Prokaryotic cold shock proteins (CSPs) are considered to play an important role in the transcriptional and translational regulation of gene expression, possibly by acting as transcription anti-terminators and “RNA chaperones”. They bind with high affinity to single-stranded nucleic acids. Here we report the binding epitope of TmCsp from Thermotoga maritima for both single-stranded DNA and RNA, using heteronuclear 2D NMR spectroscopy. At “physiological” growth temperatures of TmCsp (≥ 343 K), all oligonucleotides studied have dissociation constants between 1.6 ((dT)7) and 25.2 ((dA)7) μM as determined by tryptophan fluorescence quenching. Reduction of the temperature to 303 K leads to a pronounced increase of affinity for thymidylate (dT)7 and uridylate (rU)7 heptamers with dissociation constants of 4.0 and 10.8 nM, respectively, whereas the weak binding of TmCsp to cytidylate, adenylate, and guanylate heptamers (dC)7, (dA)7, and (dT)7 is almost unaffected by temperature. The change of affinities of TmCsp for (dT)7 and (rU)7 by approximately 3 orders of magnitude shows that it represents a cold chock sensor that switches on the cold shock reaction of the cell. A temperature dependent conformational switch of the protein is required for this action. The binding epitope on TmCsp for the ssDNA and RNA heptamers is very similar and comprises β-strands 1 and 2, the loop β1–β2 as well as the loops connecting β3 with β4 and β4 with β5. Besides the loop regions, surprisingly, mainly the RNA-binding motif RNP1 is involved in ssDNA and RNA binding, while only two amino acids, H28 and W29, of the postulated RNA-binding motif RNP2 interact with the uridylate and thymidylate homonucleotides, although a high affinity in the nanomolar range is achieved. This is in contrast to the binding properties of other CSPs or cold shock domains, where RNP1 as well as RNP2 are involved in binding. TmCsp takes up a unique position since it is the only one which possesses a tryptophan residue instead of a usually highly conserved phenylalanine or tyrosine residue at the end of RNP2. NMR titrations suggest that neither (dT)7 nor (rU)7 represent the full binding motif and that non-optimal intercalation of W29 into these oligonucleotides blocks the access of the RNP2 site to the DNA or RNA. NMR-experiments with (dA)7 suggest an interaction of W29 with the adenine ring. Full binding seems to require at least one single purine base well-positioned within a thymine- or uracil-rich stretch of nucleic acids.
Read full abstract