A leaky integrate-and-fire computational model based on the connectome of the entire adult Drosophila brain reveals insights into sensorimotor processing.

The forthcoming assembly of the adult Drosophila melanogaster central brain connectome, containing over 125,000 neurons and 50 million synaptic connections, provides a template for examining sensory processing throughout the brain. Here, we create a leaky integrate-and-fire computational model of the entire Drosophila brain, based on neural connectivity and neurotransmitter identity, to study circuit properties of feeding and grooming behaviors. We show that activation of sugar-sensing or water-sensing gustatory neurons in the computational model accurately predicts neurons that respond to tastes and are required for feeding initiation. Computational activation of neurons in the feeding region of the Drosophila brain predicts those that elicit motor neuron firing, a testable hypothesis that we validate by optogenetic activation and behavioral studies. Moreover, computational activation of different classes of gustatory neurons makes accurate predictions of how multiple taste modalities interact, providing circuit-level insight into aversive and appetitive taste processing. Our computational model predicts that the sugar and water pathways form a partially shared appetitive feeding initiation pathway, which our calcium imaging and behavioral experiments confirm. Additionally, we applied this model to mechanosensory circuits and found that computational activation of mechanosensory neurons predicts activation of a small set of neurons comprising the antennal grooming circuit that do not overlap with gustatory circuits, and accurately describes the circuit response upon activation of different mechanosensory subtypes. Our results demonstrate that modeling brain circuits purely from connectivity and predicted neurotransmitter identity generates experimentally testable hypotheses and can accurately describe complete sensorimotor transformations.

A connectome of the Drosophila central complex reveals network motifs suitable for flexible navigation and context-dependent action selection.

Flexible behaviors over long timescales are thought to engage recurrent neural networks in deep brain regions, which are experimentally challenging to study. In insects, recurrent circuit dynamics in a brain region called the central complex (CX) enable directed locomotion, sleep, and context- and experience-dependent spatial navigation. We describe the first complete electron microscopy-based connectome of the Drosophila CX, including all its neurons and circuits at synaptic resolution. We identified new CX neuron types, novel sensory and motor pathways, and network motifs that likely enable the CX to extract the fly's head direction, maintain it with attractor dynamics, and combine it with other sensorimotor information to perform vector-based navigational computations. We also identified numerous pathways that may facilitate the selection of CX-driven behavioral patterns by context and internal state. The CX connectome provides a comprehensive blueprint necessary for a detailed understanding of network dynamics underlying sleep, flexible navigation, and state-dependent action selection.

Building a functional connectome of the Drosophila central complex.

The central complex is a highly conserved insect brain region composed of morphologically stereotyped neurons that arborize in distinctively shaped substructures. The region is implicated in a wide range of behaviors and several modeling studies have explored its circuit computations. Most studies have relied on assumptions about connectivity between neurons based on their overlap in light microscopy images. Here, we present an extensive functional connectome of Drosophila melanogaster's central complex at cell-type resolution. Using simultaneous optogenetic stimulation, calcium imaging and pharmacology, we tested the connectivity between 70 presynaptic-to-postsynaptic cell-type pairs. We identified numerous inputs to the central complex, but only a small number of output channels. Additionally, the connectivity of this highly recurrent circuit appears to be sparser than anticipated from light microscopy images. Finally, the connectivity matrix highlights the potentially critical role of a class of bottleneck interneurons. All data are provided for interactive exploration on a website.

Neural signatures of dynamic stimulus selection in Drosophila.

Many animals orient using visual cues, but how a single cue is selected from among many is poorly understood. Here we show that Drosophila ring neurons-central brain neurons implicated in navigation-display visual stimulus selection. Using in vivo two-color two-photon imaging with genetically encoded calcium indicators, we demonstrate that individual ring neurons inherit simple-cell-like receptive fields from their upstream partners. Stimuli in the contralateral visual field suppressed responses to ipsilateral stimuli in both populations. Suppression strength depended on when and where the contralateral stimulus was presented, an effect stronger in ring neurons than in their upstream inputs. This history-dependent effect on the temporal structure of visual responses, which was well modeled by a simple biphasic filter, may determine how visual references are selected for the fly's internal compass. Our approach highlights how two-color calcium imaging can help identify and localize the origins of sensory transformations across synaptically connected neural populations.

Angular velocity integration in a fly heading circuit.

Many animals maintain an internal representation of their heading as they move through their surroundings. Such a compass representation was recently discovered in a neural population in the Drosophila melanogaster central complex, a brain region implicated in spatial navigation. Here, we use two-photon calcium imaging and electrophysiology in head-fixed walking flies to identify a different neural population that conjunctively encodes heading and angular velocity, and is excited selectively by turns in either the clockwise or counterclockwise direction. We show how these mirror-symmetric turn responses combine with the neurons' connectivity to the compass neurons to create an elegant mechanism for updating the fly's heading representation when the animal turns in darkness. This mechanism, which employs recurrent loops with an angular shift, bears a resemblance to those proposed in theoretical models for rodent head direction cells. Our results provide a striking example of structure matching function for a broadly relevant computation.

A neural command circuit for grooming movement control.

Animals perform many stereotyped movements, but how nervous systems are organized for controlling specific movements remains unclear. Here we use anatomical, optogenetic, behavioral, and physiological techniques to identify a circuit in Drosophila melanogaster that can elicit stereotyped leg movements that groom the antennae. Mechanosensory chordotonal neurons detect displacements of the antennae and excite three different classes of functionally connected interneurons, which include two classes of brain interneurons and different parallel descending neurons. This multilayered circuit is organized such that neurons within each layer are sufficient to specifically elicit antennal grooming. However, we find differences in the durations of antennal grooming elicited by neurons in the different layers, suggesting that the circuit is organized to both command antennal grooming and control its duration. As similar features underlie stimulus-induced movements in other animals, we infer the possibility of a common circuit organization for movement control that can be dissected in Drosophila.

Central neural circuitry mediating courtship song perception in male Drosophila.

Animals use acoustic signals across a variety of social behaviors, particularly courtship. In Drosophila, song is detected by antennal mechanosensory neurons and further processed by second-order aPN1/aLN(al) neurons. However, little is known about the central pathways mediating courtship hearing. In this study, we identified a male-specific pathway for courtship hearing via third-order ventrolateral protocerebrum Projection Neuron 1 (vPN1) neurons and fourth-order pC1 neurons. Genetic inactivation of vPN1 or pC1 disrupts song-induced male-chaining behavior. Calcium imaging reveals that vPN1 responds preferentially to pulse song with long inter-pulse intervals (IPIs), while pC1 responses to pulse song closely match the behavioral chaining responses at different IPIs. Moreover, genetic activation of either vPN1 or pC1 induced courtship chaining, mimicking the behavioral response to song. These results outline the aPN1-vPN1-pC1 pathway as a labeled line for the processing and transformation of courtship song in males.

An excitatory GABA loop operating in vivo.

While it has been proposed that the conventional inhibitory neurotransmitter GABA can be excitatory in the mammalian brain, much remains to be learned concerning the circumstances and the cellular mechanisms governing potential excitatory GABA action. Using a combination of optogenetics and two-photon calcium imaging in vivo, we find that activation of chloride-permeable GABAA receptors in parallel fibers (PFs) of the cerebellar molecular layer of adult mice causes parallel fiber excitation. Stimulation of PFs at submaximal stimulus intensities leads to GABA release from molecular layer interneurons (MLIs), thus creating a positive feedback loop that enhances excitation near the center of an activated PF bundle. Our results imply that elevated chloride concentration can occur in specific intracellular compartments of mature mammalian neurons and suggest an excitatory role for GABAA receptors in the cerebellar cortex of adult mice.

A multilevel multimodal circuit enhances action selection in Drosophila.

Natural events present multiple types of sensory cues, each detected by a specialized sensory modality. Combining information from several modalities is essential for the selection of appropriate actions. Key to understanding multimodal computations is determining the structural patterns of multimodal convergence and how these patterns contribute to behaviour. Modalities could converge early, late or at multiple levels in the sensory processing hierarchy. Here we show that combining mechanosensory and nociceptive cues synergistically enhances the selection of the fastest mode of escape locomotion in Drosophila larvae. In an electron microscopy volume that spans the entire insect nervous system, we reconstructed the multisensory circuit supporting the synergy, spanning multiple levels of the sensory processing hierarchy. The wiring diagram revealed a complex multilevel multimodal convergence architecture. Using behavioural and physiological studies, we identified functionally connected circuit nodes that trigger the fastest locomotor mode, and others that facilitate it, and we provide evidence that multiple levels of multimodal integration contribute to escape mode selection. We propose that the multilevel multimodal convergence architecture may be a general feature of multisensory circuits enabling complex input-output functions and selective tuning to ecologically relevant combinations of cues.

Estimating background-subtracted fluorescence transients in calcium imaging experiments: a quantitative approach.

Calcium imaging has become a routine technique in neuroscience for subcellular to network level investigations. The fast progresses in the development of new indicators and imaging techniques call for dedicated reliable analysis methods. In particular, efficient and quantitative background fluorescence subtraction routines would be beneficial to most of the calcium imaging research field. A background-subtracted fluorescence transients estimation method that does not require any independent background measurement is therefore developed. This method is based on a fluorescence model fitted to single-trial data using a classical nonlinear regression approach. The model includes an appropriate probabilistic description of the acquisition system's noise leading to accurate confidence intervals on all quantities of interest (background fluorescence, normalized background-subtracted fluorescence time course) when background fluorescence is homogeneous. An automatic procedure detecting background inhomogeneities inside the region of interest is also developed and is shown to be efficient on simulated data. The implementation and performances of the proposed method on experimental recordings from the mouse hypothalamus are presented in details. This method, which applies to both single-cell and bulk-stained tissues recordings, should help improving the statistical comparison of fluorescence calcium signals between experiments and studies.

Measuring the firing rate of high-resistance neurons with cell-attached recording.

Cell-attached recording is extensively used to study the firing rate of mammalian neurons, but potential limitations of the method have not been investigated in detail. Here we perform cell-attached recording of molecular layer interneurons in cerebellar slices from rats and mice, and we study how experimental conditions influence the measured firing rate. We find that this rate depends on time in cell-attached mode, on pipette potential, and on pipette ionic composition. In the first minute after sealing, action currents are variable in shape and size, presumably reflecting membrane instability. The firing rate remains approximately constant during the first 4 min after sealing and gradually increases afterward. Making the pipette potential more positive leads to an increase in the firing rate, with a steeper dependence on voltage if the pipette solution contains K(+) as the main cation than if it contains Na(+). Ca(2+) imaging experiments show that establishing a cell-attached recording can result in an increased somatic Ca(2+) concentration, reflecting an increased firing rate linked to an increase in the pipette-cell conductance. Pipette effects on cell firing are traced to a combination of passive electrical coupling, opening of voltage- and Ca(2+)-sensitive K(+) channels (BK channels) after action potentials, and random activation of voltage-insensitive, presumably mechanosensitive, cationic channels. We conclude that, unless experimental conditions are optimized, cell-attached recordings in small neurons may report erroneous firing rates.

Making neurophysiological data analysis reproducible: why and how?

Reproducible data analysis is an approach aiming at complementing classical printed scientific articles with everything required to independently reproduce the results they present. "Everything" covers here: the data, the computer codes and a precise description of how the code was applied to the data. A brief history of this approach is presented first, starting with what economists have been calling replication since the early eighties to end with what is now called reproducible research in computational data analysis oriented fields like statistics and signal processing. Since efficient tools are instrumental for a routine implementation of these approaches, a description of some of the available ones is presented next. A toy example demonstrates then the use of two open source software programs for reproducible data analysis: the "Sweave family" and the org-mode of emacs. The former is bound to R while the latter can be used with R, Matlab, Python and many more "generalist" data processing software. Both solutions can be used with Unix-like, Windows and Mac families of operating systems. It is argued that neuroscientists could communicate much more efficiently their results by adopting the reproducible research paradigm from their lab books all the way to their articles, thesis and books.

Somatic calcium level reports integrated spiking activity of cerebellar interneurons in vitro and in vivo.

We examined the relationship between somatic Ca²âº signals and spiking activity of cerebellar molecular layer interneurons (MLIs) in adult mice. Using two-photon microscopy in conjunction with cell-attached recordings in slices, we show that in tonically firing MLIs loaded with high-affinity Ca²âº probes, Ca²âº-dependent fluorescence transients are absent. Spike-triggered averages of fluorescence traces for MLIs spiking at low rates revealed that the fluorescence change associated with an action potential is small (1% of the basal fluorescence). To uncover the relationship between intracellular Ca²âº concentration ([Ca²âº](i)) and firing rates, spikes were transiently silenced with puffs of the GABA(A) receptor agonist muscimol. [Ca²âº](i) relaxed toward basal levels following a single exponential whose amplitude correlated to the preceding spike frequency. The relaxation time constant was slow (2.5 s) and independent of the probe concentration. Data from parvalbumin (PV)-/- animals indicate that PV controls the amplitude and decay time of spike-triggered averages as well as the time course of [Ca²âº](i) relaxations following spike silencing. The [Ca²âº](i) signals were sensitive to the L-type Ca²âº channel blocker nimodipine and insensitive to ryanodine. In anesthetized mice, as in slices, fluorescence traces from most MLIs did not show spontaneous transients. They nonetheless responded to muscimol iontophoresis with relaxations similar to those obtained in vitro, suggesting a state of tonic firing with estimated spiking rates ranging from 2 to 30 Hz. Altogether, the [Ca²âº](i) signal appears to reflect the integral of the spiking activity in MLIs. We propose that the muscimol silencing strategy can be extended to other tonically spiking neurons with similar [Ca²âº](i) homeostasis.

Activation of metabotropic glutamate receptors induces periodic burst firing and concomitant cytosolic Ca2+ oscillations in cerebellar interneurons.

Little is known about the generation of slow rhythms in brain neuronal circuits. Nevertheless, a few studies, both from reconstituted systems and from hippocampal slices, indicate that activation of metabotropic glutamate receptors (mGluRs) could generate such rhythms. Here we show in rat cerebellar slices that after either release of glutamate by repetitive stimulation, or direct stimulation of type 1 mGluRs, molecular layer interneurons exhibit repetitive slow Ca(2+) transients. By combining cell-attached patch-clamp recording with Ca(2+) imaging, we show that the regular Ca(2+) transients (mean frequency, 35 mHz induced by 2 microm quisqualate in the presence of ionotropic glutamate receptor blockers) are locked with bursts of action potentials. Nevertheless, the Ca(2+) transients are not blocked by tetrodotoxin, indicating that firing is not necessary to entrain oscillations. The first Ca(2+) transient within a train is different in several ways from subsequent transients. It is broader than the subsequent transients, displays a different phase relationship to associated spike bursts, and exhibits a distinct sensitivity to ionic and pharmacological manipulations. Whereas the first transient appears to involve entry of Ca(2+) ions through transient receptor potential channel-like channels and secondarily activated L-type Ca(2+) channels, subsequent transients rely mostly on an exchange of Ca(2+) ions between the cytosol and D-myo-inositol-1,4,5-triphosphate-sensitive intracellular Ca(2+) stores. The slow, highly regular oscillations observed in the present work are likely to drive pauses in postsynaptic Purkinje cells, and could play a role in coordinating slow oscillations involving the cerebello-olivar circuit loop.

Repetitive firing of rat cerebellar parallel fibres after a single stimulation.

The excitatory postsynaptic currents (EPSCs) evoked in Purkinje cells (PCs) by stimulating parallel fibres (PFs) usually show a single peak, but EPSCs with multiple peaks (polyphasic EPSCs) can be observed in slices from animals older than 15 days. The EPSCs remain polyphasic when the postsynaptic current is reduced (either by reducing the intensity of the PF stimulation or by adding AMPA receptor antagonists) and when the PC membrane potential is made positive. Thus the late peaks are not due to postsynaptic active currents generated in the imperfectly clamped PC, and must arise from repetitive action potentials in the PF. Extracellular recordings from granule cell (GC) somata showed that a single PF stimulation can elicit a doublet or a train of action potentials. Both the late action potentials recorded in the GCs and the late peaks of the polyphasic EPSCs recorded in the PCs were reduced or abolished by paired-pulse stimulation of the PF or by bath application of the GABA(A) agonist muscimol. The late action potentials in the GCs were also suppressed by local application of muscimol around the cell body. We propose that after a single stimulation of a PF, the antidromic invasion of the ascending axon and the granule cell can trigger a doublet or a burst of action potentials which back-propagate into the PF (except for the first, which finds the PF still in its refractory period). The repetitive activation of the PF by a single stimulation could play a role in the induction of long-term depression.