SUMMARY Following synthesis, integral membrane proteins dwell in the endoplasmic reticulum (ER) for variable periods that are typically rate limiting for plasma membrane delivery. In neurons, the ER extends for hundreds of microns as an anastomosing network throughout highly branched dendrites. However, little is known about the mobility, spatial scales, or dynamic regulation of cargo in the dendritic ER. Here we show that membrane proteins, including AMPA-type glutamate receptors, rapidly diffuse within the continuous network of dendritic ER but are confined by increased ER complexity at dendritic branch points and near dendritic spines. The spatial range of receptor mobility is rapidly restricted by type I mGluR signaling through a mechanism involving protein kinase C (PKC) and the ER protein CLIMP63. Moreover, local zones of ER complexity compartmentalize ER export and correspond to sites of new dendritic branches. Thus, local control of ER complexity spatially scales secretory trafficking within elaborate dendritic arbors.

At synapses in organisms ranging from fly to human, a decrease in postsynaptic neurotransmitter receptor function elicits a homeostatic increase in presynaptic release that restores baseline synaptic efficacy. This process, termed presynaptic homeostasis, requires a retrograde, trans-synaptic signal of unknown identity. In a forward genetic screen for homeostatic plasticity genes we identified multiplexin. Multiplexin is the Drosophila homologue of Collagen XV/XVIII, a matrix protein that can be proteolytically cleaved to release Endostatin, an anti-angiogenesis signaling factor. Here we demonstrate that Multiplexin is required for normal calcium channel abundance, presynaptic calcium influx and neurotransmitter release. Remarkably, Endostatin has a specific activity, independent of baseline synapse development that is required for the homeostatic modulation of presynaptic calcium influx and neurotransmitter release. Our data support a model in which proteolytic release of Endostatin signals trans-synaptically, acting in concert with the presynaptic CaV2.1 calcium channel, to promote presynaptic homeostasis.

SUMMARY Ion channel gene expression can vary substantially among neurons of a given type, even though neuron-type-specific firing properties remain stable and reproducible. The mechanisms that modulate ion channel gene expression and stabilize neural firing properties are unknown. In Drosophila, we demonstrate that loss of the Shal potassium channel induces the compensatory rebalancing of ion channel expression including, but not limited to, the enhanced expression and function of Shaker and slowpoke. Using genomic and network modeling approaches combined with genetic and electrophysiological assays, we demonstrate that the transcription factor Krüppel is necessary for the homeostatic modulation of Shaker and slowpoke expression. Remarkably, Krüppel induction is specific to the loss of Shal, not being observed in five other potassium channel mutants that cause enhanced neuronal excitability. Thus, homeostatic signaling systems responsible for rebalancing ion channel expression can be selectively induced after the loss or impairment of a specific ion channel.

The homeostatic modulation of neurotransmitter release, termed presynaptic homeostatic potentiation (PHP), is a fundamental type of neuromodulation, conserved from Drosophila to human, that stabilizes information transfer at synaptic connections throughout the nervous system. Here we demonstrate that α2δ-3, an auxiliary subunit of the presynaptic calcium channel, is required for PHP. The α2δ gene family has been linked to chronic pain, epilepsy, autism and the action of two psychiatric drugs, gabapentin and pregabalin. We demonstrate that loss of α2δ-3 blocks both the rapid induction and sustained expression of PHP due to a failure to potentiate presynaptic calcium influx and the RIM-dependent readily-releasable vesicle pool. These deficits are independent of α2δ-3-mediated regulation of baseline calcium influx and presynaptic action potential waveform. α2δ proteins reside at the extracellular face of presynaptic release sites throughout the nervous system, an ideal site to mediate rapid, trans-synaptic homeostatic signaling in health and disease.

Presynaptic homeostatic plasticity stabilizes information transfer at synaptic connections in organisms ranging from insect to human. By analogy with principles of engineering and control theory, the molecular implementation of PHP is thought to require postsynaptic signaling modules that encode homeostatic sensors, a set point, and a controller that regulates transsynaptic negative feedback. The molecular basis for these postsynaptic, homeostatic signaling elements remains unknown. Here, an electrophysiology-based screen of the Drosophila kinome and phosphatome defines a postsynaptic signaling platform that includes a required function for PI3K-cII, PI3K-cIII and the small GTPase Rab11 during the rapid and sustained expression of PHP. We present evidence that PI3K-cII localizes to Golgi-derived, clathrin-positive vesicles and is necessary to generate an endosomal pool of PI(3)P that recruits Rab11 to recycling endosomal membranes. A morphologically distinct subdivision of this platform concentrates postsynaptically where we propose it functions as a homeostatic controller for retrograde, trans-synaptic signaling.

Highlights d Multiple SAGA complex genes are essential for presynaptic homeostasis d The SAGA complex acts specifically within glia during presynaptic homeostasis d Glial SAGA regulates the extracellular matrix to stabilize synaptic homeostasis d Glutamate receptor function is linked to a systemic, glial, epigenetic response

Background Single-cell RNA-Sequencing (scRNA-Seq) has provided single-cell level insights into complex biological processes. However, the high frequency of gene expression detection failures in scRNA-Seq data make it challenging to achieve reliable identification of cell-types and Differentially Expressed Genes (DEG). Moreover, with the explosive growth of single-cell data using 10x genomics protocol, existing methods will soon reach the computation limit due to scalability issues. The single-cell transcriptomics field desperately need new tools and framework to facilitate large-scale single-cell analysis. Results In order to improve the accuracy, robustness, and speed of scRNA-Seq data processing, we propose a generalized zero-inflated negative binomial mixture model, “JOINT,” that can perform probability-based cell-type discovery and DEG analysis simultaneously without the need for imputation. JOINT performs soft-clustering for cell-type identification by computing the probability of individual cells, i.e. each cell can belong to multiple cell types with different probabilities. This is drastically different from existing hard-clustering methods where each cell can only belong to one cell type. The soft-clustering component of the algorithm significantly facilitates the accuracy and robustness of single-cell analysis, especially when the scRNA-Seq datasets are noisy and contain a large number of dropout events. Moreover, JOINT is able to determine the optimal number of cell-types automatically rather than specifying it empirically. The proposed model is an unsupervised learning problem which is solved by using the Expectation and Maximization (EM) algorithm. The EM algorithm is implemented using the TensorFlow deep learning framework, dramatically accelerating the speed for data analysis through parallel GPU computing. Conclusions Taken together, the JOINT algorithm is accurate and efficient for large-scale scRNA-Seq data analysis via parallel computing. The Python package that we have developed can be readily applied to aid future advances in parallel computing-based single-cell algorithms and research in various biological and biomedical fields.

No abstract