Funded Research

Growth charts for height and weight have been a cornerstone of pediatric medicine for centuries – they quantify the normal range of physical development and allow physicians to detect clinically relevant abnormalities. The wide availability of imaging data and computational tools has spurred initiatives to develop similar normative models for brain development and ageing. These models have great potential to accelerate clinical discovery, and to improve diagnosis and disease prediction for individual patients.

‍
The cerebellum plays a prominent role in cognitive and motor development – nonetheless, a system that can deliver accurate age-dependent norms for the cerebellum is currently not available. The goal of this grant is to develop such a system. Using our expertise in image processing and normative modelling, we will develop and validate a system that characterizes the anatomical development of the cerebellum across the entire lifespan and that works robustly with different types of imaging data. The resulting system will allow users to quantify changes in cerebellar anatomy in individuals and small clinical studies. Having a common reference system will accelerate our understanding of the cerebellar role in developmental and neuropsychiatric conditions, and will enable clinicians to improve individual prognoses, informing individual treatments and prevention.

The cerebellum plays a key role in human brain development. This grant will develop an openly available system that quantifies cerebellar development at high resolution, enabling the sensitive detection of cerebellar abnormalities in individuals and small clinical studies.

We will use a multidisciplinary approach to characterize neuropathological, metabolic, and neuroimmune alterations caused by prematurity-induced insult in the developing human cerebellum. We will leverage on a unique set of cerebellar tissue samples obtained from a comprehensive cohort of 98 neonates and utilize cutting-edge experimental approaches to define cellular and molecular abnormalities induced by each insult of prematurity in the developing cerebellum. We will integrate spatial transcriptomics and cellular analysis in developing human cerebellum to pursue rigorous neuropathological and genetic investigation, in order to identify molecular alterations underlying specific cellular changes in response to MIA and HX. The two research teams at Children’s National Hospital and Johns Hopkins University School of Medicine are currently collaborating on a variety of projects on cerebellar development and injury and will provide complementary skills (CN: neonatal brain injury, human neonatal neuropathology, brain bioenergetics, and cerebellar molecular physiology; JHU: computational spatial transcriptomics and neuroimmunology) to address the specific aims of the present proposal.

Currently available commercial 1H head radiofrequency (RF) coils for 7T do not adequately image the cerebellum, resulting in an incomplete understanding of cerebellar structure, function and metabolism. Our groups (UWO, UTSW) have experience building coils for multi-site studies, and currently available head and neck coil designs from our group (UTSW) already outperform commercially available coils with respect to cerebellar imaging.Building upon our experience we will optimize the head and neck RF coil design further to ensure excellent coverage and SNR in the cerebellum, enable to investigate the integration of the cerebellum within the central nervous system and allow for anatomical, functional and metabolic MRI studies of cerebrum and cerebellum. It is planned to harmonize the design to ensure compatibility across four different 7T MRI scanner configurations and to build a total of 4 coils (2 for Philips systems by UTSW; 2 for Siemens systems by UWO) for the two sites involved in this project and the two consulting sites. Coils will be distributed to all sites for subsequent neuroimaging studies.

To develop a consensus coil suitable for imaging the entire brain, including the cerebellum, at 7T for use in adults as well as children and adolescents 8-20 years old. The RF coil design shall allow integration with Philips and Siemens 7T MRI scanners.

Head-fixed mouse experiments are crucial for understanding cerebellar function because theyallow for precise neural activity monitoring/manipulation along with movement tracking thus connecting neuralcircuits to animal behavior. Findings from such experiments have been instrumental in pinpointing phenotypes/biomarkers linked to cerebellar dysfunction. However, the complexity and time-consuming nature of head-fixed behavior experiments, alongside a lack of standardization for their use across labs, curtail theirutility for high-throughput screening, which is critical for rapid therapeutic development. This limitation highlights a technological shortfall that hampers progress in treating cerebellar disorders. By automating the most time-consuming steps associated with the training and assessment of mice performing head-fixed tasks, high-throughput behavioral screens use can be conducted at a nearly unlimited scale. While there has been a recent effort to develop automated head-fixed behavioral methods, they fall short of the standard we aim to achieve in a variety of ways: they only support training of a single behavior, are bespoke for single lab use, and they offer no sophisticated method of data curation and analysis. To support high-throughput screening of therapeutics for mice modeling cerebellar dysfunction, the device must be capable of assessing multiple modes of animal behavior using a reliable commercial-grade system that is easy to use and maintain.

To create a modular platform facilitating automated behavior experiments in head-fixed mice thataccommodates a range of tasks across domains of brain function in which the cerebellum plays a role.

One of the major goals of The Raynor Cerebellum Project is to develop novel tools capable of discerning cerebellar development in a clinical setup. We will collect two sets of anatomical images from each subject at 3T using imaging parameters: 1) similar to the historical images at 14 and 16 years old and 2) consistent with the state-of-the-art resolution (0.8mm3). These images will enable the development of machine learning approaches to reconstruct synthetic high-resolution images from the available low-resolution clinical data, allowing consistent estimations of gray and white matter volumes from both types of images. The data will also serve as calibration data to analyze these trajectories within the already developed normative models by the Raynor Cerebellar Growth Chart team. Furthermore, since MRI stands to gain substantially from improvements in signal-to-noise ratio (SNR) and resolution at 7T. 7T MR images and MR spectroscopy (MRS) will be utilized in delineating cerebellar abnormalities.

This project aims to develop an efficient method for delivering therapeutic genes to Purkinje cells throughout the primate cerebellum. Leveraging the anatomical projection of most Purkinje cell axons to the deep cerebellar nuclei, we will use AAV2retro viral vectors with different promoters and fluorescent reporters to comprehensively map viral transduction patterns following targeted deep cerebellar nuclear injections in macaques. These preclinical data will demonstrate the efficacy of this approach for delivering therapeutic genes to Purkinje cells, paving the way for treating cerebellar disorders.

This groundbreaking research represents a critical step towards developing precise genetic therapies for cerebellar disorders by establishing a novel, targeted approach to gene delivery in primate neural circuits.

This study will start to address the urgent need to determine if volumetric changes in the cerebellum of children born very preterm reflect evolution following static injury. This project will also facilitate the development of automated, user-independent, quantitative assessments of the cerebellum, which will enable structure-function correlations and contribute to the development of treatments and neuroprotective strategies for motor, cognitive, and neuropsychiatric challenges facing people with cerebellar disorders.

My laboratory for the past 30 years has identified many new genetic causes of ataxia (1-8). Here, we propose to focus on familial cases of early onset ataxia in which clinical genetic tests have failed to identify a causative mutation. We propose to include a few already ascertained families as well as recruit new early-onset familial cases. We will perform research-based whole genome sequencing, as some mutations are missed in the 1-2% of DNA covered by clinical whole exome sequencing. Through our established bioinformatic pipeline and collaboration with experts in DNA structural abnormality analyses, we propose to analyze 10 families with ataxia, and hypothesize that in several of these familial cases, we will identify a likely causative gene mutation, either a mutation in an already known gene that was missed by WES, or mutations in novel genes. The mutations will be molecularly confirmed, and their effects studied in patients’ blood or fibroblast-derived cell lines. Once candidate mutations in novel ataxia genes have been found, we will identify other researcher studying cases with mutations in the same gene through “GeneMatcher”, and collaborators studying the pathways implicated by the candidate proteins.

At the end of two years of funding, our work will result in the identification of several novel genes involved in familial, early-onset ataxia, confirm the mutation and their molecular effects, and have identified additional families with mutations in the same gene and collaborators to initiate further studies such as the development of animal models.

‍