Establishing an in vitro disease model for Fabry Disease using patient specific induced pluripotent stem cell-derived sensory neurons

Abstract

Fabry disease (FD) is an X-linked lysosomal storage disorder caused by deficiency of the α-galactosidase A (GLA), leading to intracellular accumulations of globotriaosylceramide (Gb3). Acral burning pain, which can be triggered by heat, fever or physical activity is an early hallmark of FD and greatly reduces patients’ quality of life. The pathophysiology of FD pain is unknown and research is hindered by the limited in vivo availability of suitable human biomaterial. To overcome this obstacle, we generated induced pluripotent stem cells (iPSC) from one female and two male patients with a differing pain phenotype, and developed a refined differentiation protocol for sensory neurons to increase reliability and survival of these neurons, serving as an in vitro disease model. Neurons were characterized for the correct neuronal subtype using immunocytochemistry, gene expression analysis, and for their functionality using electrophysiological measurements. iPSC and sensory neurons from the male patients showed Gb3 accumulations mimicking the disease phenotype, whereas no Gb3 depositions were detected in sensory neurons derived from the female cell line, likely caused by a skewed X-chromosomal inactivation in favor of healthy GLA. Using super-resolution imaging techniques we showed that Gb3 is localized in neuronal lysosomes of male patients and in a first experiment using dSTORM microscopy we were able to visualize the neuronal membrane in great detail. To test our disease model, we treated the neurons with enzyme replacement therapy (ERT) and analyzed its effect on the cellular Gb3 load, which was reduced in the male FD-lines, compared to non-treated cells. We also identified time-dependent differences of Gb3 accumulations, of which some seemed to be resistant to ERT. We also used confocal Ca2+ imaging to investigate spontaneous neuronal network activity, but analysis of the dataset proofed to be difficult, nonetheless showing a high potential for further investigations. We revealed that neurons from a patient with pain pain are more easily excitable, compared to cells from a patient without pain and a healthy control. We provide evidence for the potential of patient-specific iPSC to generate a neuronal in vitro disease model, showing the typical molecular FD phenotype, responding to treatment, and pointing towards underlying electrophysiological mechanisms causing different pain phenotypes. Our sensory neurons are suitable for state-of-the-art microscopy techniques, opening new possibilities for an in-depth analysis of cellular changes, caused by pathological Gb3 accumulations. Taken together, our system can easily be used to investigate the effect of the different mutations of GLA on a functional and a molecular level in affected neurons.