Dissecting the Effect of BCATm Inhibition in Novel Model Systems of Methylmalonic Acidemia

Abstract

Methylmalonic acidemia (MMA) is an organic aciduria primarily caused by deficiency of methylmalonyl-CoA mutase (MMUT) that impairs the conversion of methylmalonyl-CoA to succinyl-CoA. MMUT deficiency leads to accumulation of methylmalonic acid, propionylcarnitine (C3), 2-methylcitrate (MCA), and other metabolites. MMA has an incidence of 1 in 50,000 to 100,000 in the United States. It generally in the newborn period with lethargy, failure to thrive, and metabolic decompensation that may lead to coma or death. MMA is associated with significant morbidity and mortality, including multisystemic sequalae of kidney dysfunction and neurological impairment. Standard of care is limited to strict restriction of dietary protein and supportive care during episodes of metabolic decompensation. In severe cases, orthotopic liver or combined liver/kidney transplantation is used to confer a measure of metabolic stability, however is not curative. Thus, there is a clear need for better therapies to improve patient outcomes and quality of life.The development of robust and relevant model systems is critical to the investigation of disease progression and evaluation of novel therapeutics. We developed several new mouse models of MMA that encompass the genetic and phenotypic heterogeneity of the disorder. We utilized CRISPR-Cas9 gene editing techniques to generate the p.E686* allele based on the human p.E688* variant as well as a novel insertion/deletion allele, p.L690Ins. Homozygosity for the p.E686* allele resulted in neonatal lethality and a complete loss of MMUT expression in the liver, synonymous with the mut0 subtype of MMA. Mice homozygous for the p.L690Ins allele experienced significant post-weaning lethality, failure to thrive, increased disease metabolites, decreased MMUT expression, and ultrastructural changes in hepatic and renal mitochondria. Additionally, we produced two compound heterozygous models with previously generated p.M698K and p.R106C missense alleles, the Mmutp.M698K/p.E686* and Mmutp.R106C/p.M698K mice. These mice both manifested with elevated MMA and MCA. When fed a 70% casein or precursor-enriched (PE) diet, both models exhibited a failure to thrive phenotype and, on PE diet, significant mortality. The differences in severity of disease presentation between these two models may give further insight into the interallelic effects across different protein domains and mutation type. Of the described models, Mmutp.L690Ins/p.L690Ins mice in particular show many common clinical manifestations of MMA, as well as long term complications such as kidney dysfunction and degenerative morphological changes in the liver. As a potential upstream target for therapeutic intervention, we investigated the effect of a small molecule branched-chain amino transferase (BCAT) inhibitor in human MMA hepatocytes and the Mmutp.L690Ins/p.L690Ins MMA mouse model. Metabolic flux analyses confirmed robust BCAT inhibition, with reduction of labeling of proximal and distal BCAA catabolites in MMA hepatocytes. In vivo experiments confirmed the BCAT inhibition but only the proximal catabolites were reduced, while distal BCAA catabolites, disease markers and clinical symptoms persisted. Our study showcases the importance of using multiple models not only to evaluate a potential drug target, but also to understand the underlying pathology of metabolic disorders.