Erratum: SLP-2 interacts with Parkin in mitochondria and prevents mitochondrial dysfunction in Parkin-deficient human iPSC-derived neurons and Drosophila (Human Molecular Genetics (2017) 26:13 (2412-2425) DOI: 10.1093/hmg/ddx132)

The authors of the article "SLP-2 interacts with Parkin in mitochondria and prevents mitochondrial dysfunction in Parkin-deficient human iPSC-derived neurons and Drosophila" would like to report the incorrect description of an iPSC line in the paper. Authentication analysis at Coriell Institute for Medical Research (Camden, New Jersey) revealed that the cell line named iPS-HFF in the article was not derived from newborn foreskin fibroblasts (as reported in Table S2) but from cells of a female patient with osteogenesis imperfecta (GM17602/GM17604), carrier of a heterozygous missense mutation (p.G700C; NM_000089) in the COL1A2 gene. Repeated karyotype analysis performed for the line named iPS-HFF showed a normal female karyotype, which means that the male karyotype displayed in Figure S4B is mislabeled and was not derived from iPS-HFF. Differentiated dopaminergic neurons from the line named iPS-HFF were examined for mitochondrial complex I activity (Figure 4A) and mitochondrial morphology (Figure 4B) as one of in total three analyzed control iPSC lines compared to Parkin mutant lines. Importantly, both experiments display robust phenotypes and are replicative analysis of results shown in Parkin knockdown SH-SY5Y cells: Loss-of-Parkin leads to decreased complex I activity (Figure 2B) and an increased amount of fragmented mitochondria (Figure 2D), while overexpression of SLP-2 rescues both phenotypes. This confirms that the analyzed data in Figure 4 are not influenced by the COL1A2 mutation. Furthermore, mutations in the COL1A2 gene have not been associated with neurodegenerative diseases and the symptoms of neurological movement disorders have not been described for osteogenesis imperfecta. Therefore, the fact that this particular cell line is from a diseased individual with osteogenesis imperfecta rather than a healthy control does not alter the interpretation of our results as pertaining to rescue of mitochondrial phenotypes as described in the manuscript. The authors would like to apologize for this mistake.