Supplemental Information. Natural RNA Polymerase Aptamers. Regulate Transcription in E. coli

Transcription

1 Molecular Cell, Volume 67 Supplemental Information Natural RNA Polymerase Aptamers Regulate Transcription in E. coli Nadezda Sedlyarova, Philipp Rescheneder, Andrés Magán, Niko Popitsch, Natascha Rziha, Ivana Bilusic, Vitaly Epshtein, Bob Zimmermann, Meghan Lybecker, Vitaly Sedlyarov, Renée Schroeder, and Evgeny Nudler

2 SUPPLEMENTAL INFORMATION Table S1. Full list of E. coli RAPs from RNAP SELEX, cycle 7 (related to Figure 1). Columns: RAP ID: RAP identification number Genomic coordinates: genomic location of the corresponding peak Strand: strand information for RAP location (+ positive, - negative) Categorization: true/false values indicate RAP category. Based on the genomic location we classify RAPs into several categories: 5pUTR (corresponding sequence aligns within any annotated 5 UTR), 3pUTR (corresponding sequence aligns within any annotated 3 UTR), intragenic (corresponding sequence aligns within the annotated gene, excluding cases of 5 - and 3 UTR), antisense (corresponding sequence aligns on the strand opposite to the annotated gene) and intergenic (corresponding sequence in between the annotated genes). Mapped Bases: number of mapped bases within peak interval (within RAP) Table S2. List of E. coli RAPs with at least 1 stable 3 end within their sequence extended 75 nt downstream (related to Figure 2). Columns: RAP ID: RAP identification number Genomic coordinates: genomic location of the corresponding peak Strand: strand information for RAP location (+ positive, - negative) Categorization: true/false values indicate RAP category. Based on the genomic location we classify RAPs into several categories: 5pUTR (corresponding sequence aligns within any annotated 5 UTR), 3pUTR (corresponding sequence aligns within any annotated 3 UTR), intragenic (corresponding sequence aligns within the annotated gene, excluding cases of 5 - and 3 UTR), antisense (corresponding sequence aligns on the strand opposite to the annotated gene) and intergenic (corresponding sequence in between the annotated genes). Mapped Bases: number of mapped bases within peak interval (within RAP) Table S3. List of E. coli antisense RAPs being expressed under normal growth conditions in rich medium (related to Figure 6). Columns: RAP ID: RAP identification number Coordinate min/coordinate max: genomic location of RAP (of the corresponding peak) Strand: strand information for RAP location (+ positive, - negative) Opposite gene: annotated gene opposite to the antisense RAP

3 Table S4. List of oligonucleotide sequences used in this study (related to STAR Methods, Key Resource Table). ID Sequence qrt-pcr primers lacz _FW AGCGCGATCCCGTCGTTTTACA lacz _RV CAGGCTGCGCAACTGTTGGG GFP_FW TGGAGAGGGTGAAGGTGAT GFP_RV AGCATTGAACACCATAAGTCAAAG gapa_fw GCACCACCAACTGCCTGGCT gapa_rv CGCCGCGCCAGTCTTTGTGA nadd_beforerap_fw ACAGGCTCTGTTTGGCGGCAC nadd_beforerap_rv CGCCAGCGTTTCCACGGGT nadd_afterrap_fw AGCGAACAGCGTGCAGCGTA nadd_afterrap_rv TGCGCAGTGTAAGAGGGGGC Northern hybridization probes P1: before RAP TTAACCAGTAACAACAGAATTCTAGCCC P2: after RAP TTATTATCTAGAGGATCCCCGGGTGCATT Some of the inserts / RAPs tested in reporter constructs (see also Table 1) GTCACAATCATCCCTAATAATGTTCCTCCGCATCGTCCC RAP #15 (insert for reporter) 15 mut (insert for reporter) GTCACAATCATCCCTGGTAATGTACGACCATGACGTCCC 15rev (insert for reporter) RAP 1086 RAP 2667 RAP 7768 rut (canonical Rho-utilization site sequence) Templates for in vitro transcription #15 (Figures 3 and S3) GGGACGATGCGGAGGAACATTATTAGGGATGATTGTGAC GCGTTTCAAATGCGCATCAACCTGCCACACTCCCCCACA TCAACCTGATCGTGCCAGGACCATTCACACTCCAGTCCCAGTTC ATGT CACCATAGTACGACCACACGGCGGCCACCCCCATGG CCCTCAACGACCCCTTCCTTCTCCCCATCGCTACCTCATATCCG CACCTCCTCAAACGCTACCTCGACCAGCCTCCCTCCC tccagatcccgaaaatttatcaaaaagagtattgacttaaagtctaacctataggatactt acagccatcgagagggccgggctagaattctgttgtta CTGGTTAACCTGAAACGCCAGTCTGCCCATACGCCA CTGCGTGTCACAATCATCCCTAATAATGTTCCTCC GCATCGTCCCCCCATGGTTTCGCGTTGACCAGGGGT GTTGGAGCGCCATTGATATCAACCAAAAATTCGCGA CGCTACGCGTCCTCAATAGCCGTGCCATCGGGGTCG AGAGTGCACCCGGGGATCCTCTAGA #1086 tccagatcccgaaaatttatcaaaaagagtattgacttaaagtctaacctataggatactt acagccatcgagagggccgggctagaattctgttgtta CTGGTTAACCTGAAACGCCAGTCTGCCCATACGCCA CTGCGTGCGTTTCAAATGCGCATCAACCTGCCACA CTCCCCCACACCCATGGTTTCGCGTTGACCAGGGGT GTTGGAGCGCCATTGATATCAACCAAAAATTCGCGA CGCTACGCGTCCTCAATAGCCGTGCCATCGGGGTCG AGAGTGCACCCGGGGATCCTCTAGA

4 #2667 tccagatcccgaaaatttatcaaaaagagtattgacttaaagtctaacctataggatactt acagccatcgagagggccgggctagaattctgttgtta CTGGTTAACCTGAAACGCCAGTCTGCCCATACGCCA CTGCGTTCAACCTGATCGTGCCAGGACCATTCACA CTCCAGTCCCAGTTCATGTCCCATGGTTTCGCGTT GACCAGGGGTGTTGGAGCGCCATTGATATCAACCAA AAATTCGCGACGCTACGCGTCCTCAATAGCCGTGCC ATCGGGGTCGAGAGTGCACCCGGGGATCCTCTAGA #7768 tccagatcccgaaaatttatcaaaaagagtattgacttaaagtctaacctataggatactt acagccatcgagagggccgggctagaattctgttgtta CTGGTTAACCTGAAACGCCAGTCTGCCCATACGCCA CTGCGTCACCATAGTACGACCACACGGCGGCCAC CCCCATGGTTTCGCGTTGACCAGGGGTGTTGGAGCG CCATTGATATCAACCAAAAATTCGCGACGCTACGCG TCCTCAATAGCCGTGCCATCGGGGTCGAGAGTGCAC CCGGGGATCCTCTAGA 15 mut (Figure S3) T7A1 promoter fused with 20 nts for in vitro transcription tccagatcccgaaaatttatcaaaaagagtattgacttaaagtctaacctataggatactt acagccatcgagagggccgggctagaattctgttgtta CTGGTTAACCTGAAACGCCAGTCTGCCCATACGCCA CTGCGTGTCACAATCATCCCTGGTAATGTACGACC ATGACGTCCCCCCATGGTTTCGCGTTGACCAGGGGT GTTGGAGCGCCATTGATATCAACCAAAAATTCGCGA CGCTACGCGTCCTCAATAGCCGTGCCATCGGGGTCG AGAGTGCACCCGGGGATCCTCTAGA tccagatcccgaaaatttatcaaaaagagtattgacttaaagtctaacctataggatactt acagccatcgagagggacacggcgaat nadd RAP 15 (Figure S5D) nadd reverse 15 (Figure S5D) GCCAACACTTGTCACTACTtccagatcccgaaaatttatcaaaaagagtattg acttaaagtctaacctataggatacttacagccatcgagagggacacggcga ATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTTGATCCGG TGCACTATGGTCATCTAAAACCCGTGGAAACGCTGGCGAATTT GATTGGTCTGACGCGGGTCACAATCATCCCTAATAATGTTCC TCCGCATCGTCCCCAGCCGGAAGCGAACAGCGTGCAGCGTAA ACACATGCTTGAACTGGCGATTGCCGACAAGCCATTATTTACTC TTGATGAACGCGAGCTAAAGCGCAATGCCCCCTCTTCCATGCCC GAAGGTTATGT GCCAACACTTGTCACTACTtccagatcccgaaaatttatcaaaaagagtattg acttaaagtctaacctataggatacttacagccatcgagagggacacggcga ATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTTGATCCGG TGCACTATGGTCATCTAAAACCCGTGGAAACGCTGGCGAATTT GATTGGTCTGACGCGGGGGACGATGCGGAGGAACATTATTA GGGATGATTGTGACCAGCCGGAAGCGAACAGCGTGCAGCGTA AACACATGCTTGAACTGGCGATTGCCGACAAGCCATTATTTACT CTTGATGAACGCGAGCTAAAGCGCAATGCCCCCTCTTCCATGCC CGAAGGTTATGT nad-gfp translational fusions see also Figure 5 (here are the inserts introduced to pgfp vector between EcoRI and NcoI unique sites)

5 nad(rap 15)-GFP GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGGTCACAATCATCCCTA ATAATGTTCCTCCGCATCGTCCCCAGCCGGAAGCGAACAGC GTGCAGCGTAGTAAAGGAGAAGAACTTTTCACTGGAGTTGTCC CAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCACAAATTT TCTGTCAGTGGAGAGGGTGAAGGTGATGCAACATACGGAAAAC TTACCCTTAAATTTATTTGCACTACTGGAAAACTACCTGTT nad(rev 15)-GFP GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGGGGACGATGCGGAGG AACATTATTAGGGATGATTGTGACCAGCCGGAAGCGAACAG CGTGCAGCGTAGTAAAGGAGAAGAACTTTTCACTGGAGTTGT CCCAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCACAAAT TTTCTGTCAGTGGAGAGGGTGAAGGTGATGCAACATACGGAAA ACTTACCCTTAAATTTATTTGCACTACTGGAAAACTACCTGTT nad(rut)-gfp GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGCCCTCAACGACCCCTT CCTTCTCCCCATCGCTACCTCATATCCGCACCTCCTCAAAC GCTACCTCGACCAGCCTCCCTCCCCAGCCGGAAGCGAACAG CGTGCAGCGTAGTAAAGGAGAAGAACTTTTCACTGGAGTTGT CCCAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCACAAAT TTTCTGTCAGTGGAGAGGGTGAAGGTGATGCAACATACGGAAA ACTTACCCTTAAATTTATTTGCACTACTGGAAAACTACCTGTT nad(rap 683)-GFP GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGTTAAACGAATGTCTGC ATATATTGTGGCGTATTCGCTTTGCCCTGCATCTGCAGCCG GAAGCGAACAGCGTGCAGCGTAGTAAAGGAGAAGAACTTTT CACTGGAGTTGTCCCAATTCTTGTTGAATTAGATGGTGATGTTA ATGGGCACAAATTTTCTGTCAGTGGAGAGGGTGAAGGTGATGC AACATACGGAAAACTTACCCTTAAATTTATTTGCACTACTGGAA AACTACCTGTT nad(rap 1510)-GFP GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGCAACCCGTATTCCTCG AAGTAGTGGATGAAAGCTATCGTCACAAcCAGCCGGAAGCG AACAGCGTGCAGCGTAGTAAAGGAGAAGAACTTTTCACTGGA GTTGTCCCAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCA CAAATTTTCTGTCAGTGGAGAGGGTGAAGGTGATGCAACATAC GGAAAACTTACCCTTAAATTTATTTGCACTACTGGAAAACTACC TGTT nad(rap 2136)-GFP GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGCACTTCACCATCGATC CTTCCCGCATTAAACAACATGTCCGTCAGCCGGAAGCGAAC AGCGTGCAGCGTAGTAAAGGAGAAGAACTTTTCACTGGAGTT GTCCCAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCACAA ATTTTCTGTCAGTGGAGAGGGTGAAGGTGATGCAACATACGGA

6 AAACTTACCCTTAAATTTATTTGCACTACTGGAAAACTACCTGT T nad(rap 10243)-GFP GGCTTATTCCCTAACTAACTAAAGATTAACTTTATAAGGAGGA AAAACATatgAAATCTTTACAGGCTCTGTTTGGCGGCACCTTT GATCCGGTGCACTATGGTCATCTAAAACCCGTGGAAACGCT GGCGAATTTGATTGGTCTGACGCGGACCAAACCAACGTACA GAGCGTACAAGCCAACACAGCCGGAAGCGAACAGCGTGCA GCGTAGTAAAGGAGAAGAACTTTTCACTGGAGTTGTCCCAATT CTTGTTGAATTAGATGGTGATGTTAATGGGCACAAATTTTCTGT CAGTGGAGAGGGTGAAGGTGATGCAACATACGGAAAACTTACC CTTAAATTTATTTGCACTACTGGAAAACTACCTGTT

7 SUPPLEMENTAL FIGURES Figure S1. Identification of RAPs, natural RNAP-binding RNA aptamers, within the E. coli genome (related to Figure 1). (A) Common MEME-predicted motif for 12% of RAPs. (B) Exemplary screen shot of deep sequencing results of genomic SELEX. Read coverage for the peak representing intragenic RAP-15. Zoomed-in fragment of nadd gene with read coverage and tested sequence of intragenic RAP-15 (shown as blue arrow). (C) Checking RAP interaction with E. coli RNAP. RNAP binding activity of several individual iraps. 0.4 nm 32 P-labeled in vitro synthesized RNA of each type was used in the gel shift experiment with nm of RNAP (as was previously performed for the whole pool of aptamers in (Windbichler et al., 2008)). Binding reaction was performed for 15 minutes in 40 mm Tris-HCl, ph 8.0; 10 mm MgCl2, 100 mm NaCl, followed by separation on a nondenaturing 4% TBE polyacrylamide gel at 4 C.

8 Figure S2. RAPs activity in the GFP reporter system (related to Figure 2). (A) Reporter constructs. Left panel: pgfp construct with the GFP reporter only. Right panel: transcriptional GFP-based fusion prap#-gfp used to test the effect of RAPs (shown as grey rectangle) in cis. cp (grey triangle) indicates the location of a constitutive promoter. Grey arrow depicts the transcription start site; RBS is the ribosome binding site.

9 (B) RAP effects on transcription as detected by realtime fluorescence. E. coli cells transformed with empty construct (pgfp) and two vectors with different RAPs (#15 and #393) grew in LB for more than 11 hours while cell optical density (left) and fluorescence intensities (right; shown in relative units, RU) were monitored simultaneously. Grey dashed lines indicate late exponential and stationary growth phases. For each time point values represent means ±SD, n = 3. (C) Levels of GFP expressed from constructs with different iraps (labeled with numbers) normalized to the amount of bacterial cells (Fluorescence Intensities, nu) measured in two different growth phases late exponential (after 5 hours growth; upper panel) and stationary (after 10 hour growth; lower panel). Grey line indicates fluorescence intensity (in %) relatively to the empty pgfp construct. RAP-393 has no inhibiting effect on GFP expression; it was used as a control. For each time point values represent means ±SD, n = 3. (D)-(E) Nucleotide distribution within 24 tested inhibitory RAPs. (D) Average nucleotide frequency for tested iraps. Values represent means ±SD. (E) Nucleotide frequency for each tested irap.

10 Figure S3. (A)-(B) 3 RACE analysis for the RAP-15-containing transcripts (related to Figures 2 and 3). (A) Alignment of sequences obtained by 3 RACE: RAP-containing transcripts, representing the products of the preliminary RAP-dependent transcription termination. The location of RAP is shown in red rectangle, the location of 3 RACE primers is marked with blue rectangles and labeled accordingly. (B) 3 RACE analysis for the RAP-15-containing transcripts before and after BCM treatment. 3 RACE PCR product resulting from two biological replicates were separated on 6.5 % polyacrylamide gel. The prevailing short transcript with RAP-15 is labeled with ST. The nonspecific amplification product is marked with an asterisk. The addition of BCM (Rho inhibitor) to the bacterial culture leads to the pronounced accumulation of longer RAPcontaining transcripts (marked with blue bar), at the same time decreasing the levels of the ST transcript (pointed with red arrows).

11 (C)-(E) Mutational analysis of irap-15 (related to Figures 2 and 3). (C) RAP-15 sequence translated into the corresponding protein sequence. RAP-15 is located within nadd, encoding the conserved region of nicotinate-mononucleotide adenylyltransferase. The introduced mutations are within the sequences encoding 2 out of 6 conserved catalytic amino acids (shown in green). (D) Upper panel: schematic of the lacz-based construct used to test the effect of RAPs placed upstream of the reporter (grey rectangle; genomic context as in the reporter from Figure 2B). Schematic of the template used for in vitro transcription with tested RAPs (grey rectangle) is shown below. Lower panel: β-galactosidase activity with different RAP-LacZ reporters. Numbering on the bottom indicates the insert (RAP or mutant) in the reporter construct. Values represent means ±SD, n = 3. (E) Representative run-off transcription assay. Pre-formed elongation complexes were chased with NTPs in the absence of Rho (lanes 1, 4), in the presence of Rho (lanes 2, 5,) or with Rho and NusG (lanes 3, 6). The products of Rho-dependent termination (present in RAP-15- containing template, but not in its mutated version) are marked with red bars. Blue arrow points to the run-off products. (F) irap-15 demonstrates moderate affinity to Rho. RNA polymerase binding activity of 29 nt irap-15 (left panel) and its mutant variant 15-mut (right panel, see also Figure S3 D-F). 2 nm 32 P-labeled in vitro synthesized RNA of each type was used in the gel shift experiment with nm of Rho (calculated for Rho as a hexamer). Binding reaction was performed for 10 minutes in the reaction buffer: 40 mm Tris- HCl, ph 8.0; 10 mm MgCl2, 50 mm NaCl, followed by the separation on a non-denaturing 6% polyacrylamide gel at 4 C.

12 Figure S4. Effect of BCM on iraps activity in the GFP reporter system (related to Figure 3 and Table 1). (A) A schematic of the reporter construct used to test the effect of RAPs (grey rectangle) placed upstream of the GFP reporter gene. P (grey triangle) indicates the location of a constitutive bacterial RNAP promoter. RBS is the ribosome-binding site. (B), (D) E. coli GFP plate assay. Upper panels: E. coli cells transformed with GFP reporter plasmids containing different iraps were grown on LB agar plates (no bicyclomycin) followed by fluorescence intensity measurements (GFP mode). E. coli cells with GFP reporter without RAP insert are marked as "No RAP". RAP-393 (left panel) is an example of

13 a non-inhibitory RAP, as it does not lead to GFP signal decrease (see also Figure S2). Lower panels: same plates were captured under visible light showing the bacterial density of strains with RAP-GFP fusions (Light mode). (C), (E) E. coli GFP plate assay. Upper panels: E. coli cells transformed with GFP reporter plasmids containing different iraps were grown on LB agar plates supplemented with bicyclomycin (8-10 µg/ml) followed by fluorescence intensity measurements (GFP mode). E. coli cells with GFP reporter without RAP insert are marked as "No RAP". RAP-393 (left panel) is an example of a non-inhibitory RAP as it does not lead to GFP signal decrease (see also Figure S2). Lower panels: same plates were captured under visible light showing the bacterial density of strains with RAP-GFP fusions (Light mode).

14 Figure S5. (A)-(C) Conservation of irap-15 sequences among Enterobacteriaceae (related to Figures 2 and 4). (A) Multiple Sequence Alignment built with Clustal Omega (Li et al., 2015; Sievers et al., 2011) for RAP-15-like sequences from representatives of bacterial genera Shigella, Citrobacter, Enterobacter, Salmonella and Cedecea, demonstrating minimum 77% identity

15 to the original RAP-15 from E. coli. The resulting alignment was colored accordingly using MView tool (Li et al., 2015). (B) Neighbor-joining tree without distance corrections was built on the basis of multiple alignments from (A). (C) The in cis effect of RAP-15-like sequence from Citrobacter, one of the most divergent sequences from the original E. coli RAP-15 tested in pgfp-based system in E. coli. Upper panel: transcriptional fusion prap-gfp used to test the effect of RAPs (grey rectangle). cp (grey triangle) indicates the location of a constitutive promoter. Grey arrow depicts the transcription start site; RBS, the ribosome-binding site. Middle and lower panels: representative results from the GFP plate assay. DH5a cells transformed with no RAP construct or prap-gfp grew on LB agar plate over 16 hours and the fluorescence intensity was measured (GFP mode, middle panel). The same plate was also captured under visible light (Light mode, lower panel). RAP-15-like element from Citrobacter retains its irap activity in E. coli. (D) irap-15 promotes RNAP pausing (related to Figures 4 and 5). Left panel: schematic of a RAP-containing template used for the in vitro transcription assay. T7A1 promoter (grey arrow) is fused to the nadd-fragment encoding RAP-15 (nadd RAP- 15) or its reverse complement sequence, rev-15, as a control (nadd rev-15). Right panel: representative single-round runoff assay (in solution) utilizing the template with RAP-15 (lanes 7-12) and the control template rev-15 (lanes 9-16). Transcripts were analyzed 20, 40, 60, 120, 300 and 600 seconds after initiation of transcription. To determine the location of RAP-15 (indicated) sequencing was performed using 3 dntps (lanes 13 16). Transcript accumulation suggests that RAP-15 induces specific local pausing (compare lanes 1-4 vs. 7-10; marked with red arrows). In contrast to termination, pausing results in initial accumulation of a shorter transcript that subsequently disappears with time as the RNAP escapes and continues elongation to produce a full-length run-off transcript (blue bar).

16 Figure S6. (A) Ribosome occupancy profiles in the proximity of iraps (related to Figure 5). Ribosome occupancy data (RiboSeq coverage (Li et al., 2012)) shown relatively to the location of iraps (blue arrow/shaded in blue) within translated regions. Maximal peak of ribosome stalling upstream of irap is marked with red asterisk.

17 (B)-(C) iraps activity in the GFP translational fusions (related to Figure 5). (B) Fluorescence intensity signal range for different nad(rap#)-gfp fusions (schematically depicted on the upper panel) measured after 6 (late exponential) and 10 hours (stationary phase) of growth in rich media. Relative fluorescence values normalized to the number of cells per sample (nu). In contrast to the conventional rut sequence, iraps retain down regulating activity within the translated region. Values represent means ±SD, n 3. (C) Translated protein sequence fragments for constructs nad(rap#)-gfp from (B). Importantly, there are no additional stop codons introduced to the designed translational fusion constructs.

18 Figure S7. (A)-(B) Examples of iraps with stable 3 ends in proximity (related to Figure 3). Exemplary screen shots of deep sequencing results of genomic SELEX combined with the

19 massive 3 ends identification across E. coli genome. Mapped reads form peaks, representing RAPs (tracks 1-2 from top) or identified 3 end signals (tracks 3-4 from top). Stable 3 ends identified with custom algorithm are depicted on a separate track (Stable 3 ends). (A) Data for sense intragenic RAP-1510 and corresponding 3 end formation: zoomed-in fragment of bola gene with RAP-1510 and downstream 3 ends. (B) Data for antisense RAP and corresponding 3 end formation: zoomed-in fragment of antisense RAP (encoded on the strand opposite to nusa) and downstream 3 ends. (C)-(D) BCM-mediated effects of antisense iraps on the E. coli transcriptome profile (related to Figure 6). Exemplary screen shots of total RNA deep sequencing without and with BCM (Sedlyarova et al., 2016) with marked location of the antisense iraps. In both cases treatment with BCM affects the read ratio of positive and negative strands, resulting in the suppressed levels of sense transcription upon upregulation of the antisense transcript.