Purkinje cell vulnerability induced by diffuse traumatic brain injury is linked to disruption of long-range neuronal circuits

Cerebellar dysfunction is commonly observed following traumatic brain injury (TBI). While direct impact to the cerebellum by TBI is rare, cerebellar pathology may be caused by indirect injury via cortico-cerebellar pathways. To address the hypothesis that degeneration of Purkinje cells (PCs), which constitute the sole output from the cerebellum, is linked to long-range axonal injury and demyelination, we used the central fluid percussion injury (cFPI) model of widespread traumatic axonal injury in mice. Compared to controls, TBI resulted in early PC loss accompanied by alterations in the size of pinceau synapses and levels of non-phosphorylated neurofilament in PCs. A combination of vDISCO tissue clearing technique and immunohistochemistry for vesicular glutamate transporter type 2 show that diffuse TBI decreased mossy and climbing fiber synapses on PCs. At 2 days post-injury, numerous axonal varicosities were found in the cerebellum supported by fractional anisotropy measurements using 9.4 T MRI. The disruption and demyelination of the cortico-cerebellar circuits was associated with poor performance of brain-injured mice in the beam-walk test. Despite a lack of direct input from the injury site to the cerebellum, these findings argue for novel long-range mechanisms causing Purkinje cell injury that likely contribute to cerebellar dysfunction after TBI.

Deep Learning Reveals Cancer Metastasis and Therapeutic Antibody Targeting in the Entire Body

Reliable detection of disseminated tumor cells and of the biodistribution of tumor-targeting therapeutic antibodies within the entire body has long been needed to better understand and treat cancer metastasis. Here, we developed an integrated pipeline for automated quantification of cancer metastases and therapeutic antibody targeting, named DeepMACT. First, we enhanced the fluorescent signal of cancer cells more than 100-fold by applying the vDISCO method to image metastasis in transparent mice. Second, we developed deep learning algorithms for automated quantification of metastases with an accuracy matching human expert manual annotation. Deep learning-based quantification in 5 different metastatic cancer models including breast, lung, and pancreatic cancer with distinct organotropisms allowed us to systematically analyze features such as size, shape, spatial distribution, and the degree to which metastases are targeted by a therapeutic monoclonal antibody in entire mice. DeepMACT can thus considerably improve the discovery of effective antibody-based therapeutics at the pre-clinical stage. VIDEO ABSTRACT.