Homeostatic plasticity restricts changes in neural activity to allow for a stable yet plastic nervous system. At the synapse, homeostatic regulation maintains synaptic efficacy when there is a disruption of postsynaptic neurotransmitter sensitivity or excitability (Turrigiano, 2012; Davis, 2006). At the Drosophila neuromuscular junction (NMJ), when glutamate receptor sensitivity is decreased, either genetically or pharmacologically, the motoneuron potentiates neurotransmitter release to precisely offset changes in receptor sensitivity. This restores muscle excitation within minutes. The molecular mechanisms that underlie this form of synaptic homeostasis are beginning to be discovered, in large part due to an ongoing forward genetic screen for mutations that block synaptic homeostasis (Dickman and Davis, 2009; Müller et al., 2011). This thesis work focuses on the examination of two candidate mutations identified by this genetic screen. These were chosen for study because they completely blocked synaptic homeostasis and were likely to disrupt genes encoding ion channel subunits.

In chapter 2 we provide evidence that two Degenerin/Epithelial Sodium Channel (DEG/ENaC) subunits, pickpocket11 (ppk11) and pickpocket16 (ppk16), are necessary for synaptic homeostasis. These genes are required for both the rapid induction and sustained expression of synaptic homeostasis. Mutations in ppk11 and ppk16 do not disrupt baseline synaptic transmission or NMJ development. ppk11 and ppk16 reside in a single locus and are coregulated as an operon-like genetic unit, with increased transcription accompanying prolonged expression of synaptic homeostasis. We use pharmacology to show that DEG/ENaC channel conductance is continuously required for the potentiation of release during homeostasis. Lastly we show that DEG/ENaC channel function is necessary to enhance action-potential evoked presynaptic calcium influx specifically during synaptic homeostasis, but not baseline release. Taken together this identifies PPK11 and PPK16 as components of the molecular pathway that potentiates release during synaptic homeostasis and we present a model for how PPK11 and PPK16 enhance calcium influx during synaptic homeostasis.

In chapter 3 we examine a third gene encoding a DEG/ENaC channel subunit, pickpocket 18 (ppk18), which is in the same genetic locus as ppk11 and ppk16. We found that despite the proximity to ppk11 and ppk16, ppk18 is transcribed independently from the other two genes, and the quantity of ppk18 transcript is unchanged during the sustained expression of synaptic homeostasis. Furthermore, mutations in ppk18 do not block the rapid induction of synaptic homeostasis, indicating that it does not act in concert with ppk11 and ppk16 to potentiate release during synaptic homeostasis.

In chapter 4 we examine a separate mutation that was identified in the same forward genetic screen for mutations that block synaptic homeostasis. This mutation resides in a complicated locus containing four genes: Ionotropic Receptor 75a (IR75a), Ionotropic Receptor 75b (IR75b), Ionotropic Receptor 75c (IR75c), and GABA and glycine-like receptor of Drosophila (GRD). We use a combination of electrophysiology, quantitative PCR, immunostaining and genetics to examine the requirement of different genes in this locus for the homeostatic potentiation of release. We show that multiple mutations within this genomic region block synaptic homeostasis, and that the distribution of postsynaptic glutamate receptors is altered by a mutation within this locus, which may be an underlying cause. We are unable to identify the responsible gene, although our data points to GRD as the most promising candidate. This work provides a basis for future investigation into this locus.

Brain function is remarkably stable throughout the many years of adult life even in the face of continuous internal and external change. Homeostatic plasticity is hypothesized to support this stability by constraining neuronal activity within a range appropriate for a neuron’s participation in neural circuits and meaningful information transfer. Presynaptic homeostatic potentiation (PHP) is one form of homeostatic plasticity acting in both peripheral and central synapses and is conserved from insects to humans. Here we describe the function of an innate immune receptor, PGRP-LC, and its downstream signaling pathway in controlling PHP. Innate immunity is known to function in the brain where it modifies the structure of synapses and circuits during development, as well responding to and controlling neuronal activity during adulthood. This is the first evidence implicating innate immunity in presynaptic function. First, we show that PGRP-LC functions as a presynaptic receptor during the induction and expression of PHP, through modulation of the readily releasable pool of synaptic vesicles. Next, we interrogate the canonical innate immune signaling pathway downstream of PGRP-LC, testing four downstream genes for function in PHP. In so doing, we find evidence that this pathway is reorganized to support the specialized structure and function of neurons, and the needs of PHP. Last, we characterize in detail the function of a kinase in the pathway, Tak1, in controlling synaptic vesicle release probability. Our data lead us to propose a model where Tak1 acts to stabilize synaptic vesicles close to the active zone by inhibiting a de-priming or de-docking process. Therefore, we have implicated a novel kinase in regulation of neurotransmitter release. We expect that innate immune signaling through PGRP-LC then modulates the function of Tak1 during PHP to achieve potentiation of vesicle release. Overall, our data speak to the importance of innate immune signaling in neurons specifically, as they undergo changes during their otherwise normal day-to-day activity.

Nerve cells contact each other at synapses and communicate via synaptic transmission. Once they form synapses are not static, but continue to grow during development, plasticity and disease. Combining genetic and cell biological tools at the Drosophila neuromuscular junction (NMJ), I isolated a number of mutations and signaling pathways regulating morphological and functional synapse growth. In my introduction I define the terms morphological and functional synapse growth as well as review our current knowledge of these topics. In the first chapter, I characterize the nerve-wracked mutation, which disrupts synaptic bouton formation. Second, I describe bad hair day, a novel gene encoding a chromatin-remodeling factor that is involved in suppression of satellite bouton formation. Next, I demonstrate that the Drosophila NF-kappa B pathway regulates GluR abundance via a novel, cytoplasmic mechanism. Finally, I suggest that cytoplasmic NF-kappaB/Dorsal acts upstream of Pelle kinase, which regulates insertion of GluR into the synaptic membrane. These findings reveal some of the molecular mechanisms in place during synapse growth, and may have implications for understanding synaptic plasticity and disease.

The magnitude and timing of neural activity precisely determines adaptive responses to stimuli. Neural activity is subject to extensive transformation between sensation and motor execution, and this transformation itself is subject to change due to the rewiring of the nervous system during development and learning. In this thesis I present data exploring how synapses between neural circuit elements might be homeostatically regulated, such that activity levels are maintained to ensure effective functioning of the nervous system.

Inhibition of postsynaptic glutamate receptors at the Drosophila neuromuscular junction (NMJ) initiates a compensatory increase in presynaptic release termed synaptic homeostasis. This ensures that muscle depolarization stays constant despite diminished postsynaptic function, and this process of synaptic homeostasis may be important for maintaining proper NMJ function throughout development. While BMP signaling has been proposed to mediate the retrograde signal that controls synaptic homeostasis at this synapse, BMP signaling is also necessary for normal synaptic growth and stability. It remains unknown whether BMPs function as instructive retrograde signals that directly modulate presynaptic transmitter release. Here we demonstrate that the BMP receptor (Wit) and ligand (Gbb) are necessary for the induction of synaptic homeostasis, but that Gbb does not function as an instructive retrograde signal. Rather our data indicate that Wit and Gbb function via the downstream transcription factor Mad, and that Mad-mediated signaling is continuously required to gate the expression of synaptic homeostasis in motoneurons.

Forms of synaptic plasticity in the central nervous system such as long-term potentiation (LTP) increase synaptic activity in a synapse- and cell-specific fashion. Although network-wide excitation triggers compensatory homeostatic changes, whether vertebrate neurons initiate homeostatic synaptic changes in response to cell-autonomous increases in excitation has not been examined. We cell-autonomously excited rodent CA1 pyramidal neurons and find that a compensatory postsynaptic depression of both AMPAR and NMDAR function results. Elevated calcium influx through L-type calcium channels leads to activation of a pathway involving CaM kinase kinase and CaM kinase 4 that induces synaptic depression of AMPAR and NMDAR responses. The synaptic depression of AMPARs but not of NMDARs requires protein synthesis and the GluA2 AMPAR subunit.

Here live imaging and genetic manipulations are combined to visualize and quantitatively characterize a population of dynamic pioneer microtubules within the Drosophila neuromuscular junction (NMJ). First, a GFP-tagged microtubule plus-end binding protein, EB1-GFP, was used to qualitatively and quantitatively characterize the growth and invasion of dynamic axonal and synaptic microtubule plus-ends in vivo. Axonal microtubules were found to be uniformly oriented. EB1-GFP movement in the axons was processive and constrained to a straight path. Interestingly, plus-end growth was found to be significantly slower in sensory (3.02 µm/minute ± 0.12) compared to motor (4.12 µm/min ± 0.038) axons, and is significantly faster within the NMJ (4.66 µm/min ± 0.10) compared to the motor axon. Unlike axonal movement, plus-end movement at the synapse was of mixed orientation. Genetic and imaging analysis revealed the movement of dynamic plus-ends to be largely independent of the Map1b-like protein Futsch, though the dynein/dynactin complex was shown to be required for normal plus-end growth. EB1-GFP labeled puncta are often observed exploring nascent and newly formed synaptic regions, an exploratory behavior that may play a role in synapse development. Supporting movies are presented as supplemental data. This tool was next applied to the study of the formin protein, diaphanous (dia), a gene identified in a screen for mutations effecting growth at the NMJ. Diaphanous is present both pre- and postsynaptically however, genetic rescue and interaction experiments support the conclusion that dia is necessary presynaptically for normal NMJ growth. Formin family proteins are potent regulators of the cytoskeleton and defects in cytoskeletal organization or regulation could correlate with the synaptic growth defect. Live imaging of dynamic microtubules and actin revealed defects in both cytoskeletal elements at the synapse. Finally, genetic evidence is presented demonstrating Dia functions downstream of the presynaptic receptor tyrosine phosphatase Dlar and the Rho-type GEF trio to control NMJ growth. Based upon the established function of DRFs as Rho-GTPase dependent regulators of the cell cytoskeleton, a model is proposed in which Diaphanous links receptor tyrosine phosphatase signaling at the plasma membrane to growth-dependent modulation of the synaptic actin and microtubule cytoskeletons.

Presynaptic homeostatic plasticity (PHP) compensates for impaired postsynaptic neurotransmitter receptor function through a rapid, persistent adjustment of neurotransmitter release, an effect that can exceed 200%. An unexplained property of PHP is the preservation of short-term plasticity (STP), thereby stabilizing activity-dependent synaptic information transfer. This dissertation aims to understand the mechanisms that stabilize STP during PHP. We use a combination of electrophysiology, genetics, and biochemistry at the Drosophila NMJ to answer this question. We demonstrate that the dramatic potentiation of presynaptic release during PHP is achieved while simultaneously maintaining a constant ratio of primed to super-primed synaptic vesicles, thereby preserving STP. Mechanistically, genetic, biochemical and electrophysiological evidence argue that a constant ratio of primed to super-primed synaptic vesicles is achieved by the concerted action of three proteins: Unc18, Syntaxin1A and RIM. Our data support a model based on the regulated availability of Unc18 at the presynaptic active zone, a process that is restrained by Syntaxin1A and facilitated by RIM. As such, regulated vesicle priming/super-priming enables PHP to stabilize both synaptic gain and the activity-dependent transfer of information at a synapse.