Glucose transporter type 1 deficiency syndrome (MIM 606777, GLUT1 DS) is a rare neurologic disorder with a considerable clinical heterogeneity including early-onset seizures, intellectual disability, neurodevelopmental problems, ataxia and microcephaly, caused by a defective transport of glucose into the brain, due to mutations in the SCL2A1 gene, coding for GLUT1. Most cases are sporadic, with an autosomal dominant mode of inheritance. Several pathogenic mutations have been described but no clear genotype-phenotype correlation has been defined. For patients with GLUT1 defects, a high-fat and calorie-reduced, i.e., the ketogenic diet, is administered to control symptoms of seizures. The ketogenic diet should be started early making an early diagnosis of crucial importance.
The diagnosis is based on neurological examination and neuropsychological testing, and is
confirmed with biochemical (glucose levels in cerebrospinal fluid and serum) and molecular (SLC2A1 sequencing) approaches. However, the clinical diagnosis is challenging because of the lack of distinctive clinical manifestations, and a spectrum of symptoms appearing and changing over the years.
The purpose of our study is to improve the diagnostic yield of genetic test, to bring new insights into the GLUT1 DS molecular pathology and to clarify genotype-phenotype correlations.
We aim at implementing data analysis workflow of genetic test based on SLC2A1 sequencing to analyze variants in regions not routinely screened or classified as neutral in standard analyses. We also aimed at characterizing the functional effects of coding and non-coding mutations using advanced in silico (molecular dynamics simulations) and in vitro (expression/localization/glucose transport) approaches.
Efforts aimed at disclosing molecular functions of GLUT1 and its dysfunctions in pathogenesis are a fundamental prerequisite that will provide information with direct impact on diagnosis, prognosis, counseling, and patient management.
GLUT1 deficiency syndrome is a relatively new and underdiagnosed neurologic disorder with a wide spectrum of clinical symptoms and with varying age of onset. Currently the different clinical manifestations and the pathophysiology mechanisms have not been yet elucidated.
The diagnosis is challenging because of the lack of distinctive clinical manifestations, and a spectrum of symptoms appearing and changing over the years. Therefore, it is of crucial importance to better define the heterogeneity that characterizes the wide clinical picture in order to clarify genotype-phenotype correlations with a direct impact on diagnosis, prognosis and on therapy.
The PI and components of the project have already recruited cases with phenotypes that are classified as GLUT1 DS based on clinical and neurological examinations and with phenotypes that partially overlap (e.g., epilepsy, paroxismal movement disorders, and hemiplegic migrain), thereby confirming the wide spectrum of this syndrome. The detailed clinical assessment of these and further cases will allow us to recruit a cohort that could be analyzed through clinical, biochemical, and molecular studies.
A significant portion of cases with clinical and biochemical (glucose levels) features suggestive of glucose transport defects, are negative for SLC2A1 mutations pointing to potential pathogenetic mechanisms, such as SLC2A1 regulatory mechanisms, pathogenic mutations in non-coding regions, and further causative disease genes. Very recently, the reported identification of synonymous pathogenic mutations (Tacik et al. 2014), somatic variants (Takahashi et al. 2017), non-coding mutations, i.e. in intron (Liu et al. 2016) and 5'UTR (Willemsen et al. 2017), and copy number variants involving the SLC2A1 gene (Tsuchida N. et al. 2018) have confirmed this hypothesis. However, the lack of a workflow of data analysis to disclose the occurrence of this type of variants in routine diagnostic workflows is still lacking. The development of an integrated pipeline of NGS data analysis that is based not only on standard criteria but taking into account also further potential pathogenic variants (i.e. genomic, somatic, non-coding, silent changes), could increase the diagnostic yield with a direct impact on diagnosis, prognosis and especially on therapy as in this context the availability of a diet (i.e. the ketogenic diet) could represent an effective therapeutic approach to improve at least some symptoms. The PI has already developed a workflow of NGS data analysis that allows disclosing the occurrence of potentially pathogenic variants, in the framework of analyzing molecular bases of rare Mendelian diseases, with the aim of improving diagnostic rates of undiagnosed diseases.
The identification of coding and non-coding variants raises the challenging task of a reliable functional effect prediction that could allow a plausible classification of pathogenicity.
To date, the American College of Medical Genetics guidelines represent the primary reference to classify DNA variants. However, some of the used criteria, rely on computational tools to evaluate the pathogenicity of a variant that includes both individual scoring systems and meta-predictors. Those approaches have some limitations, such as moderate accuracy, low specificity, and over-prediction, which could sometimes give conflicting results. In this context the use of a molecular dynamics simulation approach could represent a valuable and affordable tool to support and integrate existing criteria, providing a prediction on the effect that a missense variant could exert on the kinetics of the glucose transport mediated by GLUT1 and providing valuable hints to experimental functional studies.
To date, only preliminary studies have been accomplished to study the functional role of GLUT1 mutants on the pathogenesis of human disease; therefore functional assays to explore the mechanism involving this glucose carrier could fill this gap, providing clues on the differential effect that exerts, when mutated, on GLUT1 DS related developmental processes. Moreover, the identification of new pathogenic missense mutations and also a new class of pathogenic mutations and their related pathogenic mechanisms could shed light on new aspects of pathogenesis and pave the way for the development of new therapeutic strategies.
Therefore, we suggest that efforts aimed at disclosing molecular functions of GLUT1 and its dysfunctions in pathogenesis are a fundamental prerequisite that will provide molecular information with direct impact on diagnosis, prognosis, counseling, and patient management.