The neocortex is functionally organized into layers. Layer four receives the densest bottom up sensory inputs, while layers 2/3 and 5 receive top down inputs that may convey predictive information. A subset of cortical somatostatin (SST) neurons, the Martinotti cells, gate top down input by inhibiting the apical dendrites of pyramidal cells in layers 2/3 and 5, but it is unknown whether an analogous inhibitory mechanism controls activity in layer 4. Using high precision circuit mapping, in vivo optogenetic perturbations, and single cell transcriptional profiling, we reveal complementary circuits in the mouse barrel cortex involving genetically distinct SST subtypes that specifically and reciprocally interconnect with excitatory cells in different layers: Martinotti cells connect with layers 2/3 and 5, whereas non-Martinotti cells connect with layer 4. By enforcing layer-specific inhibition, these parallel SST subnetworks could independently regulate the balance between bottom up and top down input.

Gliomas synaptically integrate into neural circuits1,2. Previous research has demonstrated bidirectional interactions between neurons and glioma cells, with neuronal activity driving glioma growth1–4 and gliomas increasing neuronal excitability2,5–8. Here we sought to determine how glioma-induced neuronal changes influence neural circuits underlying cognition and whether these interactions influence patient survival. Using intracranial brain recordings during lexical retrieval language tasks in awake humans together with site-specific tumour tissue biopsies and cell biology experiments, we find that gliomas remodel functional neural circuitry such that task-relevant neural responses activate tumour-infiltrated cortex well beyond the cortical regions that are normally recruited in the healthy brain. Site-directed biopsies from regions within the tumour that exhibit high functional connectivity between the tumour and the rest of the brain are enriched for a glioblastoma subpopulation that exhibits a distinct synaptogenic and neuronotrophic phenotype. Tumour cells from functionally connected regions secrete the synaptogenic factor thrombospondin-1, which contributes to the differential neuron–glioma interactions observed in functionally connected tumour regions compared with tumour regions with less functional connectivity. Pharmacological inhibition of thrombospondin-1 using the FDA-approved drug gabapentin decreases glioblastoma proliferation. The degree of functional connectivity between glioblastoma and the normal brain negatively affects both patient survival and performance in language tasks. These data demonstrate that high-grade gliomas functionally remodel neural circuits in the human brain, which both promotes tumour progression and impairs cognition.

The connectivity patterns of excitatory and inhibitory microcircuits are fundamental to computation in the neocortex. Highly specific excitatory projections form a stereotyped microcircuit linking the six cortical layers, but it is unclear whether inhibitory circuits are structured according to a similar layer-based logic or instead wire up non-selectively across the different layers. Understanding principles of inhibitory wiring is critical, since they constrain the computational operations that cortical inhibition can perform. If subnetworks of inhibitory neurons target specific functional components of cortical circuits (e.g. cortical input and output layers), these targets could be independently modulated, enabling a richer repertoire of inhibitory computations. Here we use one and two photon optogenetic circuit mapping techniques to demonstrate that two distinct subtypes of spatially intermingled Layer 5 (L5) somatostatin (SOM) interneurons form exquisitely selective and complementary intracortical circuits. One subtype connects predominantly with L4 and L6 -the primary cortical input layers, while a second subtype connects nearly exclusively with L2/3 and L5 -the primary cortical 2 output layers. This highly specific architecture suggests that separate SOM networks could differentially modulate processing at the input and output stages of the neocortical microcircuit.