Spatial regulation of an E3 ubiqitin ligase specifies neuronal connectivity through selective synapse elimination.

How are extracellular synaptogenic signals converted into an intracellular synaptic assembly program, resulting in synapses forming at the right location and with the appropriate density? Theoretically, stereotyped synaptic connectivity can arise both from precise recognition between appropriate partners during synaptogenesis and from selective synapse elimination after an initial phase of exuberant synapse formation. The molecular mechanisms that underlie selective synapse removal are largely unknown. We decided to address this question by using the HSNL neuron in C. elegans, in which synapse formation is mediated by the transmembrane molecules SYG-1 and SYG-2. In syg-1 and syg-2 mutants, HSNL fails to form synapses with its normal synaptic partners at the Primary Synaptic Region (PSR), and instead forms ectopic synapses onto abnormal targets at the Secondary Synaptic Region (SSR) (Fig. 3-1). Guidepost cell-expressed SYG-2 clusters SYG-1 at the PSR, thus promoting synaptogenesis at this site.

Using time-lapse experiments, we showed that during early development, numerous synapses form at both the PSR and SSR. Then, within hours the synapses in SSR are eliminated, while the synapses in PSR are maintained and strengthened. We further showed that the speed and completeness of the synaptic elimination is dependent on the dosage of SYG-1 in HSNL. In syg-1 mutants, SSR synapses fail to be eliminated, while in animals overexpressing syg-1, the SSR synapses are eliminated faster. In a yeast two-hybrid screen with the cytosolic tail of SYG-1 as a bait, we identified SKR-1 as a binding partner of SYG-1. SKR-1 is the homolog of vertebrate Skp1, an integral subunit of the SCF E3 ubiquitin ligase complex. We found that SYG-1 binds to SKR-1 directly and inhibits assembly of the SCF complex, thereby protecting nearby synapses from destruction. Since the SCF complex is present along the entire HSNL axon but SYG-1 is only localized at PSR, the SCF activity is low near PSR and high at SSR. Thus, synapses persist at PSR and are eliminated at SSR. Our data suggest that subcellular regulation of ubiquitin-mediated protein degradation contributes to precise synaptic connectivity through selective synapse elimination.

Since the mutant phenotype of sel-10, the SCF complex F-box protein, is not as strong as that of SYG-1, we are currently performing screens to look for enhancers of sel-10. We are also interested in understanding the targets of the SCF complex in the synapse formation. One potential target we are currently testing is SYD-2/¢G\-liprin, a presynaptic molecule that is essential for the assembly of HSN synapses.

Fig. 3-1 Amodel for the selective elimination of synapses in HSNL.