Search bioRxiv⌕ Search

Biology subjects

Schwarzbach, E.

Publications and source records attributed to Schwarzbach, E..

3 recordsLinked to original sources

Human Peripheral Nerve-on-a-Chip on a Multiwell Microelectrode Array as a Scalable Preclinical Neurotoxicity Assay

Reliable human-relevant models of peripheral nerve function remain a critical unmet need in preclinical drug development, particularly for predicting neurotoxicity and bridging the gap to clinical translation. Here, we introduce a next generation Nerve-on-a-Chip, PNS-3D organoids, as a novel human-cell-based 3D peripheral nerve microphysiological system (MPS) that recapitulates key functional and structural features of native nerves, including long-distance axonal outgrowth, physiological myelination, and clinically translational population level electrophysiology. The platform integrates iPSC-derived human sensory neurons and primary human Schwann cells within a spatially organized 3D environment, coupled to a custom embedded electrode array that enables high-content, longitudinal, and clinically translatable functional assessments. As a proof of concept, we evaluated the platforms predictive power using vincristine, a chemotherapeutic agent known to cause chemotherapy-induced peripheral neuropathy (CIPN). PNS-3D organoids captured dose- and time-dependent deficits in nerve conduction velocity, compound action potential amplitude, and axonal degeneration, with IC values in line with human clinical exposures-- outperforming traditional 2D cultures and in vivo benchmarks. Transcriptomic and morphological analyses further revealed neuron-specific degeneration consistent with axonopathy. These results validate the platform as a human-relevant, clinically translation, and scalable solution, enabling mechanistic safety assessment and drug discovery for neurotoxic and neuroprotective therapeutics.

pharmacology and toxicology↗

Exploring Neural Dynamics in the Auditory Telencephalon of Crows using Functional Ultrasound Imaging

Crows, known for advanced cognitive abilities and vocal communication, rely on intricate auditory systems. While the neuroanatomy of corvid auditory pathways is partially explored, the underlying neurophysiological mechanisms are largely unknown. This study used functional ultrasound imaging (fUSi) to investigate sound-induced cerebral blood volume (CBV) changes in the field L complex of the crows auditory telencephalon, a functional analogue to the mammalian auditory cortex. FUSi revealed frequency-specific CBV responses, showing a tonotopic organization within the field L complex, with low frequencies in the posterior-dorsal region and high frequencies in the anterior-ventral region, similar to neuronal patterns reported in other songbirds. The responses were more robust when the crows were awake compared to when they were anesthetized. Machine learning analyses showed fUSi signals could be used to classify sound types accurately. Shorter stimuli reliably triggered transient CBV increases, while longer sounds resulted in variable responses, including negative deflections. This variability in CBV responses suggests a delineation of subregions within the field L complex with the central region (L2) as the primary processing hub and outer regions (L1, L3) integrating auditory information. These findings highlight the potential of fUSi for providing high-resolution insights into functional systems in corvids, enabling future exploration of task-related cognitive dynamics.

neuroscience↗

Deep learning and automated Cell Painting reveal Parkinson's disease-specific signatures in primary patient fibroblasts

Drug discovery for diseases such as Parkinsons disease are impeded by the lack of screenable cellular phenotypes. We present an unbiased phenotypic profiling platform that combines automated cell culture, high-content imaging, Cell Painting, and deep learning. We applied this platform to primary fibroblasts from 91 Parkinsons disease patients and matched healthy controls, creating the largest publicly available Cell Painting image dataset to date at 48 terabytes. We use fixed weights from a convolutional deep neural network trained on ImageNet to generate deep embeddings from each image and train machine learning models to detect morphological disease phenotypes. Our platforms robustness and sensitivity allow the detection of individual-specific variation with high fidelity across batches and plate layouts. Lastly, our models confidently separate LRRK2 and sporadic Parkinsons disease lines from healthy controls (receiver operating characteristic area under curve 0.79 (0.08 standard deviation)), supporting the capacity of this platform for complex disease modeling and drug screening applications.

cell biology↗