GBA2-deficiency and the effect on sphingolipid metabolism in hereditary spastic paraplegia type 46

Abstract

Bi-allelic, pathogenic variants in the GBA2 gene cause hereditary spastic paraplegia subtype 46, complicated by cerebellar ataxia. The GBA2 gene encodes the non-lysosomal ß-glucosidase, also known as GBA2. This enzyme is involved in the sphingolipid metabolism and catalyzes the breakdown of glucosylceramide into glucose and ceramide.

To assess consequences of the disturbed sphingolipid metabolism, we have reprogrammed patient-derived fibroblasts into induced pluripotent stem cells and differentiated these induced pluripotent stem cells into cortical neurons. These patient-derived cells carry three different pathogenic variants: Patient 1 carries compound heterozygous variants, comprising of a missense variant in exon 7 and a nonsense variant in the last exon, exon 17. Two siblings are the carriers of the third pathogenic variant studied in this work. Both patients carry a bi-allelic premature stop codon variant in exon 4. As a part of the thesis, I generated isogenic controls by using the CRISPR-Cas9 technique and thereby restored the enzymatic activity of GBA2 in these patient-derived cells. This technology allows creating isogenic controls by the correction of the disease-causing variants while preserving the genetic background of the donor cells. To reveal differences in the sphingolipid metabolism in a disease-specific background and a disease relevant cell type, I differentiated patient-derived induced pluripotent stem cells into cortical neurons and analyzed these neurons via lipid mass spectrometry.

Two lipid classes were increased in patient-derived cortical neurons, these lipids belong either to the class of dihydro-monohexosylceramides or monohexosylceramides. To verify the relevance of these results in vivo, we analyzed biofluids, collected from the same patients we obtained fibroblast for reprogramming induced pluripotent stem cells. Six different hexosylceramide species were measured in plasma and cerebrospinal fluid samples. Two hexosylceramide species were increased in cerebrospinal fluid samples and one hexosylceramide species was elevated in the plasma samples. The same species were elevated in iPSC-derived neurons, demonstrating the relevance of our cell-based disease model. As a proof of principle study, I treated patient-derived cortical neurons with different compounds to decrease hexosylceramide levels in these cells. One compound was Miglustat, an approved treatment for substrate reduction therapy in Gaucher disease. This glucosylceramidase inhibitor decreased hexosylceramide levels in cortical neurons originated from two different patients after a four-day treatment. These results improve the understanding of implicated mechanisms in this rare subtype of hereditary spastic paraplegia and present a strategy to lower the hexosylceramide burden in patient-derived cortical neurons.