Skip to main content
Open Biology logoLink to Open Biology
. 2018 Feb 14;8(2):180010. doi: 10.1098/rsob.180010

MHF1-2/CENP-S-X performs distinct roles in centromere metabolism and genetic recombination

Sonali Bhattacharjee, Fekret Osman, Laura Feeney, Alexander Lorenz, Claire Bryer, Matthew C Whitby
PMCID: PMC5830538  PMID: 29445032

Open Biol. 3, 130102. (Published online 11 September 2013). (doi:10.1098/rsob.130102)

We wish to retract the research paper ‘MHF1-2/CENP-S-X performs distinct roles in centromere metabolism and genetic recombination’. We have discovered that six of the fission yeast strains used in this study (MCW5895, MCW5932, MCW6001, MCW6141, MCW6142 and MCW6152) contain a fml1ΔC1-603::natMX4 mutation instead of the stated fml1AAA::natMX4 mutation. When we re-analysed strains with the correct fml1AAA::natMX4 mutation, we were unable to reproduce these reported observations: an increase in meiotic crossovers (figure 4d; electronic supplementary material, table S1); a reduction in RTS1-induced direct repeat recombination (figure 5c); a change in Mhf1-GFP nuclear localization (figure 6); and an increase in the number and length of mitotic bridges (figure 7c,d). Moreover, a fml1AAA::natMX4 strain exhibits only very modest hypersensitivity to MMS, which is significantly less than shown in figure 3b. We have also discovered that a fml1ΔC1-603::natMX4 mutant exhibits a similar hypersensitivity to genotoxins as a fml1Δ strain, and not the intermediate level of sensitivity shown in figure 3b. However, we were able to reproduce the result that a fml1AAA::natMX4 mutant exhibits an increase in mitotic crossovers (figure 4b) and confirm that a C-terminal fragment of Fml1, containing the AAA mutation, has a weakened interaction with Mhf1–Mhf2 in vitro (figure 2g). Data from the analysis of the fml1AAA::natMX4 mutant were used to support our conclusion that a direct physical interaction between Fml1 and Mhf1–Mhf2 is needed to promote Fml1's activities in DNA repair and recombination. In our recent work, we have been able to validate this conclusion using a mutation that more severely weakens the interaction between Fml1 and Mhf1–Mhf2 than the AAA mutation (Neo J.P.S., Wong I.N., Osman F. and Whitby M.C. 2016, unpublished data). We have also confirmed that the other data in our paper are robust and reproducible and the central conclusions of the paper remain true. We sincerely apologize for any inconvenience that publication of the data pertaining to the fml1AAA::natMX4 mutant has caused for others.

Sonali Bhattacharjee

One Bungtown Road, Cold Spring Harbor, NY 11724

bhattacharjee@cshl.edu

Fekret Osman

Laura Feeney

Memorial Sloan Kettering Cancer Center, 430 East 67th Street, New York, NY 10065

feeneyl@mskcc.org

Alexander Lorenz

University of Aberdeen, Institute of Medical Sciences, Aberdeen AB25 2ZD

a.lorenz@abdn.ac.uk

Claire Bryer

Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, B15 2TT

C.Bryer@bham.ac.uk

Matthew C. Whitby

University of Oxford, South Parks Road, Oxford OX1 3QU

matthew.whitby@bioch.ox.ac.uk

Prof. David M. Glover FRS

Editor-in-Chief, Open Biology


Articles from Open Biology are provided here courtesy of The Royal Society

RESOURCES