Summary
Nature has retracted a 2014 study that reported miRNAs and the NMD pathway regulate ribosomal frameshifting in CCR5 mRNA. The journal says a cryptic splice site created artefactual reporter variants, and follow-up assays found no detectable frameshifting with a revised reporter.
Nature has retracted a 2014 paper that reported ribosomal frameshifting in CCR5 mRNA was regulated by microRNAs and the nonsense-mediated mRNA decay (NMD) pathway. In a retraction note published on 1 October 2026, the journal’s editors said they no longer had confidence in data supporting one of the study’s key conclusions.
The concern centres on the reporter used to detect programmed −1 ribosomal frameshifting. In this process, a ribosome shifts back by one nucleotide while reading an mRNA, changing how it groups the sequence into codons from that point onward.
A splice site confounded the reporter assay
The original study placed a CCR5-derived sequence in a dual-luciferase reporter: an experimental system that uses the output of two light-producing enzymes to study gene expression and recoding. In that construct, transcription was driven by an SV40 promoter, and an artificial intron was present.
The retraction note says further investigation found a cryptic 3′ splice acceptor site in the CCR5-derived sequence. A splice acceptor is a sequence that can direct the removal of part of an RNA molecule during processing. Here, its activity generated unintended splice variants. The editors say this splicing artefact could account for the apparent frameshifting signal.
The authors attempted to reproduce the result with an improved reporter in which the SV40 promoter was replaced by a strong CMV promoter and the artificial intron was removed. In that system, the CCR5-derived sequence did not produce detectable levels of programmed −1 frameshifting. The note says these findings have also been replicated by at least two other groups.
What the retraction changes
The original paper’s proposed link between CCR5 frameshifting, microRNAs and NMD depended on the reported frameshifting activity. The editors say the concerns about the reporter data undermine the study’s main conclusion, and the paper is now retracted. The follow-up result is specific to the revised reporter assay: it found no detectable frameshifting signal with that construct.
The authors did not agree unanimously with the retraction. Ashton Trey Belew, Sharmishtha Musalgaonkar, Vivek M. Advani and Sergey O. Sulima agreed; Arturas Meskauskas, Wojciech K. Kasprzak, Bruce A. Shapiro and Jonathan D. Dinman disagreed.