgf signalling would lead to the phosphorylation of MAPK/ERK that would then regulate autophagy.
In summary, the authors conclude that autophagy would act as a prerequisite for the regeneration of the zebrafish tail fin through the regulation of the reorganization and remodelling of the cytoplasmic compartment during cell dedifferentiation.
Zebrafish fins are composed of multiple bony rays each of which is comprised of multiple bony segments separated by joints. As in all vertebrates these joints must be formed at precise positions to ensure the flexibility of the skeletal system. During regeneration (as well as during growth), there is a sequential addition of new bony segments and joints in a proximal to distal direction. Therefore, the youngest tissue is located most distally. Not much is known about the genes involved in joint morphogenesis in this context. Now a paper from the laboratory of M. Kathryn Iovine (http://www.ncbi.nlm.nih.gov/pubmed/24278401) has characterized a pathway involved in joint regeneration during zebrafish fin regeneration.
Previous reports have suggested that the transcription factor even-skipped 1 (evx1), an eve-related homeobox gene, is required for joint formation during regeneration. Genes that are expressed during joint formation are expressed in discrete groups of cells located within the lateral population of skeletal precursor cells. Based on the expression pattern the authors found that three genes dlx5a (distal-less homeobox-5a), mmp9 (matrix metalloproteinase 9) and col10a1b showed a similar expression than evx1 in this lateral mesenchymal compartment of the ray. In order to characterize the relationship between these three genes and evx1, they first checked how their expression was affected in evx1 mutants. Whereas the expression of col10a10b was not affected in evx1 mutants, dlx5 and mmp9 were significantly reduced, although not completely abolished. These results suggested that dlx5 and mmp9 were expressed downstream of evx1. Next, the authors characterized the function of dlx5 and mmp9 on fin regeneration. Morpholino-mediated knockdowns of these genes resulted in increased ray segment length suggesting that dlx5 and mmp9 are necessary for correct joint placement.
During fin ray regeneration genes required at early stages of differentiation are expressed in more distal regions, whereas proximal regions include late differentiation genes. Evx1 was expressed in the most distal domain of the skeletal precursor cells, consistent with this gene acting earlier. On the other side, dlx5a was expressed more proximally than evx1 and mmp9 was expressed even more proximally than evx1 and dlx5a. These results suggested a linear pathway initiated by evx1 followed by dlx5a and then followed by mmp9. This was further supported by the observation that mmp9 expression was reduced in dlx5a knockdowns. In contrast dlx5a expression was not affected after mmp9 knockdown. Similarly evx1 expression was not affected in dlx5a or mmp9 knockdowns.
This same laboratory had previously reported that the activity of Cx43 (a gap junction protein) suppresses joint formation. Cx43 is expressed in the medial compartment adjacent to the lateral population of skeletal precursor cells. Mutants for this gene are characterized by short segments because of premature joints. Accordingly, and as predicted, the expression of evx1, dlx5 and mmp9 was initiated more distally (meaning sooner) in regenerating fins of Cx34 mutants. This premature expression of these genes would account for the short segments observed in Cx34 mutants.
In summary, the authors propose a model in which Cx43 would inhibit evx1 which is itself required to sequentially activate dlx5a and mmp9 for proper joint formation.
Salamanders display amazing regenerative capacities that include the ability to regrow new limbs. Previous results have indicated that blastema cells are originated mainly by the dedifferentiation of pre-existing tissues into cells that re-enter the cell cycle and proliferate to form the missing structures. However, and because salamanders possess muscle satellite cells positive for Pax7 it was not clear the relative contribution of mature myofibers and those satellite cells into the regenerated limb skeletal muscle. Now, a recent paper from the laboratories of András Simon and Elly Tanaka (http://www.ncbi.nlm.nih.gov/pubmed/24268695) provides more clear evidence of the cellular origin of the regenerated muscle fibers in two different salamanders: Notophtalmus viridescens (newt) and Ambystoma mexicanun (axolotl).
Remarkably, by using a Cre–loxP-based genetic fate mapping of muscle fibers during regeneration they found fundamental differences in the cellular origin of the regenerated muscle in these two species of salamanders. Two weeks after co-electroporating the upper arm of newts with the different constructs needed to specifically label the mature muscle fibers (positive for myosin heavy chain (MHC)) with YFP (without labelling the Pax7+ satellite cells), those limbs were amputated. The authors found YFP+ cells all along the regenerated limb except the digit tips, indicating a direct contribution of the pre-existing muscle into the regenerated myofibers. Detailed analyses of those blastemas showed that in their distal regions they contained YFP+ cells that were negative for MHC pointing out to a dedifferentiation of the muscle cells to form mononuclear blastema cells. Moreover those blastema cells had re-entered the cell cycle as indicated by labelling with PCNA and the incorporation of the nucleotide analog Edu. Importantly, Edu was never incorporated in YFP+/MHC+ or MHC+ cells suggesting that cell cycle re-entry occurred after the fragmentation and dedifferentiation of muscle fibers into blastema cells.
In contrast, when they used the same strategy to label the mature muscle fibers of the axolotl and trace them upon amputation, no YFP+ cells were found within the blastema or in the regenerated limb. Although some morphological changes of the pre-existing fibers were seen at the amputation plane, these results suggested that the pre-existing muscle fibers did not contribute to muscle regeneration in axolotls. Previous results from the laboratory of Elly Tanaka had shown that GFP+ muscle fibers and satellite cells contribute to muscle regeneration in axolotls, but without being able to distinguish their relative contributions. Here, they found that of 834 GFP+ positive blastema cells, 809 expressed Pax7. Also, these GFP+ were cycling as they incorporated EdU. Taking in account these two results (lack of YFP+ cells and presence of GFP+/Pax7+ cells within the blastema) the authors conclude that were the Pax7+ satellite cells from the mature limb the ones that gave rise to the proliferative muscle progenitors required for regeneration.
Finally, they analysed whether those differences between newts and axolotls could be explained by the neotenic nature of axolotls. They metamorphosed axolotls and analysed limb regeneration in them, following the same methodological approach. Similarly, muscle regeneration in this context did not seem to depend on the dedifferentiation of pre-existing muscle fibers. Also, and complementarily, Pax7+ cells were not found either in larval newt limb blastemas (in agreement with the results obtained in adult newt blastemas).
In summary, this study characterizes at the cellular level the origin of the regenerated muscle fibers in two different species of salamanders. Remarkably, these two species use very different strategies to achieve the same final goal: skeletal muscle regeneration. These results are a beautiful and clear example of how different animals (even relatively closed phylogenetically) use different strategies to form the same cell type within the regenerative blastema. This flexibility and diversity of successful strategies may be of special relevance for the field of regenerative medicine.