RPM-1 ( R egulator of P re-synaptic M orphology) is a large, evolutionarily conserved protein that functions in the organization/generation of pre-synaptic termini in C. elegans . In a typical presynaptic terminal, synaptic vesicles are clustered orderly surrounding the electron-dense active zone. Loss of function mutations in
rpm-1 cause alterations in the size and overall arrangement of presynaptic termini (Zhen et al., 2000). RPM-1 is localized to a region called the 'periactive zone', referring to a subsynaptic domain that is adjacent to synaptic vesicle clusters and the active zone. RPM-1 has several conserved functional modules including a GEF and a Ring-finger ubiquitin E3 ligase domain. To elucidate the molecular mechanism of RPM-1's action, we have carried out genetic, biochemical, and cell biological experiments. Thus far, our analysis has revealed a role for RPM-1 as an E3 ubiquitin ligase in the regulation of a novel MAP Kinase signaling cascade. We performed a suppressor screen for
rpm-1. Analysis of a set of suppressor mutations led to the identification of three genes,
dlk-1,
mkk-4, and
pmk-3 (Nakata, et. al. , 2003 International Worm Meeting). DLK-1 is the C. elegans ortholog of the Dual-Leucine zipper bearing kinase, a member of the mixed lineage group of mitogen activated protein (MAP) kinase kinase kinases. MKK-4 is a member of the MKK4 family of MAP kinase kinases. PMK-3 is one of the three
p38-like MAP kinases in C. elegans . Genetic analysis supports that these kinases act in a linear pathway with
dlk-1 upstream of
mkk-4 , and
mkk-4 acting upstream of
pmk-3. Loss of function in any kinase gene suppresses loss of function in
rpm-1. In contrast, over-expression or constitutive activation of the MAP kinase cascade phenocopies the effect of loss of
rpm-1 function. The function of this MAP kinase cascade is required in presynaptic neurons. Moreover, functional DLK-1::GFP is localized to synapses, and is more abundant in
rpm-1 mutants than in wild type animals. These in vivo studies support a conclusion that RPM-1 acts as a negative regulator of the
dlk-1/mkk-4/pmk-3 pathway. One possible mechanism by which this may occur is through the E3 ligase activity of RPM-1. We tested this by co-expressing the ring finger domain of RPM-1 with DLK-1 in HEK293T cells, and observed that the ring finger of RPM-1 was capable of enhancing ubiquitination of DLK-1. Based on our findings, we propose that the negative regulation of the MAP kinase pathway by RPM-1 at the periactive zone may create a subsynaptic 'boundary', which facilitates the formation and stabilization of pre-synaptic architecture. To further test this hypothesis and to gain a deep understanding of 'periactive zone' signaling mechanism in synapse formation and function, we are continuing to isolate and characterize
rpm-1 suppressor mutations, in combination with candidate gene approaches. In addition, we are addressing how the synaptic localization of RPM-1 is regulated by immunocytochemical studies in mutants that exhibit altered synaptic architecture. We have also undertaken a structure/function analysis of RPM-1, and have identified a region of RPM-1 that appears to be important for its synaptic localization. Such information will be used to identify additional proteins associated with RPM-1, thereby further dissecting the signaling pathways by which RPM-1 regulates synapse development.