Inhibition of Expression in Escherichia coli of a Virulence Regulator MglB of Francisella tularensis Using External Guide Sequence Technology

External guide sequences (EGSs) have successfully been used to inhibit expression of target genes at the post-transcriptional level in both prokaryotes and eukaryotes. We previously reported that EGS accessible and cleavable sites in the target RNAs can rapidly be identified by screening random EGS (rEGS) libraries. Here the method of screening rEGS libraries and a partial RNase T1 digestion assay were used to identify sites accessible to EGSs in the mRNA of a global virulence regulator MglB from Francisella tularensis, a Gram-negative pathogenic bacterium. Specific EGSs were subsequently designed and their activities in terms of the cleavage of mglB mRNA by RNase P were tested in vitro and in vivo. EGS73, EGS148, and EGS155 in both stem and M1 EGS constructs induced mglB mRNA cleavage in vitro. Expression of stem EGS73 and EGS155 in Escherichia coli resulted in significant reduction of the mglB mRNA level coded for the F. tularensis mglB gene inserted in those cells.


Introduction
Francisella tularensis is a Gram-negative, facultative intracellular bacterium that is the etiological agent of tularemia [1]. Tularemia is primarily a disease of the Northern hemisphere that is spread by blood-sucking insects or acquired from contact with infected animals such as rabbits or rodents. In addition, F. tularensis is a category A agent of biowarfare/biodefense due to its high infectivity and lethality when inhaled [2]. Survival and replication inside macrophages of hosts is important for the virulence of F. tularensis. The MglA and MglB proteins encoded in an operon regulate transcription of a particular set of genes, some of which are required for intramacrophage growth and virulence of F. tularensis [2][3][4][5][6].
Inhibition of gene expression at the RNA level has been achieved by base-pairing EGSs with mRNAs of specific genes to form a complex mimicking the natural substrates of RNase P [7][8][9][10][11]. RNase P is a ubiquitous enzyme and its prime physiological role is to cleave tRNA precursors (ptRNAs) to generate the 59 termini of mature tRNA molecules [12]. RNase P can specifically cleave any RNA in vitro and in vivo, provided that the target RNA is hybridized to an EGS that is complementary to the targeted region [13,14]. EGSs have been designed to convert antimicrobial resistance to sensitivity [8,11], inhibit bacterial viability, and decrease secretion of virulence factors involved in host cell invasion in Gram-negative bacteria [9,10].
In Escherichia coli, RNase P consists of M1 RNA and C5 protein. The M1 RNA can execute the enzymatic function in vitro in the absence of the C5 protein. Depending on whether or not M1 RNA is covalently linked with an EGS, EGSs are classified as M1 EGSs and stem EGSs, respectively.
Here we describe application of stem and M1 EGSs to induce F. tularensis mglB mRNA cleavage by RNase P in E. coli. Stem and M1 EGSs targeting positions 73, 148, and 155 cleaved the mglB mRNA in vitro. Expression of stem EGSs targeting positions 73 and 155 resulted in a 50,60% reduction of the mglB mRNA level in E. coli.
A similar successful method has been described to test EGSs on different pathogens [15].

Results
Mapping in vitro of accessible sites in the mglB mRNA Many RNAs spontaneously self-assemble into highly folded structures. EGS accessible sites within target RNAs are determined to facilitate the design of specific EGSs. Several methods have been successfully used to identify accessible sites on various target RNAs [16,17]. In this study, the rapid rEGS assay was used to screen for stem EGSs that selectively cleave the target mglB mRNA. An effective stem EGS candidate should meet the following requirements: an EGS-directed cleavage reaction occurs only if the target RNA is incubated with both rEGS and the RNase P holoenzyme; the amount of cleavage product increases as the amount of rEGS increases. Figure 1A suggests that an EGS targets the position 155 ( Fig. 1A and 1B).
The 59 end nucleotide of ptRNA cleavage products of RNase P is guanine (G) in most cases. The rEGSs consisting of the random 14-nucleotide sequence (N 14 ) and a cytosine (C) were designed to bind to a G nucleotide in the target RNA. Interestingly the nucleotide at position 155 in the mglB RNA is a C rather than a G. In principle, the nucleotide next to the CACCA sequence in rEGS(s) might be G, which base pairs with the nucleotide C at the position 155 of the mglB sequence.
As described previously for screening rEGSs targeting the vjbR mRNA of Brucella melitensis [17], both rEGSe and rEGSx libraries were used (Fig. 1A). The only difference between rEGSe and rEGSx is the shorter length of the 39 end of rEGSx mRNA that resulted from the BstNI digestion of the DNA fragments for in vitro transcription. Consistent with previous report (17), the rEGSx library gave much stronger cleavage than the rEGSe library (Fig. 1A).
The partial RNase T1 digestion assay is normally used to locate the exposed G nucleotides in the target RNA [18]. The signal strength of bands in the T1 lane in Figure 1A indicates the accessibility of the G nucleotides in the mglB mRNA ( Fig. 1A and 1B). Although not identified by the rEGS assay, positions 52, 73, and 148 could be EGS accessible sites ( Fig. 1A and 1B).
Cleavage in vitro of the mglB mRNA by specific EGSs Stem EGSs and M1 EGSs targeting the potential cleavage sites, 52, 73, 148, and 155 (rEGS), were respectively made and tested in cleavage assays in vitro. To follow the sequence of rEGSs, N 14 CACCA, we designed EGS155R assuming the targeted position 155 to be a G nucleotide. EGS155, which perfectly base-pairs with the target mglB sequence, was also designed. Figures 2A and 2B indicate that stem EGSs targeting positions 73, 148, and 155 were able to cleave the mglB mRNA. Increase of the EGSs concentration resulted in increase of the amount of cleavage products ( Fig. 2A and 2B). Failure in base-pairing of the nucleotide immediately next to ACCA in EGS155R with the target mglB sequence resulted in loss of cleavage activity (Fig. 2B). The cleavage sites were confirmed as shown in Figure 1A by 59 endlabeling the mglB mRNA, cleavage assays in vitro, and then separation by denaturing PAGE with alkali-treated mglB mRNA and partial RNase T1 digested mglB mRNA as size markers (data not shown). Figures 3A and 3B show that higher concentration of Mg 2+ (100 mM) induced the M1 EGSs in the absence of RNase P holoenzyme to give the same cleavage patterns as the correspond- Figure 1. Mapping of accessible sites in the target mglB mRNA by RNase T1 and rEGS libraries. A. Polyacrylamide gel electrophoresis of cleavage products of 59end labeled mglB mRNA generated by RNase P holoenzyme and rEGSe or rEGSx libraries. The RNase P holoenzyme was reconstituted by 10 nM M1 RNA and 100 nM C5 protein. mglB mRNA cleavage products were separated on 8% polyacrylamide/7 M urea gels together with an alkali ladder and a partial RNase T1 digest of the same RNA. Arrows indicate positions selected for further analyses. The black filled triangles represent decreasing concentrations of rEGSs (1000-, 100-, and 10-fold molar excess to the mglB mRNA) incubated with RNase P holoenzyme. Lane OH represents the alkali ladder. Lanes T1 and M1+C5 represent reaction with RNase T1 and RNase P holoenzyme, respectively. The difference between rEGSe and rEGSx is the length of 39 end of rEGS mRNA resulted from the BstNI digestion of the DNA fragments for in vitro transcription. The numbers indicate the positions of cleavage relative to the mglB translational initiation site. B. Complexes between the mglB mRNA and EGS52, EGS73, EGS148, and EGS155. The secondary structure of the partial mglB sequence used for the assay was created by the Mfold program [21]. Signal strength is according to the RNase T1 assay. doi:10.1371/journal.pone.0003719.g001 ing stem EGSs in the presence of E. coli RNase P holoenzyme ( Fig. 2A, 2B, 3A, and 3B). Consistent with the results of stem EGSs, only M1 EGSs targeting positions 73, 148, and 155 cleaved the mglB mRNA. Increase of the M1 EGSs concentration resulted in increase of the amount of cleavage products ( Fig. 3A and 3B). M1 EGS52 and M1 EGS155R could not cleave the mglB mRNA ( Fig. 3B and data not shown).
Transcription of mglB was under the control of the rnpB promoter and transcription of EGSs under the control of the IPTG-inducible T7 promoter (data not shown). To avoid the mutual influence of the two promoters, mglB and EGSs were transcribed in opposite directions. Figures 4A and 4B show that IPTG-induced expression of stem EGS73 and EGS155 resulted in reduction of the mglB mRNA level to 40,50% of that without the EGSs in E. coli cells. Surprisingly, the mglB mRNA level increased in the presence of leaky expression of stem EGS148 for unknown reasons but increase of EGS148 transcription still resulted in reduction of the mglB mRNA level ( Fig. 4A and 4B). None of the Activities of stem EGS155 and EGS155R. B. Activities of stem EGS52, EGS73, and EGS148. If needed, 10 nM M1 RNA and 100 nM C5 protein in buffer PA were added to reconstitute the RNase P holoenzyme. The mglB mRNA cleavage products were separated on 8% polyacrylamide/ 7 M urea gels along with the mglB mRNA alone or the mglB mRNA with the RNase P holoenzyme. Internally labeled pSupS1 ptRNA was used as a positive control to check for activity of the reconstituted RNase P holoenzyme. EGSs were added in 100-, 50-, and 10-fold molar excess to the mglB mRNA, denoted by black triangles. doi:10.1371/journal.pone.0003719.g002

Discussion
We successfully identified two EGSs, EGS73 and EGS155, which induced reduction of the mRNA level of the global virulence regulator MglB from F. tularensis both in vitro and in E. coli. In a previous report, stem EGS117 and EGS166 were identified to cleave the mRNA of a virulence regulator VjbR from B. melitensis in vitro [17]. We subsequently found that stem EGS117 reduced the vjbR mRNA level in E. coli by 80% and stem EGS166 by 100% compared with that in the absence of the two EGSs (data not shown). Thus, it is possible that if the EGSs, EGS73 and EGS155 for mglB of F. tularensis and EGS117 and EGS166 for vjbR of B. melitensis, are delivered to the pathogenic bacteria, they will reduce their target mRNA levels so that virulence of the bacteria will be reduced.
The method of screening rEGS libraries has successfully been used to identify accessible and cleavable sites in the target RNA [17]. All specific EGSs that we have designed so far based on the results of the rEGS method are able to cleave the target mRNAs [17]. The efficiency of the rEGS method is better than that of any other method that has been used, including the partial RNase T1 digestion assay. However, in some cases where we searched for rEGS, we have been unable to identify any EGS-induced cleavable site or EGS accessible sites that were identified by other methods [15,17]. There are two ways to solve the problem. One is to combine the results of the rapid rEGS method and the partial RNase T1 digestion assay to design specific EGSs, as we did in this study. The other is to make a transcript of the target gene that is much larger than the one used in this case so that the whole landscape of the targeted gene is examined.
The nucleotide G is preferred to lie immediately next to the natural ptRNA cleavage site and base-pairs with the nucleotide C before the ACCA sequence in the 39 end of the ptRNA cleavage product [17]. When we originally designed the rEGS libraries, we used the nucleotide C after the N 14 sequence and before the ACCA sequence to base-pair with the nucleotide G in target RNAs [17]. Interestingly, the nucleotide C at the position 155 of the mglB sequence identified by the rEGS method does not follow this rule. However, difference in base-pairing of a single nucleotide in EGS155 and EGS155R with the nucleotide G significantly affects EGS-induced RNase P cleavage activity ( Fig. 2A), suggesting the importance of base-pairing between EGSs and the nucleotide next to the cleavage site. The interaction between the nucleotide T at the position 156 of the mglB sequence and the nucleotide A next to the CACCA sequence in EGS155R may be too weak to form a mRNA-EGS complex to induce RNase P cleavage activity.
While we have outlined a safe method to test the effect of EGSs on genes from pathogenic bacteria in this report, we also note that shuttle plasmids that contain the successful EGSs have been constructed and are now in the process of being tested in the actual pathogens in BL3 facilities of the Armed Forces Insitute of Pathology. The efficiency of EGS inactivation of expression of target genes can be enhanced by putting two EGSs attacking different sites on the target mRNA on a shuttle plasmid. Indeed, two EGSs attacking the same site can also be used. Agents that can be injected into mice or administered to them by nasal inhalation will be used in these experiments. Chemically modified EGSs that can be taken up by bacteria (Avi Biopharma, Corvallis, Oregon) should be used in this respect and might be successful. Clinical application of these EGSs will rely on the results of the mouse experiments.

Strains and growth conditions
Bacterial strains used are listed in Table 1. E. coli cells were grown at 37uC in Luria-Bertani (LB) broth or on LB agar. E. coli  (2) EGS was used as a negative control. The plasmid vector pMlac-mglB with (+) mglB but without (2) EGS was used as a positive control. The plasmids pMlac-mglB-EGS73, pMlac-mglB-EGS148, and pMlac-mglB-EGS155 carried both (+) mglB and EGS73, EGS148, and EGS155, respectively. Total RNAs were prepared from E. coli BL21(DE3) cells in the absence (2) or presence (+) of IPTG to induce expression of the EGSs listed. The RNA samples were separated on 2% agarose gels. The mglB mRNA was probed with 59 end labeled oligonucleotide RNA155 and the 5S rRNA with 5S1 on the same membrane. The experiments were independently done twice and only a typical figure was shown. B. Average of the relative mglB mRNA level in E. coli. The ratios of the mglB mRNA to the 5S rRNA for cells carrying different plasmids were calculated. The ratio for cells carrying pMlac-mglB was chosen as 100% to calculate the relative mglB mRNA level in other cells listed. doi:10.1371/journal.pone.0003719.g004 DH5a was routinely used for subcloning and plasmid propagation. E. coli BL21(DE3) was used for protein expression. Ampicillin or carbenicillin (100 mg/ml) was used for plasmid selection and maintenance in E. coli.

Preparation of the target mglB mRNA
Restricted genomic DNA of F. tularensis was obtained from the Armed Forced Institute of Pathology, Washington, DC. The template for the mglB mRNA of F. tularensis was prepared as described previously [17]. To map accessible sites in the mglB mRNA, the 59 fragment was amplified by PCR using primers mglB F (59-ATGGCTATGCTTAGAGCATATG-39) and mglB R (59-TCATCATCTTCGCCTTCATT-39). The PCR fragment was sub-cloned into the pGEM-T vector, generating the plasmid pGEMglB. To reduce the length of the vector sequence between the T7 promoter in the pGEM-T vector and the inserted DNA fragment, the plasmid pGEMglB was digested with the restriction enzymes ApaI and SacII, purified from an agarose gel using Ultrafree-MC columns, blunt ended using DNA polymerase I Klenow fragment, and then re-ligated using T4 DNA ligase. The resulting plasmid pGEDMglB was digested with the restriction enzymes FokI and TaqI, ethanol-precipitated and dissolved in distilled water, and used as the template for in vitro transcription using T7 RNA polymerase [17]. For internally labeled transcripts, [a-32 P]GTP was used during the transcription reaction. For the mapping of cleavage sites, transcripts were labeled at the 59 end using [c-32 P]ATP and T4 polynucleotide kinase.

Mapping of accessible sites in the mglB mRNA
The rEGS assay for the mglB mRNA was performed as described previously [17]. Partial RNase T1 reaction was performed as previously described (8). After the reaction, the mRNA was separated on an 8% polyacrylamide sequencing gel that contained 7 M urea. Partial alkaline ladder digests were carried out as previously described [17]. Cleavage sites were determined by comparing rEGS-mediated and partial RNase T1 cleavage products to partial alkaline ladder results. The yeast suppressor precursor tRNA Ser (SupS1) was internally labeled with [a-32 P]GTP and used as a substrate in control reaction [15,19].

Preparation of stem and M1 EGSs against the mglB mRNA
Specific EGSs were designed based on the rEGS assay and partial RNase T1 assay according to previous reports [17,18]. The specific EGSs used for in vitro cleavage assays were phosphorylated at their 59 ends, annealed together by heating at 95uC for 5 min and cooled down slowly. The annealed oligos had the restriction enzymes PstI and BstEII overhangs at the 59 and 39 ends, respectively. These oligos used for EGS52, EGS73, EGS148, EGS155, and EGS155R were 52 The hybridized EGS oligos were inserted into a pUC19-based vector containing the T7 promoter, pUCT7/AEFRNAHHT7T, pre-digested with PstI and BstEII, generating the plasmids pUCT7mglBEGS52, pUCT7mgl-BEGS73, pUCT7mglBEGS148, pUCT7mglBEGS155, and pUCT7mglBEGS155R. To create the templates for M1 EGSs, the hybridized EGS oligos were inserted into a pUC19-based vector containing the T7 promoter and an M1, pUCT7/M1AEFR-NAHHT7T, pre-digested with PstI and BstEII, generating the plasmids pUCT7M1mglBEGS52, pUCT7M1mglBEGS73, pUCT7M1mglBEGS148, pUCT7M1mglBEGS155, and pUCT7M1mglBEGS155R. These constructs were sequenced and the plasmids digested with BstEII, ethanol-precipitated and dissolved in distilled water, and used as templates for in vitro transcription.

EGS activity assays in vitro
EGS activity assays were performed according to a previous report [17]. The EGS in 100-, 50-or 10-fold molar excess was incubated with 10 nM of 59 end-labeled or internally labeled target mRNAs in PA buffer (50 mM Tris-Cl, pH 7.5, 10 mM MgCl 2 , 100 mM NH 4 Cl) for 5 min at room temperature. Subsequently, 10 nM of M1 RNA and 100 nM of C5 protein were added to the mixture and further incubated for 30 min at 37uC. For reactions with M1 or M1 EGS RNA alone, 100 nM of M1 or M1 EGS RNA was used in PA100 buffer (PA buffer containing 100 mM MgCl 2 ). Reactions in 10 ml volume were performed at 37uC for 30 min, and terminated by adding 10 ml of 10 M urea/10% phenol solution. The product was separated on 5% polyacrylamide that contained 7 M urea gels.

Construction of plasmids for the EGS activity assay in E. coli
The plasmids contain the platform expressing both a target mRNA and a specific EGS by using the rnpB promoter for the target mglB mRNA and the T7 promoter for various EGSs. An additional 'C' was introduced into the -35 region of lacZ on the plasmids in order to inactivate transcription from the lacZ promoter by making a StuI site [15].
To make the mutagenesis in the -35 region of lacZ in the plasmids, the HindIII-AflIII fragment, which carries the mutated lacZ promoter, from the plasmid pMlac was cut and inserted into the plasmid pKB283-mglB to create pMlac-mglB. The HindIII-AflIII fragment was also inserted into the plasmid pKB28359HHC5EGS139HHM1T to create pKB283-Mlac.

Northern analysis
Freshly transformed cells were used for Northern analysis. BL21(DE3) competent cells (150 ml) were mixed with 0.1 mg of the appropriate plasmid and transformed at 42uC for 90 sec. The transformed cells were cultured overnight at 37uC on an LB plate with ampicillin. It is essential to use freshly transformed cells: when the cells were cultured from the glycerol stock of transformed cells, the expression of EGS was poor very often. A colony was picked from the LB plate and was cultured overnight in LB media containing carbenicillin at 37uC.
The overnight culture was diluted 1:100 and incubated at 37uC to OD600 = 0.6. The cells were split into two tubes and 2 mM (final concentration) IPTG was added into one of the tubes. The cells were further incubated for one hour and total RNA was prepared as described previously [20]. The RNAs (4 mg) were separated on a 2% agarose gel.