Expression of small RNAs of Bordetella pertussis colonizing murine tracheas

Abstract We performed RNA sequencing on Bordetella pertussis, the causative agent of whooping cough, and identified nine novel small RNAs (sRNAs) that were transcribed during the bacterial colonization of murine tracheas. Among them, four sRNAs were more strongly expressed in vivo than in vitro. Moreover, the expression of eight sRNAs was not regulated by the BvgAS two‐component system, which is the master regulator for the expression of genes contributing to the bacterial infection. The present results suggest a BvgAS‐independent gene regulatory system involving the sRNAs that is active during B. pertussis infection.

Bordetella pertussis causes whooping cough, a contagious respiratory disease that has been resurging recently despite high vaccination coverage. 1,2 This organism produces multiple virulence factors, including toxins and adhesins, the expression of which is largely regulated by the BvgAS two-component system, consisting of the sensor kinase BvgS and response regulator BvgA. 3 At 37°C in standard Bordetella media, the BvgAS system activates the transcription of a set of genes (Bvg-activated genes) including various virulence genes. Conversely, this system is inactivated at temperatures lower than 26°C or in the presence of MgSO 4 (40-50 mM) or nicotinic acid (10-20 mM), and B. pertussis eventually does not express the Bvg-activated genes. The former bacterial state is called the Bvg + phase, and the latter is the Bvg − phase. The BvgAS system is considered to play a major role in the expression of genes involved in the pathogenesis of B. pertussis; however, recent in vivo studies found that several Bvg-activated genes were repressed in B. pertussis colonizing the respiratory tracts of mice. 4,5 van Beek et al. also reported that approximately 30% of all genes were differentially expressed between in vitro and in vivo conditions. 4 Furthermore, a B. pertussis clinical strain, the BvgAS system of which was dysfunctional due to a spontaneous mutation in the bvgS gene, was isolated from a pertussis patient. 6 These findings suggest that a complex mechanism, besides the BvgAS system, is involved in the regulation of the bacterial gene expression during the course of infection.
Bacterial small RNAs (sRNAs) are functional noncoding RNA molecules that range between 50 and 500 nucleotides in length. 7 Previous studies identified numerous sRNAs in various pathogenic and commensal bacteria using a computational analysis and laboratorybased techniques, such as microarrays, Northern blotting, and RNA sequencing (RNA-seq). [8][9][10][11] Most sRNAs posttranscriptionally upregulate or downregulate downstream gene expression by affecting the stability and translational efficiency of target messenger RNAs (mRNAs) through base pairing with them. 12,13 A wide variety of physiological processes, including metabolism, stress responses, and the expression of virulence genes, are regulated by sRNAs. [14][15][16][17][18] In B. pertussis, many types of sRNAs have been identified or predicted by an in silico analysis and RNA-seq on the bacteria grown in vitro. 9,11,14 However, it currently remains unclear whether B. pertussis sRNAs are involved in the regulation of in vivo gene expression, which is associated with the establishment of bacterial infection. In the present study, we performed in vivo RNA-seq on B. pertussis colonizing the murine tracheas and identified novel sRNAs that were strongly expressed during colonization.
In vivo expression of sRNAs were analyzed by RNAseq using tracheas of three mice independently infected with B. pertussis-type strain 18323. This organism was grown at 37°C on Bordet-Gengou agar (Becton Dickinson, Franklin Lakes, NJ) containing 1% HIPOLY-PEPTON (Nihon Pharmaceutical, Tokyo, Japan), 1% glycerol, 15% defibrinated horse blood, and 10 µg/mL ceftibuten (BG plate). The bacteria recovered from the colonies on BG plates were suspended in Stainer-Scholte (SS) broth 19 to obtain an OD 650 of 0.2, and cultured at 37°C for 14 hr with shaking. Bacterial CFUs were estimated from OD 650 values according to the following equation: 1 OD 650 = 3.3 × 10 9 CFU/mL. Seven-week-old male C57BL/6J mice (CLEA Japan, Osaka, Japan) were anesthetized with a mixture of medetomidine (Kyoritsu Seiyaku, Tokyo, Japan), midazolam (Teva Takeda Pharma, Nagoya, Japan), and butorphanol (Meiji Seika Pharma, Yokohama, Japan) at final doses of 0.3, 2, and 5 mg/kg body weight, respectively, and intranasally inoculated with B. pertussis 18323 (1 × 10 7 CFU) in 50 μL of SS medium using a micropipette with a needle-like tip. On Day 4 after inoculation, mice were killed with pentobarbital, and the tracheas were excised and frozen in liquid nitrogen. Total RNA was extracted from the tracheas with TRIzol Reagent (Thermo Fisher Scientific, Waltham, MA), treated with RNase-Free DNase (Takara Bio, Shiga, Japan), and then purified with the PureLink RNA Mini Kit (Thermo Fisher Scientific) according to the manufacturer's instructions. Bacterial and murine ribosomal RNAs (rRNAs) were simultaneously depleted from the total RNA using the Ribo-Zero rRNA Removal Kit for Human/Mouse/Rat and Gram-Negative Bacteria (Illumina, San Diego, CA). The quality and quantity of RNA samples were assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). Reverse transcription was performed with the rRNA-depleted RNA, SuperScript III Reverse Transcriptase (Thermo Fisher Scientific), and Random Primer N9 (Takara Bio), and double-stranded DNA was then synthesized using DNA polymerase I (Klenow fragment [3′-5′ exo-]; New England Biolabs, Ipswich, MA). The resultant complementary DNA (cDNA) was sheared to approximately 600 bp fragments using Covaris S220 (Covaris, Woburn, MA) and purified with Agencourt AMPure XP beads (Beckman Coulter, Miami, FL). Libraries of the cDNA fragments were then prepared with the KAPA Library Preparation Kit (Kapa Biosystems, Wilmington, MA) and TruSeq adapters (Illumina), and sequenced with a HiSeq 2500 (Illumina) to obtain 101 bp single-end reads. The sequenced reads were mapped to the genomic DNA of B. pertussis 18323 (GenBank: NC_018518.1) using CLC Genomics Workbench, version 8.0.3 (CLC bio, Waltham, MA). All animal experiments were approved by the Animal Care and Use Committee of the Research Institute for Microbial Disease, Osaka University, and conducted according to the Regulations on Animal Experiments at Osaka University. The numbers of total sequenced reads were 54, 143, and 137 million, and 0.06%, 0.72%, and 0.04% of the reads in each sample were aligned to the genome sequence of B. pertussis 18323. A large portion of the reads aligned to the bacterial genome corresponded to protein-, rRNA-, and transfer RNA-coding sequences (95.5%, 99.9%, and 97.3%), whereas the residual reads were aligned to the intergenic regions: the numbers of reads were 1180, 1316, and 836, respectively. We predicted that these noncoding sequences located in the intergenic regions are potential sRNA sequences. Among these sRNA candidates, we selected nine novel sRNAs, for which the number of sequenced reads was more than 20 counts, and designated them B. pertussis sRNA (Bpr) 1-9 according to a previous study 9 (Table 1). Homologous sRNAs to Bpr1-9 were not found in the public databases including BLAST (https://blast.ncbi.nlm.nih.gov/Blast. cgi) and sRNAMap (http://srnamap.mbc.nctu.edu.tw).
In vitro and in vivo expression of Bpr1-9 were compared by qRT-PCR analyses. Total RNA was extracted and purified from the tracheas of mice independently infected with B. pertussis Tohama, a vaccine strain, and two clinical strains (BP139 and BP143 gifted from K. Kamachi, National Institute for Infectious Diseases) 20 in the same manner as B. pertussis 18323. Total RNA was also prepared from the four strains of B. pertussis and Bvg +and Bvg − -locked mutants derived from B. pertussis 18323 grown in vitro using the PureLink RNA Mini Kit and RNase-Free DNase according to the manufacturer's instructions. The Bvg +and Bvg − -locked mutants, which constitutively express the Bvg + and Bvg − phenotypes, respectively, were constructed by the sitedirected mutagenesis of BvgS to replace Arg with His at position 570 and to delete the region of amino acid positions from 542 to 1020, respectively, 21 using double-crossover homologous recombination as described previously. 22 In brief, the plasmids bvgS-C3-pABB-CRS2-Gm and ΔbvgS-pABB-CRS2-Gm 22 were introduced into Escherichia coli DH5α λpir, and then transconjugated into B. pertussis 18323 by triparental conjugation with the helper strain E. coli HB101 harboring pRK2013, 23 Table 2 under the following conditions: initial denaturation at 95°C for 10 min, and 40 cycles at 95°C for 15 s and 60°C for 1 min. qRT-PCR analyses revealed that the expression levels of Bpr4, 5, 8, and 9 in B. pertussis 18323 colonizing murine tracheas were significantly higher (118-, 64-, 9-, and 6-fold, respectively) than those in in vitro-cultured bacteria (Figure 1a). By contrast, no significant differences in the expression of Bpr1-3, 6, or 7 were observed between in vitro and in vivo conditions. Similar results were obtained with B. pertussis Tohama and two clinical strains (Figure 1b). The in vitro expression levels of Bpr1-7 and 9 in B. pertussis 18323 were largely unaffected by the absence or presence of 40 mM MgSO 4 . In addition, the Bvg +and Bvg − -locked mutants equally expressed these sRNAs (Figure 1c). By contrast, the expression of Bpr8 was negligible in B. pertussis 18323 wild-type grown in the presence of 40 mM MgSO 4 (i.e. Bvg − phase condition) and the Bvg − -locked mutant. These results indicate that the expression of Bpr1-7 and 9 is independent of the BvgAS regulatory system, whereas that of Bpr8 is BvgAS dependent.
The presence of Bpr4, 5, 8, and 9 in B. pertussis 18323 was confirmed by rapid amplification of cDNA end (RACE) and Northern blotting. For the determination of the transcription start and termination sites of the Bpr, 5′-and 3′-RACE were performed using a SMARTer RACE 5′/3′ Kit (Takara Bio) according to the manufacturer's instructions. In brief, total RNA was extracted from in vitro-cultured B. pertussis 18323 and polyadenylated by poly(A) polymerase (New England Biolabs). After reverse transcription by SMARTScribe Reverse Transcriptase, the resultant cDNA was used as a template for PCR with Universal primer and each bpr-specific primers ( Table 2). The PCR products were then cloned into linearized pRACE and five individual clones were sequenced. The precise transcription start and termination sites of Bpr4 For production of digoxigenin (DIG)-labeled RNA probes, partial antisense strands of bpr and recA genes were amplified from B. pertussis 18323 using appropriate primers (Table 2), and cloned into the downstream of T7 promoter on pSPT18 (Sigma-Aldrich). The resulting plasmids were linearized with SalI, and DIG-labeled RNA probes were pertussis 18323 was subjected to electrophoresis in a 1.5% denaturing formaldehyde agarose gel, transferred to a positively charged membrane (Hybond-N+; GE Healthcare, Piscataway, NJ), and UV cross-linked to the membrane. The membrane was then independently incubated with DIG-labeled RNA probes for each Bpr and recA, respectively, followed by alkaline phosphate-conjugated sheep anti-DIG immunoglobulin G, and visualized with CDP-Star. Northern blotting using the RNA probes for Bpr4, 8, and 9 detected a single band, whereas Bpr5 migrated as two bands ( Figure 2). The mobility of each Bpr corresponded to that estimated from its length determined by RACE. sRNAs regulate the expression of genes involved in a wide variety of physiological processes in bacteria, including the adaptation to host environments and virulence. [14][15][16][17][18] In B. pertussis, 14 types of sRNAs designated as BprA-N were identified by an in silico analysis and Northern blotting 9 ; however, these sRNAs have not yet been characterized. Recent studies performed an RNA-seq analysis using B. pertussis grown in vitro and identified an sRNA designated as RgtA (repressor of glutamate transport) that was found to reduce the translation of BP3831, a periplasmic amino acidbinding protein of an ABC transporter, by base pairing with the 5′ untranslated region of BP3831 mRNA. 11,14 Although this protein is related to the transport of glutamate, it currently remains unclear whether RgtA is involved in the pathogenesis of B. pertussis. In the present study, we identified nine types of novel sRNAs that were strongly expressed during the bacterial colonization, and demonstrated that the expression of four types of sRNAs (Bpr4, 5, 8, and 9) was stronger in vivo than in vitro. To the best of our knowledge, this is the first study to identify the in vivo strongly expressed sRNAs of B. pertussis. Bpr4, 5, 8, and 9, which were strongly expressed in vivo, may be involved in regulating the expression of genes necessary for the bacterial colonization or infection. In Salmonella enterica serovar Typhimurium, PinT, a PhoP-induced sRNA, was shown to be upregulated by up to 100-fold during the infection, and regulated the expression of the invasion-associated effectors and virulence genes required for intracellular survival. 17 Li et al. also reported that Ysr170, a strongly expressed sRNA in Yersinia pestis invading host cells, contributed to the bacterial intracellular survival. 18 These findings support our hypothesis that the sRNAs strongly induced during infection are involved in the adaptation and/or pathogenesis of B. pertussis. In addition, we found that the expression of Bpr1-7 and 9 was not regulated by the BvgAS system. Although the BvgAS system was previously considered to be the master virulence regulator in B. pertussis, 3 recent studies demonstrated that the expression profiles of Bvg-regulated genes were largely different between in vitro and in vivo conditions, 4,5 suggesting a complex mechanism that regulates in vivo gene expression. Several groups reported the PlrSR two-component system and BspR/BtrA, an anti-σ factor, as accessory regulatory systems downstream of BvgAS, which may play a part in this complex gene regulatory system in vivo. [24][25][26][27] Besides these regulators, the sRNAs identified in the present study may function as another regulator for gene expression during B. pertussis infection. Further research is currently in progress in our laboratory to identify genes whose expression is regulated by the sRNAs and elucidate the mechanisms by which the sRNAs regulate the gene expression.