Jaidar, O. | bioRxiv Science

bioRxiv2021

Enhancing motor learning by increasing stability of newly formed dendritic spines in motor cortex

Dendritic spine dynamics of Layer 5 Pyramidal neurons (L5PNs) are thought to be physical substrates for motor learning and memory of motor skills and altered spine dynamics are frequently correlated with poor motor performance. Here we describe an exception to this rule by studying mice lacking Paired immunoglobulin receptor B (PirB-/-). Using chronic two-photon imaging of primary motor cortex (M1) of PirB-/-;Thy1-YFP-H mice, we found a significant increase in the survival of spines on apical dendritic tufts of L5PNs, as well as increased spine formation rates and spine density. Surprisingly and contrary to expectations, adult PirB-/- mice learn a skilled reaching task more rapidly compared to wild type (WT) littermate controls. Conditional excision of PirB from forebrain pyramidal neurons in adult mice replicated these results. Furthermore, chronic imaging of L5PN dendrites throughout the learning period revealed that the stabilization of learning-induced newly formed spines is significantly elevated in PirB-/- mice. The degree of survival of newly formed spines in M1 yielded the strongest correlation with task performance, suggesting that this increased spine stability is advantageous and can translate into enhanced acquisition and maintenance of motor skills. Notably, inhibiting PirB function acutely in M1 of adult WT mice throughout training increases the survival of spines formed during early training and enhances motor learning. These results suggest that increasing the stability of newly formed spines is sufficient to improve long-lasting learning and motor performance and demonstrate that there are limits on motor learning that can be lifted by manipulating PirB, even in adulthood.

neuroscience↗

bioRxiv2020

Ultra-sensitive measurement of brain penetration mechanics and blood vessel rupture with microscale probes

Microscale electrodes, on the order of 10-100 m, are rapidly becoming critical tools for neuroscience and brain-machine interfaces (BMIs) for their high channel counts and spatial resolution, yet the mechanical details of how probes at this scale insert into brain tissue are largely unknown. Here, we performed quantitative measurements of the force and compression mechanics together with real-time microscopy for in vivo insertion of a systematic series of microelectrode probes as a function of diameter (7.5-100 m and rectangular Neuropixels) and tip geometry (flat, angled, and electrochemically sharpened). Results elucidated the role of tip geometry, surface forces, and mechanical scaling with diameter. Surprisingly, the insertion force post-pia penetration was constant with distance and did not depend on tip shape. Real-time microscopy revealed that at small enough lengthscales (<25 m), blood vessel rupture and bleeding during implantation could be entirely avoided. This appears to occur via vessel displacement, avoiding capture on the probe surface which led to elongation and tearing for larger probes. We propose a new, three-zone model to account for the probe size dependence of bleeding, and provide mechanistic guidance for probe design. Significance StatementMicroscale neural probes are central to next-generation brain-machine interfaces, yet how they physically penetrate living brain remains poorly quantified. Using a high-sensitivity force sensor integrated with real-time microscopy, we measured in vivo force-displacement and visualized vascular responses for microwires (7.5-100 m) and Neuropixels. We find that once the brains protective pia membrane is breached, insertion force remains essentially constant with depth, while pia puncture force and pre-penetration compression scale linearly with probe diameter. Real-time imaging reveals a sub-25 m regime in which blood vessels are displaced rather than ruptured. These results motivate a three-zone model of vessel capture versus displacement and provide actionable mechanical design rules for low-trauma, high-density neural interfaces.

neuroscience↗