Characterization of the Smk: A riboswitch that binds S-adenosylmethionine

Riboswitches are conserved RNA sequences that regulate downstream gene expression via changes in the RNA structure. Riboswitches usually respond to an environmental stimulus, typically binding of an effector molecule, that causes a structural rearrangement without additional trans-acting factors. The structural rearrangement can cause or prevent the formation of an intrinsic transcriptional terminator to regulate the gene at the level of premature transcription termination, or it can modulate the accessibility of the ribosome binding site to regulate at the level of translation initiation. Previous work has shown that there are multiple classes of riboswitches that bind S-adenosylmethionine (SAM) and regulate genes involved in biosynthesis of methionine, cysteine and SAM. The S box (or SAM-I) riboswitches are found in a wide range of bacteria while the SAM-II riboswitch is found primarily in Proteobacteria. Both riboswitches recognize SAM in a similar manner, but they are very different in terms of sequence, secondary and tertiary structure. This demonstrates that it is possible for bacteria to evolve very different regulatory RNAs to respond to the same metabolite.;We discovered a third SAM binding riboswitch motif (called the S MK box or SAM-III) upstream of the metK gene (encoding SAM synthetase) in many Lactobacillales species. The SMK box RNA was shown to bind SAM and discriminate against the closely related analog S-adenosylhomocysteine (SAH). Through enzymatic probing, it was determined that SAM causes a structural rearrangement in the RNA that results in sequestration of part of the Shine-Dalgarno (SD) region. This observation was confirmed by X-ray crystallography, which also showed that part of the SD directly contacts SAM. A translational fusion of the Enterococcus faecalis metK leader to lacZ was made and introduced into Bacillus subtilis. When the B. subtilis cells were grown under conditions in which SAM pools are elevated, beta-galactosidase activity decreased. In contrast, a transcriptional fusion showed no effect. This suggested that the SMK box riboswitch down-regulates gene expression at the level of translation in response to SAM. This observation was supported by the fact that SAM inhibits binding of Escherichia coli 30S subunits to SMK box RNA in vitro. Mutant S MK box sequences that are deficient in SAM binding in vitro showed no SAM dependent effect on 30S subunit binding and no regulation of lacZ expression when SAM pools were modulated.;Reverse transcription and quantitative PCR techniques were utilized in order to analyze properties of the metK transcript in vivo in a native organism, E. faecalis. The abundance of metK transcript in E. faecalis cells was found to be unchanged during growth under conditions resulting in high or low SAM pools. This supported the model that regulation of metK by the SMK box is not at the level of transcription. The half-life (t1/2) of the metK transcript was determined to be ∼3 min regardless of SAM pools. In contrast, the t 1/2 of the interaction between SAM and the SMK box in vitro is 7.8 sec (A. Smith, unpublished results). Thus, the interaction of the SMK box with its effector was shown to be of much shorter duration than the persistence of the transcript in the cell, suggesting that the regulatory effect and structural changes resulting from SAM binding are reversible.;Attempts were made to demonstrate SMK box-mediated regulation in E. faecalis by utilizing a translational metK-gusA fusion and direct measurement of the activity of the metK gene product, SAM synthetase. Neither method showed regulation of gene expression due to SAM. The basis for these observations is unknown. It is possible that there are other factors involved in regulation of the metK gene in E. faecalis that are masking the effect of the SMK box in vivo.