Synaptogenesis: Wnt and TGF-β Take Centre Stage

Synapse formation is fundamental to the development and function of the brain, and it is thought to be of central importance in learning and memory. Assembly of synapses requires coordination between the neurotransmitter-releasing presynaptic neuron and the signal-receiving postsynaptic cell. In recent years, major advances have shown how presynaptic and postsynaptic components interact during synaptogenesis. Crosstalk between presynaptic and postsynaptic cells is essential for the proper apposition of synaptic components (for review see [1]). The identity of secreted signalling molecules involved in this crosstalk is now becoming clearer with the findings [2–5] that signalling via Wnt and TGF-β occurs at the neuromuscular synapse. Wnts are secreted signalling molecules implicated in many important development processes, notably embryonic patterning, tissue polarity and cell movement [6]. Recent work has shown that Wnts also regulate axonal behaviour and synapse formation. The first evidence that Wnts regulate synapse formation came from studies in Wnt-7a mutant mice, showing a delay in cerebellar synaptic maturation [7]. In cultured neurons, Wnt-7a has been shown to regulate axonal microtubule organisation and the clustering of presynaptic proteins [7]. Packard et al. [2] have now reported strong evidence that Wnt signalling plays a role in the assembly and growth of the Drosophila neuromuscular junction. Packard et al. [2] found that Wingless (Wg) — the prototypical Drosophila Wnt — and its receptor are localised at the neuromuscular junction during larval development [2]. Wg is secreted by the motoneuron and accumulates at both the presynaptic and postsynaptic terminals. Consistent with this, the Wg receptor DFz-2 is localised on both presynaptic and postsynaptic membranes. These findings suggest that Wg is an anterograde — that is, forward — signal regulating the presynaptic and postsynaptic cells. This contrasts with the mouse Wnt-7a, which functions as a retrograde — backward — signal to regulate presynaptic differentiation [7]. Using a thermosensitive allele of wg (wgts) to bypass the requirement of Wg function during early development, Packard et al. [2] were able to show that secreted Wg is required for the normal growth and maturation of the neuromuscular junction. This is the first evidence that Wnt signalling plays a role at the neuromuscular junction. Wg signalling appears to regulate the shape, size and ultrastructure of the neuromuscular junction. The wgts mutant fly has significantly decreased fewer synaptic boutons (presynaptic terminals) [2]. These boutons are larger than normal, with an irregular morphology. They have fewer ‘active zones’ — subcellular structures involved in neurotransmitter release — and mitochondria. Conversely, overexpression of wg in motoneurons was found to increase the number of synaptic boutons. The wgts mutant also exhibits a postsynaptic phenotype [2]. These defects suggest that Wg might signal to both sides of the synapse. Alternatively, Wg might directly signal to just one or other terminal — either presynaptic or postsynaptic — to elicit secondary changes on the other side of the synapse. Overproduction of a dominant-negative form of the receptor Dfz2 in muscles mimics the wg phenotype [2]. At first sight this would seem to suggest that Wg signals to the postsynaptic terminal; however, overproduction of wild-type DFz-2 in muscles also results in a wg phenotype, arguing that postsynaptic DFz-2 may act as a Wg ‘sink’ that modulates the level of Wg in the synaptic cleft. These findings raise the possibility that DFz2 may have a novel function as a negative regulator of Wg signalling at the postsynaptic membrane (Figure 1). But further experiments are needed to test the direct function of Wg and the role of DFz-2 on each side of the synapse. How does Wg regulate the shape and size of boutons? Lack of wg function induces changes in microtubule organisation at the presynaptic terminal, as revealed by staining for Futsch [2]. Futsch is a homologue of the microtubule-associated protein MAP-1B which regulates microtubule organisation at synaptic boutons [8]. The wgts mutant exhibits an increase in unbundled, splayed and punctate Futschlabelled microtubules, suggesting that the cytoskeleton in wgts boutons is more dynamic than normal [2]. Just like the wgts mutant flies, futsch mutants also exhibit a decreased number of boutons with an increased bouton size, associated with changes in microtubule organisation [8]. These results suggest that Wg regulates microtubules at synaptic boutons by affecting Futsch function. These findings are consistent with studies on Wnts and MAP-1B in developing mouse axons. Wnt-7a signalling in mouse was shown to lead to increased microtubule stability and decreased MAP-1B phosphorylation (and possibly binding to microtubules) through inhibition of GSK-3β [9]. GSK-3β is a serine/ threonine protein kinase which directly phospho-rylates MAP-1B to maintain microtubules in a dynamic state Current Biology, Vol. 13, R60–R62, January 21, 2003, ©2003 Elsevier Science Ltd. All rights reserved. PII S0960-9822(02)01429-X