Human Molecular Genetics, 2015, 24, 1211–1224. doi: 10.1093/hmg/ddu532.
The Authors report the following error regarding this knockin mouse model of spinocerebellar ataxia type 3 (SCA3):
In further examining the genetics of the knockin mouse model of SCA3 described in this publication, the Authors identified that this mouse harbors a tandem duplicate of the targeting vector which contains the CAG repeat expansion of exon 10 in Atxn3. Thus, the genetic map of this atypical SCA3 knockin mouse in Figure 1 is incorrect, and the correct genetic map is provided in the figure below (Figure 1). The identification of this duplication raised the possibility that the aberrant splicing of the Atxn3 transcript and prominent ataxin-3 aggregation in this atypical SCA3 knockin mouse model results from effects of the duplication rather than the expanded CAG repeat alone. A follow-up manuscript describing the generation of a corrected SCA3 knockin mouse and its consequences on Atxn3 splicing and ataxin-3 aggregation has been accepted for publication in HMG (Ramani et al 2017, accepted for publication in HMG). The original conclusion in the published report that the CAG repeat expansion directly alters Atxn3 splicing in this knockin mouse is not supported by the newest findings from the corrected knockin mouse.
Figure 1A.
Generation of the SCA3 knockin mouse resulted in tandem duplicate insertion of the targeting vector. We designed a targeting vector containing murine Atxn3 intron 9, CAG repeat-expanded exon 10, and intron 10 for homologous recombination to replace the same region of wild-type murine Atxn3. During homologous recombination, the targeting vector inserted twice in tandem, resulting in a founder mouse harboring two copies of mutant exon 10 separated by intron 10, a PGK-driven thymidine kinase (PGK-TK), and intron 9. The neomycin (Neo) cassette was removed by FLP/FRT recombination. This resulting atypical SCA3 knockin mouse is characterized in this publication.
The remaining results in this publication, however, are accurate, including the characterization of this atypical SCA3 knockin mouse and the assessment of ATXN3 splicing in different SCA3 mouse models and patient fibroblasts. Importantly, the human ATXN3 gene naturally undergoes the mis-splicing of Atxn3 described in this manuscript. This atypical SCA3 knockin mouse model exhibits critical features of disease and, as detailed in the follow-up manuscript, continues to provide insight into SCA3 pathogenesis (Ramani et al 2017, submitted). This atypical SCA3 knockin mouse model was used in one other publication by Zeng et al. and is currently housed at Jackson Laboratories (B6(Cg)-Atxn3tm1Hlp/J).
The Authors apologize for any confusion caused by this error.