bioRxiv Subject Collection: Neuroscience's Journal

Distributed coding of space and valence in the medial temporal lobe
The hippocampus has been associated with spatial information processing (OKeefe, 1976). However, it is also not clear whether such ensemble coding of spatial information extends to other brain regions in the medial temporal lobe. Hence, this study uses various classification techniques to attempt to decode spatial and valence information utilizing publicly available rodent single-unit data (Girardeau et al., 2017). We found that the spatial information is encoded sparsely and distributed among many neurons in the hippocampus. Additionally we found that both spatial and valence information could be decoded in the hippocampus, the amygdala, and the piriform cortex. We also compared the strengths and weaknesses of four different classifiers (LDA, KNN, SVM, NB) theoretically and in practice with this dataset.

Sensory History Drives Adaptive Neural Geometry in LP/Pulvinar-Prefrontal Cortex Circuits
Prior expectations guide attention and support perceptual filtering for efficient processing during decision-making. Here we show that during a visual discrimination task, mice adaptively use prior stimulus history to guide ongoing choices by estimating differences in evidence between consecutive trials (| Delta Dir |). The thalamic lateral posterior (LP)/pulvinar nucleus provides robust inputs to the Anterior Cingulate Cortex (ACC), which has been implicated in selective attention and predictive processing, but the function of the LP-ACC projection is unknown. We found that optogenetic manipulations of LP-ACC axons disrupted animals' ability to effectively estimate and use information across stimulus history, leading to | Delta Dir |-dependent ipsilateral biases. Two-photon calcium imaging of LP-ACC axons revealed an engagement-dependent low-dimensional organization of stimuli along a curved manifold. This representation was scaled by | Delta Dir | in a manner that emphasized greater deviations from prior evidence. Thus, our work identifies the LP-ACC pathway as essential for selecting and evaluating stimuli relative to prior evidence to guide decisions.

Representational learning by optimization of neural manifolds in an olfactory memory network
Higher brain functions depend on experience-dependent representations of relevant information that may be organized by attractor dynamics or by geometrical modifications of continuous "neural manifolds". To explore these scenarios we analyzed odor-evoked activity in telencephalic area pDp of juvenile and adult zebrafish, the homolog of piriform cortex. No obvious signatures of attractor dynamics were detected. Rather, olfactory discrimination training selectively enhanced the separation of neural manifolds representing task-relevant odors from other representations, consistent with predictions of autoassociative network models endowed with precise synaptic balance. Analytical approaches using the framework of manifold capacity revealed multiple geometrical modifications of representational manifolds that supported the classification of task-relevant sensory information. Manifold capacity predicted odor discrimination across individuals, indicating a close link between manifold geometry and behavior. Hence, pDp and possibly related recurrent networks store information in the geometry of representational manifolds, resulting in joint sensory and semantic maps that may support distributed learning processes.

Haploinsufficiency of lysosomal enzyme genes in Alzheimers disease
There is growing evidence suggesting that the lysosome or lysosome dysfunction is associated with Alzheimers disease (AD). Pathway analysis of post mortem brain-derived proteomic data from AD patients shows that the lysosomal system is perturbed relative to similarly aged unaffected controls. However, it is unclear if these changes contributed to the pathogenesis or are a response to the disease. Consistent with the hypothesis that lysosome dysfunction contributes to AD pathogenesis, whole genome sequencing data indicate that heterozygous pathogenic mutations and predicted protein-damaging variants in multiple lysosomal enzyme genes are enriched in AD patients compared to matched controls. Heterozygous loss-of-function mutations in the palmitoyl protein thioesterase-1 (PPT1), alpha-L-iduronidase (IDUA), {beta}-glucuronidase (GUSB), N-acetylglucosaminidase (NAGLU), and galactocerebrosidase (GALC) genes have a gene-dosage effect on Abeta40 levels in brain interstitial fluid in C57BL/6 mice and significantly increase Abeta plaque formation in the 5xFAD mouse model of AD, thus providing in vivo validation of the human genetic data. A more detailed analysis of PPT1 heterozygosity in 18-month-old mice revealed changes in alpha-, beta-, and gamma-secretases that favor an amyloidogenic pathway. Proteomic changes in brain tissue from aged PPT1 heterozygous sheep are consistent with both the mouse data and the potential activation of AD pathways. Finally, CNS-directed, AAV-mediated gene therapy significantly decreased Abeta plaques, increased life span, and improved behavioral performance in 5xFAD/PPT1+/- mice. Collectively, these data strongly suggest that heterozygosity of multiple lysosomal enzyme genes represent risk factors for AD and may identify precise therapeutic targets for a subset of genetically-defined AD patients.

Nonlinear brain connectivity from neurons to networks: quantification, sources and localization.
Since the first studies in functional connectivity, Pearson's correlation has been the primary tool to determine relatedness between the activity of different brain locations. Over the years, concern over the information neglected by correlation pushed toward using different measures accounting for non-linearity. However, some studies suggest that, at the typical observation scale, a linear description of the brain captures a vast majority of the information. Therefore, we measured the fraction of information that would be lost using a linear description and which regions would be affected the most. We considered fMRI, EEG, iEEG, and single unit spikes to assess how the observation scale impacts the amount of non-linearity. We observe that the information loss is reduced for modalities with large temporal or spatial averaging (fMRI and EEG) and gains relevance on more fine descriptions of the activity (iEEG and single unit spikes). We conclude that for most human applications, Pearson's correlation coefficient adequately describes pairwise interactions in time series from current recording techniques.

Predicting whole-brain neural dynamics from prefrontal cortex fNIRS signal during movie-watching
Functional near-infrared spectroscopy (fNIRS) offers a portable, cost-effective alternative to functional magnetic resonance imaging (fMRI) for non-invasively measuring neural activity. However, fNIRS measurements are limited to cortical regions near the scalp, missing important medial and deeper brain areas. We introduce a predictive model that maps prefrontal fNIRS signals to whole-brain fMRI activity during movie-watching. By aligning neural responses to a common audiovisual stimulus, our approach leverages shared dynamics across imaging modalities to map fNIRS signals to broader neural activity patterns. We scanned participants with fNIRS and utilized a publicly available fMRI dataset of participants watching the same TV episode. The model was trained on the first half of the episode and tested on a held-out participant watching the second half to assess cross-individual and cross-stimulus generalizability. The model significantly predicted fMRI time courses in 66 out of 122 brain regions, including in areas otherwise inaccessible to fNIRS. The predicted fMRI time course also replicated intersubject functional connectivity patterns and retained semantic information about the movie content. Our publicly available model enables researchers to infer broader neural dynamics from localized fNIRS data, offering new opportunities for studying the neural basis of complex cognitive processes during naturalistic tasks.

Inhibition of Notch Signaling Attenuates Epileptic Discharges in the Adolescent Rat Brain after Status Epilepticus Induction
Background Notch signaling plays a critical role in neuroregeneration after injuries such as those caused by status epilepticus (SE). Objective To explore the effects of Notch signaling on epileptogenesis and the underlying mechanisms in adolescent rat brains in the acute phase after SE induction. Methods N-[N-(3,5-difluorophenacetyl)-l-alanyl)]-S-phenylglycine t-butyl ester (DAPT), which indirectly inhibits Notch, was injected into rats during the acute phase after SE induction to inhibit Notch signaling. Electroencephalogram (EEG) was used to observe spontaneous recurrent seizures. Differences in the synaptic structures of the hippocampus were observed by transmission electron microscopy. Nissl staining and Timm staining were used to observe the loss of hippocampal neurons and sprouting of mossy fibers, respectively, in the hippocampus at 28 days after SE. Results EEG illustrated that DAPT treatment reduced the severity of epileptic discharges after SE induction. Transmission electron microscopy revealed reductions in the presynaptic membrane active band length and postsynaptic membrane dense matter thickness in the CA1 region of the hippocampus. Meanwhile, Nissl staining demonstrated that DAPT treatment reduced the loss of hippocampal neuronal cell degeneration, and the hippocampal structure was repaired to a certain extent. Meanwhile, Timm staining illustrated that DAPT treatment did not affect mossy fiber sprouting (MFS) after SE induction. Conclusion Inhibiting Notch signaling reduced EEG epileptic activity, attenuated synaptic damage, and partially restored the hippocampal neuronal structure. However, it did not alter MFS after SE induction.

Mapping of dI3 neuron sensorimotor circuits across the cervical and lumbar spinal cord
From the fine control of hand movements to the dynamic corrective adjustments during locomotion, spinal circuits integrate descending supraspinal and sensory inputs to modulate diverse motor functions. The integration of such a wide range of signals across the spinal cord is primarily mediated by propriospinal interneurons. In this study, we investigate the connectivity of a population of propriospinal interneurons marked by the expression of Isl1, called dI3 neurons. These dI3s integrate supraspinal and sensory signals to facilitate many important functions, such as hand grasp, locomotion, and motor recovery after spinal cord injury; however, we have a limited understanding of how subpopulations of dI3s modulate network activity across the spinal cord to contribute to these behaviors. Their functional connectivity to motor circuits across the cervical and lumbar spinal cord was assessed through optogenetic activation of dI3s localized in different spinal segments. Our data demonstrates that cervical and lumbar dI3 subpopulations can form local, commissural, intersegmental, and long propriospinal pathways. Furthermore, dI3 subpopulations can be tonically stimulated to elicit locomotor activity. These extensive projection patterns of dI3s across the cervical and lumbar spinal cord suggest that dI3 subpopulations can modulate the activity of multiple motor networks within their respective spinal cord segments or across distant forelimb and hindlimb segments to facilitate a wide variety of motor functions.

Enhancing information extraction from field potentials in electrophysiology studies
Multi-channel recordings from the brain serve as the primary approach to address various mechanistic and behavioral questions about neural activity and the underlying circuitry. While electrodes can detect neural activity at a distance from their source generators, activity from other concurrently active neural sources is also volume conducted to the electrodes, thus creating linear dependencies among the channels. These dependencies pose challenges in discerning specific communication patterns between different neuronal populations, and their behavioral implications. In this study, we demonstrate the capability of a marked point process (MPP) representation of channel activity, focusing on oscillatory bursts, to curtail the effects of volume conduction noise. By characterizing the localized spectral information within oscillatory bursts, we achieved a 45% average reduction in channel correlations across three recording modalities of field potentials (electroencephalography, electrocorticography, and local field potentials). We further provide evidence that the implemented sparse representation preserves both behavioral and causal information in the signal. We illustrate our findings with two examples: 1) retention of finger-level movement information in field potentials recorded from humans, demonstrated using a simple online classifier, and 2) retention of top-down connectivity information between the prefrontal and motor cortices of behaving rats. Overall, our results underscore the novelty of using a marked point process representation of oscillatory bursts to concisely encode behavioral and connectivity information while attenuating the effects of volume conduction from the causal signal sources.

Simultaneous loss of CAMK2A and CAMK2B reveals endogenous in vivo substrates
Ca2+/calmodulin-dependent protein kinase 2 (CAMK2) plays a critical role in calcium signaling. Recent gene knockout studies show that CAMK2A and CAMK2B can have distinct roles yet also partially compensate for each other in yet unknown brain functions. In order to provide insight into potential novel CAMK2 functions, we performed parallel phosphoproteomic analyses on non-stimulated cortex tissue from inducible Camk2a and Camk2b double knockout (Camk2af/f;Camk2bf/f;CAG-CreESR) mice and from wild type mice. A total of 5622 phosphorylated peptides derived from 2080 proteins were identified. Phosphorylation at serine/threonine residues in 130 proteins were downregulated in the double knockout mice, including residues in 113 proteins that have not previously been identified as potential CAMK2 substrates. Comparison of amino acid sequences surrounding the downregulated phosphorylation residues provided new insights into the CAMK2-substrate consensus sequences in vivo. This dataset provides an important resource for future studies examining novel roles for CAMK2 in the brain.

Compensating for a sensorimotor delay requires a memory buffer, a state observer, and a predictor
Effective motor control requires sensory feedback about our body's state. However, adjusting motor commands based on sensory feedback is seriously complicated by the time delay between the state of the body at feedback generation and the arrival of the feedback-informed motor commands in the body's muscles, the so-called sensorimotor delay (SMD). In this paper, I investigate the effect of a SMD on the control performance of a computational system that represents the central nervous system. I specify three SMD compensation mechanisms that operate the level of this computational system: memory-informed state estimation, prediction, and output gain adjustment. I use formal analysis and computer simulations of balance control, both free standing and while riding a bicycle. This study provides strong evidence for the claim that SMD compensation requires a memory buffer of efference copies, a state observer, and a predictor. In addition, for prediction to be an effective SMD compensation mechanism, it needs (1) the output of a state observer as its initial condition and (2) a memory buffer of efference copies that must be convolved with a set of weights that is to be learned. For output gain adjustment to make a substantial contribution to SMD compensation, memory-informed state estimation and prediction must be imperfect.

Dynamic Changes in Chloride Homeostasis Coordinate Midbrain Inhibitory Network Activity during Reward Learning
The ability to associate environmental stimuli with positive outcomes is a fundamental form of learning. While extensive research has focused on the response profiles of midbrain dopamine neurons during associative learning, less is known about learning-mediated changes in the afferents that shape their responses. We demonstrate that during critical phases of learning, anion homeostasis in midbrain GABA neurons - a primary source of input to dopamine neurons - is disrupted due to downregulation of the chloride transporter KCC2. This alteration in GABA neurons preferentially impacted lateral mesoaccumbal dopamine pathways and was not observed after learning was established. At the network level, learning-mediated KCC2 downregulation was associated with enhanced synchronization between individual GABA neurons and increased dopamine responses to reward-related stimuli. Conversely, enhancing KCC2 function during learning reduced GABA synchronization, diminished relevant dopamine signaling, and prevented cue-reward associations. Thus, circuit-specific adaptations in midbrain GABA neurons are crucial for forming new reward-related behaviors.

Profiling Human iPSC-Derived Sensory Neurons for Analgesic Drug Screening Using a Multi-Electrode Array
Chronic pain is a major global health issue, yet effective treatments are limited by poor translation from preclinical studies to humans. To address this, we developed a high-content screening (HCS) platform for analgesic discovery using hiPSC-derived nociceptors. These cells were cultured on multi-well micro-electrode arrays to monitor activity, achieving nearly 100% active electrodes by week two, maintaining stable activity for at least two weeks. After maturation (28 days), we exposed the nociceptors to various drugs, assessing their effects on neuronal activity, with excellent assay performance (Z' values >0.5). Pharmacological tests showed responses to analgesic targets, including ion channels (Nav, Cav, Kv, TRPV1), neurotransmitter receptors (AMPAR, GABA-R), and kinase inhibitors (tyrosine, JAK1/2). Transcriptomic analysis confirmed the presence of these drug targets, although expression levels varied compared to primary human dorsal root ganglion cells. This HCS platform facilitates the rapid discovery of novel analgesics, reducing the risk of preclinical-to-human translation failure.

Evaluation of near-infrared light therapy for the treatment of neurodegenerative diseases: Limited penetration depth into the brain likely hinders efficacy
Background: Near-infrared (NIR) light therapy is used to treat various musculoskeletal disorders. It has been proposed that transcranial NIR light treatment may also be beneficial for Alzheimer's disease (AD). However, the ability of NIR light to penetrate the scalp and skull efficiently and induce cytoprotective responses in the brain parenchyma has not been sufficiently examined so far. This study aimed to evaluate whether the amount of NIR light that can penetrate through the human skull can cause a biological effect. Methods: Three commercially available devices (a medical laser emitting light at a wavelength of 905 nm and two LED helmets operating at wavelengths of 810 nm and 1070 nm, respectively) were used to measure the NIR light transmittance through human post-mortem skulls with a thermal power sensor. Furthermore, the biological effects of the fraction of light power that passed through the skull were investigated in a human neuronal cell line and in C. elegans. Results: The 905 nm laser achieved transmittances of up to 0.31% (173 u W/cm2) of its input power, and the LED helmets 0.71% (41 u W/cm2; 810 nm) and 0.45% (19 u W/cm2; 1070 nm) of their respective input powers. NIR light exposure at a power density of 134 mW/cm2 was sufficient to activate mitochondrial metabolism in cultured human neurons and C. elegans, as demonstrated by increased cytochrome c oxidase activity and induction of mitochondrial chaperones. However, this stimulatory effect was no longer observed when the applied power density was reduced to 2.5 mW/cm2. Conclusions: More than 99% of the NIR light emitted by the investigated devices was either absorbed or scattered by the human skull. The residual NIR light that would reach underlying brain structures was too weak to elicit biological effects. In conclusion, NIR light treatment is unlikely to be effective to treat brain diseases such as AD due to the low penetrability of the skull.

The infant brain rapidly entrains to visual statistical regularities during stimulus exposure
Statistical learning (SL) has been studied quite extensively in infancy. Still, most evidence relies on post-exposure behavioural tasks whose directionality (familiarity vs. novel effects) may not be straightforward to predict nor to interpret. In addition, these tasks do not tell anything about the online learning dynamics and may be influenced by memory effects. In this work, we investigated online SL mechanisms by tracking neural entrainment to visual regularities in a group of 4- to 6-month-old infants exposed to a stream of shapes presented at 6 Hz. Shapes were either organized in doublets or presented randomly. Results revealed neural entrainment at the base frequency of 6 Hz and harmonics did not differ across conditions, confirming that infants attended to the stimuli similarly across conditions. On the other hand, entrainment at the doublet frequency of 3 Hz and harmonics varied across conditions and trials. Infants showed greater occipital entrainment to the doublet frequency in the deterministic condition than in the random one, especially over the first trials of exposure. This suggests that the brain can detect visual regularities from early infancy. Further, this sensitivity emerged early over the exposure period and did not show a learning curve. Hence, considering its time course and the brain regions involved, neural entrainment at the doublet frequency seems to reflect a bottom-up detection mechanism rather than a learning process. These findings are crucial to better understand how infants extract regularities during stimulus exposure and what neural entrainment can reveal in a visual SL task.

Inter- and Intrahemispheric Sources of Vestibular Signals to V1
Head movements are sensed by the vestibular organs. Unlike classical senses, signals from vestibular organs are not conveyed to a dedicated cortical area but are broadcast throughout the cortex. Surprisingly, the routes taken by vestibular signals to reach the cortex are still largely uncharted. Here we show that the primary visual cortex (V1) receives real-time head movement signals -- direction, velocity, and acceleration -- from the ipsilateral pulvinar and contralateral visual cortex. The ipsilateral pulvinar provides the main head movement signal, with a bias toward contraversive movements (e.g. clockwise movements in left V1). Conversely, the contralateral visual cortex provides head movement signals during ipsiversive movements. Crucially, head movement variables encoded in V1 are already encoded in the pulvinar, suggesting that those variables are computed subcortically. Thus, the convergence of inter- and intrahemispheric signals endows V1 with a rich representation of the animal's head movements.

Neural evidence for decision-making underlying attractive serial dependence
Recall of stimuli is biased by stimulus history, variously manifested as an attractive bias toward or repulsive bias from previous stimuli (i.e., serial dependence). It is unclear when attractive vs repulsive biases arise and if they share neural mechanisms. A recent model of attractive serial dependence proposes a two-stage process in which adaptation causes a repulsive bias during encoding that is later counteracted by an attractive bias at the decision-making stage in a Bayesian-inference-like manner. Neural evidence exists for a repulsive bias at encoding, but evidence for the attractive bias during the response period has been more elusive. We recently [1] showed that while different stimuli in trial history exerted different (attractive or repulsive) serial biases on behavioral reports, during encoding the neural representation of the current item was always repulsively biased. Here we assessed whether this discrepancy between neural and behavioral effects is resolved during subsequent decision-making. Multivariate decoding of magnetoencephalography data during working memory recall showed a neural distinction between attractive and repulsive biases: an attractive neural bias emerged only late in recall. But stimuli that created a repulsive bias on behavior led to a repulsive neural bias early in the recall phase, suggesting that it had already been incorporated earlier. Our results suggest that attractive (but not repulsive) serial dependence arises during decision-making, and that priors that influence post-perceptual decision-making are updated by the previous trial's target, but not by other stimuli.

Challenges and promises in optimising a non-clinical protocol of intracerebroventricular human neural stem cell transplantation in ALS
Background and aims: Neural stem cell (NSC) transplantation holds promising therapeutic potential for neurodegenerative disorders like amyotrophic lateral sclerosis (ALS). However, pre-clinical studies and early-phase clinical trials have faced challenges hindering the effective clinical translation of this approach. Crucial hurdles include the side-effects of prolonged immunosuppression, concerns regarding cell origin and transplantation dosage, identification of the most appropriate therapeutic window, and invasiveness of surgical procedures. Here, we show challenges and promises in optimizing a non-clinical protocol to assess safety and efficacy of human NSC (hNSC) intracerebroventricular (ICV) transplantation for ALS. Methods: We evaluated the safety of administering up to 1x106 hNSCs in immunodeficient mice and assessed their potential efficacy in reducing ALS hallmarks employing the SOD1G93A mouse model. Both, transient (15 days) and prolonged immunosuppression regimens, at low (15 mg/kg) and high (30 mg/kg) doses, were tested along with two different cell dosages (3x105 and 1x106). Results: Bilateral ICV injection of up to 1x106 hNSCs proved to be safe, with no evidence of tumor formation. At 40 days post-transplantation, hNSCs induced a trend toward delaying motor decline and reducing spinal cord (SC) microgliosis when transplanted under prolonged high-dose (30 mg/kg) immunosuppression. Conclusions: Our study suggests that: (i) a bilateral ICV transplantation of 1x106 hNSCs is safe and non-tumorigenic in immunodeficient hosts; (ii) sustained high-dose immunosuppression is essential for ensuring cell survival in immunocompetent mice; and (iii) hNSC transplantation may provide therapeutic benefits in ALS by delaying motor decline and reducing microgliosis. This study also highlights persisting hurdles that need to be further addressed, such as the aggressive murine immune response to exogenous cells.

Quantitative mapping of cerebrovascular reactivity amplitude and delay with breath-hold BOLD fMRI when end-tidal CO2 quality is low
Cerebrovascular reactivity (CVR), the ability of cerebral blood vessels to dilate or constrict in order to regulate blood flow, is a clinically useful measure of cerebrovascular health. CVR is often measured using a breath-hold task to modulate blood CO2 levels during an fMRI scan. Measuring end-tidal CO2 (PETCO2) with a nasal cannula during the task allows CVR amplitude to be calculated in standard units (vascular response per unit change in CO2, or %BOLD/mmHg) and CVR delay to be calculated in seconds. The use of standard units allows for normative CVR ranges to be established and for CVR comparisons to be made across subjects and scan sessions. Although breath-holding can be successfully performed by diverse patient populations, obtaining accurate PETCO2 measurements requires additional task compliance; specifically, participants must breathe exclusively through their nose and exhale immediately before and after each breath hold. Meeting these requirements is challenging, even in healthy participants, and this has limited the translational potential of breath-hold fMRI for CVR mapping. Previous work has focused on using alternative regressors such as respiration volume per time (RVT), derived from respiratory belt measurements, to map CVR. Because measuring RVT does not require additional task compliance from participants, it is a more feasible measure than PETCO2. However, using RVT does not produce CVR in standard units. In this work, we explored how to achieve CVR maps, in standard units, when breath-hold task PETCO2 data quality is low. First, we evaluated whether RVT could be scaled to units of mmHg using a subset of PETCO2 data of sufficiently high quality. Second, we explored whether a PETCO2 timeseries predicted from RVT using deep learning allows for more accurate CVR measurements. Using a dense-mapping breath-hold fMRI dataset, we showed that both rescaled RVT and rescaled, predicted PETCO2 can be used to produce maps of CVR amplitude and delay in standard units with strong absolute agreement to ground-truth maps. However, the rescaled, predicted PETCO2 regressor resulted in superior accuracy for both CVR amplitude and delay. In an individual with regions of increased CVR delay due to Moyamoya disease, the predicted PETCO2 regressor also provided greater sensitivity to pathology than RVT. Ultimately, this work will increase the clinical applicability of CVR in populations exhibiting decreased task compliance.

Microglia-Dependent and Independent Modulation of Brain Lipid Metabolism in Alzheimer's Disease Revealed by Pharmacological and Genetic Microglial Depletion
Abnormal lipid metabolism in Alzheimer's disease (AD) was first documented by Alois Alzheimer in his early observations of patients with a then unrecognized brain disease, which he noted was characterized by a significant presence of "adipose inclusions" or "lipoid granules". Despite this early recognition, until recently the significance of abnormal lipid metabolism in AD has been largely overlooked by the scientific community for decades, highlighting a critical gap in our understanding of this complex disease. In the past decade, numerous loci and genes with genome-wide significant evidence of affecting AD risk have been reported. Notably, a significant portion of these AD risk genes are either preferentially or exclusively expressed by microglia in the brain and/or code for enzymes that directly or indirectly regulate lipid metabolism. This suggests a major, yet uncharacterized, role of microglia in modulating brain lipid metabolism under AD pathological conditions. In our study, we dissected microglia-dependent and independent regulation of lipid metabolism in an AD-like mouse model of amyloidosis, 5xFAD, taking advantage of pharmacological and genetic interventions to eliminate microglia. Using multidimensional mass spectrometry-based shotgun lipidomics (MDMS-SL), we identified overt changes in a number of AD-associated lipids (ADALs) in postmortem patient brains and mouse models of amyloidosis. This included bis(monoacylglycerol)phosphate (BMP), a lipid class enriched in endosomal/lysosomal compartments, and the two most abundant classes of lysophospholipids: lysophosphatidylcholine (LPC) and lysophosphatidylethanolamine (LPE), which are commonly associated with inflammation. Our findings revealed that microglial depletion prevented the accumulation of arachidonic acid-containing BMP species, which are associated with lysosomal activation induced by amyloidosis via a mechanism that involves progranulin, coded by AD risk gene GRN, as shown by targeted transcriptomics, immunoblotting, and immunofluorescence. Surprisingly, AD-associated LPC and LPE accumulation was not driven by microglia. Instead, LPC accumulation correlated with astrocytic activation, while LPE accumulation seems to be associated with oxidative stress. In summary, we uncovered novel microglia-dependent and independent mechanisms that drive lipid dysregulation in AD. These findings may be mechanistically linked with the early glial lipoid deposits described by Dr. Alzheimer.

Behavioral risk models explain locomotor and balance changes when walking at virtual heights
Walking in daily life requires humans to adapt to environments that can influence fear of falling and anxiety about a potential fall. In such environments, individuals may adopt compensatory locomotor and balance changes to maintain a constant expected risk function equal to the product of the probability of some event (e.g., a fall) and the cost of that event (e.g., injury or death). Here, we tested whether locomotor behaviors broadly align with this risk model in two experiments with height-related threats in immersive virtual reality. In Experiment 1, we examined how individuals change their locomotor trajectory while walking along a straight high-elevation walkway. In Experiment 2, we examined how individuals change trajectory and balance control during curved walking where the location of high elevation threat varied. Participants adopted two behaviors that decreased their probability of falling off the edge and aligned with the risk-based model: participants altered their proximity to perceived threats that pose high costs (e.g., a high-elevation ledge), and decreased mediolateral center of mass velocity when that was not possible. Taken together, our results suggest that individuals alter locomotor behavior to change the probability of falling based on the perceived cost of that fall.

Mapping the impact of DUSP6 in an ADHD iPSC derived-dopaminergic neuronal model
Although the gene DUSP6 has been implicated as a risk gene for ADHD in recent GWAS studies, its functional role in the aetiology of the condition remains poorly understood. DUSP6 is reported to regulate dopaminergic neurotransmission by decreasing available synaptic dopamine, suggesting a potential mechanism by which DUSP6 may confer risk to ADHD. In this study, we employed CRISPR-Cas9 to knockout DUSP6 in induced pluripotent stem cells (iPSCs) derived from an individual with ADHD. These isogenic engineered cells were then differentiated into highly homogeneous dopaminergic neurons, including both heterozygous (Het) and homozygous (Hom) knockout lines, along with parental control lines, to assess changes in dopaminergic neurotransmission at both cellular and transcriptomic levels. Enzyme Linked Immunosorbent Assay (ELISA) analyses showed a dose-dependent trend of increasing dopamine levels, at extracellular level, in the knockout lines. RNA sequencing further supported this finding, showing that downregulation of DUSP6 lead to the upregulation of differentially expressed genes (DEGs) associated with the regulation of dopamine secretion and synaptic functions critical for dopamine signalling. These DEGs appear to encode components of the synaptic machinery essential for effective dopamine transmission or act as regulators of dopaminergic neurotransmission. Additionally, RNA sequencing uncovered other potential biological mechanisms through which DUSP6 may contribute to increase the risk for ADHD, including effects on neurogenesis, extracellular matrix-associated processes, lipid metabolism, and sex-specific gene expression. Furthermore, we identified overlaps between DUSP6 knockout DEGs and those associated with other neuropsychiatric disorders, including major depression (MDD), bipolar disorder (BD), and schizophrenia (SCZ), suggesting shared genetic pathways potentially influenced by DUSP6. Together, this study provides deeper insights into the molecular underpinnings of the role played by DUSP6 in the genetic aetiology of ADHD and its broader implications across related neuropsychiatric conditions.

Therapeutic window for treatment of prion disease defined by timed depletion of Prion protein
There is no effective treatment preventing the progression of neurodegenerative diseases such as prion and Alzheimer's diseases. These fatal diseases of the central nervous system, involve progressive accumulation of a misfolded protein long before overt clinical signs of disease. Removal of prion protein early in the pathological process appears to halt the progression however, it is not known whether intervention at later disease stages could be effective. We investigated the potential for intervention throughout the course of prion disease, by developing a mouse model in which Prnp expression can be manipulated in a tissue specific and time dependent manner. Depleting Prnp from neuronal populations in CNS throughout the preclinical phase substantially prolonged incubation. The pathology was dramatically altered to a pattern of astrocytic associated prion deposition. However once overt clinical symptoms of disease were apparent Prnp depletion did not alter disease progression. This study establishes a wide window for intervention, and suggests timely treatment could delay the onset of clinical disease potentially well beyond the lifetime of an individual.

Targeted activation of microglial PPARdelta reprograms immunometabolism and enhances insulin sensitivity in diet-induced obesity.
Microglia play a crucial role in maintaining neuronal health through phagocytosis, a function that becomes compromised during diet-induced obesity and is associated with altered lipid metabolism. Previous research demonstrated that disrupting lipid metabolism in microglia, such as through lipoprotein lipase deficiency, impairs their phagocytic function and exacerbates obesity, glucose dysregulation, and hypothalamic neuron dysfunction. This study investigated whether enhancing lipid metabolism via peroxisome proliferator-activated receptor delta (PPARdelta) activation could counteract obesity-related metabolic disturbances. Thermal proteome profiling identified GW0742 as the most potent PPARdelta ligand among those tested. GW0742 enhanced microglial phagocytosis, reduced inflammation, and shifted energy metabolism towards glycolysis over oxidative phosphorylation. Targeted delivery of GW0742 using nanoparticles (NPs-GW0742) to microglia in the mediobasal hypothalamus of obese rats significantly improved insulin sensitivity without affecting body weight or food intake. Enhanced microglial activation was evidenced by increased soma size and coverage. These findings underscore the importance of microglial lipid metabolism in systemic glucose regulation and highlight the potential of PPARdelta-targeted therapies to mitigate hypothalamic inflammation and improve metabolic health in obesity.

Delta phase-dependent modulation of temporal predictions by parietal transcranial alternating current stimulation
Background: Previous research has shown that temporal prediction processes are associated with phase resets of low-frequency delta oscillations in a network of parietal, sensory and frontal areas during non-rhythmic sensory stimulation. Transcranial alternating current stimulation (tACS) modulates perceptually relevant brain oscillations in a frequency and phase-specific manner, allowing the assessment of their functional qualities in certain cognitive functions like temporal prediction. Objective: We addressed the relation between oscillatory activity and temporal prediction by using tACS to manipulate brain activity in a sinusoidal manner. This enables the investigation of the relevance of low-frequency oscillations' phase for temporal prediction. Methods: Delta tACS was applied over the left and right parietal cortex in two separate unimodal and crossmodal experiments. Participants judged either the visual or the tactile reappearance of a uniformly moving visual stimulus, which shortly disappeared behind and occluder. Using an intermittent electrical stimulation protocol, tACS was applied with six different phase shifts relative to sensory stimulation in both experiments while participants performed a temporal prediction task. Additionally, a computational model was developed and analysed to elucidate oscillation-based functional principles for the generation of temporal predictions. Results: Only in the unimodal experiment, the application of delta tACS resulted in a phase-dependent modulation of temporal prediction performance. By considering the effect of sustained tACS in the computational model, we demonstrate that the entrained dynamics can phase-specifically modulate temporal prediction accuracy. Conclusion: Our findings suggest that phase-specific neuromodulation through delta tACS can influence temporal prediction accuracy in a unimodal context. This provides support to the notion that low-frequency delta oscillations are of causal relevance for temporal prediction.

Connectome-based models of feature selectivity in a cortical circuit
Feature selectivity, the ability of neurons to respond preferentially to specific stimulus configurations, is a fundamental building block of cortical functions. Various mechanisms have been proposed to explain its origins, differing primarily in their assumptions about the connectivity between neurons. Some models attribute selectivity to structured, tuning-dependent feedforward or recurrent connections, whereas others suggest it can emerge within randomly connected networks when interactions are sufficiently strong. This range of plausible explanations makes it challenging to identify the core mechanisms of feature selectivity in the cortex. We developed a novel, data-driven approach to construct mechanistic models by utilizing connectomic data-synaptic wiring diagrams obtained through electron microscopy-to minimize preconceived assumptions about the underlying connectivity. With this approach, leveraging the MICrONS dataset, we investigate the mechanisms governing selectivity to oriented visual stimuli in layer 2/3 of mouse primary visual cortex. We show that connectome-constrained network models replicate experimental neural responses and point to connectivity heterogeneity as the dominant factor shaping selectivity, with structured recurrent and feedforward connections having a noticeable but secondary effect in its amplification. These findings provide novel insights on the mechanisms underlying feature selectivity in cortex and highlight the potential of connectome-based models for exploring the mechanistic basis of cortical functions.

Disrupted activation of pain networks during attack imagery in patients with episodic migraine
Migraine is a prevalent and disabling brain disorder characterized by recurrent headache attacks alternating with pain-free periods. Patients with migraine exhibit altered brain activation in response to experimental nociceptive stimuli compared to healthy controls, suggesting impaired pain inhibition mechanisms. However, the relevance of these alterations to the underlying pathophysiology of migraine attacks remains unclear. To explore this relationship, we propose a novel pain imagery paradigm aimed to induce brain activation patterns associated with migraine attacks specifically during pain-free. The task required patients to alternate between imagining a severe migraine attack and being headache free, while healthy controls imagined severe physical head pain (e.g., toothache) and pain relief. We collected fMRI data from 14 episodic migraine without aura patients in the interictal phase and 15 healthy controls performing this task. Both patients and controls activated pain-related brain areas during imagery of pain relative to pain relief. Moreover, controls also activated the medial pain system associated with pain inhibition and attentional modulation of pain via top-down pathways. In contrast, patients significantly deactivated these areas, namely the anterior cingulate cortex and dorsolateral prefrontal cortex. Collectively, our findings indicate altered functioning of pain networks in migraine patients, suggesting a disease-related dysregulation of pain inhibition. Eventually, the proposed attack imagery paradigm may provide a promising alternative to studies of pain mechanisms in migraine research.

Multimodal laminar characterization of visual areas along the cortical hierarchy
Understanding the relationship between brain structure and function is a central goal in neuroscience. While post-mortem studies using microscopic techniques have provided detailed insights into the brain's cytoarchitectonic and myeloarchitectonic patterns, linking these structural findings to functional outcomes remains challenging. Magnetic resonance imaging (MRI) has emerged as a powerful non-invasive tool for studying both structure and function, but discrepancies in spatial resolution between structural and functional imaging, especially in layer-fMRI, complicate the interpretation of functional results. In this study, we explore how visual cortical hierarchy relates to microscopic and mesoscopic laminar features. Focusing on visual areas that span progressive hierarchical levels, V1, V2, V3, and hMT+, we apply a multimodal approach combining post-mortem histology, post-mortem and in-vivo quantitative MRI (qMRI), and resting-state layer-fMRI. Using the open-access post-mortem AHEAD dataset, which integrates histological and qMRI contrasts from the same brain samples, we bridge microscopic observations with qMRI data. In parallel, we incorporate high-resolution qR2* MRI and resting-state layer-fMRI from the same participant, allowing for a comparative analysis of laminar profiles across cortical depth. For computing laminar profiles, we developed an analysis pipeline that bridges histology images, mesoscopic qMRI, and layer-fMRI. Our findings highlight parvalbumin laminar profiles (reflecting interneuron parvalbumin density) as the most discriminative feature for differentiating brain areas. Additionally, we report laminar quantitative T2* (as 1/R2*) profiles from post-mortem and in-vivo data, together with T2*-weighted resting-state layer-fMRI, all of which exhibit a similar overall shape across modalities. Using our methodological framework, a similar laminar characterization can be extended to study other brain regions. Generative models for layer fMRI will benefit from incorporating these new empirical microstructural (parvalbumin) and physical quantitative (qR2*) data, leading to more area-specific and accurate models.

GABA and glycine synaptic release on axotomized motoneuron cell bodies promotes motor axon regeneration
Motor axon regeneration after traumatic nerve injuries is a slow process that adversely influences patient outcomes because muscle reinnervation delays result in irreversible muscle atrophy and suboptimal axon regeneration. This advocates for investigating methods to accelerate motor axon growth. Electrical nerve stimulation and exercise both enhance motor axon regeneration in rodents and patients, but these interventions cannot always be easily implemented. A roadblock to uncover novel therapeutic approaches based on the effects of activity is the lack of understanding of the synaptic drives responsible for activity-mediated facilitation of axon regeneration. We hypothesized that the relevant excitatory inputs facilitating axon regrowth originate in GABA/glycine synapses which become depolarizing after downregulation of the potassium chloride cotransporter 2 in motoneurons following axotomy. To test this, we injected tetanus toxin (TeTx) in the tibialis anterior (TA) muscle of mice to block the release of GABA/glycine specifically on TA motoneurons. Thereafter, we axotomized all sciatic motoneurons by nerve crush and analyzed the time-courses of muscle reinnervation in TeTx-treated (TA) and untreated (lateral gastrocnemius, LG) motoneurons. Muscle reinnervation was slower in TA motoneurons with blocked GABA/glycine synapses, as measured by recovery of M-responses and anatomical reinnervation of neuromuscular junctions. Post-hoc immunohistochemistry confirmed the removal of the vesicular associated membrane proteins 1 and 2 by TeTx activity, specifically from inhibitory synapses. These proteins are necessary for exocytotic release of neurotransmitters. Therefore, we conclude that GABA/glycine neurotransmission on regenerating motoneurons facilitates axon growth and muscle reinnervation and discuss possible interventions to modulate these inputs on regenerating motoneurons.

Dendritic synaptic integration modes under in vivo-like states
The neural code remains undiscovered and understanding synaptic input integration under in vivo-like conditions is just the initial step toward unraveling it. Synaptic signals generate fast dendritic spikes through two main modes of temporal summation: coincidence detection and integration. In coincidence detection, dendrites fire only when multiple incoming signals arrive in rapid succession, whereas integration involves summation of postsynaptic potentials over longer periods with minimal membrane leakage. This process is influenced by ionic properties, especially as the membrane potential approaches the firing threshold, where inactivating currents play a critical role. However, the modulation of temporal summation by these currents under in vivo-like conditions has not been thoroughly studied. In our research, we used computer simulations to investigate how three inactivating currents - A-type potassium, T-type calcium, and transient sodium - affect temporal summation. We found that calcium and sodium currents promote integrative behavior in dendrites, while potassium currents enhance their ability to act as coincidence detectors. By adjusting the levels of these currents in dendrites, neurons can flexibly switch between integration and coincidence detection modes, providing them with a versatile mechanism for complex tasks like multiplexing. This flexibility could be key to understanding how neural circuits process information in real time.

Coming up short: generative network models fail to accurately capture long-range connectivity
Generative network models (GNMs) have been proposed to identify the mechanisms/constraints that shape the organisation of the connectome. These models most commonly parameterise the formation of inter-regional axonal connections using a trade-off between connection cost and some measure of topological complexity or functional value. Despite its simplicity, GNMs can generate synthetic networks that capture many topological properties of empirical brain networks. However, current models often fail to capture the spatial embedding-i.e., the topography-of many such properties, such as the anatomical location of network hubs. In this study, we investigate a diverse array of generative network model formulations and find that none can accurately capture empirical patterns of long-range connectivity. We demonstrate that the spatial embedding of long-range connections is critical in defining hub locations and that it is precisely these connections that are poorly captured by extant models. We further show how standard metrics used for model optimisation and evaluation fail to capture the true correspondence between synthetic and empirical brain networks. Overall, our findings demonstrate common failure modes of GNMs, identify why these models do not fully capture brain network organisation, and suggest ways the field can move forward to address these challenges.

Microglia States are Susceptible to Senescence and Cholesterol Dysregulation in Alzheimer's Disease
Cellular senescence is a major contributor to aging-related degenerative diseases, including Alzheimer's disease (AD). However, much less is known about the key cell types and pathways driving mechanisms of senescence in the brain. We hypothesized that dysregulated cholesterol metabolism is central to cellular senescence in AD. We analyzed whole transcriptomic data and utilized single-cell RNA seq integration techniques to unveil the convoluted cell-type-specific and sub-cell-type-state-specific senescence pathologies in AD using both ROSMAP and Sea-AD datasets. We identified that microglia are central components to AD-associated senescence phenotypes in ROSMAP snRNA-seq data (982,384 nuclei from postmortem prefrontal cortex of 239 AD and 188 non-AD) among non-neuron cell types. We identified that homeostatic, inflammatory, phagocytic, lipid processing and neuronal surveillance microglia states were associated with AD-associated senescence in ROSMAP (152,459 microglia nuclei from six regions of brain tissue of 138 early AD, 79 late AD and 226 control subjects) and in Sea-AD (82,486 microglia nuclei of 42 dementia, 42 no dementia and 5 reference subjects) via integrative analysis, which preserves the meaningful biological information of microglia cell states across the datasets. We assessed top senescence-associated bioprocesses including mitochondrial, apoptosis, oxidative stress, ER stress, endosomes, and lysosomes systems. Specifically, we found that senescent microglia have altered cholesterol-related bioprocesses and dysregulated cholesterol. We discovered three gene co-expression modules representing the specific cholesterol-related senescence transcriptomic signatures in postmortem brains. To validate these findings, the activation of specific cholesterol associated senescence transcriptomic signatures was assessed using integrative analysis of snRNA-seq data from iMGs (microglia induced from iPSCs) exposed to myelin, Abeta, and synaptosomes (56,454 microglia across two replicates of untreated and four treated groups). In vivo cholesterol-associated senescence transcriptomic signatures were preserved and altered after treatment with AD pathological substrates in iMGs. This study provides the first evidence that dysregulation of cholesterol metabolism in microglia is a major driver of senescence pathologies in AD. Targeting cholesterol pathways in senescent microglia is an attractive strategy to slow down AD progression.

Evaluation of Brain Age as a Specific Marker of Brain Health
Brain age is a powerful marker of general brain health. Furthermore, brain age models are trained on large datasets, thus giving them a potential advantage in predicting specific outcomes - much like the success of finetuning large language models for specific applications. However, it is also well-accepted in machine learning that models trained to directly predict specific outcomes (i.e., direct models) often perform better than those trained on surrogate outcomes. Therefore, despite their much larger training data, it is unclear whether brain age models outperform direct models in predicting specific brain health outcomes. Here, we compare large-scale brain age models and direct models for predicting specific health outcomes in the context of Alzheimer's Disease (AD) dementia. Using anatomical T1 scans from three continents (N = 1,848), we find that direct models outperform brain age models without finetuning. Finetuned brain age models yielded similar performance as direct models, but importantly, did not outperform direct models although the brain age models were pretrained on 1000 times more data than the direct models: N = 53,542 vs N = 50. Overall, our results do not discount brain age as a useful marker of general brain health. However, in this era of large-scale brain age models, our results suggest that small-scale, targeted approaches for extracting specific brain health markers still hold significant value.

Dual-timescale motor circuit dynamics underlies fast head exploratory behavior and efficient locomotion
Caenorhabditis elegans exhibits complex head exploratory behavior in natural environments. We quantified these movements and examined the motor circuits responsible for their intricate dynamics. Using variational mode decomposition, we distinguished between fast casts and slow bends. Slow bends backpropagate along the body, whereas fast casts show phase-specific timing and influence directional bias during forward movement. Combinatorial ablations of three types of cholinergic motor neurons, coupled with dynamical systems analysis of subsequent behaviors, revealed their distinct and overlapping roles. RMD contribute to head casts; SMD maintain bending states; SMB and SMD enable slow rhythmic bending and head-body coupling; RMD, SMD, and SMB together constitute the head central pattern generator (CPG) that drives forward locomotion. We propose a computational model with dual-timescale proprioceptive feedback to reproduce fast casts and slow bends and demonstrate how phase-specific head casts enhance roaming efficiency by optimizing movement speed. These findings highlight how different excitatory motor neurons control distinct and complementary aspects of head bending dynamics while working synergistically to maximize locomotion efficiency.

ELAVL3 regulates splicing of RNAs encoding synaptic signaling proteins in D1 and D2 striatal medium spiny neurons.
The neuronal RNA-binding protein (RBP) family nELAVL regulates key neuronal processes by binding directly to target RNA transcripts. In this study, we demonstrate that ELAVL3 is the predominant nELAVL paralog expressed in D1 and D2 medium spiny neurons of the striatum. To investigate its function, we developed ELAVL3 cTag-crosslinking and immunoprecipitation (CLIP) to generate RBP-RNA interaction maps from these neurons. By integrating data from ELAVL3-cTag and Elavl3 knockout mice, we identified distinct regulatory effects of ELAVL3 on alternative splicing of its target transcripts. Notably, ELAVL3 modulates splicing of transcripts encoding proteins critical for glutamate and dopamine receptor signaling. These findings underscore the role of ELAVL3 in RNA-mediated regulation of molecular pathways essential for medium spiny neuron function in the striatum.

Molecular Polymorphism of tau aggregates in Pick's disease
Tau protein plays a central role in many neuropathies. The trajectory by which tau spreads through neural networks is disease-specific but the events driving progression are unknown. This is due in part to the challenge of characterizing tau aggregates in situ. We address that challenge using in situ micro-x-ray diffraction (micro XRD) and micro-X-ray fluorescence (XRF) to examine tau lesions in the brain of a 79-year-old male with dementia. Neuropathological examination revealed classical forms of tau in the hippocampal formation: extensive Pick bodies in the granular layer; modest numbers of neurofibrillary tangles and dystrophic neurites in the CA4 and hilus. Micro XRD indicated that Pick bodies are low in fibril content, whereas neurofibrillary lesions within adjacent tissue exhibit far greater density of fibrillar tau. XRF demonstrated elevated levels of zinc, calcium and phosphorous in all tau-containing lesions whereas sulfur deposition was greatest in lesions exhibiting high fibrillar content. Correlation of lesion morphology with anatomical localization, tau fibrillation and differential elemental accumulation suggests tau fibrils generate biochemically distinct microenvironments that influence lesion morphology, tau seed formation and spreading.