CA2636427A1 - Method for identifying small rnas - Google Patents

Abstract

The present invention relates to a method for identifying small RNAs in genome. More particularly, the present invention is concerned with the design of a new in silico strategy which is able to identify known and new sRNAs in the genome of bacteria, such as Escherichia coli strains.

Description

METHOD FOR IDENTIFYING SMALL RNAS
FIELD OF THE INVENTION

The present invention relates to small RNAs and, in particular, to in silico method for identifying small RNAs.

BRIEF DESCRIPTION OF THE PRIOR ART

In Eubacterial cell metabolism, a growing number of cell regulatory pathways were subject to control by small ribonucleic acids. These non messenger, non transfer and non ribosomal RNAs were also called small RNAs (sRNA) or non coding RNAs (ncRNA) or regulatory RNAs (regRNA) because they are small molecules and not often translated. These ones are involved in regulation of gene expression at the post transcriptionnal level by modulating mRNA
traductability and stability (Gottesman 2005), protein catalytic activity (Pichon & Felden 2007) and protein synthesis quality control (Gillet & Felden 2001) as well as in the general cell metabolism (Gottesman 2005) and in the control of virulence expression (Romby et al. 2006). Small RNA genes are often located in the core genome of bacteria (Wassarman et al. 2001) but could be in the mobile genetic elements including transposable elements, plasmids and phages (Brantl 2007 for a review), and in pathogenicity islands (Pichon & Felden 2005).
In the case of protein coding genes, several software had been developed and showed a strong accuracy in open reading frame characterization. But identification of sRNA genes with in silico methodologies was a more recent quest (Pichon &
Felden 2008). At the end of the twenty's century, known sRNAs of the Escherichia coli (E. co/i)bacterium were identified "by chance" during in vitro or in vivo experiments. In consequence of, sRNA diversity and importance for the eubacterial phylum were unexplored although they were already known as global regulators of the cell metabolism (Wassarman et al. 1999) and controlled of virulence genes (Novick et al. 1993). First in silico experimentations used to identify new sRNA genes

2 in bacterial genome exploited two main hypotheses, reported for the first time in the pioneering work of Karen Wassarman and colleagues (Wassarman et a/. 2001).
They hypothesised that sRNA genes were conserved among closed relative bacteria genome, especially in the inter-genic regions (IGR) and found 17 new sRNAs in the E. coli genome. Since, numerous in silico methods based on comparative genomics (Argaman et al. 2001, Rivas et al. 2001, Pichon & Felden 2003, Axmann et al.
2005, Livny et al. 2005), statistics/probabilities analysis (Carter et al. 2001, Schattner 2002, Saetom et al. 2005, Yachie et al. 2006, Wang et al. 2006) and RNA secondary structure analysis (Rivas et al. 2001, Uzilov et al. 2007) had emerged with variable effectiveness (Pichon & Felden 2008 for a review). In spite of several efforts, exiting methods could be perfectible and some strategies had not been tested.
Identification of sRNA secondary structures by sRNA genefinders had been used to but displayed weak effectiveness (Rivas et al. 2000) or no in vivo validation (Uzilov et al. 2006). Past studies about RNA secondary structures elucidation had been showed that full or partial structure conservation of sRNAs between closed relative bacteria was not fully mediated by nucleic sequence conservation and some slightly different sequences could fold into the same RNA structure thanks to the appearance of compensatory mutations (Chen et al. 1999) (Figure 1A&B).
Following Kimura's definition (Kimura 1985), by compensatory mutations we mean a pair of mutations at different nucleotide sites that may be individually deleterious but are neutral in appropriate combinations. Thus, individual mutations occurring at nucleotide sites involved in Watson Crick pairs within RNA stems are expected to be deleterious if they break up the pairing of double strand structures; however, fitness can be restored, when a second, compensatory mutation occurs at the appropriate position on the opposite strand of the stem and reestablishes the pairing. Few RNA
structure finder algorithms identified compensatory mutations, a strong phylogenetic proof of the existence of a stem structure and could be a necessary step in the finding of a general sRNA genefinder algorithm as proposed in the past.

3 SUMMARY
The present invention concerns an in silico method for identifying small RNAs. More particularly, the present invention relates to a method for the identification of small RNAs, inside the sequence of a first nucleic acid molecule, comprising the following steps:
A) Preparation of a test sequence dataset of candidate sequences comprising:
i) identification of termination sites of transcription elongation complexes in said first nucleic acid molecule, ii) elimination of all termination sites whose sequences are located near the stop codon of an open reading frame with the same nucleic acid strand orientation, iii) gathering of nucleic acid sequences comprising the sequence, between 50 to 500 nucleotides, upstream each remaining termination sites after step (ii) plus the termination site sequences themselves;

B) Identification of at least one nucleic acid sequence sharing significant sequence identities for each sequence from the test sequence dataset, C) Multi alignment of all nucleic acid sequences sharing significant identities with a sequence from the test sequences dataset, D) Identification of those nucleic acid sequences having each one at least two partially conserved stem loops (small RNAs) and further identification of at least one putative compensatory mutation in each stem, E) Analysis of the putative expression of each candidate small RNA genes.

4 BRIEF DESCRIPTION OF THE FIGURES

Figure 1: Definition of a compensatory mutation. (A) Multi-alignment of a theorical sRNA. (B) Secondary structures from the previous sRNA showing compensatory mutations.

Figure 2: Workflow of sRNA genefinder based on MASR and CSSR utilization.
Figure 3: Observation of 7 candidates that are positively expressed with the 5S gene as expression reference.

Figure 4 illustrates the sequence of 10 small RNAs, identified in the genome of E. coli strain 55989.

DETAILED DESCRIPTION OF AN EMBODIMENT OF THE INVENTION

The inventors described the design and validation of a new in silico strategy which was able to identify known and new sRNAs in the genome of some Escherichia coli strains. The inventors implement a package of two softwares which are able to analyze automatically a set of sequences ended by a predicted rho-independent terminator. The program combined comparative genomic and RNA
structure prediction based on identification of compensatory mutations especially in the stem of rho-independent terminators which led to the discovery of new sRNA
genes without taking care of its genome locus. This method used less genetic dependent features than any other in silico identification methodologies and enabled rapid and efficient analysis of any bacterial genomes. As previously showed for Staphylococcus aureus (Pichon et al. 2005) and Salmonella typhymurium (Pfeiffer et al. 2007), the inventors have identified new sRNA genes in the pathogenicity islands of EAEC and ExPEC strains which can be used to control a present virulence of E.
coli.

Materials & Methods Nucleic sequences

5 All available genomic sequences of Enterobacterae were obtained from Genbank database (http://www.ncbi.nlm.nih.gov/genbank/) and listed in Table 1.
All know sRNAs previously identified in Eubacteria core genomes were listed in Table 2.
The sequence from these sRNAs was get from genome sequence where it was first characterized. All bioinformatics data and biochimical experiements available in particular 5' and 3' rapid amplification of chimeric ends (RACE), primer extension analysis, and Northern blot were used for characterization of each sRNA
primary transcript ends, promoter and terminator localisation. At default, when no sufficient data were available, bioinformatic predictions of sRNA sequences were used.

Summary of the MASR program Multiple alignement of Small RNAs (MASR) software was composed of two perl script whose functions were to select and align putative sRNA sequences.
Sequences suspected to contained sRNA gene (called the reference sequence) was analysed with the FASTA 3 program (Pearson 2000) or any identities search programs (e.g. BLAST) against a database containing DNA sequences (e.g.
genomic sequence). The resulting hit sequences was analysed by a first script called PPSA (Post Process of Simple Alignments) in order to characterize the best significant hit results in function of 25 different parameters alone or in combination, including e-value, size of the hit sequence and percentage of conservation.
The PPSA script generated an output file specific of the MASR (Multi Alignment of Small RNAs) script. The principle of MASR was based on the following affirmations;
when the kth nucleotide of the reference sequence noted Nk and the ith nucleotide include in the alignment area of the sequence j (noted N;j) were at the same position in a simple alignment (generated by FASTA3), they must be aligned in the multiple alignment. When the Nijth nucleotide had no nucleotide match in the reference

6 sequence, a gap was created.

Summary of the CSSR program The Covariation Search in Small RNAs (CSSR) software is designed to search structurally conserved stem loops in a set of aligned sequences (make by MASR
program). When one of them is found, it searched for compensatory mutations into and discarded all fully conserved stems. The CSSR program requires one or more RNA secondary structure generated by any kind of prediction softwares or get available structural data. The software generates a file containing all stem loops with putative compensatory mutation locations aligned along the reference sequence.
All Watson Crick pairs which change to or from a GU weak pair are added to the results as a putative site for appearance of compensatory mutations.

Identification of rho-independent terminators Rho-independent terminator localisation were predicted with the RNAMotif program (Macke et al. 2001) using the previously described method of Elena Lesnik and co-workers with modifications (Lesnik et al. 2001). In order to calculated score, the free Gibbs energy (oG 37) of the RNA stems were calculated with the efn2 subroutine of the RNAMotif program. The OG 37 of the RNA:DNA hybrid duplex of the predicted poly uracyl tail of the terminator and its corresponding genomic DNA
sequences was calculated with the Melting 4 program (Le Novere 2001) and use of the nearest neighbour thermodynamic parameters for RNA:DNA duplex formation (Sugimoto et al. 1995). The score for all the terminators were calculated as described (Lesnik et al. 2001) and those with a score > -4.0 kcal/mol were removed.
In silico identification of smali RNAs in E. coli genomic sequences.

A reference sequence dataset composed of the sequence of predicted rho-independent terminator (see upper section) with their 200 upstream nucleotides were

7 analyzed for sequence conservations against a genome sequences database (Table 1) with the FASTA 3.4 program (Pearson 2000). Hit results were analyzed with PPSA in order to exclude those with an e-value greater than 0.0001. The MASR
program was applied to transform individual alignment in a multi alignment with it default parameters. The global RNA structures of reference sequences were predicted with the Mfold 3.2 program (Mathews et al. 1999). The CSSR program combined the multi alignments of MASR and RNA secondary structure predictions in order to find at least two conserved structures which harboured compensatory mutations including the rho-independent terminator stem.
Evaluation of efficiency of the in silico method.

84 sRNAs had been characterized in the genome of the E. coli MG1655 strain and approximatively 70% of them used rho-independent termination of the transcription (Table 2). Efficiency of MASR/CSSR was evaluated by calculating the percentage of rho-independent terminated sRNAs found by them in the E. coli MG1655 genome by applying the upper section workflow (Figure 2).

Bacterial strain and growth conditions All E. coli strains (Table 3) were cultivated in liquid Luria Bertani Broth (LB) or M9 medium supplemented with 0.4% of sodium pyruvate.

RNA sample preparation Small precultures of the E. coli 55989 (CNCM deposit number 1-3144) strain was done in 40 ml of LB medium in a 100 ml erlen for 19h (overnight) without antibiotic selection at 37 C under constant 120 rpm agitation. Culture medium was removed by centrifugation 4000 rpm, 10 min at room temperature and resuspended in 40 ml of fresh LB or M9 pyruvate media. Eriens with 400 ml of LB or M9 pyruvate media were inoculated with 8 ml (1/50t" dilution) of LB or M9 pyruvate preculture,

8 respectively, and incubated at 37 C, 120 rpm. E. coli total RNAs was isolated at exponential (sample E) and stationary phase (sample S) when DO60o reached 0.52 and 4.29 (24h), respectively, for LB medium cultures and reached 0.51 and 1.28 (24h) for M9 pyruvate medium cultures.
Total RNAs were isolated immediately with Trizol (lnvitrogen) according to the manufacturer protocol excepted bacteria were harvested by centrifugation at rpm for 5 min at room temperature for avoiding cold shock stress. After all protocol steps, traces of ethanol were removed by air dry and RNA samples were resuspended with Tris buffer (10 mM Tris HCI pH 7.5). RNA samples were treated two times with 30 units of FPLC pure DNase I (Amersham) in 10 mM Tris HCI pH
7.5, 6 mM MgC12 buffer for 1h30 at 37 C in order to digest DNA contaminants, phenoUchloroform extracted and ethanol precipited. RNA samples were resuspended with Tris buffer, quantified by UV spectrophotometry and checked for putative degradations on 2% agarose gel. Genomic DNA contaminations were analyzed by PCR amplification of the multicopy 5S ribosomal gene with the 5S.Fw and 5S.RT primers (Table 4). PCR were done with 2 units of Taq polymerase (QBiogen) in 1X supplied buffer, 200 pM dNTP (QBiogen), 200 nM of each primers (Sigma Proligo) and 10 pg of RNA samples. DNA contaminations were checked in a 2% agarose gel and considered unsignificant if no PCR products were observed by ethidium bromide staining.

Semiquantitative RT-PCR

Chimeric DNA (cDNA) were synthetized from lOpg of total RNA with 200 units of Superscript III reverse transcriptase enzyme (Invitrogen) at 55 C for 1 hour with 2 pmol of specific primer (named gene.RT) (Sigma Proligo) (Table 4) and after enzyme was heat inactivated according to supplier protocol. PCR amplification of cDNA
was done with 0.4 unit of Taq polymerase (Qbiogen), 100 nM of each primers (named gene.RT and gene.Fw), 200 pM of dNTP, 2 NI of the RT reaction with thermal cycling of 94 C, 3 minutes, 40 cycles of 94 C, 30 sec.; 55 C, 30 sec.; 72 C for 30 sec. and a final extension of 72 C, 7 minutes. Identical reactions were done for same RNA

9 samples with the tmRNA.Fw and tmRNA.RT primers used as positive expression control. PCR products were analysed with a 2% agarose gel by ethidium staining.
Northern blot hybridization Northern blot membrane preparation and hybridization were done as previously described (Pichon & Felden 2005). Shortly, poly acrylamide gel electrophoresis of RNA samples were done in urea denaturing 8% bis-tris polyacrylamide (Sigma) gel and electro-transferred to Zeta probe GT membrane (Biorad). Membrane were hybridized with 32P 5' end labelled oligonucleotides with ExpressHyb solution (Clontech) as described (Pichon & Felden 2005). Hybridization of the TmRNA
molecules with the TmRNA.RT primer was used as positive expression control.
Results Identification of small RNAs.

Comparative genomic of sRNA sequences showed that compensatory mutations may be observed in the rho-independent terminator (RIT) stem of two related sequences (Figure 1). The inventors hypothetized such phenomenum could be observed in any RIT structures. The inventors applied the workflow described in Figure 2 on the genomes of straon of E. coli from the Table 3. The inventors focused on the E. coli 55989 results and took 9 candidates (Table 5) randomly and done detection of putative sRNA transcript by RT-PCR. The inventors observed that 7 candidates are positively expressed (Figure3) with the 5S gene as expression reference.

Accuracy of the in silico approach.
The inventors gathered sRNA gene coordinates and presence for rho-independent terminator (RIT) data from a set of publications (Wassarman et al.
2001, Agarmann et al. 2001, Rivas et al. 2001, Chen et al. 2002, Tjaden et al.
2002, Vogel et al. 2003, Kawano et al. 2005, Yachie et al. 2006) and their corresponding sequences from E. coli MG1655 genome. Among the 84 non redundant sRNAs, 58 5 have a confirmed or a predicted RIT according to available data, i.e. 69,0%
of the sRNA genes used this kind of termination as previously reported. The inventors applied the method of identifying sRNAs of the present invention (see upper paragraph; Figure 2) except E. coli 55989 genome is replaced by E. coli MG1655 and show that 47 (81.0%) of the RIT are predicted with the present method. The

10 unpredicted ones correspond to terminators which do not have a perfect stem loop (Lesnik et al. 2001). If the imperfect model for RIT prediction (tolerating one maximum mismatch and bulged stem) is used, the inventors found 10 of the 11 others RIT which increased the success rate of prediction system to 98.3%. The inventors assessed for presence of compensatory mutations in the 57 predicted RIT
and show that 49 candidates (82,7%) correspond to the inventors' selection criterion.
The inventors hypothetized that absence of covariation are caused by low sequence homologies in this region of the genome. As same the inventors applied sRNA
analyse of the E. coli 55989 genome and found identical results as same as E.
coli MG1655 except that the c0293, sokW and rdlC sRNA genes were absents from the E. coli 55989 genome.

11 Table 1: Genomic sequences used in this study.

Bacteria Strain Accession Database Buchnera aphidicola Bp NC_004545 Genbank Buchnera aphidicola Cc NC_008513 Genbank Buchnera aphidicola Sg NC_004061 Genbank Buchnera aphidicola APS NC_002528 Genbank Enterobacter 638 NC_009436 Genbank Erwinia carotovora SCRI1043 NC_004547 Genbank Escherichia coli MG1655 NC_000913 Genbank Escherichia coli EDL933 NC_002655 Genbank Escherichia coli Sakai NC_002695 Genbank Escherichia coli CFT073 NC_004431 Genbank Escherichia coli S88 N.A. Coliscope Escherichia coli UMN026 N.A. Coliscope Escherichia coli ED1a N.A. Coliscope Escherichia coli IAI1 N.A. Coliscope Escherichia coli IAI39 N.A. Coliscope Escherichia coli 55989 N.A. Coliscope Escherichia coli 042 N.A. Sanger Escherichia coli W3110 AC_000091 Genbank Escherichia coli APEC01 NC_008563 Genbank Escherichia coli UT189 NC_007946 Genbank Escherichia coli 536 NC008253 Genbank Escherichia fergusonnii ATCC35469 N.A. Coliscope Photorhabdus luminescens TTO1 NC_005126 Genbank Salmonella enterica Ty2 NC_004631 Genbank Salmonella enterica CT18 NC_003198 Genbank Salmonella enterica SC-B67 NC_006905 Genbank Salmonella enterica ATCC9150 NC_006511 Genbank Salmonella typhimurium LT2 NC_003197 Genbank Shigella boydii Sb227 NC_007613 Genbank Shigella dysenteriae Sd197 NC_007606 Genbank Shigella flexneri 301 NC_004337 Genbank Shigella flexneri 2457T NC_004741 Genbank Shigella flexneri 8401 NC_008258 Genbank Shigella sonnei Ss046 NC_007384 Genbank Soda/is glossinidius morsitans NC_007712 Genbank Wigglesworthia glossinidia Gb NC_004344 Genbank Yersinia enterocolitica 8081 NC_008800 Genbank Yersinia pestis C092 NC_003143 Genbank Yersinia pestis Antiqua NC_008150 Genbank Yersinia pestis 91001 NC_005810 Genbank Yersinia pestis KIM NC_004088 Genbank Yersinia pestis NepaI516 NC_008149 Genbank Yersinia pestis Pestoides F NC 009381 Genbank

12 Table 2: List of know sRNA genes from Escherichia coli species.

sRNA Gene sRNA synonym Localization Previous Next Strand Strain name gene gene 4.5S ffs 475611 475786 ybaZ ybaA <>> MG1655 6S ssrS 3053956 3054187 zapA ygfA >>> MG1655 C0067 c0067 238411 238623 yafT yafU >>< MG1655 C0293 c0293 1195889 1196009 icd ymfD > > < MG1655 C0299 c0299 1229852 1229930 hlyE umuD <>> MG1655 C0343 c0343 1407387 1407461 ydaN dbpA >>> MG1655 C0362 c0362 1549943 1550423 cueO gcd ><< MG1655 C0465 c0465 1970719 1970843 tar cheW <>< MG1655 C0614 c0614 2651472 2651558 sseA 1s128 ><> MG1655 C0664 c0664 2833077 2833189 norW hypF > > < MG1655 C0719 c0719 3119303 3119648 yghK glcB <>< MG1655 CrpTic crpTic 3483855 3484014 yhfA crp <<> MG1655 CsrB csrB 2922178 2922581 yqcC syd <<< MG1655 CsrC csrC SraK / RyiB / Tpk2 / IS198 4049020 4049304 yihA yihl <>> MG1655 DicF dicF 1647406 1647458 rzpQ dicB >>> MG1655 DsrA dsrA 2023242 2023370 dsr8 yedP > MG1655 DsrB dsrB 2022661 2022867 rcsA dsrA ><< MG1655 GadY gadY IS183 3662852 3662991 gadW gadX <>< MG1655 GcvB gcvB IS145 2940683 2940923 gcvA ydll <> < MG1655 GImY gImY Tke1 / SroF 2689177 2689389 yfhK purL <<< MG1655 GImZ glmZ SraJ / K19 / RyiA 3984419 3984665 aslA hemY <>< MG1655 IS128 is128 2651506 2651734 C0614 ryfA <>> MG1655 Isf isf 1019490 1019890 ompA sulA <>< MG1655 IsrA isrA IS061 1403680 1403866 abgR ydaL ><< MG1655 IsrB isrB IS092 1985865 1986059 yecJ yecR <<> MG1655 IsrC isrC IS102 2069307 2069538 yeeP flu >>> MG1655 IstR istR 3851141 3851316 ivbL tisA <<> MG1655 M1 rnpB 10Sb / M1 3268199 3268650 yhaC yhaK ><< MG1655 MicA micA SraD 2812790 2812901 luxS gshA <>< MG1655 MicC micC Tke8 / IS063 1435110 1435253 ompN yjcD ><> MG1655 MicF micF 2311070 2311198 ompC rcsD <>> MG1655 NC092 nc092 3069272 3069486 fbaA pgk <> < MG1655 OmrA omrA RygA / T59 / PAIR2 2974124 2974246 aas gal <<> MG1655 OmrB omrB RygB / T59 / PAIR2 / SraE 2974326 2974440 aas ga/ <<> MG1655 OxyS oxyS 4156301 4156455 argH oxyR > < > MG1655 RdIA rdlA 1268511 1268615 IdrA ldrB <> < MG1655 RdIB rdlB 1269046 1269150 IdrB ldrC <> < MG1655 RdIC rdlC 1269581 1269685 /drC chaA <> < MG1655 RdID rdlD 3698124 3698228 ldrD yhjV < MG1655 RprA rprA IS083 1768361 1768501 ydiK ydiL >>> MG1655 RseX rseX 2031637 2031763 yedR yedS <>> MG1655 Rtt rtt RttR / RtV1 1286289 1286459 dppA proK <<< MG1655 RybA rybA 852175 852263 ybiL mntR > < > MG1655 RybB rybB P25 887199 887314 ybjK ybjL ><< MG 1655 RybC rybC SroB 506393 506511 ybaK ybaP <>< MG1655 RybD rybD 764212 764373 sucD mngR > > < MG 1655 RydB rydB Tpe7 / IS082 1762737 1762804 sufA ydiH <<< MG1655 RydC rydC 1489466 1489562 cybB ydcA > < > MG 1655 RyeA ryeA SraC / Tpke79 / IS091 1921041 1921362 pphA yebY <>< MG1655

13 sRNA Gene sRNA synonym Localization Previous Next Strand Strain name gene gene RyeB ryeB Tpke79 1921188 1921308 pphA yebY <<< MG1655 RyeC ryeC Tp11 / QUAD1a 2151299 2151475 yegL yegM <>> MG1655 RyeD ryeD Tpe60 / QUAD1 b 2151634 2151803 yegL yegM <>> MG1655 RyeE ryeE 2165079 2165224 yegQ orgK >>< MG1655 RyeF RyeF 1956465 1956584 torY cutC <<< MG1655 RyfA ryfA Tp1 / PAIR3 2651828 2652180 /s128 sseB >>< MG1655 RyfB ryfB 2698078 2698435 yfhL ryfC > < > MG 1655 RyfC ryfC 2698505 2698620 ryfB acpS <> < MG1655 RyfD ryfD 2732175 2732343 clpB yfiH <<< MG1655 RygC rygC T27 / QUAD1c 3054837 3055016 ygfA serA >>< MG1655 C0730 / QUAD1d I IS156 / 3192738 3192922 yqiK -faE > < <
RygD rygD Tp8 MG1655 RygE rygE QUAD1 3193114 3193297 yqiK rfaE ><< MG1655 RyhA ryhA SraH 3348564 3348722 elbB arcB <>< MG1655 RyhB ryhB Sral / IS176 3578945 3579075 yhhX yhhY <<> MG1655 RyjA ryjA SraL 4275946 4276124 soxR yjcD ><> MG1655 RyjB ryjB 4525965 4526089 sgcA sgcQ <> < MG1655 SgrS sgrS RyaA 77331 77593 sgrR setA <>> MG1655 SokA sokA -3718623 3720136 hokA insJ <>> MG1655 SokB sokB 1490107 1490205 hokB trg <>> MG1655 SokC sokC Sof 16917 17012 mokC nhaA < MG1655 SokE sokE 606956 607051 ydbK hokE <<> MG1655 SokW sokW 2777339 2777409 mokW Z3118 <>< 0157 SokX sokX 2885339 2885429 ygcB cysH <>< MG1655 Spot42 spf IS197 4047922 4048032 polA yihA > > < MG1655 SraA sraA T15 457949 458104 clpX Ion > < > MG1655 SraB sraB Pke20 1145812 1145980 yceF yceD <>> MG1655 SraF sraF Tpk1 / IS160 3236396 3236583 ygjR ygjT >>> MG1655 SraG sraG P3 3309247 3309420 pnp rps0 <> < MG1655 SroH sroH 4188342 4188545 htrC thiH > < < MG1655 SymR symR RyjC 4577822 4577953 yjiW hsdS <> < MG1655 T44 t44 Tff 189676 189860 map rpsB <> > MG1655 TmRNA ssrA 10Sa RNA / M2 2753571 2754056 smpB intA >>> MG1655 Tp2 tp2 122857 123023 pdhR aceE > < > MG1655 Tpkell tpkell 14077 14444 dnaK dnaJ >>> MG1655 Tpke70 tpke70 2494609 2494649 ddg yfdZ > < < MG1655 Notes :
sokA gene in E. coli MG1655 strain is interupted by an IS150 insertion.

14 Table 3: Strains and plasmids used in this study.

z Name Pathotype Origin Origin Pg.
Relevant Reference genotype Strains E. coli MG1655 N.A. Com. fecal A hf , Blattnet et al. 1997 E. coli UT189 UPEC Cys. urine B2 Chen et al. 2006 E. coli IAI39 UPEC urine D Picard et al. 1999 E. coli CFT073 UPEC Pyl. urine B2 Welch et al. 2002 E. coliAL862 Unknow blood Lalioui etal. 2001 E. coli 55989 EAEC Diarrhea fecal B1 Bernier et a/. 2002 E. coli 536 UPEC Pyl. urine B2 Hochhut et al. 2006 Plasmids BR322 cat, bla Lab. Collection Human clinical isolates: Commensal (Com.), Cystitis (Cys.), Pyelonephritis (Pyl.) and Laboratory Strain (Ls,).

2 Phylogenetic group.

Table 4: Oligo used in this study.

Small Primer RNA Name Sequence PCR Size (bp) Positive control 5S 5S.Fw 5'-GGTGGTCCCACCTGACC-3' 101 5S. RT 5'-ATGCCTGGCAGTTCCCTACT-3' TmRNA tmRNA.Fw 5'-TCTGGATTCGACGGGATTTG-3' 193 tmRNA.RT 5'-CGCGATCTCTTTTGGGTTTG-3' Small RNA candidates of E. coli 55989 213 213.Fw 5'-CACCGCCAGGAAGGTGTAT-3' 105 213.RT 5'-ACACATGCAGGGCGTCTAAC-3' 597.Fw 5'-TACCGCTTACGTTGAGAGCA-3' 597 597.RT 5'-TAAAACAAAACCCGCCGTAG-3' 105 1069 1069. Fw 5'-TAGGAGATCAGCCCGTCAAG-3' 127 1069. RT 5'-CGGGGCATTTTTGTACAGGT-3' 1079.Fw 5'-TGCGTTAGTGTTTTTTTGCC-3' 1079 1079. RT 5'-AAAAATCCCGCAGTGATCG-3' 97 1325.Fw 5'-GAGAGAATCTATTGAAGTGCATGG-3' 1325 1325. RT 5'-GGGCAATCAGCGAGTAGGTA-3' 101 1663.Fw 5'-CGCTGTGTGAAATACGGATG-3' 1663 1663. RT 5'-ACCTTAGCAACCGATTGACG-3' 100 1757. Fw 5'-TAAATACAGCCCCAGCCATT-3' 1757 1757. RT 5'-CCTGACGGGTGAAATGAATAA-3' 101 2037 2037.Fw 5'-AGTAGTTGGCTTTGGGGTGA-3' 88 2037. RT 5'-GAATTGACTTTGGCGGTGAC-3' 2046.Fw 5'-AATTCGCAGGACCGTGATAC-3' 2046 2046. RT 5'-CGCCTCATTCATGTTCTGGT-3' 117 2531.Fw 5'-CTTCGCGGTCTCTTTTCCTC-3' 2531 2531.RT 5'-CGCGAGCGCGCCATTG-3' 92 Small RNA candidates of E. coli AL862 10 RIG10.Fw 5'-ATCAGTCCGTTGTGTGCAAT-3' 117 RIG10.RT 5'--CGATCGATAAAACAGGTATCG3' RIG11.Fw 5'-TCAGCATTCAGTGCAGGAAC-3' 11 RIG11.RT 5'-CTGCCGGGAAGAATCATAAA-3' 110 RIG12.Fw 5'-AGTTCCAGCCTGCGACTTT-3' 12 RIGI2.RT 5'-TTTCAGGGAAGCTGGTATCC-3' 147 RIG14.Fw 5'-GGCATGATGAGAACGCAGTA-3' 14 RIGI4.RT 5'-CGCTCAGACGGATGCTTAAT-3' 121 53 RIG53.Fw 5'-AATGTAAGTGTAAACTGAGTGCCGTA-3' 100 RIG53.RT 5'-ATGTTCCATAACAGACGTCCAC-3' vpe.Fw 5'-TTAATTAATGTGATGATTGTCG-3' vpe vpe.RT 5'-TAGCGCTATCACAAAGATTG-3' 71 Table 5: Expressed sRNA in ExPEC.

Flanking Flanking Strand Souches Localisation ORF ORF orientation sRNA candidates 14 AL862 16870 17298 0RF00049 ORF00051 > > >
53 AL862 56273 56806 afaD afaE > < >
213 55989 300904 301451 Hyp. Prot. Hyp. Prot. < < <
597 55989 822633 822919 Hyp. Prot. Hyp. Prot. > < >
1069 55989 1430311 1430726 Rz Hyp. Prot. > < >
1079 55989 1448272 1448613 I 3 > < >
1325 55989 1785228 1785593 tolA Hyp. Prot. < > <
1663 55989 2176098 2176920 D Hyp. Prot. > > <
1757 55989 2305688 2305907 Hyp. Prot. Hyp. Prot. > > >
2037 55989 2699641 2700054 Hyp. Prot. N < < >
2046 55989 2705703 2706231 Hyp. Prot. cscB > > <
2531 55989 3343375 3343569 Hyp. Prot. Hyp. Prot. < < >
sRNA used as positive expression control 5S AL862 N.A. N.A. N.A. N.A. N.A.
ARNtm AL862 N.A. N.A. N.A. N.A. N.A.

ARNtm 55989 2978731 2979092 smpB EC55_2909 > > >
N.A. : Not Applicable.

References Antal, M., Bordeau, V., Douchin, V., Felden, B. (2005) A small bacterial RNA
regulates a putative ABC transporter. J Biol Chem 280, 7901-7908.
Argaman, L., Hershberg, R., Vogel, J., Bejerano, G., Wagner, E.G.H., Margalit, H.
and Altuvia, S. (2001) Novel small RNA-encoding genes in the intergenic regions of Escherichia coli. Curr Biol 11, 941-950.

Axmann, I., Kensche, P., Vogel, J., Kohl, S., Herzel, H. and Hess, W. (2005) Identification of cyanobacterial non-coding RNAs by comparative genome analysis.
Genome 8iol. 6, R73.

Blattner FR, Plunkett G 3rd, Bloch CA, Perna NT, Burland V, Riley M, Collado-Vides J, Glasner JD, Rode CK, Mayhew GF, Gregor J, Davis NW, Kirkpatrick HA, Goeden MA, Rose DJ, Mau B, Shao Y. (1997) The complete genome sequence of Escherichia coli K-12. Science, 277, 1453-1474.

Brantl, S. (2007) Regulatory mechanisms employed by cis-encoded antisense RNAs. Curr Opin Microbiol 10, 102-109.

Carter, R., Dubchak, I. and Holbrook, S. (2001) A computational approach to identify genes for functional RNAs in genomic sequences. Nucleic Acid Res 29, 3928-3938.
Chen, Y., Carlini, D.B., Baines, J.F., Parsch, J., Braverman, J.M., Tanda, S.
and Stephan, W. (1999) RNA secondary structure and compensatory evolution. Genes Genet Syst. 74, 271-286.

Chen, S., Lesnik, E.A., Hall, T.A., Sampath, R., Griffrey, R.H., Ecker, D.J., and Blyn, L.B. (2002) A bioinformatics based approach to discover small RNA genes in the Escherichia coli genome. BioSystems 65, 157-177.

Chen SL, Hung CS, Xu J, Reigstad CS, Magrini V, Sabo A, Blasiar D, Bieri T, Meyer RR, Ozersky P, Armstrong JR, Fulton RS, Latreille JP, Spieth J, Hooton TM, Mardis ER, Hultgren SJ, Gordon JI. (2006) Identification of genes subject to positive selection in uropathogenic strains of Escherichia coli: a comparative genomics approach. Proc Natl Acad Sci USA, 103, 5977-5982.
Gottesman, S. (2005) Micros for microbes: non-coding regulatory RNAs in bacteria.
Trends in Genet 21, 399-404.

Hochhut B, Wilde C, Balling G, Middendorf B, Dobrindt U, Brzuszkiewicz E, Gottschalk G, Carniel E, Hacker J. (2006) Role of pathogenicity island-associated integrases in the genome plasticity of uropathogenic Escherichia coli strain 536. Mol Microbiol, 61, 584-595.

Lalioui, L., and C. Le Bougu6nec. 2001. afa-8 Gene cluster is carried by a pathogenicity island inserted into the tRNAPhe of human and bovine pathogenic Escherichia coli isolates. Infect. lmmun., 69, 937-948.

Le Novere, N. (2001). MELTING, computing the melting temperature of nucleic acid duplex. Bioinformatics 17: 1226-1227.
Lesnik, E. A., Sampath, R., Levene, H. B., Henderson, T. J., McNeil, J. A. and Ecker, D. J. (2001) Prediction of rho-independent transcriptional terminators in Escherichia coli. Nucleic Acids Res 29, 3583-3594.

Livny, J., Fogel, M. A., Davis, B. M. and Waldor, M. K. (2005) sRNAPredict: an integrative computational approach to identify sRNAs in bacterial genomes.
Nucleic Acids Res 33, 4096-4105.

Macke, T. J., Ecker, D. J., Gutell, R. R., Gautheret, D., Case, D. A. and Sampath, R.
(2001) RNAMotif, an RNA secondary structure definition and search algorithm.
Nucleic Acids Res 29, 4724-4735.

Mathews, D. H, Sabina, J., Zuker, M. and Turner, D. H. (1999) Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J Mol Biol 288, 911-940.
Pearson, W.R. (2000). Flexible sequence similarity searching with the FASTA3 program package. Methods Mol Biol 132, 185-219.

Pfeiffer, V., Sittka, A., Tomer, R., Tedin, K., Brinkmann, V. and Vogel, J.
(2007) A
small non-coding RNA of the invasion gene island (SPI-1) represses outer membrane protein synthesis from the Salmonella core genome. Mol. Microbiol.
66, 1174-1191.

Picard, B., Garcia, J.S., Gouriou, S., Duriez, P., Brahimi, N., Bingen, E., Elion, J. and Denamur, E. (1999) The link between phylogeny and virulence in Escherichia coli extraintestinal infection. Infect lmmun., 67, 546-553.

Pichon, C. and Felden, B. (2003) Intergenic Sequence Inspector: Searching and identifying bacterial RNAs. Bioinformatics 19, 1707-1709.
Pichon, C. and Felden B. (2005) Small RNA genes expressed from Staphylococcus aureus genomic and pathogenicity islands with specific expression among pathogenic strains. Proc Natl Acad Sci USA 102, 14249-14254.

Pichon, C., Felden, B. (2007) Proteins that interact with bacterial small RNA
regulators. FEMS Microbiol Rev 31, 614-625, Review.

Pichon, C. and Felden, B. (2008) Small RNA genes identifications and mRNA
targets predictions in bacteria. Bioinformatics, review. Submitted.

5 Rivas,E. and Eddy,S.R. (2000) Secondary structure alone is generally not statistically significant for the detection of noncoding RNAs. Bioinformatics, 16, 583-605.

Rivas, E., Klein, R. J., Jones, T. A. and Eddy, S. R. (2001) Computational 10 identification of noncoding RNAs in E. coli by comparative genomics. Curr Biol 11, 1369-1373.

Romby, P., Vandenesch, F. and Wagner, E. G. (2006) The role of RNAs in the regulation of virulence-gene expression. Curr Opin Microbiol 9, 229-236.
Saetrom, P., Sneve, R., Kristiansen, K.I., Snove, 0., Grunfeld, T., Rognes, T.
and Seeberg, E. (2005) Predicting non-coding RNA genes in Escherichia coli with boosted genetic programming. Nucleic Acid Res 33, 3263-3270.

Schattner, P. (2002) Searching for RNA genes using base composition statistics.
Nucleic Acid Res 30: 2076-2082.

Tjaden, B., Saxena, R.M., Stolyar, S., Haynor, D., Kolker, E. and Rosenow, C.
(2002) Transcriptome analysis of Escherichia coli using high-density oligonucleotide probe arrays. Nucleic Acid Res 30, 3732-3738.

Uzilov, A., Keegan, J., and Mathews, D. (2006) Detection of non-coding RNAs on the basis of predicted secondary structure formation free energy change. BMC
Bioinformatics 7, 173-203.
Wang, C., Ding, C., Meraz, R. and Holbrook, S. (2006) PSoL: a positive sample only learning algorithm for finding non-coding RNA genes. Bioinformatics 22, 2590-2596.
Wassarman, K., Repoila, F., Rosenow, C., Storz, G. and Gottesman, S. (2001) Identification of novel small RNAs using comparative genomics and microarrays.
Genes & Dev 15, 1637-1651.

Wassarman, K. M. and Saecker, R. M. (2006) Synthesis-mediated release of a small RNA inhibitor of RNA polymerase. Science 314, 1601-1603.

Welch, R. A., V. Burland, G. Plunkett III, P. Redford, P. Roesch, D. Rasko, E.
L.
Buckles, S. R. Liou, A. Boutin, J. Hackett, D. Stroud, G. F. Mayhew, D. J.
Rose, S.
Zhou, D. C. Schwartz, N. T. Perna, H. L. Mobley, M. S. Donnenberg, and F. R.
Blattner. 2002. Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli. Proc. Natl. Acad. Sci. USA, 99, 17024.

Yachie, N., Numata, K., Saito, R., Kanai, A. and Tomita, M. (2006) Prediction of non-coding and antisense RNA genes in Escherichia coli with Gapped Markov Model.
Gene, 372, 171-181.
Zuo, Y. and Deutscher, M. P. (2002) Mechanism of action of RNase T. I.
Identification of residues required for catalysis, substrate binding, and dimerization. J
Biol Chem 277, 50155-50159.

Claims (11)

CLAIMS:

1 A method for the identification of small RNAs, inside the sequence of a first nucleic acid molecule, comprising the following steps :
A) Preparation of a test sequence dataset of candidate sequences comprising:
i) identification of termination sites of transcription elongation complexes in said first nucleic acid molecule, ii) elimination of all termination sites whose sequences are located near the stop codon of an open reading frame with the same nucleic acid strand orientation, iii) gathering of nucleic acid sequences comprising the sequence, between 50 to 500 nucleotides, upstream each remaining termination sites after step (ii) plus the termination site sequences themselves;

B) Identification of at least one nucleic acid sequence sharing significant sequence identities for each sequence from the test sequence dataset, C) Multi alignment of all nucleic acid sequences sharing significant identities with a sequence from the test sequences dataset, D) Identification of those nucleic acid sequences having each one at least two partially conserved stem loops (small RNAs) and further identification of at least one putative compensatory mutation in each stem, E) Analysis of the putative expression of each candidate small RNA genes.

2 The method according to Claim 1 wherein the sequence of the first nucleic acid molecule is a partial or a full genome sequence.

3 The method according to Claim 1 wherein the sequence of the first nucleic acid molecule is a partial or a full sequence of a plasmid.

4 The method according to Claim 1 wherein the termination sites of the transcription elongation complexes comprise at least one stem loop structure.

The method according to Claim 1 wherein the termination sites of step A (ii) are located between the nucleotides -20 to +60 around the stop codon.

6 The method according to Claim 4 wherein the termination sites are bacterial rho-independent terminators.

7 The method according to claim 1 wherein the identification at step B
comprises the identification of sequence identities between the sequences of the test sequence dataset and those of a database of nucleic acid sequences.

8 The method according to claim 1 wherein the alignment of step D comprises the alignment of the sequences identified in step B with the Multiple Alignment of Small RNAs (MASR) software.

9 The method according to Claim 1 wherein the identification of compensatory mutations in step D comprise the following steps :
1) Search for putative stem loops of a tested sequence whose stems are at least 4 nucleotides long and no more than 18 nucleotides long, 2) Identification of stems whose sequences and structures are partially conserved between the tested sequence and sequences sharing sequence identities with them (those found according to claim 1, B), 3) Identification of two complementary nucleotides in the stems of the tested sequence which are both not conserved in other sequences (those found according to claim 1, B) but corresponding two nucleotides are complementary.

24 The method according to Claim 9 wherein the identification of compensatory mutations in step D is carried out with Covariation Search in Small RNAs (CSSR) software.

11 The method according to Claim 1, wherein the analysis of the putative expression of each candidate small RNA genes in step E comprises the transcriptome analysis by microarray, reverse transcription of RNA and amplification by polymerase chain reaction (RT-PCR), Northern blot hybridization.