3D snapshot microscopy enables fast volumetric imaging by capturing a 3D volume in a single 2D camera image and performing computational reconstruction. Fast volumetric imaging has a variety of biological applications such as whole brain imaging of rapid neural activity in larval zebrafish. The optimal microscope design for this optical 3D-to-2D encoding is both sample- and task-dependent, with no general solution known. Deep learning based decoders can be combined with a differentiable simulation of an optical encoder for end-to-end optimization of both the deep learning decoder and optical encoder. This technique has been used to engineer local optical encoders for other problems such as depth estimation, 3D particle localization, and lensless photography. However, 3D snapshot microscopy is known to require a highly non-local optical encoder which existing UNet-based decoders are not able to engineer. We show that a neural network architecture based on global kernel Fourier convolutional neural networks can efficiently decode information from multiple depths in a volume, globally encoded across a 3D snapshot image. We show in simulation that our proposed networks succeed in engineering and reconstructing optical encoders for 3D snapshot microscopy where the existing state-of-the-art UNet architecture fails. We also show that our networks outperform the state-of-the-art learned reconstruction algorithms for a computational photography dataset collected on a prototype lensless camera which also uses a highly non-local optical encoding.

Aquaporin-9, an aquaglyceroporin present in diverse tissues, is unique among aquaporins because it is not only permeable to water, urea and glycerol, but also allows passage of larger uncharged solutes. Single particle analysis of negatively stained recombinant rat aquaporin-9 revealed a particle size characteristic of the tetrameric organization of all members of the aquaporin family. Reconstitution of aquaporin-9 into two-dimensional crystals enabled us to calculate a projection map at 7 A resolution. The projection structure indicates a tetrameric structure, similar to GlpF, with each square-like monomer forming a pore. A comparison of the pore-lining residues between the crystal structure of GlpF and a homology model of aquaporin-9 locates substitutions in these residues predominantly to the hydrophobic edge of the tripathic pore of GlpF, providing first insights into the structural basis for the broader substrate specificity of aquaporin-9.

The brain receives information about the direction of object motion from several types of retinal ganglion cells (RGCs). On-Off direction-selective (DS) RGCs respond preferentially to stimuli moving quickly in one of four directions and provide a significant (but difficult to quantify) fraction of RGC input to the SC. On DS RGCs, in comparison, respond preferentially to stimuli moving slowly in one of three directions and are thought to only target retinorecipient nuclei comprising the accessory optic system, e.g., the medial terminal nucleus (MTN). To determine the fraction of SC-projecting RGCs that exhibit direction selectivity, and the specificity with which On-Off and On DS RGCs target retinorecipient areas, we performed optical and electrophysiological recordings from RGCs retrogradely labeled from the mouse SC and MTN. We found, surprisingly, that both On-Off and On DS RGCs innervate the SC; collectively they constitute nearly 40% of SC-projecting RGCs. In comparison, only On DS RGCs project to the MTN. Subsequent experiments revealed that individual On DS RGCs innervate either the SC or MTN and exhibit robust projection-specific differences in somatodendritic morphology, cellular excitability, and light-evoked activity; several projection-specific differences in the output of On DS RGCs correspond closely to differences in excitatory synaptic input the cells receive. Our results reveal a robust projection of On DS RGCs to the SC, projection-specific differences in the response properties of On DS RGCs, and biophysical and synaptic mechanisms that underlie these functional differences.

During low-frequency firing, action potentials actively invade the dendrites of CA1 pyramidal neurons. At higher firing rates, however, activity-dependent processes result in the attenuation of back-propagating action potentials, and propagation failures occur at some dendritic branch points. We tested two major hypotheses related to this activity-dependent attenuation of back-propagating action potentials: (1) that it is mediated by a prolonged form of sodium channel inactivation and (2) that it is mediated by a persistent dendritic shunt activated by back-propagating action potentials. We found no evidence for a persistent shunt, but we did find that cumulative, prolonged inactivation of sodium channels develops during repetitive action potential firing. This inactivation is significant after a single action potential and continues to develop during several action potentials thereafter, until a steady-state sodium current is established. Recovery from this form of inactivation is much slower than its induction, but recovery can be accelerated by hyperpolarization. The similarity of these properties to the time and voltage dependence of attenuation and recovery of dendritic action potentials suggests that dendritic sodium channel inactivation contributes to the activity dependence of action potential back-propagation in CA1 neurons. Hence, the biophysical properties of dendritic sodium channels will be important determinants of action potential-mediated effects on synaptic integration and plasticity in hippocampal neurons.

Selective transcription of human mitochondrial DNA requires a transcription factor (mtTF) in addition to an essentially nonselective RNA polymerase. Partially purified mtTF is able to sequester promoter-containing DNA in preinitiation complexes in the absence of mitochondrial RNA polymerase, suggesting a DNA-binding mechanism for factor activity. Functional domains, required for positive transcriptional regulation by mtTF, are identified within both major promoters of human mtDNA through transcription of mutant promoter templates in a reconstituted in vitro system. These domains are essentially coextensive with DNA sequences protected from nuclease digestion by mtTF-binding. Comparison of the sequences of the two mtTF-responsive elements reveals significant homology only when one sequence is inverted; the binding sites are in opposite orientations with respect to the predominant direction of transcription. Thus mtTF may function bidirectionally, requiring additional protein-DNA interactions to dictate transcriptional polarity. The mtTF-responsive elements are arrayed as direct repeats, separated by approximately 80 bp within the displacement-loop region of human mitochondrial DNA; this arrangement may reflect duplication of an ancestral bidirectional promoter, giving rise to separate, unidirectional promoters for each strand.

In mammals, cerebellar neurons are categorized as glutamatergic or GABAergic, and are derived from progenitors that express the proneural genes atoh1 or ptf1a, respectively. In zebrafish, three atoh1 genes, atoh1a, atoh1b, and atoh1c, are expressed in overlapping but distinct expression domains in the upper rhombic lip (URL): ptf1a is expressed exclusively in the ventricular zone (VZ). Using transgenic lines expressing fluorescent proteins under the control of the regulatory elements of atoh1a and ptf1a, we traced the lineages of the cerebellar neurons. The atoh1+ progenitors gave rise not only to granule cells but also to neurons of the anteroventral rhombencephalon. The ptf1a+ progenitors generated Purkinje cells. The olig2+ eurydendroid cells, which are glutamatergic, were derived mostly from ptf1a+ progenitors in the VZ but some originated from the atoh1+ progenitors in the URL. In the adult cerebellum, atoh1a, atoh1b, and atoh1c are expressed in the molecular layer of the valvula cerebelli and of the medial corpus cerebelli, and ptf1a was detected in the VZ. The proneural gene expression patterns coincided with the sites of proliferating neuronal progenitors in the adult cerebellum. Our data indicate that proneural gene-linked neurogenesis is evolutionarily conserved in the cerebellum among vertebrates, and that the continuously generated neurons help remodel neural circuits in the adult zebrafish cerebellum.

The Drosophila mushroom body (MB) is a key associative memory center that has also been implicated in the control of sleep. However, the identity of MB neurons underlying homeostatic sleep regulation, as well as the types of sleep signals generated by specific classes of MB neurons, has remained poorly understood. We recently identified two MB output neuron (MBON) classes whose axons convey sleep control signals from the MB to converge in the same downstream target region: a cholinergic sleep-promoting MBON class and a glutamatergic wake-promoting MBON class. Here, we deploy a combination of neurogenetic, behavioral, and physiological approaches to identify and mechanistically dissect sleep-controlling circuits of the MB. Our studies reveal the existence of two segregated excitatory synaptic microcircuits that propagate homeostatic sleep information from different populations of intrinsic MB "Kenyon cells" (KCs) to specific sleep-regulating MBONs: sleep-promoting KCs increase sleep by preferentially activating the cholinergic MBONs, while wake-promoting KCs decrease sleep by preferentially activating the glutamatergic MBONs. Importantly, activity of the sleep-promoting MB microcircuit is increased by sleep deprivation and is necessary for homeostatic rebound sleep (i.e., the increased sleep that occurs after, and in compensation for, sleep lost during deprivation). These studies reveal for the first time specific functional connections between subsets of KCs and particular MBONs and establish the identity of synaptic microcircuits underlying transmission of homeostatic sleep signals in the MB.

The formation of functional neuronal circuits relies on accurate migration and proper axonal outgrowth of neuronal precursors. On the route to their targets migrating cells and growing axons depend on both, directional information from neurotropic cues and adhesive interactions mediated via extracellular matrix molecules or neighbouring cells. The inactivation of guidance cues or the interference with cell adhesion can cause severe defects in neuronal migration and axon guidance. In this study we have analyzed the function of the MAM domain containing glycosylphosphatidylinositol anchor 2A (MDGA2A) protein in zebrafish cranial motoneuron development. MDGA2A is prominently expressed in distinct clusters of cranial motoneurons, especially in the ones of the trigeminal and facial nerves. Analyses of MDGA2A knockdown embryos by light sheet and confocal microscopy revealed impaired migration and aberrant axonal outgrowth of these neurons; suggesting that adhesive interactions mediated by MDGA2A are required for the proper arrangement and outgrowth of cranial motoneuron subtypes.

Vocal learning in songbirds requires a basal ganglia circuit termed the anterior forebrain pathway (AFP). The AFP is not required for song production, and its role in song learning is not well understood. Like the mammalian striatum, the striatal component of the AFP, Area X, receives dense dopaminergic innervation from the midbrain. Since dopamine (DA) clearly plays a crucial role in basal ganglia-mediated motor control and learning in mammals, it seems likely that DA signaling contributes importantly to the functions of Area X as well. In this study, we used voltammetric methods to detect subsecond changes in extracellular DA concentration to gain better understanding of the properties and regulation of DA release and uptake in Area X. We electrically stimulated Ca(2+)- and action potential-dependent release of an electroactive substance in Area X brain slices and identified the substance as DA by the voltammetric waveform, electrode selectivity, and neurochemical and pharmacological evidence. As in the mammalian striatum, DA release in Area X is depressed by autoinhibition, and the lifetime of extracellular DA is strongly constrained by monoamine transporters. These results add to the known physiological similarities of the mammalian and songbird striatum and support further use of voltammetry in songbirds to investigate the role of basal ganglia DA in motor learning.