Bisected graph matching improves automated pairing of bilaterally homologous neurons from connectomes.

Graph matching algorithms attempt to find the best correspondence between the nodes of two networks. These techniques have been used to match individual neurons in nanoscale connectomes-in particular, to find pairings of neurons across hemispheres. However, since graph matching techniques deal with two isolated networks, they have only utilized the ipsilateral (same hemisphere) subgraphs when performing the matching. Here, we present a modification to a state-of-the-art graph matching algorithm that allows it to solve what we call the bisected graph matching problem. This modification allows us to leverage the connections between the brain hemispheres when predicting neuron pairs. Via simulations and experiments on real connectome datasets, we show that this approach improves matching accuracy when sufficient edge correlation is present between the contralateral (between hemisphere) subgraphs. We also show how matching accuracy can be further improved by combining our approach with previously proposed extensions to graph matching, which utilize edge types and previously known neuron pairings. We expect that our proposed method will improve future endeavors to accurately match neurons across hemispheres in connectomes, and be useful in other applications where the bisected graph matching problem arises.

Generative network modeling reveals quantitative definitions of bilateral symmetry exhibited by a whole insect brain connectome.

Comparing connectomes can help explain how neural connectivity is related to genetics, disease, development, learning, and behavior. However, making statistical inferences about the significance and nature of differences between two networks is an open problem, and such analysis has not been extensively applied to nanoscale connectomes. Here, we investigate this problem via a case study on the bilateral symmetry of a larval Drosophila brain connectome. We translate notions of 'bilateral symmetry' to generative models of the network structure of the left and right hemispheres, allowing us to test and refine our understanding of symmetry. We find significant differences in connection probabilities both across the entire left and right networks and between specific cell types. By rescaling connection probabilities or removing certain edges based on weight, we also present adjusted definitions of bilateral symmetry exhibited by this connectome. This work shows how statistical inferences from networks can inform the study of connectomes, facilitating future comparisons of neural structures.

The connectome of an insect brain.

Brains contain networks of interconnected neurons and so knowing the network architecture is essential for understanding brain function. We therefore mapped the synaptic-resolution connectome of an entire insect brain (Drosophila larva) with rich behavior, including learning, value computation, and action selection, comprising 3016 neurons and 548,000 synapses. We characterized neuron types, hubs, feedforward and feedback pathways, as well as cross-hemisphere and brain-nerve cord interactions. We found pervasive multisensory and interhemispheric integration, highly recurrent architecture, abundant feedback from descending neurons, and multiple novel circuit motifs. The brain's most recurrent circuits comprised the input and output neurons of the learning center. Some structural features, including multilayer shortcuts and nested recurrent loops, resembled state-of-the-art deep learning architectures. The identified brain architecture provides a basis for future experimental and theoretical studies of neural circuits.

High-throughput automated methods for classical and operant conditioning of Drosophila larvae.

Learning which stimuli (classical conditioning) or which actions (operant conditioning) predict rewards or punishments can improve chances of survival. However, the circuit mechanisms that underlie distinct types of associative learning are still not fully understood. Automated, high-throughput paradigms for studying different types of associative learning, combined with manipulation of specific neurons in freely behaving animals, can help advance this field. The Drosophila melanogaster larva is a tractable model system for studying the circuit basis of behaviour, but many forms of associative learning have not yet been demonstrated in this animal. Here, we developed a high-throughput (i.e. multi-larva) training system that combines real-time behaviour detection of freely moving larvae with targeted opto- and thermogenetic stimulation of tracked animals. Both stimuli are controlled in either open- or closed-loop, and delivered with high temporal and spatial precision. Using this tracker, we show for the first time that Drosophila larvae can perform classical conditioning with no overlap between sensory stimuli (i.e. trace conditioning). We also demonstrate that larvae are capable of operant conditioning by inducing a bend direction preference through optogenetic activation of reward-encoding serotonergic neurons. Our results extend the known associative learning capacities of Drosophila larvae. Our automated training rig will facilitate the study of many different forms of associative learning and the identification of the neural circuits that underpin them.

Electrophysiological validation of monosynaptic connectivity between premotor interneurons and the aCC motoneuron in the Drosophila larval CNS.

The Drosophila connectome project aims to map the synaptic connectivity of entire larval and adult fly neural networks, which is essential for understanding nervous system development and function. So far, the project has produced an impressive amount of electron microscopy data that has facilitated reconstructions of specific synapses, including many in the larval locomotor circuit. While this breakthrough represents a technical tour-de-force, the data remain under-utilised, partly due to a lack of functional validation of reconstructions. Attempts to validate connectivity posited by the connectome project, have mostly relied on behavioural assays and/or GRASP or GCaMP imaging. While these techniques are useful, they have limited spatial or temporal resolution. Electrophysiological assays of synaptic connectivity overcome these limitations. Here, we combine patch clamp recordings with optogenetic stimulation in male and female larvae, to test synaptic connectivity proposed by connectome reconstructions. Specifically, we use multiple driver lines to confirm that several connections between premotor interneurons and the anterior corner cell (aCC) motoneuron are, as the connectome project suggests, monosynaptic. In contrast, our results also show that conclusions based on GRASP imaging may provide false positive results regarding connectivity between cells. We also present a novel imaging tool, based on the same technology as our electrophysiology, as a favourable alternative to GRASP. Finally, of eight Gal4 lines tested, five are reliably expressed in the premotors they are targeted to. Thus, our work highlights the need to confirm functional synaptic connectivity, driver line specificity, and use of appropriate genetic tools to support connectome projects.SIGNIFICANCE STATEMENTThe Drosophila connectome project aims to provide a complete description of connectivity between neurons in an organism that presents experimental advantages over other models. It has reconstructed over 80 percent of the fly larva's synaptic connections by manual identification of anatomical landmarks present in serial section transmission electron microscopy (ssTEM) volumes of the larval CNS. We use a highly reliable electrophysiological approach to verify these connections, so provide useful insight into the accuracy of work based on ssTEM. We also present a novel imaging tool for validating excitatory monosynaptic connections between cells, and show that several genetic driver lines designed to target neurons of the larval connectome exhibit non-specific and/or unreliable expression.

Circuits for integrating learned and innate valences in the insect brain.

Animal behavior is shaped both by evolution and by individual experience. Parallel brain pathways encode innate and learned valences of cues, but the way in which they are integrated during action-selection is not well understood. We used electron microscopy to comprehensively map with synaptic resolution all neurons downstream of all mushroom body (MB) output neurons (encoding learned valences) and characterized their patterns of interaction with lateral horn (LH) neurons (encoding innate valences) in Drosophila larva. The connectome revealed multiple convergence neuron types that receive convergent MB and LH inputs. A subset of these receives excitatory input from positive-valence MB and LH pathways and inhibitory input from negative-valence MB pathways. We confirmed functional connectivity from LH and MB pathways and behavioral roles of two of these neurons. These neurons encode integrated odor value and bidirectionally regulate turning. Based on this, we speculate that learning could potentially skew the balance of excitation and inhibition onto these neurons and thereby modulate turning. Together, our study provides insights into the circuits that integrate learned and innate valences to modify behavior.

Recurrent architecture for adaptive regulation of learning in the insect brain.

Dopaminergic neurons (DANs) drive learning across the animal kingdom, but the upstream circuits that regulate their activity and thereby learning remain poorly understood. We provide a synaptic-resolution connectome of the circuitry upstream of all DANs in a learning center, the mushroom body of Drosophila larva. We discover afferent sensory pathways and a large population of neurons that provide feedback from mushroom body output neurons and link distinct memory systems (aversive and appetitive). We combine this with functional studies of DANs and their presynaptic partners and with comprehensive circuit modeling. We find that DANs compare convergent feedback from aversive and appetitive systems, which enables the computation of integrated predictions that may improve future learning. Computational modeling reveals that the discovered feedback motifs increase model flexibility and performance on learning tasks. Our study provides the most detailed view to date of biological circuit motifs that support associative learning.

Neural Substrates of Drosophila Larval Anemotaxis.

Animals use sensory information to move toward more favorable conditions. Drosophila larvae can move up or down gradients of odors (chemotax), light (phototax), and temperature (thermotax) by modulating the probability, direction, and size of turns based on sensory input. Whether larvae can anemotax in gradients of mechanosensory cues is unknown. Further, although many of the sensory neurons that mediate taxis have been described, the central circuits are not well understood. Here, we used high-throughput, quantitative behavioral assays to demonstrate Drosophila larvae anemotax in gradients of wind speeds and to characterize the behavioral strategies involved. We found that larvae modulate the probability, direction, and size of turns to move away from higher wind speeds. This suggests that similar central decision-making mechanisms underlie taxis in somatosensory and other sensory modalities. By silencing the activity of single or very few neuron types in a behavioral screen, we found two sensory (chordotonal and multidendritic class III) and six nerve cord neuron types involved in anemotaxis. We reconstructed the identified neurons in an electron microscopy volume that spans the entire larval nervous system and found they received direct input from the mechanosensory neurons or from each other. In this way, we identified local interneurons and first- and second-order subesophageal zone (SEZ) and brain projection neurons. Finally, silencing a dopaminergic brain neuron type impairs anemotaxis. These findings suggest that anemotaxis involves both nerve cord and brain circuits. The candidate neurons and circuitry identified in our study provide a basis for future detailed mechanistic understanding of the circuit principles of anemotaxis.

Microtubule-microtubule sliding by kinesin-1 is essential for normal cytoplasmic streaming in Drosophila oocytes.

Cytoplasmic streaming in Drosophila oocytes is a microtubule-based bulk cytoplasmic movement. Streaming efficiently circulates and localizes mRNAs and proteins deposited by the nurse cells across the oocyte. This movement is driven by kinesin-1, a major microtubule motor. Recently, we have shown that kinesin-1 heavy chain (KHC) can transport one microtubule on another microtubule, thus driving microtubule-microtubule sliding in multiple cell types. To study the role of microtubule sliding in oocyte cytoplasmic streaming, we used a Khc mutant that is deficient in microtubule sliding but able to transport a majority of cargoes. We demonstrated that streaming is reduced by genomic replacement of wild-type Khc with this sliding-deficient mutant. Streaming can be fully rescued by wild-type KHC and partially rescued by a chimeric motor that cannot move organelles but is active in microtubule sliding. Consistent with these data, we identified two populations of microtubules in fast-streaming oocytes: a network of stable microtubules anchored to the actin cortex and free cytoplasmic microtubules that moved in the ooplasm. We further demonstrated that the reduced streaming in sliding-deficient oocytes resulted in posterior determination defects. Together, we propose that kinesin-1 slides free cytoplasmic microtubules against cortically immobilized microtubules, generating forces that contribute to cytoplasmic streaming and are essential for the refinement of posterior determinants.

Role of kinesin-1-based microtubule sliding in Drosophila nervous system development.

The plus-end microtubule (MT) motor kinesin-1 is essential for normal development, with key roles in the nervous system. Kinesin-1 drives axonal transport of membrane cargoes to fulfill the metabolic needs of neurons and maintain synapses. We have previously demonstrated that kinesin-1, in addition to its well-established role in organelle transport, can drive MT-MT sliding by transporting "cargo" MTs along "track" MTs, resulting in dramatic cell shape changes. The mechanism and physiological relevance of this MT sliding are unclear. In addition to its motor domain, kinesin-1 contains a second MT-binding site, located at the C terminus of the heavy chain. Here, we mutated this C-terminal MT-binding site such that the ability of kinesin-1 to slide MTs is significantly compromised, whereas cargo transport is unaffected. We introduced this mutation into the genomic locus of kinesin-1 heavy chain (KHC), generating the Khc(mutA) allele. Khc(mutA) neurons displayed significant MT sliding defects while maintaining normal transport of many cargoes. Using this mutant, we demonstrated that MT sliding is required for axon and dendrite outgrowth in vivo. Consistent with these results, Khc(mutA) flies displayed severe locomotion and viability defects. To test the role of MT sliding further, we engineered a chimeric motor that actively slides MTs but cannot transport organelles. Activation of MT sliding in Khc(mutA) neurons using this chimeric motor rescued axon outgrowth in cultured neurons and in vivo, firmly establishing the role of sliding in axon outgrowth. These results demonstrate that MT sliding by kinesin-1 is an essential biological phenomenon required for neuronal morphogenesis and normal nervous system development.

Engineered kinesin motor proteins amenable to small-molecule inhibition.

The human genome encodes 45 kinesin motor proteins that drive cell division, cell motility, intracellular trafficking and ciliary function. Determining the cellular function of each kinesin would benefit from specific small-molecule inhibitors. However, screens have yielded only a few specific inhibitors. Here we present a novel chemical-genetic approach to engineer kinesin motors that can carry out the function of the wild-type motor yet can also be efficiently inhibited by small, cell-permeable molecules. Using kinesin-1 as a prototype, we develop two independent strategies to generate inhibitable motors, and characterize the resulting inhibition in single-molecule assays and in cells. We further apply these two strategies to create analogously inhibitable kinesin-3 motors. These inhibitable motors will be of great utility to study the functions of specific kinesins in a dynamic manner in cells and animals. Furthermore, these strategies can be used to generate inhibitable versions of any motor protein of interest.

A Genome-wide RNAi Screen for Microtubule Bundle Formation and Lysosome Motility Regulation in Drosophila S2 Cells.

Long-distance intracellular transport of organelles, mRNA, and proteins ("cargo") occurs along the microtubule cytoskeleton by the action of kinesin and dynein motor proteins, but the vast network of factors involved in regulating intracellular cargo transport are still unknown. We capitalize on the Drosophila melanogaster S2 model cell system to monitor lysosome transport along microtubule bundles, which require enzymatically active kinesin-1 motor protein for their formation. We use an automated tracking program and a naive Bayesian classifier for the multivariate motility data to analyze 15,683 gene phenotypes and find 98 proteins involved in regulating lysosome motility along microtubules and 48 involved in the formation of microtubule filled processes in S2 cells. We identify innate immunity genes, ion channels, and signaling proteins having a role in lysosome motility regulation and find an unexpected relationship between the dynein motor, Rab7a, and lysosome motility regulation.

Interplay between kinesin-1 and cortical dynein during axonal outgrowth and microtubule organization in Drosophila neurons.

In this study, we investigated how microtubule motors organize microtubules in Drosophila neurons. We showed that, during the initial stages of axon outgrowth, microtubules display mixed polarity and minus-end-out microtubules push the tip of the axon, consistent with kinesin-1 driving outgrowth by sliding antiparallel microtubules. At later stages, the microtubule orientation in the axon switches from mixed to uniform polarity with plus-end-out. Dynein knockdown prevents this rearrangement and results in microtubules of mixed orientation in axons and accumulation of microtubule minus-ends at axon tips. Microtubule reorganization requires recruitment of dynein to the actin cortex, as actin depolymerization phenocopies dynein depletion, and direct recruitment of dynein to the membrane bypasses the actin requirement. Our results show that cortical dynein slides 'minus-end-out' microtubules from the axon, generating uniform microtubule arrays. We speculate that differences in microtubule orientation between axons and dendrites could be dictated by differential activity of cortical dynein.

Pavarotti/MKLP1 regulates microtubule sliding and neurite outgrowth in Drosophila neurons.

Recently, we demonstrated that kinesin-1 can slide microtubules against each other, providing the mechanical force required for initial neurite extension in Drosophila neurons. This sliding is only observed in young neurons actively forming neurites and is dramatically downregulated in older neurons. The downregulation is not caused by the global shutdown of kinesin-1, as the ability of kinesin-1 to transport membrane organelles is not diminished in mature neurons, suggesting that microtubule sliding is regulated by a dedicated mechanism. Here, we have identified the "mitotic" kinesin-6 Pavarotti (Pav-KLP) as an inhibitor of kinesin-1-driven microtubule sliding. Depletion of Pav-KLP in neurons strongly stimulated the sliding of long microtubules and neurite outgrowth, while its ectopic overexpression in the cytoplasm blocked both of these processes. Furthermore, postmitotic depletion of Pav-KLP in Drosophila neurons in vivo reduced embryonic and larval viability, with only a few animals surviving to the third instar larval stage. A detailed examination of motor neurons in the surviving larvae revealed the overextension of axons and mistargeting of neuromuscular junctions, resulting in uncoordinated locomotion. Taken together, our results identify a new role for Pav-KLP as a negative regulator of kinesin-1-driven neurite formation. These data suggest an important parallel between long microtubule-microtubule sliding in anaphase B and sliding of interphase microtubules during neurite formation.

Breaking up isn't easy: myosin V and its cargoes need Dma1 ubiquitin ligase's help.

In this issue of Developmental Cell, Yau et al. (2014) report that degradation of cargo adapters releases yeast vacuoles and peroxisomes from myosin V (Myo2) and terminates organelle transport from the mother cell to the bud.

Zwint-1 is a novel Aurora B substrate required for the assembly of a dynein-binding platform on kinetochores.

Aurora B (AurB) is a mitotic kinase responsible for multiple aspects of mitotic progression, including assembly of the outer kinetochore. Cytoplasmic dynein is an abundant kinetochore protein whose recruitment to kinetochores requires phosphorylation. To assess whether AurB regulates recruitment of dynein to kinetochores, we inhibited AurB using ZM447439 or a kinase-dead AurB construct. Inhibition of AurB reduced accumulation of dynein at kinetochores substantially; however, this reflected a loss of dynein-associated proteins rather than a defect in dynein phosphorylation. We determined that AurB inhibition affected recruitment of the ROD, ZW10, zwilch (RZZ) complex to kinetochores but not zwint-1 or more-proximal kinetochore proteins. AurB phosphorylated zwint-1 but not ZW10 in vitro, and three novel phosphorylation sites were identified by tandem mass spectrometry analysis. Expression of a triple-Ala zwint-1 mutant blocked kinetochore assembly of RZZ-dependent proteins and induced defects in chromosome movement during prometaphase. Expression of a triple-Glu zwint-1 mutant rendered cells resistant to AurB inhibition during prometaphase. However, cells expressing the triple-Glu mutant failed to satisfy the spindle assembly checkpoint (SAC) at metaphase because poleward streaming of dynein/dynactin/RZZ was inhibited. These studies identify zwint-1 as a novel AurB substrate required for kinetochore assembly and for proper SAC silencing at metaphase.

Polo-like kinase1 is required for recruitment of dynein to kinetochores during mitosis.

Kinetochore dynein has been implicated in microtubule capture, correcting inappropriate microtubule attachments, chromosome movement, and checkpoint silencing. It remains unclear how dynein coordinates this diverse set of functions. Phosphorylation is responsible for some dynein heterogeneity (Whyte, J., Bader, J. R., Tauhata, S. B., Raycroft, M., Hornick, J., Pfister, K. K., Lane, W. S., Chan, G. K., Hinchcliffe, E. H., Vaughan, P. S., and Vaughan, K. T. (2008) J. Cell Biol. 183, 819-834), and phosphorylated and dephosphorylated forms of dynein coexist at prometaphase kinetochores. In this study, we measured the impact of inhibiting polo-like kinase 1 (Plk1) on both dynein populations. Phosphorylated dynein was ablated at kinetochores after inhibiting Plk1 with a small molecule inhibitor (5-Cyano-7-nitro-2-(benzothiazolo-N-oxide)-carboxamide) or chemical genetic approaches. The total complement of kinetochore dynein was also reduced but not eliminated, reflecting the presence of some dephosphorylated dynein after Plk1 inhibition. Although Plk1 inhibition had a profound effect on dynein, kinetochore populations of dynactin, spindly, and zw10 were not reduced. Plk1-independent dynein was reduced after p150(Glued) depletion, consistent with the binding of dephosphorylated dynein to dynactin. Plk1 phosphorylated dynein intermediate chains at Thr-89 in vitro and generated the phospho-Thr-89 phospho-epitope on recombinant dynein intermediate chains. Finally, inhibition of Plk1 induced defects in microtubule capture and persistent microtubule attachment, suggesting a role for phosphorylated dynein in these functions during prometaphase. These findings suggest that Plk1 is a dynein kinase required for recruitment of phosphorylated dynein to kinetochores.