No abstract

Biallelic pathogenic variants in PLPBP (formerly called PROSC) have recently been shown to cause a novel form of vitamin B6-dependent epilepsy, the pathophysiological basis of which is poorly understood. When left untreated, the disease can progress to status epilepticus and death in infancy. Here we present 12 previously undescribed patients and six novel pathogenic variants in PLPBP. Suspected clinical diagnoses prior to identification of PLPBP variants included mitochondrial encephalopathy (two patients), folinic acid-responsive epilepsy (one patient) and a movement disorder compatible with AADC deficiency (one patient). The encoded protein, PLPHP is believed to be crucial for B6 homeostasis. We modelled the pathogenicity of the variants and developed a clinical severity scoring system. The most severe phenotypes were associated with variants leading to loss of function of PLPBP or significantly affecting protein stability/PLP-binding. To explore the pathophysiology of this disease further, we developed the first zebrafish model of PLPHP deficiency using CRISPR/Cas9. Our model recapitulates the disease, with plpbp À/À larvae showing behavioural, biochemical, and electrophysiological signs of seizure activity by 10 days post-fertilization and early death by 16 days post-fertilization. Treatment with pyridoxine significantly improved the epileptic phenotype and extended lifespan in plpbp À/À animals. Larvae had disruptions in amino acid metabolism as well as GABA and catecholamine biosynthesis, indicating impairment of PLP-dependent enzymatic activities. Using mass spectrometry, we observed significant B6 vitamer level changes in plpbp À/À zebrafish, patient fibroblasts and PLPHP-deficient HEK293 cells. Additional studies in human cells and yeast provide the first empirical evidence that PLPHP is localized in mitochondria and may play a role in mitochondrial metabolism. These models provide new insights into disease mechanisms and can serve as a platform for drug discovery.

Advances in high-throughput DNA sequencing have revolutionized the discovery of variants in the human genome; however, interpreting the phenotypic effects of those variants is still a challenge. While several computational approaches to predict variant impact are available, their accuracy is limited and further improvement is needed. Here, we introduce ClinPred, an efficient tool for identifying disease-relevant nonsynonymous variants. Our predictor incorporates two machine learning algorithms that use existing pathogenicity scores and, notably, benefits from inclusion of normal population allele frequency from the gnomAD database as an input feature. Another major strength of our approach is the use of ClinVar-a rapidly growing database that allows selection of confidently annotated disease-causing variants-as a training set. Compared to other methods, ClinPred showed superior accuracy for predicting pathogenicity, achieving the highest area under the curve (AUC) score and increasing both the specificity and sensitivity in different test datasets. It also obtained the best performance according to various other metrics. Moreover, ClinPred performance remained robust with respect to disease type (cancer or rare disease) and mechanism (gain or loss of function). Importantly, we observed that adding allele frequency as a predictive feature-as opposed to setting fixed allele frequency cutoffs-boosts the performance of prediction. We provide pre-computed ClinPred scores for all possible human missense variants in the exome to facilitate its use by the community.