Nonlinear fusion is optimal for a wide class of multisensory tasks.

Animals continuously detect information via multiple sensory channels, like vision and hearing, and integrate these signals to realise faster and more accurate decisions; a fundamental neural computation known as multisensory integration. A widespread view of this process is that multimodal neurons linearly fuse information across sensory channels. However, does linear fusion generalise beyond the classical tasks used to explore multisensory integration? Here, we develop novel multisensory tasks, which focus on the underlying statistical relationships between channels, and deploy models at three levels of abstraction: from probabilistic ideal observers to artificial and spiking neural networks. Using these models, we demonstrate that when the information provided by different channels is not independent, linear fusion performs sub-optimally and even fails in extreme cases. This leads us to propose a simple nonlinear algorithm for multisensory integration which is compatible with our current knowledge of multimodal circuits, excels in naturalistic settings and is optimal for a wide class of multisensory tasks. Thus, our work emphasises the role of nonlinear fusion in multisensory integration, and provides testable hypotheses for the field to explore at multiple levels: from single neurons to behaviour.

Magnetic actuation of otoliths allows behavioral and brain-wide neuronal exploration of vestibulo-motor processing in larval zebrafish.

The vestibular system in the inner ear plays a central role in sensorimotor control by informing the brain about the orientation and acceleration of the head. However, most experiments in neurophysiology are performed using head-fixed configurations, depriving animals of vestibular inputs. To overcome this limitation, we decorated the utricular otolith of the vestibular system in larval zebrafish with paramagnetic nanoparticles. This procedure effectively endowed the animal with magneto-sensitive capacities: applied magnetic field gradients induced forces on the otoliths, resulting in robust behavioral responses comparable to those evoked by rotating the animal by up to 25°. We recorded the whole-brain neuronal response to this fictive motion stimulation using light-sheet functional imaging. Experiments performed in unilaterally injected fish revealed the activation of a commissural inhibition between the brain hemispheres. This magnetic-based stimulation technique for larval zebrafish opens new perspectives to functionally dissect the neural circuits underlying vestibular processing and to develop multisensory virtual environments, including vestibular feedback.

Neural assemblies uncovered by generative modeling explain whole-brain activity statistics and reflect structural connectivity.

Patterns of endogenous activity in the brain reflect a stochastic exploration of the neuronal state space that is constrained by the underlying assembly organization of neurons. Yet, it remains to be shown that this interplay between neurons and their assembly dynamics indeed suffices to generate whole-brain data statistics. Here, we recorded the activity from ∼40,000 neurons simultaneously in zebrafish larvae, and show that a data-driven generative model of neuron-assembly interactions can accurately reproduce the mean activity and pairwise correlation statistics of their spontaneous activity. This model, the compositional Restricted Boltzmann Machine (cRBM), unveils ∼200 neural assemblies, which compose neurophysiological circuits and whose various combinations form successive brain states. We then performed in silico perturbation experiments to determine the interregional functional connectivity, which is conserved across individual animals and correlates well with structural connectivity. Our results showcase how cRBMs can capture the coarse-grained organization of the zebrafish brain. Notably, this generative model can readily be deployed to parse neural data obtained by other large-scale recording techniques.

Thermal modulation of Zebrafish exploratory statistics reveals constraints on individual behavioral variability.

BACKGROUND: Variability is a hallmark of animal behavior. It contributes to survival by endowing individuals and populations with the capacity to adapt to ever-changing environmental conditions. Intra-individual variability is thought to reflect both endogenous and exogenous modulations of the neural dynamics of the central nervous system. However, how variability is internally regulated and modulated by external cues remains elusive. Here, we address this question by analyzing the statistics of spontaneous exploration of freely swimming zebrafish larvae and by probing how these locomotor patterns are impacted when changing the water temperatures within an ethologically relevant range. RESULTS: We show that, for this simple animal model, five short-term kinematic parameters - interbout interval, turn amplitude, travelled distance, turn probability, and orientational flipping rate - together control the long-term exploratory dynamics. We establish that the bath temperature consistently impacts the means of these parameters, but leave their pairwise covariance unchanged. These results indicate that the temperature merely controls the sampling statistics within a well-defined kinematic space delineated by this robust statistical structure. At a given temperature, individual animals explore the behavioral space over a timescale of tens of minutes, suggestive of a slow internal state modulation that could be externally biased through the bath temperature. By combining these various observations into a minimal stochastic model of navigation, we show that this thermal modulation of locomotor kinematics results in a thermophobic behavior, complementing direct gradient-sensing mechanisms. CONCLUSIONS: This study establishes the existence of a well-defined locomotor space accessible to zebrafish larvae during spontaneous exploration, and quantifies self-generated modulation of locomotor patterns. Intra-individual variability reflects a slow diffusive-like probing of this space by the animal. The bath temperature in turn restricts the sampling statistics to sub-regions, endowing the animal with basic thermophobicity. This study suggests that in zebrafish, as well as in other ectothermic animals, ambient temperature could be used to efficiently manipulate internal states in a simple and ethological way.

From behavior to circuit modeling of light-seeking navigation in zebrafish larvae.

Bridging brain-scale circuit dynamics and organism-scale behavior is a central challenge in neuroscience. It requires the concurrent development of minimal behavioral and neural circuit models that can quantitatively capture basic sensorimotor operations. Here, we focus on light-seeking navigation in zebrafish larvae. Using a virtual reality assay, we first characterize how motor and visual stimulation sequences govern the selection of discrete swim-bout events that subserve the fish navigation in the presence of a distant light source. These mechanisms are combined into a comprehensive Markov-chain model of navigation that quantitatively predicts the stationary distribution of the fish's body orientation under any given illumination profile. We then map this behavioral description onto a neuronal model of the ARTR, a small neural circuit involved in the orientation-selection of swim bouts. We demonstrate that this visually-biased decision-making circuit can capture the statistics of both spontaneous and contrast-driven navigation.

Whole-Brain Calcium Imaging during Physiological Vestibular Stimulation in Larval Zebrafish.

The vestibular apparatus provides animals with postural and movement-related information that is essential to adequately execute numerous sensorimotor tasks. In order to activate this sensory system in a physiological manner, one needs to macroscopically rotate or translate the animal's head, which in turn renders simultaneous neural recordings highly challenging. Here we report on a novel miniaturized, light-sheet microscope that can be dynamically co-rotated with a head-restrained zebrafish larva, enabling controlled vestibular stimulation. The mechanical rigidity of the microscope allows one to perform whole-brain functional imaging with state-of-the-art resolution and signal-to-noise ratio while imposing up to 25° in angular position and 6,000°/s2 in rotational acceleration. We illustrate the potential of this novel setup by producing the first whole-brain response maps to sinusoidal and stepwise vestibular stimulation. The responsive population spans multiple brain areas and displays bilateral symmetry, and its organization is highly stereotypic across individuals. Using Fourier and regression analysis, we identified three major functional clusters that exhibit well-defined phasic and tonic response patterns to vestibular stimulation. Our rotatable light-sheet microscope provides a unique tool for systematically studying vestibular processing in the vertebrate brain and extends the potential of virtual-reality systems to explore complex multisensory and motor integration during simulated 3D navigation.

Sensorimotor computation underlying phototaxis in zebrafish.

Animals continuously gather sensory cues to move towards favourable environments. Efficient goal-directed navigation requires sensory perception and motor commands to be intertwined in a feedback loop, yet the neural substrate underlying this sensorimotor task in the vertebrate brain remains elusive. Here, we combine virtual-reality behavioural assays, volumetric calcium imaging, optogenetic stimulation and circuit modelling to reveal the neural mechanisms through which a zebrafish performs phototaxis, i.e. actively orients towards a light source. Key to this process is a self-oscillating hindbrain population (HBO) that acts as a pacemaker for ocular saccades and controls the orientation of successive swim-bouts. It further integrates visual stimuli in a state-dependent manner, i.e. its response to visual inputs varies with the motor context, a mechanism that manifests itself in the phase-locked entrainment of the HBO by periodic stimuli. A rate model is developed that reproduces our observations and demonstrates how this sensorimotor processing eventually biases the animal trajectory towards bright regions.Active locomotion requires closed-loop sensorimotor co ordination between perception and action. Here the authors show using behavioural, imaging and modelling approaches that gaze orientation during phototaxis behaviour in larval zebrafish is related to oscillatory dynamics of a neuronal population in the hindbrain.

Rheotaxis of Larval Zebrafish: Behavioral Study of a Multi-Sensory Process.

Awake animals unceasingly perceive sensory inputs with great variability of nature and intensity, and understanding how the nervous system manages this continuous flow of diverse information to get a coherent representation of the environment is arguably a central question in systems neuroscience. Rheotaxis, the ability shared by most aquatic species to orient toward a current and swim to hold position, is an innate and robust multi-sensory behavior that is known to involve the lateral line and visual systems. To facilitate the neuroethological study of rheotaxic behavior in larval zebrafish we developed an assay for freely swimming larvae that allows for high experimental throughtput, large statistic and a fine description of the behavior. We show that there exist a clear transition from exploration to counterflow swim, and by changing the sensory modalities accessible to the fishes (visual only, lateral line only or both) and comparing the swim patterns at different ages we were able to detect and characterize two different mechanisms for position holding, one mediated by the lateral line and one mediated by the visual system. We also found that when both sensory modalities are accessible the visual system overshadows the lateral line, suggesting that at the larval stage the sensory inputs are not merged to finely tune the behavior but that redundant information pathways may be used as functional fallbacks.

Transduction channels' gating can control friction on vibrating hair-cell bundles in the ear.

Hearing starts when sound-evoked mechanical vibrations of the hair-cell bundle activate mechanosensitive ion channels, giving birth to an electrical signal. As for any mechanical system, friction impedes movements of the hair bundle and thus constrains the sensitivity and frequency selectivity of auditory transduction. Friction is generally thought to result mainly from viscous drag by the surrounding fluid. We demonstrate here that the opening and closing of the transduction channels produce internal frictional forces that can dominate viscous drag on the micrometer-sized hair bundle. We characterized friction by analyzing hysteresis in the force-displacement relation of single hair-cell bundles in response to periodic triangular stimuli. For bundle velocities high enough to outrun adaptation, we found that frictional forces were maximal within the narrow region of deflections that elicited significant channel gating, plummeted upon application of a channel blocker, and displayed a sublinear growth for increasing bundle velocity. At low velocity, the slope of the relation between the frictional force and velocity was nearly fivefold larger than the hydrodynamic friction coefficient that was measured when the transduction machinery was decoupled from bundle motion by severing tip links. A theoretical analysis reveals that channel friction arises from coupling the dynamics of the conformational change associated with channel gating to tip-link tension. Varying channel properties affects friction, with faster channels producing smaller friction. We propose that this intrinsic source of friction may contribute to the process that sets the hair cell's characteristic frequency of responsiveness.

Kinesin-8 is a low-force motor protein with a weakly bound slip state.

During the cell cycle, kinesin-8s control the length of microtubules by interacting with their plus ends. To reach these ends, the motors have to be able to take many steps without dissociating. However, the underlying mechanism for this high processivity and how stepping is affected by force are unclear. Here, we tracked the motion of yeast (Kip3) and human (Kif18A) kinesin-8s with high precision under varying loads using optical tweezers. Surprisingly, both kinesin-8 motors were much weaker compared with other kinesins. Furthermore, we discovered a force-induced stick-slip motion: the motor frequently slipped, recovered from this state, and then resumed normal stepping motility without detaching from the microtubule. The low forces are consistent with kinesin-8s being regulators of microtubule dynamics rather than cargo transporters. The weakly bound slip state, reminiscent of a molecular safety leash, may be an adaptation for high processivity.

The highly processive kinesin-8, Kip3, switches microtubule protofilaments with a bias toward the left.

Kinesin-1 motor proteins walk parallel to the protofilament axes of microtubules as they step from one tubulin dimer to the next. Is protofilament tracking an inherent property of processive kinesin motors, like kinesin-1, and what are the structural determinants underlying protofilament tracking? To address these questions, we investigated the tracking properties of the processive kinesin-8, Kip3. Using in vitro gliding motility assays, we found that Kip3 rotates microtubules counterclockwise around their longitudinal axes with periodicities of ∼1 µm. These rotations indicate that the motors switch protofilaments with a bias toward the left. Molecular modeling suggests 1), that the protofilament switching may be due to kinesin-8 having a longer neck linker than kinesin-1, and 2), that the leftward bias is due the asymmetric geometry of the motor neck linker complex.

Depolymerizing kinesins Kip3 and MCAK shape cellular microtubule architecture by differential control of catastrophe.

Microtubules are dynamic filaments whose ends alternate between periods of slow growth and rapid shortening as they explore intracellular space and move organelles. A key question is how regulatory proteins modulate catastrophe, the conversion from growth to shortening. To study this process, we reconstituted microtubule dynamics in the absence and presence of the kinesin-8 Kip3 and the kinesin-13 MCAK. Surprisingly, we found that, even in the absence of the kinesins, the microtubule catastrophe frequency depends on the age of the microtubule, indicating that catastrophe is a multistep process. Kip3 slowed microtubule growth in a length-dependent manner and increased the rate of aging. In contrast, MCAK eliminated the aging process. Thus, both kinesins are catastrophe factors; Kip3 mediates fine control of microtubule length by narrowing the distribution of maximum lengths prior to catastrophe, whereas MCAK promotes rapid restructuring of the microtubule cytoskeleton by making catastrophe a first-order random process.

Functional surface attachment in a sandwich geometry of GFP-labeled motor proteins.

Molecular motors perform work in cells by moving in an ATP-dependent manner along filamentous tracks. In vitro, the mechanical action of such motor proteins can be investigated by attaching the molecules to surfaces in the so-called gliding or bead assays. Surface attachment protocols have to be used that do not interfere with the function of the molecule. Here, we describe a sandwich protocol that preserves functionality. The protocol can be used for a large variety of proteins, in particular kinesin motor proteins that are GFP-tagged.

Microtubule dynamics reconstituted in vitro and imaged by single-molecule fluorescence microscopy.

In vitro assays that reconstitute the dynamic behavior of microtubules provide insight into the roles of microtubule-associated proteins (MAPs) in regulating the growth, shrinkage, and catastrophe of microtubules. The use of total internal reflection fluorescence microscopy with fluorescently labeled tubulin and MAPs has allowed us to study microtubule dynamics at the resolution of single molecules. In this chapter we present a practical overview of how these assays are performed in our laboratory: fluorescent labeling methods, strategies to prolong the time to photo-bleaching, preparation of stabilized microtubules, flow-cells, microtubule immobilization, and finally an overview of the workflow that we follow when performing the experiments. At all stages, we focus on practical tips and highlight potential stumbling blocks.

Studying kinesin motors by optical 3D-nanometry in gliding motility assays.

Recent developments in optical microscopy and nanometer tracking have facilitated our understanding of microtubules and their associated proteins. Using fluorescence microscopy, dynamic interactions are now routinely observed in vitro on the level of single molecules, mainly using a geometry in which labeled motors move on surface-immobilized microtubules. Yet, we think that the historically older gliding geometry, in which motor proteins bound to a substrate surface drive the motion microtubules, offers some unique advantages. (1) Motility can be precisely followed by coupling multiple fluorophores and/or single bright labels to the surface of microtubules without disturbing the activity of the motor proteins. (2) The number of motor proteins involved in active transport can be determined by several strategies. (3) Multimotor studies can be performed over a wide range of motor densities. These advantages allow for studying cooperativity of processive as well as nonprocessive motors. Moreover, the gliding geometry has proven to be most promising for nanotechnological applications of motor proteins operating in synthetic environments. In this chapter we review recent methods related to gliding motility assays in conjunction with 3D-nanometry. In particular, we aim to provide practical advice on how to set up gliding assays, how to acquire high-precision data from microtubules and attached quantum dots, and how to analyze data by 3D-nanometer tracking.

Kinesin-8 motors act cooperatively to mediate length-dependent microtubule depolymerization.

Motor proteins in the kinesin-8 family depolymerize microtubules in a length-dependent manner that may be crucial for controlling the length of organelles such as the mitotic spindle. We used single-molecule microscopy to understand the mechanism of length-dependent depolymerization by the budding yeast kinesin-8, Kip3p. We found that after binding at a random position on a microtubule and walking to the plus end, an individual Kip3p molecule pauses there until an incoming Kip3p molecule bumps it off. Kip3p dissociation is accompanied by removal of just one or two tubulin dimers (on average). Such a cooperative mechanism leads to a depolymerization rate that is proportional to the flux of motors to the microtubule end and accounts for the length dependence of depolymerization. This type of feedback between length and disassembly may serve as a model for understanding how an ensemble of molecules can measure and control polymer length.

Protein friction limits diffusive and directed movements of kinesin motors on microtubules.

Friction limits the operation of macroscopic engines and is critical to the performance of micromechanical devices. We report measurements of friction in a biological nanomachine. Using optical tweezers, we characterized the frictional drag force of individual kinesin-8 motor proteins interacting with their microtubule tracks. At low speeds and with no energy source, the frictional drag was related to the diffusion coefficient by the Einstein relation. At higher speeds, the frictional drag force increased nonlinearly, consistent with the motor jumping 8 nanometers between adjacent tubulin dimers along the microtubule, and was asymmetric, reflecting the structural polarity of the microtubule. We argue that these frictional forces arise from breaking bonds between the motor domains and the microtubule, and they limit the speed and efficiency of kinesin.

Optical trapping of coated microspheres.

In an optical trap, micron-sized dielectric particles are held by a tightly focused laser beam. The optical force on the particle is composed of an attractive gradient force and a destabilizing scattering force. We hypothesized that using anti-reflection-coated microspheres would reduce scattering and lead to stronger trapping. We found that homogeneous silica and polystyrene microspheres had a sharp maximum trap stiffness at a diameter of around 800 nm--the trapping laser wavelength in water--and that a silica coating on a polystyrene microsphere was a substantial improvement for larger diameters. In addition, we noticed that homogeneous spheres of a correct size demonstrated anti-reflective properties. Our results quantitatively agreed with Mie scattering calculations and serve as a proof of principle. We used a DNA stretching experiment to confirm the large linear range in detection and force of the coated microspheres and performed a high-force motor protein assay. These measurements show that the surfaces of the coated microspheres are compatible with biophysical assays.

Optical monitoring of neuronal activity at high frame rate with a digital random-access multiphoton (RAMP) microscope.

Two-photon microscopy offers the promise of monitoring brain activity at multiple locations within intact tissue. However, serial sampling of voxels has been difficult to reconcile with millisecond timescales characteristic of neuronal activity. This is due to the conflicting constraints of scanning speed and signal amplitude. The recent use of acousto-optic deflector scanning to implement random-access multiphoton microscopy (RAMP) potentially allows to preserve long illumination dwell times while sampling multiple points-of-interest at high rates. However, the real-life abilities of RAMP microscopy regarding sensitivity and phototoxicity issues, which have so far impeded prolonged optical recordings at high frame rates, have not been assessed. Here, we describe the design, implementation and characterisation of an optimised RAMP microscope. We demonstrate the application of the microscope by monitoring calcium transients in Purkinje cells and cortical pyramidal cell dendrites and spines. We quantify the illumination constraints imposed by phototoxicity and show that stable continuous high-rate recordings can be obtained. During these recordings the fluorescence signal is large enough to detect spikes with a temporal resolution limited only by the calcium dye dynamics, improving upon previous techniques by at least an order of magnitude.

LED illumination for video-enhanced DIC imaging of single microtubules.

In many applications high-resolution video-enhanced differential interference contrast microscopy is used to visualize and track the ends of single microtubules. We show that single ultrabright light emitting diodes from Luxeon can be used to replace conventional light sources for these kinds of applications without loss of function. We measured the signal-to-noise ratio of microtubules imaged with three different light emitting diode colours (blue, red, green). The blue light emitting diode performed best, and the signal-to-noise ratios were high enough to automatically track the ends of dynamic microtubules. Light emitting diodes as light sources for video-enhanced differential interference contrast microscopy are high performing, low-cost and easy to align alternatives to existing illumination solutions.