Aleksandra Badura

Synaptic inhibition of Purkinje cells mediates consolidation of vestibulo-cerebellar motor learning

Although feedforward inhibition onto Purkinje cells was first documented 40 years ago, we understand little of how inhibitory interneurons contribute to cerebellar function in behaving animals. Using a mouse line (PC-Δγ2) in which GABAA receptor–mediated synaptic inhibition is selectively removed from Purkinje cells, we examined how feedforward inhibition from molecular layer interneurons regulates adaptation of the vestibulo-ocular reflex. Although impairment of baseline motor performance was relatively mild, the ability to adapt the phase of the vestibulo-ocular reflex and to consolidate gain adaptations was strongly compromised. Purkinje cells showed abnormal patterns of simple spikes, both during and in the absence of evoked compensatory eye movements. On the basis of modeling our experimental data, we propose that feedforward inhibition, by controlling the fine-scale patterns of Purkinje cell activity, enables the induction of plasticity in neurons of the cerebellar and vestibular nuclei.

Genetic Dissection of the Function of Hindbrain Axonal Commissures

The Robo3 receptor controls midline crossing by axons. Deleting Robo3 in specific commissural neurons with a conditional knockout reveals their contribution to sensory and motor integration, and models human neurological conditions.

Fast GCaMPs for improved tracking of neuronal activity

The use of genetically encodable calcium indicator proteins to monitor neuronal activity is hampered by slow response times and a narrow Ca(2+)-sensitive range. Here we identify three performance-limiting features of GCaMP3, a popular genetically encodable calcium indicator protein. First, we find that affinity is regulated by the calmodulin domains Ca(2+)-chelating residues. Second, we find that off-responses to Ca(2+) are rate-limited by dissociation of the RS20 domain from calmodulins hydrophobic pocket. Third, we find that on-responses are limited by fast binding to the N-lobe at high Ca(2+) and by slow binding to the C-lobe at lower Ca(2+). We develop Fast-GCaMPs, which have up to 20-fold accelerated off-responses and show that they have a 200-fold range of K(D), allowing coexpression of multiple variants to span an expanded range of Ca(2+) concentrations. Finally, we show that Fast-GCaMPs track natural song in Drosophila auditory neurons and generate rapid responses in mammalian neurons, supporting the utility of our approach.

Raising cytosolic Cl− in cerebellar granule cells affects their excitability and vestibulo-ocular learning

Cerebellar cortical throughput involved in motor control comprises granule cells (GCs) and Purkinje cells (PCs), both of which receive inhibitory GABAergic input from interneurons. The GABAergic input to PCs is essential for learning and consolidation of the vestibulo‐ocular reflex, but the role of GC excitability remains unclear. We now disrupted the Kcc2 K‐Cl cotransporter specifically in either cell type to manipulate their excitability and inhibition by GABAA‐receptor Cl− channels. Although Kcc2 may have a morphogenic role in synapse development, Kcc2 disruption neither changed synapse density nor spine morphology. In both GCs and PCs, disruption of Kcc2, but not Kcc3, increased [Cl−]i roughly two‐fold. The reduced Cl− gradient nearly abolished GABA‐induced hyperpolarization in PCs, but in GCs it merely affected excitability by membrane depolarization. Ablation of Kcc2 from GCs impaired consolidation of long‐term phase learning of the vestibulo‐ocular reflex, whereas baseline performance, short‐term gain‐decrease learning and gain consolidation remained intact. These functions, however, were affected by disruption of Kcc2 in PCs. GC excitability plays a previously unknown, but specific role in consolidation of phase learning.

Cerebellar associative sensory learning defects in five mouse autism models

Sensory integration difficulties have been reported in autism, but their underlying brain-circuit mechanisms are underexplored. Using five autism-related mouse models, Shank3+/ΔC, Mecp2R308/Y, Cntnap2−/−, L7-Tsc1 (L7/Pcp2Cre::Tsc1flox/+), and patDp(15q11-13)/+, we report specific perturbations in delay eyeblink conditioning, a form of associative sensory learning requiring cerebellar plasticity. By distinguishing perturbations in the probability and characteristics of learned responses, we found that probability was reduced in Cntnap2−/−, patDp(15q11-13)/+, and L7/Pcp2Cre::Tsc1flox/+, which are associated with Purkinje-cell/deep-nuclear gene expression, along with Shank3+/ΔC. Amplitudes were smaller in L7/Pcp2Cre::Tsc1flox/+ as well as Shank3+/ΔC and Mecp2R308/Y, which are associated with granule cell pathway expression. Shank3+/ΔC and Mecp2R308/Y also showed aberrant response timing and reduced Purkinje-cell dendritic spine density. Overall, our observations are potentially accounted for by defects in instructed learning in the olivocerebellar loop and response representation in the granule cell pathway. Our findings indicate that defects in associative temporal binding of sensory events are widespread in autism mouse models. DOI: http://dx.doi.org/10.7554/eLife.06085.001

Fast calcium sensor proteins for monitoring neural activity

Abstract. A major goal of the BRAIN Initiative is the development of technologies to monitor neuronal network activity during active information processing. Toward this goal, genetically encoded calcium indicator proteins have become widely used for reporting activity in preparations ranging from invertebrates to awake mammals. However, slow response times, the narrow sensitivity range of Ca2+ and in some cases, poor signal-to-noise ratio still limit their usefulness. Here, we review recent improvements in the field of neural activity-sensitive probe design with a focus on the GCaMP family of calcium indicator proteins. In this context, we present our newly developed Fast-GCaMPs, which have up to 4-fold accelerated off-responses compared with the next-fastest GCaMP, GCaMP6f. Fast-GCaMPs were designed by destabilizing the association of the hydrophobic pocket of calcium-bound calmodulin with the RS20 binding domain, an intramolecular interaction that protects the green fluorescent protein chromophore. Fast-GCaMP6f-RS06 and Fast-GCaMP6f-RS09 have rapid off-responses in stopped-flow fluorimetry, in neocortical brain slices, and in the intact cerebellum in vivo. Fast-GCaMP6f variants should be useful for tracking action potentials closely spaced in time, and for following neural activity in fast-changing compartments, such as axons and dendrites. Finally, we discuss strategies that may allow tracking of a wider range of neuronal firing rates and improve spike detection.

Cerebellar granule cells acquire a widespread predictive feedback signal during motor learning

Cerebellar granule cells, which constitute half the brains neurons, supply Purkinje cells with contextual information necessary for motor learning, but how they encode this information is unknown. Here we show, using two-photon microscopy to track neural activity over multiple days of cerebellum-dependent eyeblink conditioning in mice, that granule cell populations acquire a dense representation of the anticipatory eyelid movement. Initially, granule cells responded to neutral visual and somatosensory stimuli as well as periorbital airpuffs used for training. As learning progressed, two-thirds of monitored granule cells acquired a conditional response whose timing matched or preceded the learned eyelid movements. Granule cell activity covaried trial by trial to form a redundant code. Many granule cells were also active during movements of nearby body structures. Thus, a predictive signal about the upcoming movement is widely available at the input stage of the cerebellar cortex, as required by forward models of cerebellar control.

A cerebellar learning model that reproduces the behavior of vestibulo-ocular reflex adaptation in wild-type and knock-out mice

The cerebellum is crucial for different types of motor learning. Established theories of cerebellar learning posit that the cerebellum learns by adjusting the weights of Parallel Fiber (PF) to Purkinje cells (PC) synapses, thanks to teaching signals provided by Climbing Fiber inputs. While these theories are consistent with a large body of experimental data, in particular on synaptic plasticity in PF to PC synapses, they cannot easily explain a growing body of experimental work, which seems to indicate a significant role of other sites of plasticity. Recent advances in the development of a large number in transgenic animals, as well as behavioral and electrophysiogical comparative studies between these animals and wild-type animals, have opened an unprecedented window into the mechanisms underlying learning in this structure. In particular, it has been shown that specific knock-outs are impaired selectively on difficult variants of the vestibulo-ocular reflex (VOR) adaptation task, one of the most studied cerebellar-dependent motor learning tasks. These impairments can occur even though the classical plasticity mechanisms are left untouched. These data pose significant new challenges for established models of cerebellar learning. To better understand the mechanisms of learning in the cerebellum, we built a model that can reproduce the available data on VOR adaptation, in both wild-type and transgenic animals. The model includes some of the main cell types involved in this task: granule cells (GCs), the input layer of cerebellar cortex, that receives vestibular information from the mossy fibers (MFs); Purkinje cells (PCs), as well as molecular layer interneurons (INs); and two cell populations in the medial vestibular nuclei (MVN), one excitatory and one inhibitory, that together control eye movement. The model also includes two sites of learning: the classical GC to PF plasticity site, as well as plasticity in the MF to MVN synapses. We provide a mechanistic understanding on how the system learns VOR adaptation in normal conditions, as well as how the system is impaired by specific knock-outs, which selectively suppress inhibition onto PCs, or increase the excitability of GCs. Finally, we show that our model is consistent with behavioral, as well as in vivo electrophysiological recordings.

Modeled changes of cerebellar activity in mutant mice are predictive of their learning impairments

Translating neuronal activity to measurable behavioral changes has been a long-standing goal of systems neuroscience. Recently, we have developed a model of phase-reversal learning of the vestibulo-ocular reflex, a well-established, cerebellar-dependent task. The model, comprising both the cerebellar cortex and vestibular nuclei, reproduces behavioral data and accounts for the changes in neural activity during learning in wild type mice. Here, we used our model to predict Purkinje cell spiking as well as behavior before and after learning of five different lines of mutant mice with distinct cell-specific alterations of the cerebellar cortical circuitry. We tested these predictions by obtaining electrophysiological data depicting changes in neuronal spiking. We show that our data is largely consistent with the model predictions for simple spike modulation of Purkinje cells and concomitant behavioral learning in four of the mutants. In addition, our model accurately predicts a shift in simple spike activity in a mutant mouse with a brainstem specific mutation. This combination of electrophysiological and computational techniques opens a possibility of predicting behavioral impairments from neural activity.

Normal cognitive and social development require posterior cerebellar activity

Cognitive and social capacities require postnatal experience, yet the pathways by which experience guides development are unknown. Here we show that the normal development of motor and nonmotor capacities requires cerebellar activity. Using chemogenetic perturbation of molecular layer interneurons to attenuate cerebellar output in mice, we found that activity of posterior regions in juvenile life modulates adult expression of eyeblink conditioning (paravermal lobule VI, crus I), reversal learning (lobule VI), persistive behavior and novelty-seeking (lobule VII), and social preference (crus I/II). Perturbation in adult life altered only a subset of phenotypes. Both adult and juvenile disruption left gait metrics largely unaffected. Contributions to phenotypes increased with the amount of lobule inactivated. Using an anterograde transsynaptic tracer, we found that posterior cerebellum made strong connections with prelimbic, orbitofrontal, and anterior cingulate cortex. These findings provide anatomical substrates for the clinical observation that cerebellar injury increases the risk of autism.