Descending neurons (DNs) occupy a key position in the sensorimotor hierarchy, conveying signals from the brain to the rest of the body below the neck. In Drosophila melanogaster flies, approximately 480 DN cell types have been described from electron-microscopy image datasets. Genetic access to these cell types is crucial for further investigation of their role in generating behaviour. We previously conducted the first large-scale survey of Drosophila melanogaster DNs, describing 98 unique cell types from light microscopy and generating cell-type-specific split-Gal4 driver lines for 65 of them. Here, we extend our previous work, describing the morphology of 146 additional DN types from light microscopy, bringing the total number DN types identified in light microscopy datasets to 244, or roughly 50% of all DN types. In addition, we produced 500 new sparse split-Gal4 driver lines and compiled a list of previously published DN lines from the literature for a combined list of 806 split-Gal4 driver lines targeting 190 DN types.

The brain exhibits rich oscillatory dynamics that play critical roles in vigilance and cognition, such as the neural rhythms that define sleep. These rhythms continuously fluctuate, signaling major changes in vigilance, but the widespread brain dynamics underlying these oscillations are difficult to investigate. Using simultaneous EEG and fast fMRI in humans who fell asleep inside the scanner, we developed a machine learning approach to investigate which fMRI regions and networks predict fluctuations in neural rhythms. We demonstrated that the rise and fall of alpha (8-12 Hz) and delta (1-4 Hz) power-two canonical EEG bands critically involved with cognition and vigilance-can be predicted from fMRI data in subjects that were not present in the training set. This approach also identified predictive information in individual brain regions across the cortex and subcortex. Finally, we developed an approach to identify shared and unique predictive information, and found that information about alpha rhythms was highly separable in two networks linked to arousal and visual systems. Conversely, delta rhythms were diffusely represented on a large spatial scale primarily across the cortex. These results demonstrate that EEG rhythms can be predicted from fMRI data, identify large-scale network patterns that underlie alpha and delta rhythms, and establish a novel framework for investigating multimodal brain dynamics.

Animals efficiently learn to navigate their environment. In the laboratory, naive mice explore their environment via highly structured trajectories and can learn to localize new spatial targets in as few as a handful of trials. It is unclear how such efficient learning is possible, since existing computational models of spatial navigation require far more experience to achieve comparable performance and do not attempt to explain the evolving structure of animal behavior during learning. To inform a new algorithm for rapid learning of navigational goals, we took inspiration from the reliable structure of behavior as mice learned to intercept hidden spatial targets. We designed agents that generate behavioral trajectories by controlling the speed and angular velocity of smooth path segments between anchor points. To rapidly learn good anchors, we use Bayesian inference on the history of rewarded and unrewarded trajectories to infer the probability that an anchor will be successful, and active sampling to trim hypothesized anchors. Agents learn within tens of trials to generate compact trajectories that intercept a target, capturing the evolution of behavioral structure and matching the upper limits of learning efficiency observed in mice. We further show that this algorithm can explain how mice avoid obstacles and rapidly adapt to target switches. Finally, we show that this framework naturally encompasses both egocentric and allocentric strategies for navigation.

The mammalian cerebral cortex is composed of neurons whose properties vary in a continuous fashion rather than falling into discrete cell types. In the mouse visual cortex, excitatory neurons in layer 2 and 3 (L2/3) form such a continuum along cortical depth, patterned by the graded expression of hundreds of genes. Here we sought to understand how this continuum develops and contributes to cortical wiring. Using single-nucleus multiomics (RNA- and ATAC-Seq) and spatial transcriptomics, we show that the L2/3 continuum is established in two phases. During the first postnatal week, a genetically hardwired program establishes a primitive continuum of cell identities spanning the depth of L2/3. The second program, promoted by visual experience, is later superimposed upon the preexisting continuum. This second phase is driven by activity-regulated transcription factors that drive the L2/3 depth-dependent expression of genes linked to synaptic function and plasticity. We show that neurons at different positions along the L2/3 continuum project preferentially to distinct higher visual areas and that visual deprivation disrupts targeting to some higher visual areas while sparing others. Thus, cortical continua emerge through a stepwise process in which genetic programs and sensory experience specify neuronal identity and sculpt intracortical wiring specificity.

Tremor is a common movement disorder associated with several neurodegenerative diseases, yet its mechanisms are not well understood. Using a machine learning method, FLLIT, we previously characterised gait and tremor signatures in the Drosophila model for Spinocerebellar ataxia 3 (SCA3), and found them to be analogous to human SCA3. Here, we carried out a functional screen for neuronal populations that underlie tremor, and found that dysfunction of a specific population of neurons in the ventral nerve cord (VNC) is necessary and sufficient for tremor. Adult-onset expression of mutant ATXN3 in or genetic hypo-activation of these neurons leads to tremor, indicating their important role in adult motor control. RNAseq and functional experiments showed that dysfunction of GABAergic neurons, and not other neurotransmitter populations tested, causes tremor. Finally, we identified a small subset of approximately 30 predominantly GABAergic neurons within the adult VNC that are essential for smooth walking. This study demonstrates that tremor in SCA3 flies arises from GABAergic dysfunction, and that FLLIT can be used to dissect motor control mechanisms.

Predictive coding is a theoretical framework that can explain how animals build internal models of their sensory environments by predicting sensory inputs. Predictive coding may capture either spatial or temporal relationships between sensory objects. While the original theory by Rao and Ballard, 1999 described spatial predictive coding, much of the recent experimental data has been interpreted as evidence for temporal predictive coding. Here we directly tested whether the “mismatch” neural responses in sensory cortex are due to a spatial or a temporal internal model. We adopted two common paradigms to study predictive coding: one based on virtual-reality and one based on static images. After training mice with repeated visual stimulation for several days, we performed multiple manipulations, including: 1) we introduced a novel stimulus, 2) we replaced a stimulus with a novel gray wall, 3) we duplicated a trained stimulus, or 4) we altered the order of the stimuli. The first two manipulations induced a substantial mismatch response in neural populations of up to 20,000 neurons recorded across primary and higher-order visual cortex, while the third and fourth ones did not. Thus, a mismatch response only occurred if a new spatial – not temporal – pattern was introduced.

Microscopy drives biological discovery, yet high costs limit its access to resource-limited regions. We highlight examples of successful frugal microscopes that have overcome adoption barriers, offering a roadmap to expand affordable, quantitative imaging tools and foster impactful research in resource-limited settings.

Small cell lung cancer (SCLC) is a highly aggressive type of lung cancer, characterized by rapid proliferation, early metastatic spread, frequent early relapse and a high mortality rate. Recent evidence has suggested that innervation has an important role in the development and progression of several types of cancer. Cancer-to-neuron synapses have been reported in gliomas, but whether peripheral tumours can form such structures is unknown. Here we show that SCLC cells can form functional synapses and receive synaptic transmission. Using in vivo insertional mutagenesis screening in conjunction with cross-species genomic and transcriptomic validation, we identified neuronal, synaptic and glutamatergic signalling gene sets in mouse and human SCLC. Further experiments revealed the ability of SCLC cells to form synaptic structures with neurons in vitro and in vivo. Electrophysiology and optogenetic experiments confirmed that cancer cells can receive NMDA receptor- and GABA receptor-mediated synaptic inputs. Fitting with a potential oncogenic role of neuron-SCLC interactions, we showed that SCLC cells derive a proliferation advantage when co-cultured with vagal sensory or cortical neurons. Moreover, inhibition of glutamate signalling had therapeutic efficacy in an autochthonous mouse model of SCLC. Therefore, following malignant transformation, SCLC cells seem to hijack synaptic signalling to promote tumour growth, thereby exposing a new route for therapeutic intervention.

Upon inflammation, leukocytes extravasate through endothelial cells. When they extravasate, it is generally accepted that neighboring endothelial cells disconnect. Careful examination of endothelial junctions showed a partial membrane overlap beyond VE-cadherin distribution. These overlaps are regulated by actin polymerization and, although marked by, do not require PECAM-1, nor VE-cadherin. Neutrophils prefer wider membrane overlaps as exit sites. Detailed 3D analysis of neutrophil transmigration in real time at high spatiotemporal resolution revealed that overlapping endothelial membranes form a tunnel during neutrophil transmigration. These tunnels are formed by the neutrophil lifting the membrane of the upper endothelial cell while indenting and crawling over the membrane of the underlying endothelial cell. Our work shows that endothelial cells do not simply retract upon the passage of neutrophils but provide membrane tunnels, allowing neutrophils to extravasate. This discovery defines the 3D multicellular architecture in which the paracellular transmigration of neutrophils occurs.

Optogenetic activators with red-shifted excitation spectra, such as Chrimson, have significantly advanced Drosophila neuroscience. However, until recently, available optogenetic inhibitors required shorter activation wavelengths, which don’t penetrate tissue as effectively and are stronger visual stimuli to the animal, potentially confounding behavioral results. Here, we assess the efficacy of two newly identified anion-conducting channelrhodopsins with spectral sensitivities similar to Chrimson: A1ACR and HfACR (RubyACRs). Electrophysiology and functional imaging confirmed that RubyACRs effectively hyperpolarize neurons, with stronger and faster effects than the widely used inhibitor GtACR1. Activation of RubyACRs led to circuit-specific behavioral changes in three different neuronal groups. In glutamatergic motor neurons, activating RubyACRs suppressed adult locomotor activity. In PPL1-γ1pedc dopaminergic neurons, pairing odors with RubyACR activation during learning produced odor responses consistent with synaptic silencing. Finally, activation of RubyACRs in the pIP10 neuron suppressed pulse song during courtship. Together, these results demonstrate that RubyACRs are effective and reliable tools for neuronal inhibition in Drosophila, expanding the optogenetic toolkit for circuit dissection in freely behaving animals.

Preprint: https://www.biorxiv.org/content/early/2025/06/15/2025.06.13.659144