Serotonin plays different roles across networks within the same sensory modality. Previously, we used whole-cell electrophysiology in Drosophila to show that serotonergic neurons innervating the first olfactory relay are inhibited by odorants (Zhang and Gaudry, 2016). Here we show that network-spanning serotonergic neurons segregate information about stimulus features, odor intensity and identity, by using opposing coding schemes in different olfactory neuropil. A pair of serotonergic neurons (the CSDns) innervate the antennal lobe and lateral horn, which are first and second order neuropils. CSDn processes in the antennal lobe are inhibited by odors in an identity independent manner. In the lateral horn, CSDn processes are excited in an odor identity dependent manner. Using functional imaging, modeling, and EM reconstruction, we demonstrate that antennal lobe derived inhibition arises from local GABAergic inputs and acts as a means of gain control on branch specific inputs that the CSDns receive within the lateral horn.

Dividing cells reorganize their microtubule cytoskeleton into a bipolar spindle, which moves one set of sister chromatids to each nascent daughter cell. Early spindle assembly models postulated that spindle pole-derived microtubules search the cytoplasmic space until they randomly encounter a kinetochore to form a stable attachment. More recent work uncovered several additional, centrosome-independent microtubule generation pathways, but the contributions of each pathway to spindle assembly have remained unclear. Here, we combined live microscopy and mathematical modeling to show that most microtubules nucleate at noncentrosomal regions in dividing human cells. Using a live-cell probe that selectively labels aged microtubule lattices, we demonstrate that the distribution of growing microtubule plus ends can be almost entirely explained by Augmin-dependent amplification of long-lived microtubule lattices. By ultrafast 3D lattice light-sheet microscopy, we observed that this mechanism results in a strong directional bias of microtubule growth toward individual kinetochores. Our systematic quantification of spindle dynamics reveals highly coordinated microtubule growth during kinetochore fiber assembly.

We advance two-photon microscopy for near-diffraction-limited imaging up to 850 µm below the pia in awake mice. Our approach combines direct wavefront sensing of light from a guidestar (formed by descanned fluorescence from Cy5.5-conjugated dextran in brain microvessels) with adaptive optics to compensate for tissue-induced aberrations in the wavefront. We achieve high signal-to-noise ratios in recordings of glutamate release from thalamocortical axons and calcium transients in spines of layer 5b basal dendrites during active tactile sensing.

With the development of neural recording technology, it has become possible to collect activities from hundreds or even thousands of neurons simultaneously. Visualization of neural population dynamics can help neuroscientists analyze large-scale neural activities efficiently. In this letter, Laplacian eigenmaps is applied to this task for the first time, and the experimental results show that the proposed method significantly outperforms the commonly used methods. This finding was confirmed by the systematic evaluation using nonhuman primate data, which contained the complex dynamics well suited for testing. According to our results, Laplacian eigenmaps is better than the other methods in various ways and can clearly visualize interesting biological phenomena related to neural dynamics.

We present a method combining affinity prediction with region agglomeration, which improves significantly upon the state of the art of neuron segmentation from electron microscopy (EM) in accuracy and scalability. Our method consists of a 3D U-net, trained to predict affinities between voxels, followed by iterative region agglomeration. We train using a structured loss based on MALIS, encouraging topologically correct segmentations obtained from affinity thresholding. Our extension consists of two parts: First, we present a quasi-linear method to compute the loss gradient, improving over the original quadratic algorithm. Second, we compute the gradient in two separate passes to avoid spurious gradient contributions in early training stages. Our predictions are accurate enough that simple learning-free percentile-based agglomeration outperforms more involved methods used earlier on inferior predictions. We present results on three diverse EM datasets, achieving relative improvements over previous results of 27%, 15%, and 250%. Our findings suggest that a single method can be applied to both nearly isotropic block-face EM data and anisotropic serial sectioned EM data. The runtime of our method scales linearly with the size of the volume and achieves a throughput of ~2.6 seconds per megavoxel, qualifying our method for the processing of very large datasets.

An approaching predator and self-motion toward an object can generate similar looming patterns on the retina, but these situations demand different rapid responses. How central circuits flexibly process visual cues to activate appropriate, fast motor pathways remains unclear. Here we identify two descending neuron (DN) types that control landing and contribute to visuomotor flexibility in Drosophila. For each, silencing impairs visually evoked landing, activation drives landing, and spike rate determines leg extension amplitude. Critically, visual responses of both DNs are severely attenuated during non-flight periods, effectively decoupling visual stimuli from the landing motor pathway when landing is inappropriate. The flight-dependence mechanism differs between DN types. Octopamine exposure mimics flight effects in one, whereas the other probably receives neuronal feedback from flight motor circuits. Thus, this sensorimotor flexibility arises from distinct mechanisms for gating action-specific descending pathways, such that sensory and motor networks are coupled or decoupled according to the behavioral state.

When a behavior repeatedly fails to achieve its goal, animals often give up and become passive, which can be strategic for preserving energy or regrouping between attempts. It is unknown how the brain identifies behavioral failures and mediates this behavioral-state switch. In larval zebrafish swimming in virtual reality, visual feedback can be withheld so that swim attempts fail to trigger expected visual flow. After tens of seconds of such motor futility, animals became passive for similar durations. Whole-brain calcium imaging revealed noradrenergic neurons that responded specifically to failed swim attempts and radial astrocytes whose calcium levels accumulated with increasing numbers of failed attempts. Using cell ablation and optogenetic or chemogenetic activation, we found that noradrenergic neurons progressively activated brainstem radial astrocytes, which then suppressed swimming. Thus, radial astrocytes perform a computation critical for behavior: they accumulate evidence that current actions are ineffective and consequently drive changes in behavioral states.

A neuronal population encodes information most efficiently when its activity is uncorrelated and high-dimensional, and most robustly when its activity is correlated and lower-dimensional. Here, we analyzed the correlation structure of natural image coding, in large visual cortical populations recorded from awake mice. Evoked population activity was high dimensional, with correlations obeying an unexpected power-law: the n-th principal component variance scaled as 1/n. This was not inherited from the 1/f spectrum of natural images, because it persisted after stimulus whitening. We proved mathematically that the variance spectrum must decay at least this fast if a population code is smooth, i.e. if small changes in input cannot dominate population activity. The theory also predicts larger power-law exponents for lower-dimensional stimulus ensembles, which we validated experimentally. These results suggest that coding smoothness represents a fundamental constraint governing correlations in neural population codes.

In the spindle midzone, microtubules from opposite half-spindles form bundles between segregating chromosomes. Microtubule bundles can either push or restrict chromosome movement during anaphase in different cellular contexts, but how these activities are achieved remains poorly understood. Here, we use high-resolution live-cell imaging to analyze individual microtubule bundles, growing filaments, and chromosome movement in dividing human cells. Within bundles, filament overlap length marked by the cross-linking protein PRC1 decreases during anaphase as chromosome segregation slows. Filament ends within microtubule bundles appear capped despite dynamic PRC1 turnover and submicrometer proximity to growing microtubules. Chromosome segregation distance and rate are increased in two human cell lines when microtubule bundle assembly is prevented via PRC1 knockdown. Upon expressing a mutant PRC1 with reduced microtubule affinity, bundles assemble but chromosome hypersegregation is still observed. We propose that microtubule overlap length reduction, typically linked to pushing forces generated within filament bundles, is needed to properly restrict spindle elongation and position chromosomes within daughter cells.