Molecular basis of the selective processing of short mRNA substrates by the DcpS mRNA decapping enzyme

Significance In eukoryotes, 3′ to 5′ mRNA degradation is a major pathway to reduce mRNA levels and, thus, an important means to regulate gene expression. Herein, messenger RNA (mRNA) is hydrolyzed from the 3′ end by the exosome complex, producing short capped RNA fragments, which are decapped by DcpS. Our data show that DcpS is only active on mRNA that have undergone prior processing by the exosome. This DcpS selection mechanism is conserved from yeast to humans and is caused by the inability of the enzyme to undergo structural changes that are required for the formation of a catalytically active state around long mRNA transcripts. Our work thus reveals the mechanistic basis that ensures an efficient interplay between DcpS and the exosome.
The 5′ messenger RNA (mRNA) cap structure enhances translation and protects the transcript against exonucleolytic degradation. During mRNA turnover, this cap is removed from the mRNA. This decapping step is catalyzed by the Scavenger Decapping Enzyme (DcpS), in case the mRNA has been exonucleolyticly shortened from the 3′ end by the exosome complex. Here, we show that DcpS only processes mRNA fragments that are shorter than three nucleotides in length. Based on a combination of methyl transverse relaxation optimized (TROSY) NMR spectroscopy and X-ray crystallography, we established that the DcpS substrate length-sensing mechanism is based on steric clashes between the enzyme and the third nucleotide of a capped mRNA. For longer mRNA substrates, these clashes prevent conformational changes in DcpS that are required for the formation of a catalytically competent active site. Point mutations that enlarge the space for the third nucleotide in the mRNA body enhance the activity of DcpS on longer mRNA species. We find that this mechanism to ensure that the enzyme is not active on translating long mRNAs is conserved from yeast to humans. Finally, we show that the products that the exosome releases after 3′ to 5′ degradation of the mRNA body are indeed short enough to be decapped by DcpS. Our data thus directly confirms the notion that mRNA products of the exosome are direct substrates for DcpS. In summary, we demonstrate a direct relationship between conformational changes and enzyme activity that is exploited to achieve substrate selectivity.