page 1077. The findings suggest that the axon may take morphology instructions from its postsynaptic partner.
The ends of axons branch into complex patterns that are precisely mirrored at the receiving end of a synapse at the mammalian neuromuscular junction (NMJ). On the muscle side, this complex postsynaptic pattern can be seen as a pretzel-shaped array of acetylcholine receptors (AChRs). Based on decades' worth of in vitro studies, the formation of this pretzel pattern from an oval-shaped precursor was thought to depend on the branching pattern of the apposing axon. But Kummer et al. find that when muscle cells are given the right matrix, no nerve is needed.
Matrix molecules that supported pretzel formation were fibronectin and laminin—those to which the muscle cells adhered tightly. As in vivo, their formation depends on MuSK (a kinase receptor for nerve-generated agrin) and rapsyn, which binds and clusters AChRs. The pretzel-like patterns developed through a sequence of AChR addition and remodeling steps resembling those seen in vivo. AChR receptors were seen in endocytic vesicles, suggesting that the pretzel might be formed from the original oval shape via localized endocytosis of spots of receptor-rich membrane. Whether zones of tight adherence are needed in vivo for pretzel formation is not yet clear, but laminin is abundant at NMJs.
The nerve is not without its contribution, however. In its absence, some features of the muscle side of the synapse were altered, including the distribution of matrix molecules. Future studies might test whether the muscle likewise influences axonal branching.
Branching is normal for nerves, but how muscle cells design and then form such a complex pattern is harder to imagine. Since the process can now be studied in vitro, answers may be forthcoming. ▪