M.Sc Thesis

M.Sc StudentLerer Ravital
SubjectThe Role of Kirrel3 Receptor and Rac1 Small
GTPase in Myoblast Fusion
DepartmentDepartment of Medicine
Supervisor ASSOCIATE PROF. Eyal Bengal
Full Thesis textFull thesis text - English Version


Skeletal muscles arise by fusion of precursor cells, myoblasts, into multinucleated fibers.

During myogenesis, mononucleated myoblasts withdraw from the cell cycle, initiate muscle specific gene expression, and subsequently fuse with one another to form nascent, multinucleated myofibers. This fusion process is an ordered set of specific cellular events: recognition, adhesion, alignment, and membrane union.

In Drosophila, muscle founder cells (FC) function as ‘attractants’ for surrounding fusion competent myoblasts (FCM). Several FCM's are fused with one FC to form a myotube. The attractance and adhesion of FCMs to FCs is mediated by 4 cell-type-specific transmembrane receptors containing immunoglobulin domains: Dumbfounded/Kirre (Duf) and Roughest/Irre (Rst) in the FC, and Sticks and Stones (Sns) and Hibris (Hbs) in the FCM. Duf and Rst bind directly to Sns, and these proteins are the primary mediators of myoblast adhesion.

Kirrel, the vertebrate homologue of Drosophila Duf, is also necessary for muscle precursor fusion in zebrafish. Within developing somites, Kirrel expression is localized to membranes of fusion-competent myoblasts of the fast-twitch lineage. Unlike wild-type myoblasts that form spatially arrayed syncytial multinucleated fast myofibers, those deficient in Kirrel show a significant reduction in fusion capacity.

In light of the convincing evidence of Kirrel involvement in myoblast fusion in both Drosophila and Zebrafish we hypothesized that Kirrel is also involved in mammalian muscle fusion.

In this study we found that the mammalian Kirrel3 displays two expression peaks during differentiation. The first, and more substantial one, 5 hours after inducing differentiation and the second after 72 hours. The early peak of expression precedes the fusion of C2 cells that begins at around 12 hours following transfer to differentiation medium. This pattern may indicate that Kirrel3 may function at the pre-fusion stage of differentiating myoblasts.

We also found that the expression of Kirrel3 mRNA was significantly induced following the activation of MyoD in 3T3 fibroblasts. This data indicates that Kirrel3 is a 2 transcriptional target of MyoD, and as such, may serve as part of the "differentiation program" in mammals.

It is known that Kirrel3 undergoes processing by metalloproteinases in the Kidney, which results in the shedding of its extracellular domain. In this study we found high levels of both the full length form of Kirrel3 and the processed form in early stages of differentiation, and that the processing is MyoD dependent. Therefore, MyoD may coordinate the function of Kirrel3 as well as its expression, through this processing event.

In addition, we found that the levels and processing of Kirrel3 change as a result of cell density. Kirrrel3 protein appears less stable in confluent differentiating cells than in sparse cells. This, again, may indicate the importance of Kirrel3 in the initial cell-cell recognition and attractance, and might indicate that Kirrel3 is no longer needed for the fusion process once myoblasts have established contact.

In order to investigate if the processed products have a physiological function, we constructed vectors expressing different domains of Kirrel3 and analyzed their effect on myoblast fusion. The only construct that appeared to affect myoblast fusion was the one expressing a secreted extracellular domain of Kirrel3 (ΔMC), which increased the number of myotubes containing 5 or more nuclei relative to untreated myoblasts. This result may indicates that the secreted extracellular domain signals surrounding cells towards the graded signals, which is stronger in the cells expressing this mutant, thus attracting more cells to fuse. The cellular localization of Kirrel3 during differentiation was also analyzed, by immunostaining, and was found to change during this process. However, the specific cellular localization is not yet determined, as this will require double staining with known cellular markers to specific organelles.

Knockdown of Kirrel3 in C2 myoblasts by shRNA lentivirus infection did not affect fusion, in spite the apparently successful repression of Kirrel3 expression. We believe that this data is insufficient to eliminate Kirrel3's role in mammalian myoblast fusion, and that the lack of effect could be due to the presence of other Kirrel family members (like Kirrel1) or other adhesion proteins that may function redundantly, or because the knockdown of the protein is not complete and the residual receptor is sufficient to allow normal fusion.

Another possible explanation is that the conditions used in the fusion assay masked the need of Kirrel3. Fusion of C2 myoblasts was assayed in highly crowded culture conditions. Since, we observed that the protein levels of Kirrel3 were substantially reduced under these conditions; it is possible that Kirrel3 is not needed when myoblasts already formed tight interactions with other myoblasts. Therefore, if Kirrel3 functions earlier at the initial recognition, re-performing this experiment with sparse myoblasts might influence fusion.

Cytoskeletal remodeling is an essential step in the recognition, cell-cell contacts and mechanical force that is needed to initiate the fusion process. Previous studies have shown that one of the central proteins involved in the signal to cytoskeleton remodeling in both drosophila and zebrafish is the small GTPase Rac1. In this study, we found that Rac1 is active during the initial hours of differentiation, prior to myoblast fusion. In this respect, the activity of Rac1 parallels the early expression and processing of Kirrel3. Further studies will aim to substantiate the functional and biochemical interactions between Kirrel3 and Rac1 in the fusion of mammalian myoblast cells.