L-Lactate is increasingly appreciated as a key metabolite and signaling molecule in mammals. However, investigations of the inter- and intra-cellular dynamics of L-lactate are currently hampered by the limited selection and performance of L-lactate-specific genetically encoded biosensors. Here we now report a spectrally and functionally orthogonal pair of high-performance genetically encoded biosensors: a green fluorescent extracellular L-lactate biosensor, designated eLACCO2.1, and a red fluorescent intracellular L-lactate biosensor, designated R-iLACCO1. eLACCO2.1 exhibits excellent membrane localization and robust fluorescence response. To the best of our knowledge, R-iLACCO1 and its affinity variants exhibit larger fluorescence responses than any previously reported intracellular L-lactate biosensor. We demonstrate spectrally and spatially multiplexed imaging of L-lactate dynamics by coexpression of eLACCO2.1 and R-iLACCO1 in cultured cells, and in vivo imaging of extracellular and intracellular L-lactate dynamics in mice.

High-density, integrated silicon electrodes have begun to transform systems neuroscience, by enabling large-scale neural population recordings with single cell resolution. Existing technologies, however, have provided limited functionality in nonhuman primate species such as macaques, which offer close models of human cognition and behavior. Here, we report the design, fabrication, and performance of Neuropixels 1.0-NHP, a high channel count linear electrode array designed to enable large-scale simultaneous recording in superficial and deep structures within the macaque or other large animal brain. These devices were fabricated in two versions: 4416 electrodes along a 45 mm shank, and 2496 along a 25 mm shank. For both versions, users can programmably select 384 channels, enabling simultaneous multi-area recording with a single probe. We demonstrate recording from over 3000 single neurons within a session, and simultaneous recordings from over 1000 neurons using multiple probes. This technology represents a significant increase in recording access and scalability relative to existing technologies, and enables new classes of experiments involving fine-grained electrophysiological characterization of brain areas, functional connectivity between cells, and simultaneous brain-wide recording at scale.

Cognitive maps confer animals with flexible intelligence by representing spatial, temporal, and abstract relationships that can be used to shape thought, planning, and behavior. Cognitive maps have been observed in the hippocampus, but their algorithmic form and the processes by which they are learned remain obscure. Here, we employed large-scale, longitudinal two-photon calcium imaging to record activity from thousands of neurons in the CA1 region of the hippocampus while mice learned to efficiently collect rewards from two subtly different versions of linear tracks in virtual reality. The results provide a detailed view of the formation of a cognitive map in the hippocampus. Throughout learning, both the animal behavior and hippocampal neural activity progressed through multiple intermediate stages, gradually revealing improved task understanding and behavioral efficiency. The learning process led to progressive decorrelations in initially similar hippocampal neural activity within and across tracks, ultimately resulting in orthogonalized representations resembling a state machine capturing the inherent structure of the task. We show that a Hidden Markov Model (HMM) and a biologically plausible recurrent neural network trained using Hebbian learning can both capture core aspects of the learning dynamics and the orthogonalized representational structure in neural activity. In contrast, we show that gradient-based learning of sequence models such as Long Short-Term Memory networks (LSTMs) and Transformers do not naturally produce such representations. We further demonstrate that mice exhibited adaptive behavior in novel task settings, with neural activity reflecting flexible deployment of the state machine. These findings shed light on the mathematical form of cognitive maps, the learning rules that sculpt them, and the algorithms that promote adaptive behavior in animals. The work thus charts a course toward a deeper understanding of biological intelligence and offers insights toward developing more robust learning algorithms in artificial intelligence.

Mfsd2a is the transporter for docosahexaenoic acid (DHA), an omega-3 fatty acid, across the blood brain barrier (BBB). Defects in Mfsd2a are linked to ailments from behavioral and motor dysfunctions to microcephaly. Mfsd2a transports long-chain unsaturated fatty-acids, including DHA and α-linolenic acid (ALA), that are attached to the zwitterionic lysophosphatidylcholine (LPC) headgroup. Even with the recently determined structures of Mfsd2a, the molecular details of how this transporter performs the energetically unfavorable task of translocating and flipping lysolipids across the lipid bilayer remains unclear. Here, we report five single-particle cryo-EM structures of Danio rerio Mfsd2a (drMfsd2a): in the inward-open conformation in the ligand-free state and displaying lipid-like densities modeled as ALA-LPC at four distinct positions. These Mfsd2a snapshots detail the flipping mechanism for lipid-LPC from outer to inner membrane leaflet and release for membrane integration on the cytoplasmic side. These results also map Mfsd2a mutants that disrupt lipid-LPC transport and are associated with disease.

We present an auxiliary learning task for the problem of neuron segmentation in electron microscopy volumes. The auxiliary task consists of the prediction of local shape descriptors (LSDs), which we combine with conventional voxel-wise direct neighbor affinities for neuron boundary detection. The shape descriptors capture local statistics about the neuron to be segmented, such as diameter, elongation, and direction. On a study comparing several existing methods across various specimen, imaging techniques, and resolutions, auxiliary learning of LSDs consistently increases segmentation accuracy of affinity-based methods over a range of metrics. Furthermore, the addition of LSDs promotes affinity-based segmentation methods to be on par with the current state of the art for neuron segmentation (flood-filling networks), while being two orders of magnitudes more efficient-a critical requirement for the processing of future petabyte-sized datasets.

Single molecule localization microscopy relies on the precise quantification of the position of single dye emitters in a sample. This precision is improved by the number of photons that can be detected from each molecule. Particularly recording at cryogenic temperatures dramatically reduces photobleaching and would, hence, in principle, allow the user to massively increase the illumination time to several seconds. The downside of long illuminations, however, would be image blur due to inevitable jitter or drift occurring during the illuminations, which deteriorates the localization precision. In this paper, we theoretically demonstrate that a parallel recording of the fiducial marker beads together with a fitting approach accounting for the full drift trajectory allows for largely eliminating drift effects for drift magnitudes of several hundred nanometers per frame. We showcase the method for linear and diffusional drift as well as oscillations, assuming fixed dipole orientations during each illumination.

Real-time neural signal processing is essential for brain-machine interfaces and closed-loop neuronal perturbations. However, most existing applications sacrifice cell-specific identity and temporal spiking information for speed. We developed a hybrid hardware-software system that utilizes a Field Programmable Gate Array (FPGA) chip to acquire and process data in parallel, enabling individual spikes from many simultaneously recorded neurons to be assigned single-neuron identities with 1-millisecond latency. The FPGA assigns labels, validated with ground-truth data, by comparing multichannel spike waveforms from tetrode or silicon probe recordings to a spike-sorted model generated offline in software. This platform allowed us to rapidly inactivate a region in vivo based on spikes from an upstream neuron before these spikes could excite the downstream region. Furthermore, we could decode animal location within 3 ms using data from a population of individual hippocampal neurons. These results demonstrate our system’s suitability for a broad spectrum of research and clinical applications.

One-third of the mammalian proteome is comprised of transmembrane and secretory proteins that are synthesized on endoplasmic reticulum (ER). Here, we investigate the spatial distribution and regulation of mRNAs encoding these membrane and secretory proteins (termed "secretome" mRNAs) through live cell, single molecule tracking to directly monitor the position and translation states of secretome mRNAs on ER and their relationship to other organelles. Notably, translation of secretome mRNAs occurred preferentially near lysosomes on ER marked by the ER junction-associated protein, Lunapark. Knockdown of Lunapark reduced the extent of secretome mRNA translation without affecting translation of other mRNAs. Less secretome mRNA translation also occurred when lysosome function was perturbed by raising lysosomal pH or inhibiting lysosomal proteases. Secretome mRNA translation near lysosomes was enhanced during amino acid deprivation. Addition of the integrated stress response inhibitor, ISRIB, reversed the translation inhibition seen in Lunapark knockdown cells, implying an eIF2 dependency. Altogether, these findings uncover a novel coordination between ER and lysosomes, in which local release of amino acids and other factors from ER-associated lysosomes patterns and regulates translation of mRNAs encoding secretory and membrane proteins.

The precise neural mechanisms within the brain that contribute to the remarkable lifetime persistence of memory remain unknown. Existing techniques to record neurons in animals are either unsuitable for longitudinal recording from the same cells or make it difficult for animals to express their full naturalistic behavioral repertoire. We present a magnetic voluntary head-fixation system that provides stable optical access to the brain during complex behavior. Compared to previous systems that used mechanical restraint, there are no moving parts and animals can engage and disengage entirely at will. This system is failsafe, easy for animals to use and reliable enough to allow long-term experiments to be routinely performed. Together with a novel two-photon fluorescence collection scheme that increases two-photon signal and a transgenic rat line that stably expresses the calcium sensor GCaMP6f in dorsal CA1, we are able to track and record activity from the same hippocampal neurons, during behavior, over a large fraction of animals’ lives.

A tool to map changes in synaptic strength during a defined time window could provide powerful insights into the mechanisms governing learning and memory. We developed a technique, Extracellular Protein Surface Labeling in Neurons (EPSILON), to map α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) insertion in vivo by pulse-chase labeling of surface AMPARs with membrane-impermeable dyes. This approach allows for single-synapse resolution maps of plasticity in genetically targeted neurons during memory formation. We investigated the relationship between synapse-level and cell-level memory encodings by mapping synaptic plasticity and cFos expression in hippocampal CA1 pyramidal cells upon contextual fear conditioning (CFC). We observed a strong correlation between synaptic plasticity and cFos expression, suggesting a synaptic mechanism for the association of cFos expression with memory engrams. The EPSILON technique is a useful tool for mapping synaptic plasticity and may be extended to investigate trafficking of other transmembrane proteins.