Transformation of primary sensory cortical representations from layer 4 to layer 2.

Sensory input arrives from thalamus in cortical layer (L) 4, which outputs predominantly to superficial layers. L4 to L2 thus constitutes one of the earliest cortical feedforward networks. Despite extensive study, the transformation performed by this network remains poorly understood. We use two-photon calcium imaging to record neural activity in L2-4 of primary vibrissal somatosensory cortex (vS1) as mice perform an object localization task with two whiskers. Touch responses sparsen and become more reliable from L4 to L2, with nearly half of the superficial touch response confined to ~1 % of excitatory neurons. These highly responsive neurons have broad receptive fields and can more accurately decode stimulus features. They participate disproportionately in ensembles, small subnetworks with elevated pairwise correlations. Thus, from L4 to L2, cortex transitions from distributed probabilistic coding to sparse and robust ensemble-based coding, resulting in more efficient and accurate representations.

Publisher Correction: Two dynamically distinct circuits drive inhibition in the sensory thalamus.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Two dynamically distinct circuits drive inhibition in the sensory thalamus.

Most sensory information destined for the neocortex is relayed through the thalamus, where considerable transformation occurs1,2. One means of transformation involves interactions between excitatory thalamocortical neurons that carry data to the cortex and inhibitory neurons of the thalamic reticular nucleus (TRN) that regulate the flow of those data3-6. Although the importance of the TRN has long been recognised7-9, understanding of its cell types, their organization and their functional properties has lagged behind that of the thalamocortical systems they control. Here we address this by investigating the somatosensory and visual circuits of the TRN in mice. In the somatosensory TRN we observed two groups of genetically defined neurons that are topographically segregated and physiologically distinct, and that connect reciprocally with independent thalamocortical nuclei through dynamically divergent synapses. Calbindin-expressing cells-located in the central core-connect with the ventral posterior nucleus, the primary somatosensory thalamocortical relay. By contrast, somatostatin-expressing cells-which reside along the surrounding edges of the TRN-synapse with the posterior medial thalamic nucleus, a higher-order structure that carries both top-down and bottom-up information10-12. The two TRN cell groups process their inputs in pathway-specific ways. Synapses from the ventral posterior nucleus to central TRN cells transmit rapid excitatory currents that depress deeply during repetitive activity, driving phasic spike output. Synapses from the posterior medial thalamic nucleus to edge TRN cells evoke slower, less depressing excitatory currents that drive more persistent spiking. Differences in the intrinsic physiology of TRN cell types, including state-dependent bursting, contribute to these output dynamics. The processing specializations of these two somatosensory TRN subcircuits therefore appear to be tuned to the signals they carry-a primary central subcircuit tuned to discrete sensory events, and a higher-order edge subcircuit tuned to temporally distributed signals integrated from multiple sources. The structure and function of visual TRN subcircuits closely resemble those of the somatosensory TRN. These results provide insights into how subnetworks of TRN neurons may differentially process distinct classes of thalamic information.

Recurrent interactions in local cortical circuits.

Most cortical synapses are local and excitatory. Local recurrent circuits could implement amplification, allowing pattern completion and other computations1-4. Cortical circuits contain subnetworks that consist of neurons with similar receptive fields and increased connectivity relative to the network average5,6. Cortical neurons that encode different types of information are spatially intermingled and distributed over large brain volumes5-7, and this complexity has hindered attempts to probe the function of these subnetworks by perturbing them individually8. Here we use computational modelling, optical recordings and manipulations to probe the function of recurrent coupling in layer 2/3 of the mouse vibrissal somatosensory cortex during active tactile discrimination. A neural circuit model of layer 2/3 revealed that recurrent excitation enhances sensory signals by amplification, but only for subnetworks with increased connectivity. Model networks with high amplification were sensitive to damage: loss of a few members of the subnetwork degraded stimulus encoding. We tested this prediction by mapping neuronal selectivity7 and photoablating9,10 neurons with specific selectivity. Ablation of a small proportion of layer 2/3 neurons (10-20, less than 5% of the total) representing touch markedly reduced responses in the spared touch representation, but not in other representations. Ablations most strongly affected neurons with stimulus responses that were similar to those of the ablated population, which is also consistent with network models. Recurrence among cortical neurons with similar selectivity therefore drives input-specific amplification during behaviour.

Temporal and mosaic Tsc1 deletion in the developing thalamus disrupts thalamocortical circuitry, neural function, and behavior.

Tuberous sclerosis is a developmental genetic disorder caused by mutations in TSC1, which results in epilepsy, autism, and intellectual disability. The cause of these neurological deficits remains unresolved. Imaging studies suggest that the thalamus may be affected in tuberous sclerosis patients, but this has not been experimentally interrogated. We hypothesized that thalamic deletion of Tsc1 at distinct stages of mouse brain development would produce differential phenotypes. We show that mosaic Tsc1 deletion within thalamic precursors at embryonic day (E) 12.5 disrupts thalamic circuitry and alters neuronal physiology. Tsc1 deletion at this early stage is unique in causing both seizures and compulsive grooming in adult mice. In contrast, only a subset of these phenotypes occurs when thalamic Tsc1 is deleted at a later embryonic stage. Our findings demonstrate that abnormalities in a discrete population of neurons can cause global brain dysfunction and that phenotype severity depends on developmental timing and degree of genetic mosaicism.

Genetic dissection of midbrain dopamine neuron development in vivo.

Midbrain dopamine (MbDA) neurons are partitioned into medial and lateral cohorts that control complex functions. However, the genetic underpinnings of MbDA neuron heterogeneity are unclear. While it is known that Wnt1-expressing progenitors contribute to MbDA neurons, the role of Wnt1 in MbDA neuron development in vivo is unresolved. We show that mice with a spontaneous point mutation in Wnt1 have a unique phenotype characterized by the loss of medial MbDA neurons concomitant with a severe depletion of Wnt1-expressing progenitors and diminished LMX1a-expressing progenitors. Wnt1 mutant embryos also have alterations in a hierarchical gene regulatory loop suggesting multiple gene involvement in the Wnt1 mutant MbDA neuron phenotype. To investigate this possibility, we conditionally deleted Gbx2, Fgf8, and En1/2 after their early role in patterning and asked whether these genetic manipulations phenocopied the depletion of MbDA neurons in Wnt1 mutants. The conditional deletion of Gbx2 did not result in re-positioning or distribution of MbDA neurons. The temporal deletion of Fgf8 did not result in the loss of either LMX1a-expressing progenitors nor the initial population of differentiated MbDA neurons, but did result in a complete loss of MbDA neurons at later stages. The temporal deletion and species specific manipulation of En1/2 demonstrated a continued and species specific role of Engrailed genes in MbDA neuron development. Notably, our conditional deletion experiments revealed phenotypes dissimilar to Wnt1 mutants indicating the unique role of Wnt1 in MbDA neuron development. By placing Wnt1, Fgf8, and En1/2 in the context of their temporal requirement for MbDA neuron development, we further deciphered the developmental program underpinning MbDA neuron progenitors.