Identification and Characterization of a Second Wolfram Syndrome Gene

Wolfram Syndrome (WFS) is a debilitating autosomal recessive neurodegenerative disorder characterized by juvenile onset insulin dependent diabetes mellitus (DM) and optic atrophy (OA) as well as a number of neurological and endocrine complications that result in early death due to respiratory complications. Previous research has mapped Wolfram syndrome to chromosome 4p16.1 and the disease has been attributed to mutations in the WFS1 gene affecting the WFS1 protein (wolframin), an ER membrane glycoprotein that plays an important role in the unfolded protein response (UPR) and in intracellular Ca2+ homeostasis. An additional locus for WFS on chromosome 4q22-24 was identified by linkage studies of four Jordanian Bedouin families with 16 affected individuals (El-Shanti et al., 2000). In this study, we attempted to identify the causative gene for the second WFS locus and identified a single missense mutation in a novel highly conserved iron-sulfur binding domain gene, CISD2, in the three consanguineous families of Jordanian descent from the El-Shanti et al. (2000) study (Amr et al., 2007). A G→C transversion at nucleotide 109 predicts an amino acid change from glutamic acid to glutamine (E37Q). Although the amino acid is conserved and the mutation is nonsynonymous, the missense mutation was found to disrupt messenger RNA splicing by eliminating exon 2 which results in the introduction of a premature stop codon. CISD2 is expressed in a wide variety of tissues, including those affected in WFS, the brain and pancreas. The CISD2-encoded protein, ERIS (endoplasmic reticulum intermembrane small protein) localizes to the endoplasmic reticulum but does not appear to interact directly with wolframin. Furthermore, lymphoblastoid cells from affected individuals show a significantly greater rise in intracellular calcium when stimulated with thapsigargin, compared with controls, although no difference was observed in resting concentrations of intracellular calcium. To understand the underlying pathogenesis in WFS2 patients, we examined cell death as well as known stress pathways. Cisd2 was knocked down in three cell lines derived from tissues most affected by the disease, namely rat pancreatic insulinoma cells (INS1), mouse neuroblastoma cells (N1E115), and mouse embryonic stem cell like cells (P19) which could be differentiated into neuronal cells. Cisd2 knockdown in INS1 cells shows an increase in apoptotic cell death and in the expression of the apoptotic markers CHOP and BAX, but no increase in the autophagic marker LC3-II. Assessment of the UPR in CISD2 deficient cells shows no activation of the UPR response, while Cisd2 expression in wild-type INS1 and N1E115 cells did not increase under conditions of ER stress. These findings indicate that there is an increase in apoptosis in WFS2 similar to WFS1 but the pathogenesis involves a molecular mechanism that is different than that in WFS1. Investigation of markers of oxidative stress, another major contributor to diabetes and neurodegeneration, show an increase in expression of the antioxidant enzymes Sod1 and Sod2 as well as an increase in global tyrosine nitration in INS1 Cisd2 knockdown cells compared with controls. Cell death in those cells was exacerbated with addition of known oxidative stressors, thapsigargin and paraquat compared with controls. These findings indicate that oxidative stress is a contributor to WFS2 pathogenesis, but it is not clear whether it is the primary causative factor. A recent article implicated ERIS in the BCL-2 associated inhibition of autophagy (Chang et al., 2009) and showed an increase in levels of autophagy in response to starvation in Cisd2 knockdown in a human epithelial carcinoma cell line (H1299). Starvation of INS1 Cisd2 knockdown cells did not elicit a greater autophagic response compared with controls, but did show an increase in expression of Cisd2. P19 Cisd2 knockdown cells that were differentiated into neurons by retinoic acid treatment did not show an inhibition in differentiation markers, but Cisd2 levels returned to levels similar to pre-differentiation wildtype P19 cells, which indicates that Cisd2 is upregulated during neuronal differentiation. In conclusion, the pathogensis of WFS2 can be attributed to apoptotic death of cells in affected tissues, with oxidative stress and not endoplasmic reticulum stress contributing to the development of disease, while ERIS’ relationship with BCL-2-mediated autophagy and neuronal differentiation suggest its important role in cell differentiation and survival