Signal transduction from neurons to their targets depends on specialized subcellular structures at synaptic junctions. A typical chemical synapse is composed of three compartments: a presynaptic terminal for rapid neurotransmitter release, a postsynaptic site that is rich in neurotransmitter receptors to receive the signal, and a synaptic cleft that bridges the pre- and postsynaptic partners and maintains them in precise registration. The size and organization of synapses vary widely between neuronal types and are modulated by experience. The general processes of synaptogenesis and how such processes give rise to the diversity of synapses in the nervous system are topics under intense investigation.
We use the simple nervous system of the nematode C. elegans to examine how synapses are formed. This nervous system is composed of only 302 neurons, and its entire connectivity is known at the ultrastructural level. Moreover, most of the synaptic connections can be observed in living animals using transgenic fluorescent protein markers. Our past studies have provided insights into synaptogenesis and axon guidance in the developing nervous system. Our recent work has begun to address the regulation of synapses in the mature nervous system and the ability of injured neurons to regenerate and repair damage.
Regulation of Presynaptic Architecture by the Conserved Giant Ubiquitin Ligase RPM-1
The C. elegans RPM-1 (regulator of presynaptic morphology) is a member of a conserved protein family that includes Drosophila Highwire and vertebrate Phr1, Pam, and Esrom. These large proteins contain multiple conserved motifs, including an RLD guanine nucleotide exchange factor (GEF) domain and a RING E3 ubiquitin ligase domain. They are highly expressed in the nervous systems and have been shown to regulate neuronal development at multiple steps. In mature neurons, RPM-1 is localized to the perisynaptic region, a unique region in the presynaptic terminal. In rpm-1 null mutants, many synapses exhibit an altered architecture: the ratio of synaptic vesicle to presynaptic density is reduced. In addition, some axons overextend beyond normal termination sites.
We have taken a combinatorial approach using genetics and biochemistry to dissect the signaling pathways regulated by RPM-1. We hypothesize that one reason the synapses are improperly formed in rpm-1 mutants is because RPM-1 substrates are not degraded in a timely manner. To identify such substrates, we screened among rpm-1(lf) mutants for second-site mutations that eliminate or reduce the activity of candidate RPM-1 substrates, therefore restoring synaptic defects of rpm-1(lf) to normal organization. This analysis led us to the identification of a novel MAP kinase cascade composed of the dual leucine zipperbearing MAPKKK DLK-1, the MAPKK MKK-4, and the p38-like MAP kinase PMK-3. Inactivation of these three kinases suppresses rpm-1(lf) synaptic defects, whereas elevated expression or constitutive activation of the kinases causes synaptic defects resembling rpm-1(lf). The DLK-1/MAPKKK is localized to the synaptic region, and its synaptic abundance is elevated in rpm-1 mutants. Biochemically, RPM-1 can bind DLK-1 and can stimulate ubiquitination of DLK-1. Thus, the activity of this MAP kinase cascade is negatively controlled by RPM-1 through the regulation of protein stability. This underscores the importance of ubiquitin-mediated regulation in synapse development.
Our study further reveals an in vivo function for a previously unknown MAP kinase cascade in presynaptic development. What is the signaling output of the DLK-1/MKK-4/PMK-3 cascade? In our recent studies, we have identified a fourth MAP kinase called MAPKAP2 that is activated by PMK-3. Together these four kinases control the mRNA stability of a transcription factor that is a member of the basicleucine zipper (bZip) protein family. Using live imaging and axotomy, we demonstrated that the mRNA of this bZip protein is present in axons and near synapses, and that the mRNA can be induced to be translated locally in a manner that is dependent on DLK-1. These studies reveal the potential importance of local protein synthesis in synapse formation and maintenance.
RPM-1 is a giant protein and likely functions through multiple signal transduction pathways. We have used a proteomic approach to search for proteins associated with RPM-1. One of the candidates is GLO-4, an RLD-domaincontaining GEF protein. We find that GLO-4 colocalizes with RPM-1 at the presynaptic terminals. Loss of function in glo-4 causes synaptic defects partially resembling those in rpm-1 mutants. Furthermore, we show that GLO-4 GEF acts through a Rab GTPase and the AP3 adapter proteins. The GLO/AP3 pathway is positively regulated by RPM-1 to promote vesicular trafficking through late endo-lysosomes. Thus, through a combination of genetic, biochemical, and molecular approaches, we discovered two parallel signaling pathways that function downstream of RPM-1 in synapse formation. Growing evidence supports the hypothesis that these signaling mechanisms are conserved in other organisms and in other cellular processes.
Assembly of Presynaptic Active Zones
The active zones (AZs) at the presynaptic terminals are located at the center of the synaptic vesicle (SV) exocytic zone, promote SV docking, and provide the precise registry between pre- and postsynaptic specializations. AZs display morphologically distinct ultrastructures. By screening for mutants with altered SV distribution, we discovered that the conserved protein SYD-2, a member of the alpha-liprin protein family, is a key regulator of AZ assembly. Liprins are composed of highly conserved coiled-coil and SAM domains. They were originally identified as binding partners for the LAR receptor protein-tyrosine phosphatase (RPTP). We find that C. elegans LAR is localized to presynaptic membranes and acts with the extracellular matrix protein nidogen to pattern presynaptic terminals. In syd-2(lf) mutants, AZs exhibit a fragmented appearance, causing diffused localization of SVs. In contrast, an unusual gain-of-function syd-2 mutation is able to promote synapse assembly in the absence of several upstream regulators. Using this mutation, we have dissected the complex interactions among AZ proteins. This study provides insight into the molecular network of presynaptic AZ assembly.
Regulation of Locomotion Circuit
Beyond the formation of synapses lie many unanswered questions about how a functional synapse is integrated and regulated within neuronal networks. We have used the C. elegans locomotor circuit to investigate the coordination between the excitatory and inhibitory activities. By characterizing a genetic mutant that harbors a gain-of-function mutation in a neuronal nicotinic acetylcholine receptor, ACR-2, we uncovered a regulation of excitation and inhibition balance in maintaining the sinusoidal movement of the animal. The acr-2(gf) mutant exhibits epileptic-like body contractions. Using genetics and electrophysiology, we find that the ACR-2 receptor acts in the cholinergic motor neurons to maintain their excitability. Intriguingly, hyperexcitation of the cholinergic neurons by acr-2(gf) leads to an anti-homeostatic effect on the downstream inhibitory neurons, resulting in a net overexcitation output from the motor circuit. This acr-2(gf) effect mimics the physiological state of epilepsy. The molecular nature of the acr-2(gf) mutation is nearly identical to several disease-associated mutations in human acetylcholine receptors. Our ongoing studies focus on identification of novel regulators of the network excitation and inhibition imbalance, which will provide mechanistic insight to the search for effective management of disease conditions.
Nerve Regeneration Mechanisms of C. elegans Neurons
The limited ability of mature neurons to regenerate upon injury is an enigma. Such limitations severely impede our ability to repair damaged nerves and to recover function after injury. We have developed a laser-assisted axotomy methodology to sever adult C. elegans neurons, and have observed that the injured axons exhibit neuron-type, neurite-type, and stage-dependent regenerative responses. The injured proximal axon stumps frequently re-initiate growth and elongate in an error-prone, misguided manner. Using a genetically encoded calcium sensor, we find that the regrowth is correlated with calcium transients that are triggered by axotomy.
Using genetic mutants, we have identified several intrinsic pathways that can promote or inhibit the regrowth rate. For example, elevating cAMP and protein kinase A signaling accelerates early formation of growth cones and subsequent growth rate, whereas inhibiting the DLK-1 kinase cascade, which we identified in studies of synapse formation in development, completely blocks regenerative responses in wild type and in mutants with elevated cAMP. Our studies also show that extrinsic pathways, such as the Eph receptor, are involved in axon regrowth precision. We are now performing a large-scale genetic screen to uncover novel and conserved pathways. Our studies in C. elegans will provide insights to general understanding of the key steps that a mature neuron must overcome for full functional recovery after injury.
Grants from the National Institutes of Health provided partial support for portions of this work.
As of May 30, 2012