Supplementary Material to:
Rapid NMR screening of RNA secondary structure and binding
Christina Helmlingǂ, Sara Keyhaniǂ, Florian Sochor, Boris Fürtig, Martin Hengesbach, Harald Schwalbe1,*
1Institut für Organische Chemie und Chemische Biologie, Center for Biomolecular Magnetic Resonance (BMRZ), Johann Wolfgang Goethe-Universität, Frankfurt/M., 60438, Germany
ǂJoint first authors
* To whom correspondence should be addressed. Tel:+496979829130Fax: +496979829130 Email:
S1: DMSO effect on 3’ end transcript homogeneity
The effect of DMSO on 3’ end transcript homogeneity was investigated on additional RNA sequences in order to assess the generality of the approach. For transcriptions, standard primers were used to focus on homogeneity effects caused by DMSO and not 2’-O-methylation of the primers.For this study, we chose A/U rich sequences, which caused severe amounts of non-DNA-templated nucleotide addition (>+4). First, we illustrate the effect of DMSO on 3’ end homogeneity on a series of rather short constructs (30 nt – 39 nt) to unambiguously resolve also residual n+1 transcripts (Figure S1). Transcriptions performed in the absence of DMSO clearly show heterogeneous product formation with the main products being significantly larger than the desired RNA. Incorporation of a G (35 nt. transcript) and C (39 nt. transcript) considerably reduces the transcription of additional nucleotides. Transcript 30 also shows decreased heterogeneity, which could be caused by additional C residues involved in the DNA/RNA duplex occurring prior to residue U22. Figure S1b shows a significant decrease in 3’ end heterogeneity by the addition of 20% of DMSO. In the presence of DMSO, only n+1 transcripts (< ~40%) can be detected. The formation of n+1 transcripts can presumably be further reduced by using 2’-OMe primers as shown in Figure 2. Direct comparison of constructs 31-35 transcribed in the absence and presence of DMSO shows that only construct 35 is transcribed with the correct length in the absence of DMSO. Constructs 33 and 34 show a minor fraction of n+1 product formation with the major transcripts being >+4. Construct 31 transcribes n+1 and n-1 in the absence of DMSO. Figure S1d shows a DMSO optimization of the 37 nt transcript. According to this optimizations, the largest yield and homogeneity are acquired when using 20% of DMSO in the transcription mixture.
Supplementary Fig.S120% denaturing polyacrylamide gels investigating the effect of the cosolvent DMSO on the 3’ end homogeneity of RNA transcripts containing a 13 nt A/U stretch (highlighted in red) near the 3’ end. adenaturing polyacrylamide gel of 10 RNA constructs differing in one nt length transcribed under standard conditions and bin the presence of 20% of DMSO. c direct comparison of five selected RNA constructs shown in a and b, where lanes marked with D correspond to transcriptions performed in 20% of DMSO. The loading mixture contained 50% formamide, 10% of the transcription mixture for a and 1% of the transcription mixture for b (0.2% DMSO), respectively.Gels were stained with GelRedTM.dDMSO optimization for the 37 nt. transcript. The gel was visualized by UV shadowing.
S2: Application of the method to different RNA
In order to further substantiate the improvement in 3’ end homogeneity, we show its effect for two additional larger RNA constructs, which were also transcribed from A/T rich DNA templates, in Figure S2.
Sequence: AAUGAAUAUAAAAGAAACUUAUACAGGGUAGCAUAAUGGGCUACUGACCCCGCCUUCAAACCUAUUUGGAGACUAUAAGUGAAAAACCACUCUUUAAUUAUUAAAGUUUCUUUUUAUGUCCAAAAGACAAGAAGAAA
Supplementary Fig. S2 12% denaturing polyacrylamide gels for different RNA constructs transcribed in the absence (a) and presence (b) of 20% of DMSO.The transcripts correspond to intermediates of the sequence specified above with 104-112nt transcripts highlighted in red and 134-138nt transcripts highlighted in green. The loading mixture contained 50% formamide and 50% of the transcription mixture. Gels were visualized by UV shadowing.
S3: Effect of washing volume on residual Mg2+/spermidine binding
Figure S3 illustrates the improvement in NMR spectral qualityfor dGsw80byincreasing the washing volume. Loop-loop reporter signals can be detected when washing with 60 mL of NMR buffer (U66) and the resulting spectrum shows a significant amount of minor signals, which suggests RNA heterogeneity. An increase in the washing volume to 120 mL homogenizes the mixture and the resulting NMR spectrum is comparable to HPLC purified RNA.
Supplementary Fig. S31H NMR spectra of dGsw80 obtained after washing with 60 mL and 120 mL of NMR buffer and comparison to a 1H NMR spectrum obtained from HPLC-purified dGsw80.