Phenome-wide identification of therapeutic genetic targets, leveraging knowledge graphs, graph neural networks, and UK Biobank data.

The ongoing expansion of human genomic datasets propels therapeutic target identification; however, extracting gene-disease associations from gene annotations remains challenging. Here, we introduce Mantis-ML 2.0, a framework integrating AstraZeneca's Biological Insights Knowledge Graph and numerous tabular datasets, to assess gene-disease probabilities throughout the phenome. We use graph neural networks, capturing the graph's holistic structure, and train them on hundreds of balanced datasets via a robust semi-supervised learning framework to provide gene-disease probabilities across the human exome. Mantis-ML 2.0 incorporates natural language processing to automate disease-relevant feature selection for thousands of diseases. The enhanced models demonstrate a 6.9% average classification power boost, achieving a median receiver operating characteristic (ROC) area under curve (AUC) score of 0.90 across 5220 diseases from Human Phenotype Ontology, OpenTargets, and Genomics England. Notably, Mantis-ML 2.0 prioritizes associations from an independent UK Biobank phenome-wide association study (PheWAS), providing a stronger form of triaging and mitigating against underpowered PheWAS associations. Results are exposed through an interactive web resource.

Respiratory dysfunction in a mouse model of spinocerebellar ataxia type 7.

Spinocerebellar ataxia type 7 (SCA7) is an autosomal-dominant neurodegenerative disorder caused by a CAG repeat expansion in the coding region of the ataxin-7 gene. Infantile-onset SCA7 patients display extremely large repeat expansions (>200 CAGs) and exhibit progressive ataxia, dysarthria, dysphagia and retinal degeneration. Severe hypotonia, aspiration pneumonia and respiratory failure often contribute to death in affected infants. To better understand the features of respiratory and upper airway dysfunction in SCA7, we examined breathing and putative phrenic and hypoglossal neuropathology in a knock-in mouse model of early-onset SCA7 carrying an expanded allele with 266 CAG repeats. Whole-body plethysmography was used to measure awake spontaneously breathing SCA7-266Q knock-in mice at baseline in normoxia and during a hypercapnic/hypoxic respiratory challenge at 4 and 8 weeks, before and after the onset of disease. Postmortem studies included quantification of putative phrenic and hypoglossal motor neurons and microglia, and analysis of ataxin-7 aggregation at end stage. SCA7-266Q mice had profound breathing deficits during a respiratory challenge, exhibiting reduced respiratory output and a greater percentage of time in apnea. Histologically, putative phrenic and hypoglossal motor neurons of SCA7 mice exhibited a reduction in number accompanied by increased microglial activation, indicating neurodegeneration and neuroinflammation. Furthermore, intranuclear ataxin-7 accumulation was observed in cells neighboring putative phrenic and hypoglossal motor neurons in SCA7 mice. These findings reveal the importance of phrenic and hypoglossal motor neuron pathology associated with respiratory failure and upper airway dysfunction, which are observed in infantile-onset SCA7 patients and likely contribute to their early death.

Glycogen accumulation in smooth muscle of a Pompe disease mouse model.

Pompe disease is a lysosomal storage disease caused by mutations within the GAA gene, which encodes acid α-glucosidase (GAA)-an enzyme necessary for lysosomal glycogen degradation. A lack of GAA results in an accumulation of glycogen in cardiac and skeletal muscle, as well as in motor neurons. The only FDA approved treatment for Pompe disease-an enzyme replacement therapy (ERT)-increases survival of patients, but has unmasked previously unrecognized clinical manifestations of Pompe disease. These clinical signs and symptoms include tracheo-bronchomalacia, vascular aneurysms, and gastro-intestinal discomfort. Together, these previously unrecognized pathologies indicate that GAA-deficiency impacts smooth muscle in addition to skeletal and cardiac muscle. Thus, we sought to characterize smooth muscle pathology in the airway, vascular, gastrointestinal, and genitourinary in the Gaa-/- mouse model. Increased levels of glycogen were present in smooth muscle cells of the aorta, trachea, esophagus, stomach, and bladder of Gaa-/- mice, compared to wild type mice. In addition, there was an increased abundance of both lysosome membrane protein (LAMP1) and autophagosome membrane protein (LC3) indicating vacuolar accumulation in several tissues. Taken together, we show that GAA deficiency results in subsequent pathology in smooth muscle cells, which may lead to life-threatening complications if not properly treated.

Respiratory pathology in the Optn-/- mouse model of Amyotrophic Lateral Sclerosis.

Amyotrophic Lateral Sclerosis (ALS) is a devastating neurodegenerative disorder that results in death due to respiratory failure. Many genetic defects are associated with ALS; one such defect is a mutation in the gene encoding optineurin (OPTN). Using an optineurin null mouse (Optn-/-), we sought to characterize the impact of optineurin deficiency on respiratory neurodegeneration. Respiratory function was assessed at 6 and 12 mo of age using whole body plethysmography at baseline during normoxia (FiO2: 0.21; N2 balance) and during a respiratory challenge with hypoxia and hypercapnia (FiCO2: 0.07, FiO2: 0.10; N2 balance). Histological analyses to assess motor neuron viability and respiratory nerve integrity were performed in the medulla, cervical spinal cord, hypoglossal nerve, and phrenic nerve. Minute ventilation, peak inspiratory flow, and peak expiratory flow are significantly reduced during a respiratory challenge in 6 mo Optn-/-mice. By 12 mo, tidal volume is also significantly reduced in Optn-/- mice. Furthermore, 12mo Optn-/- mice exhibit hypoglossal motor neuron loss, phrenic and hypoglossal dysmyelination, and accumulated mitochondria in the hypoglossal nerve axons. Overall, these data indicate that Optn-/- mice display neurodegenerative respiratory dysfunction and are a useful model to study the impact of novel therapies on respiratory function for optineurin-deficient ALS patients.

Motor axonopathies in a mouse model of Duchenne muscular dystrophy.

Duchenne muscular dystrophy (DMD) is a fatal neuromuscular disease caused by deleterious mutations in the DMD gene which encodes the dystrophin protein. Skeletal muscle weakness and eventual muscle degradation due to loss of dystrophin are well-documented pathological hallmarks of DMD. In contrast, the neuropathology of this disease remains understudied despite the emerging evidence of neurological abnormalities induced by dystrophin loss. Using quantitative morphological analysis of nerve sections, we characterize axonopathies in the phrenic and hypoglossal (XII) nerves of mdx mice. We observe dysfunction in these nerves - which innervate the diaphragm and genioglossus respectively - that we propose contributes to respiratory failure, the most common cause of death in DMD. These observations highlight the importance in the further characterization of the neuropathology of DMD. Additionally, these observations underscore the necessity in correcting both the nervous system pathology in addition to skeletal muscle deficits to ameliorate this disease.

Intralingual Administration of AAVrh10-miRSOD1 Improves Respiratory But Not Swallowing Function in a Superoxide Dismutase-1 Mouse Model of Amyotrophic Lateral Sclerosis.

Amyotrophic lateral sclerosis (ALS) is a fatal disease characterized by degeneration of motor neurons and muscles, and death is usually a result of impaired respiratory function due to loss of motor neurons that control upper airway muscles and/or the diaphragm. Currently, no cure for ALS exists and treatments to date do not significantly improve respiratory or swallowing function. One cause of ALS is a mutation in the superoxide dismutase-1 (SOD1) gene; thus, reducing expression of the mutated gene may slow the progression of the disease. Our group has been studying the SOD1G93A transgenic mouse model of ALS that develops progressive respiratory deficits and dysphagia. We hypothesize that solely treating the tongue in SOD1 mice will preserve respiratory and swallowing function, and it will prolong survival. At 6 weeks of age, 11 SOD1G93A mice (both sexes) received a single intralingual injection of gene therapy (AAVrh10-miRSOD1). Another 29 mice (both sexes) were divided into two control groups: (1) 12 SOD1G93A mice that received a single intralingual vehicle injection (saline); and (2) 17 non-transgenic littermates. Starting at 13 weeks of age, plethysmography (respiratory parameters) at baseline and in response to hypoxia (11% O2) + hypercapnia (7% CO2) were recorded and videofluoroscopic swallow study testing were performed twice monthly until end-stage disease. Minute ventilation during hypoxia + hypercapnia and mean inspiratory flow at baseline were significantly reduced (p < 0.05) in vehicle-injected, but not AAVrh10-miRSOD1-injected SOD1G93A mice as compared with wild-type mice. In contrast, swallowing function was unchanged by AAVrh10-miRSOD1 treatment (p > 0.05). AAVrh10-miRSOD1 injections also significantly extended survival in females by ∼1 week. In conclusion, this study indicates that intralingual AAVrh10-miRSOD1 treatment preserved respiratory (but not swallowing) function potentially via increasing upper airway patency, and it is worthy of further exploration as a possible therapy to preserve respiratory capacity in ALS patients.

The Respiratory Phenotype of Pompe Disease Mouse Models.

Pompe disease is a glycogen storage disease caused by a deficiency in acid α-glucosidase (GAA), a hydrolase necessary for the degradation of lysosomal glycogen. This deficiency in GAA results in muscle and neuronal glycogen accumulation, which causes respiratory insufficiency. Pompe disease mouse models provide a means of assessing respiratory pathology and are important for pre-clinical studies of novel therapies that aim to treat respiratory dysfunction and improve quality of life. This review aims to compile and summarize existing manuscripts that characterize the respiratory phenotype of Pompe mouse models. Manuscripts included in this review were selected utilizing specific search terms and exclusion criteria. Analysis of these findings demonstrate that Pompe disease mouse models have respiratory physiological defects as well as pathologies in the diaphragm, tongue, higher-order respiratory control centers, phrenic and hypoglossal motor nuclei, phrenic and hypoglossal nerves, neuromuscular junctions, and airway smooth muscle. Overall, the culmination of these pathologies contributes to severe respiratory dysfunction, underscoring the importance of characterizing the respiratory phenotype while developing effective therapies for patients.

The Respiratory Phenotype of Rodent Models of Amyotrophic Lateral Sclerosis and Spinocerebellar Ataxia.

Amyotrophic lateral sclerosis (ALS) and spinocerebellar ataxia (SCA) are neurodegenerative disorders that result in progressive motor dysfunction and ultimately lead to respiratory failure. Rodent models of neurodegenerative disorders provide a means to study the respiratory motor unit pathology that results in respiratory failure. In addition, they are important for pre-clinical studies of novel therapies that improve breathing, quality of life, and survival. The goal of this review is to compare the respiratory phenotype of two neurodegenerative disorders that have different pathological origins, but similar physiological outcomes. Manuscripts reviewed were identified using specific search terms and exclusion criteria. We excluded manuscripts that investigated novel therapeutics and only included those manuscripts that describe the respiratory pathology. The ALS manuscripts describe pathology in respiratory physiology, the phrenic and hypoglossal motor units, respiratory neural control centers, and accessory respiratory muscles. The SCA rodent model manuscripts characterized pathology in overall respiratory function, phrenic motor units and hypoglossal motor neurons. Overall, a combination of pathology in the respiratory motor units and control centers contribute to devastating respiratory dysfunction.