Hebbian Research Articles

A PARP14/TARG1-Regulated RACK1 MARylation Cycle Drives Stress Granule Dynamics in Ovarian Cancer Cells.

Mono(ADP-ribosyl)ation (MARylation) is emerging as a critical regulator of ribosome function and translation. Herein, we demonstrate that RACK1, an integral component of the ribosome, is MARylated on three acidic residues by the mono(ADP-ribosyl) transferase (MART) PARP14 in ovarian cancer cells. MARylation of RACK1 is required for stress granule formation and promotes the colocalization of RACK1 in stress granules with G3BP1, eIF3η, and 40S ribosomal proteins. In parallel, we observed reduced translation of a subset of mRNAs, including those encoding key cancer regulators (e.g., AKT). Treatment with a PARP14 inhibitor or mutation of the sites of MARylation on RACK1 blocks these outcomes, as well as the growth of ovarian cancer cells in culture and in vivo. To re-set the system after prolonged stress and recovery, the ADP-ribosyl hydrolase TARG1 deMARylates RACK1, leading to the dissociation of the stress granules and the restoration of translation. Collectively, our results demonstrate a therapeutically targetable pathway that controls stress granule assembly and disassembly in ovarian cancer cells.

Bioinspired Hyaluronic Acid‐Based Hydrogel Fuels Bi‐Directional Lung Organoid Maturation via PIEZO1 and ITGB1 Mediated Mechanosensation

AbstractLung diseases are one of the leading causes of global mortality. Advances in induced pluripotent stem cell (iPSC) differentiation have enabled the creation of bronchiolar and alveolar lung organoids, advancing research on lung conditions. Traditional Matrigel encapsulation, reliant on the spontaneous assembly and propagation of cells with limited external intervention, often results in variability and low reproducibility. The absence of hyaluronic acid (HA) in Matrigel, a key lung extracellular matrix component, limits bronchiolar and alveolar cell differentiation, reducing the efficacy and reproducibility of iPSC‐derived organoid generation. To address this, a novel hybrid hydrogel combining HA and 23% Matrigel, inspired by the natural lung environment, is developed. This hydrogel offers improved biochemical support and viscoelastic properties, significantly accelerating organoid development. Within eight days, the hydrogel produces uniformly sized organoids containing both bronchiolar and alveolar epithelial cells. Increased levels of active mechanosensors and transducers, including PIEZO1, Integrin, and Myosin, suggest that the hydrogel's altered viscoelasticity triggers a mechanotransduction cascade. This bioinspired hydrogel provides a robust, fast model for biomedical research, facilitating rapid drug screening, respiratory disease treatment studies, and surfactant trafficking investigations. Furthermore, it enables the exploration of underlying biomechanical mechanisms to enhance the controllability of organoid generation and maturation.

A developmental model of audio-visual attention (MAVA) for bimodal language learning in infants and robots

A social individual needs to effectively manage the amount of complex information in his or her environment relative to his or her own purpose to obtain relevant information. This paper presents a neural architecture aiming to reproduce attention mechanisms (alerting/orienting/selecting) that are efficient in humans during audiovisual tasks in robots. We evaluated the system based on its ability to identify relevant sources of information on faces of subjects emitting vowels. We propose a developmental model of audio-visual attention (MAVA) combining Hebbian learning and a competition between saliency maps based on visual movement and audio energy. MAVA effectively combines bottom-up and top-down information to orient the system toward pertinent areas. The system has several advantages, including online and autonomous learning abilities, low computation time and robustness to environmental noise. MAVA outperforms other artificial models for detecting speech sources under various noise conditions.

Mechanical Characterization of Electrolyzer Membranes and Components Under Compression

Proton-exchange membrane (PEM) water electrolysis is a promising technology for producing clean hydrogen by electrochemically splitting water when paired with renewable energy sources. A major roadblock to improving electrolyzer durability is the mechanical degradation of the cell components, which requires an understanding of their mechanical response under device-relevant conditions. However, there is a lack of studies on the mechanical characterization of the PEM and other components, as well as and their interactions. This study aims to address this gap by using a custom-designed testing apparatus to investigate the mechanics of electrolyzer components in uniaxial compression at 25 and 80 °C. Findings show stress-strain response of components have a varying degree of nonlinearity owing to their distinct deformation mechanisms and morphologies, from porous structures to polymers. These results are used to develop an expression for compressive stress-strain response of Nafion membranes and then analyze the deformation of components under applied pressure by using a 1-D spring network model of cell assembly. This work provides a new understanding of mechanical responses of the electrolyzer membrane and cell components, which can help assess material design and cell assembly strategies for improved electrolyzer durability.

Ultrahigh Pressure Generation at High Temperatures in a Walker-Type Large-Volume Press and Multiple Applications

Ultrahigh pressure generation at high temperatures is technologically challenging for large sample volumes. In this study, we successfully generated pressures of 37.3–40.4 GPa at 1900–2100 K in a Walker-type large-volume press (LVP). Expansion of the pressure range at high temperatures was achieved by adapting newly designed ZK01F tungsten carbide (WC) anvils with tapered surfaces and using cell assemblies with an ∼1 mm3 sample volume and hard materials, as well as by applying certain adjustments to the apparatus. The pressure efficiencies of the different types of WC anvils and cell assemblies were also studied. Using the above-mentioned techniques, we successfully synthesized and characterized bulk samples of nearly pure sp3-hybridized ultrahard amorphous carbon, core–shell nanocrystals with high Néel temperatures, as well as large-sized single crystals of lower-mantle minerals. The developed LVP techniques presented here could enable the exploration of the chemical and physical properties of novel materials and Earth’s interior.

Ninein domains required for its localization, association with partners dynein and ensconsin, and microtubule organization.

Ninein (Nin) is a microtubule (MT) anchor at the subdistal appendages of mother centrioles and the pericentriolar material (PCM) of centrosomes that also functions to organize MTs at noncentrosomal MT-organizing centers (ncMTOCs). In humans, the NIN gene is mutated in Seckel syndrome, an inherited developmental disorder. Here, we dissect the protein domains involved in Nin's localization and interactions with dynein and ensconsin (ens/MAP7) and show that the association with ens cooperatively regulates MT assembly in Drosophila fat body cells. We define domains of Nin responsible for its localization to the ncMTOC on the fat body cell nuclear surface, localization within the nucleus, and association with Dynein light intermediate chain (Dlic) and ens, respectively. We show that Nin's association with ens synergistically regulates MT assembly. Together, these findings reveal novel features of Nin function and its regulation of a ncMTOC.

Phase-modulated pulsing coherent acoustic tweezers for dynamic adjustable spaced particle patterning

Abstract Acoustic tweezer technique with non-contact micro-object manipulation capability has been widely utilized for particle patterning in the fields of cell arraying, target drug delivery, and functional composite fabrication. However, the manipulation resolution of acoustic tweezer device is decided by the configuration of interdigital transducers (IDTs), which is half the wavelength of surface acoustic waves (SAWs). This limitation severely reduces the manipulation versatility of acoustic tweezer devices. To address this limitation, we propose a novel approach that employs phase-modulated pulsing coherent SAWs. This method can halve the patterning spacing and allows for customization of the spacing based on specific manipulation demands. Our proposed technique increases the flexibility and manipulation resolution of acoustic tweezer device, and thus can be widely applied to reconfigurable assembly and precise manipulation of particles and cells.

Searching for long timescales without fine tuning.

Animal behavior occurs on timescales much longer than the response times of individual neurons. In many cases, it is plausible that these long timescales emerge from the recurrent dynamics of electrical activity in networks of neurons. In linear models, timescales are set by the eigenvalues of a dynamical matrix whose elements measure the strengths of synaptic connections between neurons. It is not clear to what extent these matrix elements need to be tuned to generate long timescales; in some cases, one needs not just a single long timescale but a whole range. Starting from the simplest case of random symmetric connections, we combine maximum entropy and random matrix theory methods to construct ensembles of networks, exploring the constraints required for long timescales to become generic. We argue that a single long timescale can emerge generically from realistic constraints, but a full spectrum of slow modes requires more tuning. Langevin dynamics that generates patterns of synaptic connections drawn from these ensembles involves a combination of Hebbian learning and activity-dependent synaptic scaling.

NST3 induces ectopic transdifferentiation, forming secondary walls with diverse patterns and composition in Arabidopsis thaliana.

The master transcription factor NAC SECONDARY WALL THICKENING PROMOTING FACTOR3 (NST3), also known as SND1, plays a pivotal role in regulating secondary cell wall (SCW) development in interfascicular and xylary fibers in Arabidopsis thaliana. Despite progress in understanding SCW assembly in xylem vessel-like cells, the mechanisms behind its assembly across different cell types remain unclear. Overexpressing NST3 or its homolog NST1 leads to reduced fertility, posing challenges for studying their impact on secondary wall formation. This study aimed at developing a tightly regulated dexamethasone (DEX)-inducible expression system for NST3 and NST1 to elucidate the structure and assembly of diverse SCWs. Using the DEX-inducible system, we characterized ectopically formed SCWs for their diverse patterns, mesoscale organization, cellulose microfibril orientation, and molecular composition using spinning disk confocal microscopy, field emission scanning electron microscopy (FESEM), vibrational sum-frequency generation (SFG) spectroscopy and, histochemical staining and time-of-flight secondary ion mass spectrometry (ToF-SIMS), respectively. Upon DEX treatment, NST3 and NST1 transgenic hypocotyls underwent time-dependent transdifferentiation, progressing from protoxylem-like to metaxylem-like cells. NST3-induced plants exhibited normal growth but had rough secondary wall surfaces with delaminating S2 and S3 layers. Mesoscale examination of induced SCWs in epidermal cells revealed that macrofibril thickness and orientation were comparable to xylem vessels, while wall thickness resembled that of interfascicular fibers. Additionally, induced epidermal cells formed SCWs with altered cellulose and lignin contents. These findings suggest NST3 and/or NST1 induce SCWs with shared characteristics of both xylem and fiber-like cells forming loosely arranged cell wall layers and cellulose organized at multiple angles relative to the cell growth axis and with varied cellulose and lignin abundance. This inducible system opens avenues to explore ectopic SCWs for bioenergy and bioproducts, offering valuable insights into SCW patterning across diverse cell types and developmental stages.

HfAlOx-based ferroelectric memristor for nociceptor and synapse functions.

Efficient data processing is heavily reliant on prioritizing specific stimuli and categorizing incoming information. Within human biological systems, dorsal root ganglions (particularly nociceptors situated in the skin) perform a pivotal role in detecting external stimuli. These neurons send warnings to our brain, priming it to anticipate potential harm and prevent injury. In this study, we explore the potential of using a ferroelectric memristor device structured as a metal-ferroelectric-insulator-semiconductor as an artificial nociceptor. The aim of this device is to electrically receive external damage and interpret signals of danger. The TiN/HfAlOx (HAO)/HfSiOx (HSO)/n+ Si configuration of this device replicates the key functions of a biological nociceptor. The emulation includes crucial aspects, such as threshold reactivity, relaxation, no adaptation, and sensitization phenomena known as "allodynia" and "hyperalgesia." Moreover, we propose establishing a connection between nociceptors and synapses by training the Hebbian learning rule. This involves exposing the device to injurious stimuli and using this experience to enhance its responsiveness, replicating synaptic plasticity.

Dynamic control of sequential retrieval speed in networks with heterogeneous learning rules

Temporal rescaling of sequential neural activity has been observed in multiple brain areas during behaviors involving time estimation and motor execution at variable speeds. Temporally asymmetric Hebbian rules have been used in network models to learn and retrieve sequential activity, with characteristics that are qualitatively consistent with experimental observations. However, in these models sequential activity is retrieved at a fixed speed. Here, we investigate the effects of a heterogeneity of plasticity rules on network dynamics. In a model in which neurons differ by the degree of temporal symmetry of their plasticity rule, we find that retrieval speed can be controlled by varying external inputs to the network. Neurons with temporally symmetric plasticity rules act as brakes and tend to slow down the dynamics, while neurons with temporally asymmetric rules act as accelerators of the dynamics. We also find that such networks can naturally generate separate ‘preparatory’ and ‘execution’ activity patterns with appropriate external inputs.

The glycolytic metabolite methylglyoxal covalently inactivates the NLRP3 inflammasome

The NLRP3 inflammasome promotes inflammation in disease, yet the full repertoire of mechanisms regulating its activity is not well delineated. Among established regulatory mechanisms, covalent modification of NLRP3 has emerged as a common route for the pharmacological inactivation of this protein. Here, we show that inhibition of the glycolytic enzyme phosphoglycerate kinase 1 (PGK1) results in the accumulation of methylglyoxal, a reactive metabolite whose increased levels decrease NLRP3 assembly and inflammatory signaling in cells. We find that methylglyoxal inactivates NLRP3 via a non-enzymatic, covalent-crosslinking-based mechanism, promoting inter- and intraprotein MICA (methyl imidazole crosslink between cysteine and arginine) posttranslational linkages within NLRP3. This work establishes NLRP3 as capable of sensing a host of electrophilic chemicals, both exogenous small molecules and endogenous reactive metabolites, and suggests a mechanism by which glycolytic flux can moderate the activation status of a central inflammatory signaling pathway.

Recent Advances in Dielectrophoretic Manipulation and Separation of Microparticles and Biological Cells.

Dielectrophoresis (DEP) is an advanced microfluidic manipulation technique that is based on the interaction of polarized particles with the spatial gradient of a non-uniform electric field to achieve non-contact and highly selective manipulation of particles. In recent years, DEP has made remarkable progress in the field of microfluidics, and it has gradually transitioned from laboratory-scale research to high-throughput manipulation in practical applications. This paper reviews the recent advances in dielectric manipulation and separation of microparticles and biological cells and discusses in detail the design of chip structures for the two main methods, direct current dielectrophoresis (DC-DEP) and alternating current dielectrophoresis (AC-DEP). The working principles, technical implementation details, and other improved designs of electrode-based and insulator-based chips are summarized. Functional customization of DEP systems with specific capabilities, including separation, capture, purification, aggregation, and assembly of particles and cells, is then performed. The aim of this paper is to provide new ideas for the design of novel DEP micro/nano platforms with the desired high throughput for further development in practical applications.

Hebbian spatial encoder with adaptive sparse connectivity

Biologically plausible neural networks have demonstrated efficiency in learning and recognizing patterns in data. This paper proposes a general online unsupervised algorithm for spatial data encoding using fast Hebbian learning. Inspired by the Hierarchical Temporal Memory (HTM) framework, we introduce the SpatialEncoder algorithm, which learns the spatial specialization of neurons’ receptive fields through Hebbian plasticity and k-WTA (k winners take all) inhibition. A key component of our model is a two-part synaptogenesis algorithm that enables the network to maintain a sparse connection matrix while adapting to non-stationary input data distributions. In the MNIST digit classification task, our model outperforms the HTM SpatialPooler in terms of classification accuracy and encoding stability. Compared to another baseline, a two-layer artificial neural network (ANN), our model achieves competitive classification accuracy with fewer iterations required for convergence. The proposed model offers a promising direction for future research on sparse neural networks with adaptive neural connectivity.

Spatial sensorimotor mismatch between the motor command and somatosensory feedback decreases motor cortical excitability. A transcranial magnetic stimulation-virtual reality study.

Effective control of movement predominantly depends on the exchange and integration between sensory feedback received by our body and motor command. However, the precise mechanisms governing the adaptation of the motor system's response to altered somatosensory signals (i.e., discrepancies between an action performed and feedback received) following movement execution remain largely unclear. In order to address these questions, we developed a unique paradigm using virtual reality (VR) technology. This paradigm can induce spatial incongruence between the motor commands executed by a body district (i.e., moving the right hand) and the resulting somatosensory feedback received (i.e., feeling touch on the left ankle). We measured functional sensorimotor plasticity in 17 participants by assessing the effector's motor cortical excitability (right hand) before and after a 10-min VR task. The results revealed a decrease in motor cortical excitability of the movement effector following exposure to a 10-min conflict between the motor output and the somatosensory input, in comparison to the control condition where spatial congruence between the moved body part and the area of the body that received the feedback was maintained. This finding provides valuable insights into the functional plasticity resulting from spatial sensorimotor conflict arising from the discrepancy between the anticipated and received somatosensory feedback following movement execution. The cortical reorganization observed can be attributed to functional plasticity mechanisms within the sensorimotor cortex that are related to establishing a new connection between somatosensory input and motor output, guided by temporal binding and the Hebbian plasticity rule.

Separator engineering: Assisting lithium salt dissociation and constructing LiF-rich solid electrolyte interphases for high-rate lithium metal batteries

Challenges associated with lithium dendrite growth and the formation of dead lithium significantly limit the achievable energy density of lithium metal batteries (LMBs), particularly under high operating current densities. Our innovative design employs a state-of-the-art 2500 separator featuring a meticulously engineered cellulose acetate (CA) coating (CA@2500) to suppress dendrite nucleation and propagation. The CO functional groups in CA enhances charge transfer kinetics and triggering the decomposition of lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), which leads to the formation of a more robust solid electrolyte interphase (SEI) composed primarily of LiF. Moreover, the introduction of polar functional groups in the CA enhances the separator’s hydrophilic properties, facilitating the uniform Li+ flux and creating a conductive pathway for efficient lithium migration. As a result, the CA@2500 separator exhibits a high lithium-ion transfer number (0.88) and conductivity. The lithium symmetric cell assembles with the CA@2500 separator displays a stable cycling performance over 5500 h at a current density and capacity of 10 mA cm−2 and 10 mAh cm−2, respectively. Additionally, LPF battery with CA@2500 separator shows an excellent capacity retention at 0.2 C with an average decay of 0.055 % per cycle. Moreover, a high capacity of 105 mAh g−1 is maintained after 500 cycles at 5 C with an average decay of only 0.027 % per cycle. This work achieved high stability of LMBs through simplified engineering.

Three-dimensional cellular structures for viscous and thermal energy control in acoustic and thermoacoustic applications

The multi-scale asymptotic method provides a separate description of the viscous and thermal response functions governing acoustic waves propagation through porous materials. However, these response functions are inherently interdependent for simple porous structures such as slits or tubes – their determination being directly influenced by the microstructural features of the geometry. This study aims to identify, characterize, and realize a microgeometry providing the ability to independently modify the viscous and thermal behaviours. A relative autonomy in the control of these phenomena would make it possible to regulate energy conversion processes within the structure, which can be either dissipative (e.g., sound absorption) or generative (e.g., thermoacoustic gain). Among the various microgeometries, cellular solids made by an assembly of cells with solid edges or faces, packed together so that they fill space, emerge as promising candidates for achieving such a goal. Within the interconnected cells (pores), the viscous losses are governed by the aperture sizes of the faces (throat section). On the other hand, the thermal exchanges are closely related to the interface between the fluid and the solid and consequently to the dimensions of the cells. The identified microstructure is a typical Kelvin cell-based geometry, modified to account for manufacturing constraints, and characterized by faces with well-defined opening ratio and thickness. This work presents a model, experimentally validated, to predict the transport parameters of such cellular solids. It provides a valuable tool for designing microstructures having the attributes to independently tune thermal and viscous effects for specific application requirements.

Hebbian Descent: A Unified View on Log-Likelihood Learning.

This study discusses the negative impact of the derivative of the activation functions in the output layer of artificial neural networks, in particular in continual learning. We propose Hebbian descent as a theoretical framework to overcome this limitation, which is implemented through an alternative loss function for gradient descent we refer to as Hebbian descent loss. This loss is effectively the generalized log-likelihood loss and corresponds to an alternative weight update rule for the output layer wherein the derivative of the activation function is disregarded. We show how this update avoids vanishing error signals during backpropagation in saturated regions of the activation functions, which is particularly helpful in training shallow neural networks and deep neural networks where saturating activation functions are only used in the output layer. In combination with centering, Hebbian descent leads to better continual learning capabilities. It provides a unifying perspective on Hebbian learning, gradient descent, and generalized linear models, for all of which we discuss the advantages and disadvantages. Given activation functions with strictly positive derivative (as often the case in practice), Hebbian descent inherits the convergence properties of regular gradient descent. While established pairings of loss and output layer activation function (e.g., mean squared error with linear or cross-entropy with sigmoid/softmax) are subsumed by Hebbian descent, we provide general insights for designing arbitrary loss activation function combinations that benefit from Hebbian descent. For shallow networks, we show that Hebbian descent outperforms Hebbian learning, has a performance similar to regular gradient descent, and has a much better performance than all other tested update rules in continual learning. In combination with centering, Hebbian descent implements a forgetting mechanism that prevents catastrophic interference notably better than the other tested update rules. When training deep neural networks, our experimental results suggest that Hebbian descent has better or similar performance as gradient descent.

Retinal neuroanatomy of two emerging model organisms, the spiny mouse (Acomys dimidiatus) and the Mongolian gerbil (Meriones unguiculatus)

Current research using animal models to investigate retinal cell biology and model retinal degenerative diseases largely utilize small mammals that are nocturnal and lack the ability to restore lost vision. In contrast, the Mongolian gerbil (Meriones) is a diurnal rodent with good photopic vision, and the spiny mouse (Acomys) is a small desert-dwelling rodent with remarkable regenerative capabilities. The goal of this study was to identify antibodies that detect retinal cell classes in Meriones and Acomys, and to describe the retinal anatomy of these two species in comparison to outbred laboratory mice (Mus musculus). Immunohistochemistry was performed on retinal sections with antibodies for various retinal cell types. Sections were imaged by light, fluorescence, and confocal microscopy. Cell density, morphology, and placement were compared between species qualitatively and quantitatively. Our analyses revealed a classic assembly of retinal cells in Meriones and Acomys, with a few deviations compared to Mus. Meriones displayed the highest density of cones and Acomys the lowest. A higher density of bipolar cell bodies in the proximal portion of the inner nuclear layer was observed in both Acomys and Meriones compared to Mus, and both species exhibited an increase in amacrine cell density compared to Mus. Our results provide a foundation for future research into the visual system adaptations of these interesting species.

Top-Down Priors Disambiguate Target and Distractor Features in Simulated Covert Visual Search.

Several models of visual search consider visual attention as part of a perceptual inference process, in which top-down priors disambiguate bottom-up sensory information. Many of these models have focused on gaze behavior, but there are relatively fewer models of covert spatial attention, in which attention is directed to a peripheral location in visual space without a shift in gaze direction. Here, we propose a biologically plausible model of covert attention during visual search that helps to bridge the gap between Bayesian modeling and neurophysiological modeling by using (1) top-down priors over target features that are acquired through Hebbian learning, and (2) spatial resampling of modeled cortical receptive fields to enhance local spatial resolution of image representations for downstream target classification. By training a simple generative model using a Hebbian update rule, top-down priors for target features naturally emerge without the need for hand-tuned or predetermined priors. Furthermore, the implementation of covert spatial attention in our model is based on a known neurobiological mechanism, providing a plausible process through which Bayesian priors could locally enhance the spatial resolution of image representations. We validate this model during simulated visual search for handwritten digits among nondigit distractors, demonstrating that top-down priors improve accuracy for estimation of target location and classification, relative to bottom-up signals alone. Our results support previous reports in the literature that demonstrated beneficial effects of top-down priors on visual search performance, while extending this literature to incorporate known neural mechanisms of covert spatial attention.