Involvement of the Response Regulator Ctra in the Extracellular DNA Production of the Marine Bacterium Rhodovulum Sulfidophilum

J. Gen. Appl. Microbiol., 64, 103–107 (2018) doi 10.2323/jgam.2017.08.003 „2018 Applied Microbiology, Molecular and Cellular Biosciences Research Foundation

Full Paper

Involvement of the response regulator CtrA in the extracellular DNA production of the marine bacterium Rhodovulum sulfidophilum

(Received June 16, 2017; Accepted August 14, 2017; J-STAGE Advance publication date: March 12, 2018) Hiroyuki Komatsu,1,a Junya Yamamoto,1,a Hiromichi Suzuki,1,† Nobuyoshi Nagao,1,b Yuu Hirose,1 Takashi Ohyama,2,3 So Umekage,1 and Yo Kikuchi1,2,* 1 Department of Environmental and Life Sciences, Toyohashi University of Technology, Toyohashi, Japan 2 Major in Integrative Bioscience and Biomedical Engineering, Graduate School of Science and Engineering, Waseda University, Tokyo, Japan 3 Department of Biology, Faculty of Education and Integrated Arts and Sciences, Waseda University, Tokyo, Japan

here for the free nucleic acids present in the culture The marine bacterium Rhodovulum sulfidophilum supernatant, but not those bound to any cells, vesicles, or is a nonsulfur phototrophic bacterium, which is particles, even those which are categorized as “extracel- known to produce extracellular nucleic acids in lular”. Using this property, we developed a method for soluble form in culture medium. In the present the extracellular production of homogeneous recombinant paper, constructing the response regulator ctrA- RNA molecules, such as RNA aptamers (reviewed by deficient mutant of R. sulfidophilum, we found that Kikuchi, 2010; Suzuki et al., 2010, 2011; Umekage et al., this mutation causes a significant decrease in the 2012) and short hairpin RNAs (Nagao et al., 2014) by the extracellular DNA production. However, by the in- introduction of an appropriate RNA expression plasmid troduction of a plasmid containing the wild type into the cell. This approach has been proposed as a possi- ctrA gene into the mutant, the amount of extracel- ble method in industrial technology to produce RNA medi- lular DNA produced was recovered. This is the first cines in the future (Kikuchi, 2010; Kikuchi et al., 2010). and clear evidence that the extracellular DNA pro- For such medical applications, extensive genetical and duction is actively controlled by the CtrA in R. physiological elucidation of this bacterium is important, sulfidophilum. especially from the viewpoint of the rigorous production demands of medicines. Accordingly, the complete genome Key Words: ctrA; extracellular DNA; gene trans- sequences of several strains of this bacterium have been fer agent; marine phototrophic bacterium; determined by us and others (Guzman et al., 2017; Masuda Rhodovulum sulfidophilum et al., 2013; Nagao et al., 2015a). During the course of this study, we also found that R. sulfidophilum produces a gene transfer agent-like particle Introduction (GTA-like particle) (Nagao et al., 2015b). Gene transfer agents (GTAs) were first discovered in Rhodobacter The marine bacterium Rhodovulum sulfidophilum is a capsulatus (basionym Rhodopseudomonas capsulata) by nonsulfur phototrophic bacterium, which can grow under Marrs (1974), but have now been described from various both aerobic-dark and anaerobic-light conditions (Hiraishi prokaryotic species (Lang and Beatty, 2007, 2010). GTAs and Ueda, 1994; Kikuchi, 2010). This bacterium also is are shaped like bacteriophage particles, but are different known to produce extracellular nucleic acids in soluble from bacteriophages in many ways. GTA particles do not form in culture medium (Ando et al., 2006; Suzuki et al., contain their own genome coding their structure, but they 2009a, b). The term “nucleic acids in soluble form” is used have short DNA fragments randomly cut from the host

*Corresponding author: Yo Kikuchi, Major in Integrative Bioscience and Biomedical Engineering, Graduate School of Science and Engineering, Waseda University, 2-2 Wakamatsu-cho, Shinjuku-ku, Tokyo 162-8480, Japan. E-mail: [email protected] aThese authors contributed equally to this work. bPresent address: IDAC Theranostics, Inc., Y-lab., Yokohama 230-0045, Japan. †Deceased 25 July 2016. None of the authors of this manuscript has any financial or personal relationship with other people or organizations that could inappropriately influence their work. 104 KOMATSU et al. genome. These DNA fragments are used for genetic ex- phoresis, the gel was stained with ethidium bromide. Band change between host cells by a mechanism similar to intensities were evaluated using image gauge software phage-mediated generalized transduction. The structural (Fuji Film, Japan). The band at the position around 23 kbp genes of GTAs are present in the host bacterial genome on 0.7 or 1% agarose gel electropherogram was used for like the prophage genes of lysogen. However, GTAs are extracellular DNA. not inducible by mitomycin C (Marrs, 1974) and do not Construction of the ctrA mutant of R. sulfidophilum. A form plaques (Solioz and Marrs, 1977). The expression of ctrA mutant, strain No. 3 was constructed from DSM 1374 GTA genes studied to date is controlled by cellular regu- by transduction using the suicide vector containing an in- latory systems (Brimacombe et al., 2013; Schaefer et al., activated ctrA gene by the insertion of a Kanamycin re- 2002) and functions as truly gene transfer agents for the sistant gene (Km). First, the ctrA coding region of 583 bp host cells. GTAs play a role in lateral gene transfer in na- (base number from 2,691,164 to 2,691,746: the numbers ture and, thus, affect the evolution of prokaryotic genomes. from the complete genome sequence of Masuda et al., Lang and Beatty (2000) reported that the response regu- 2013) was inserted into SmaI site of pGEM3Z to yield lator CtrA is necessary for the expression of the GTAs of pGEM-ctrA. Since the ctrA has a NruI site (base number Rba. capsulatus. CtrA were first found in the bacterium from 2,691,412 to 2,691,417, Masuda et al., 2013) at al- Caulobacter crescentus as an essential two-component most the center of the gene, the Km was inserted to this signal transduction protein (Quon et al., 1996). In C. site (the blunt end cutting site, 2,691,414) to disrupt the crescentus, CtrA is essential for viability and acts as a ctrA. We call this plasmid, pGEM-ctrA-Km. Using the master regulator of the cell cycle (reviewed by Skerker pGEM-ctrA-Km as a template, a DNA sequence having and Laub, 2004), but the homologous protein of Rba. the 5 -ctrA-Km-ctrA-3 (disrupted ctrA gene) was ampli- capsulatus has different roles. CtrA of Rba. capsulatus is ¢ ¢ fied by PCR. This sequence was inserted into the site be- not essential and is not involved in the cell cycle (Mercer tween XhoI and NotI of the plasmid vector pUTmini- et al., 2010). In a previous paper, we reported that R. Tn5Km (Biomedal Co.) to yield pUT-ctrA-Km. As this sulfidophilum produces GTA-like particles and that this plasmid requires pir for replication, this plasmid cannot production is controlled by the gene ctrA (Nagao et al., l be maintained in R. sulfidophilum and then function as a 2015b). Here, we further considered that not only the GTA- suicide vector in this bacterium. The plasmid pUT-ctrA- like particles, but also the extracellular soluble nucleic acid Km containing E. coli S17- pir was co-cultivated with R. production, may be controlled by CtrA. At present, it has l sulfidophilum, and 12 colonies of Km-resistant R. not been obvious what gene(s) are involved in the extra- sulfidophilum were selected as candidates of the ctrA mu- cellular soluble nucleic acid production of R. tants. Among these 12 candidates, we finally obtained a sulfidophilum. Here, we report the construction of a ctrA- strain (No. 3) as a ctrA mutant. The ctrA gene of this mu- deficient mutant of R. sulfidophilum and show the signifi- tant was disrupted by single crossover event on the ge- cant decrease of the amount of extracellular DNA produced nome with pUT-ctrA-Km. This was confirmed by colony by the mutant. This is the first evidence that CtrA is in- PCR, and Southern blotting analyses of the genome using volved in the extracellular soluble DNA production in R. probes specific for ctrA and Km. Also, with Northern analy- sulfidophilum. ses using the ctrA specific probes, no mature ctrA mRNA from this strain, No. 3, could be detected; instead, a longer Materials and Methods mRNA of over 1.5 kb containing the Km sequence could be detected. We concluded that the wild type ctrA gene Bacterial strains and growth conditions. The purple pho- was completely replaced by the sequence 5 -ctrA-Km- totrophic marine alphaproteobacterium, Rhodovulum ¢ ctrA-3 in strain No. 3. sulfidophilum DSM 1374 and a ctrA mutant, strain No. 3, ¢ were used throughout this study. Strain No. 3 was con- Construction of the plasmid for ctrA expression. To constructed from DSM 1374 by displacement of the ctrA gene firm the CtrA function for extracellular DNA production, with a disrupted version of ctrA using a suicide plasmid we planned to test whether the amount of the extracellu- vector (see below). Cultivation was performed essentially lar DNA can be recovered by introduction of a plasmid as described (Suzuki et al., 2010). The strains were grown containing the wild type ctrA gene into the strain No. 3. anaerobically at 25∞C in a 1.5-ml or 50-ml tube filled with For this experiment, the plasmid containing the wild type PYS medium (Nagashima et al., 1997) containing 2% (wt/ ctrA gene was constructed. First the wild type ctrA gene vol) NaCl under incandescent illumination (about 3,000 including the promoter region was amplified from the ge- lux). Kanamycin was used at a concentration of 30 mg/ml nome DNA of R. sulfidophilum by the PCR technique us- for the strains containing the Kanamycin resistant gene. ing the forward primer, 5¢-GAATTCGCCAGT- Transformation of R. sulfidophilum was carried out by the TGGAAGAAGGCG-3¢, and the reverse primer, 5¢- method described previously (Suzuki et al., 2011). TTATTAGGCGCCGAGCGCGAAGGAGCCC-3¢. The 5¢ end of the forward primer corresponds to the base number Analysis of cell growth and extracellular nucleic acid 2,690,877 of the genome and the 5 end of the reverse production. Cell growth and extracellular nucleic acid ¢ primer does to the base number 2,691,820 of the genome. production were analyzed as previously described (Ando The base numbers are from Masuda et al. (2013). The et al., 2006). Cell growth was evaluated by measuring the amplified DNA fragment of 944 bp contains a long 5 UTR turbidity of the culture medium at 600 nm. Amounts of ¢ of the ctrA gene (244 bp) which is thought to include the extracellular DNA and RNA produced was calculated from promoter region of this gene. This fragment was inserted the band intensity of gel electrophoresis. After electro- CtrA and extracellular DNA production 105

Fig. 2. Recovery of extracellular DNA production in the ctrA mutant upon the introduction of the ctrA gene containing plasmid. The extracellular DNA sample was prepared from 1 ml of supernatant of each 48 h culture, and subjected to 1% agarose gel electrophoresis. DNA was detected by ethidium bromide staining. Lanes WT, the DNA sample from the wild type culture; delta c, sample from the culture of the ctrA mutant; T, sample from the culture of the mutant strain trans- formed by the ctrA containing plasmid pBBR RESO-ctrA; M, size marker, phage lambda DNA digested by HindIII. The sizes (in kbp) are shown to the left of the panel.

Assay for the recovery of extracellular nucleic acid-producing function in ctrA mutant by transformation of the wild type ctrA gene. Three strains, wild type DSM 1374, the ctrA mutant, and the ctrA mutant harboring the wild Fig. 1. Production of extracellular soluble DNA by the wild type and type ctrA gene containing plasmid, were cultivated for 48 the ctrA-deficient mutant of R. sulfidophilum. h. After centrifugation of each culture to remove the cells, A and B, electrophoretic analysis of extracellular soluble DNA from 1 ml of supernatant was transferred to the new tube. The culture medium of the wild type (A) and ctrA mutant (B). The DNA extracellular nucleic acids from each strain were precipi- sample was prepared from the culture supernatant of 500 ml at each cultivation time point, and subjected to 0.5% agarose gel electrophore- tated by isopropanol. Each precipitate was quantitatively sis. DNA was detected by ethidium bromide staining. Lanes 1 and 2, solubilized with sterilized water. These samples were ap- DNA samples from 0 h culture; lanes 3 and 4, from 17 h culture; lanes plied to 1% agarose gel electrophoresis. After electro- 5 and 6, from 27 h culture; lanes 7 and 8, from 38 h culture; lanes 9 and phoresis, the gel was stained with ethidium bromide. Band 10, from 50 h culture; lanes 11 and 12, from 63 h culture; lane 13, from 72 h culture. M, size marker, phage lambda DNA digested by HindIII. intensities were evaluated as described above. The sizes (in kbp) are shown to the right of the panel. Arrows, DNA bands used for quantitative analysis (see below). C, time course curve Results and Discussion of cell growth and extracellular soluble DNA production of the wild type (closed symbols) and ctrA mutant (open symbols). Squares, cell Mutation of the ctrA gene causes a significant decrease growth which was evaluated by measuring the turbidity of the culture medium at 600 nm; circles, extracellular soluble DNA. The intensities in the extracellular soluble DNA production of the DNA bands on the electropherogram in A and B (arrows in A and First, we compared the amounts of extracellular soluble B) were used for the evaluation of the amount of extracellular DNA DNA of the two strains, R. sulfidophilum DSM 1374 (wild production. Error bars indicate the respective standard deviations which type) and its ctrA-deficient mutant, strain No. 3. Figures were calculated from the results of three independent experiments. 1A and B show electrophoretic analyses of the time course of the extracellular soluble DNA production of the wild type and strain No. 3, respectively. Since direct photo- metric measurement of extracellular soluble nucleic acids into the BglII site of the plasmid vector pBBR RESO in the culture supernatant gave unreliable results, prob- (MoBiTec). The cleaved BglII site of this plasmid was ably because of co-production of extracellular proteins and treated with a Klenow fragment to obtain the blunt ends, polysaccharides, the bands around 23 kbp (arrows of Figs. then the insertion of the wild type ctrA gene DNA was 1A and B) are chosen for estimation of the extracellular achieved by blunt end ligation with T4 DNA ligase. We soluble DNA amounts. Using the intensity values of these call this plasmid, pBBR RESO-ctrA which contains a full- bands, the amounts of the soluble DNAs produced were length of the wild type ctrA gene. calculated as described in Materials and Methods, and the 106 KOMATSU et al. legend of Fig. 1. Using these values, the time course curves Relationship between the 4.5 kb DNA band and the gene of the extracellular soluble DNA productions with cell transfer agent-like particle growths are shown in Fig. 1C. The wild type strain In a previous paper (Nagao et al., 2015b), we reported maximally produced DNA of about 0.12 mg/ml, whereas that R. sulfidophilum produces the GTA-like particle which the ctrA mutant produced 0.05 mg/ml. These values may is thought to be involved in genetic exchange between host be underestimated because the values are derived from cells. We also mentioned that the particle production is only the bands of the electropherograms as described also controlled by the gene ctrA (Nagao et al., 2015b). above, but these values may be useful to show reliable Although the relationship between the extracellular solu- time course curves. Most importantly, these data clearly ble DNA and the GTA-like particle of this organism is not showed that the ctrA gene is necessary for the regular yield clear at present, both may be used for genetic exchange of the extracellular soluble DNA production. Also, it is between cells. R. sulfidophilum also produces the DNA revealed that CtrA in R. sulfidophilum may not be essen- fragment of 4.5 kb as an extracellular soluble DNA (Fig. tial for viability, as there is not a significant difference 1A, lanes 9–13 and Fig. 2, lane WT). These may be the between the growth curves of the two strains (Fig. 1C). genome fragments which would be incorporated in, or have Similar effects of CtrA has been noted in Rhodobacter leaked from, the GTA-like particles (Nagao et al., 2015b). capsulatus. Mercer et al. (2010) reported that a loss of the The band of 4.5 kb is visible in lane T of Fig. 2, although CtrA did not affect growth phase regulation in Rba. it is faint. This means that the production of this fragment capsulatus. Functions of the ctrA gene of R. sulfidophilum of 4.5 kb is also controlled by the gene ctrA, even if this may be similar to that of Rba. capsulatus. controlling process is not exactly the same as that of the extracellular DNA production. Recovery of the amount of extracellular DNA in the ctrA mutant by the introduction of the wild type ctrA gene Future application and environmental nucleic acids To further confirm the involvement of the ctrA gene in Previously, we developed a method for extracellular the extracellular DNA production, the effect of the intro- production of artificially designed, functional RNAs (RNA duction of the wild type ctrA gene into the mutant strain aptamers and short hairpin RNAs) in the culture medium No. 3 was tested. The wild type ctrA gene containing plas- using R. sulfidophilum (Kikuchi, 2010; Kikuchi et al., mid, pBBR RESO-ctrA, was constructed as described in 2010; Nagao et al., 2014; Suzuki et al., 2010, 2011; Materials and Methods. The plasmid pBBR RESO-ctrA Umekage et al., 2012). This method has been proposed contains a full-length of the wild type ctrA gene. This plas- for the industrial production of RNA medicines. Elucida- mid was introduced into the mutant strain No. 3 by a heat- tion of the control mechanism, especially the up-regula- shock transformation method (Suzuki et al., 2011). As tion mechanism of extracellular nucleic acids production shown in Fig. 2, the amount of extracellular DNA was re- by the ctrA gene may now be important to obtain a high covered by the introduction of the plasmid into the mu- yield of the artificial RNA products. In the present paper, tant (Fig. 2, lane T). The intensity of the band around 23 we concentrate mainly on DNA production: the behavior kbp of lane T of Fig. 2 is similar to that of the band of the of RNA production is almost the same as that of DNA. wild type (Fig. 2, lane WT). The amounts of extracellular Extracellular DNA is detected ubiquitously throughout soluble DNAs were calculated from the intensities of the the environment, such as in seawater and soil (Lorenz and bands around 23 kbp, as described in Materials and Meth- Wackernagel, 1994; Tani and Nasu, 2010; Vlassov et al., ods. The values of the wild type (Fig. 2, lane WT), the 2007). Environmental DNAs, especially high molecular mutant (lane delta c), and the transformant by pBBR weight DNAs of bacterial origin, are, at least in part, de- RESO-ctrA (lane T), were determined to be 0.11 mg/ml, rived from the active production by a similar stimulation 0.02 mg/ml, and 0.10 mg/ml, respectively. From these val- mechanism as shown in the present paper. This extracel- ues, the recovery of the function can be estimated to be lular DNA production is an important phenomenon for almost 90%. The copy number of the plasmid pBBR, the genetic exchange, not only within species, but also may original vector of pBBR RESO-ctrA, in E. coli has been contribute to dynamic and global genetic diversity. reported to be about 30 (Antoine and Locht, 1992). Al- though we do not know the copy number of pBBR RESO- Conclusions ctrA in R. sulfidophilum at present, it is presumed that the transformant of this experiment has multiple genes of the We have demonstrated that the amount of the extracel- wild type ctrA. It may be supposed that there is no gene lular soluble DNA of R. sulfidophilum is controlled by the dosage effect in this case. It probably indicates that the gene ctrA. Although the precise mechanism for the extra- ctrA gene does not directly affect the extracellular DNA cellular nucleic acid production of this bacterium is not production but indirectly controls it through downstream yet known, we have provided a clue for its elucidation. gene(s). The ctrA mutant still produces small amounts of extra- Acknowledgments cellular DNA as visible in Fig 2 lane delta c and Fig. 1B. It may be natural to suppose that these are derived from This work was supported in part by a grant-in-aid for Scientific Re- search to Y. Kikuchi (Grant-in Aid for Scientific Research No. 25252017) some unknown basal mechanism for extracellular DNA from the Japan Society of the Promotion of Science (JSPS), and by the production. It is possible that this mechanism may be Institute for Fermentation, Osaka, to S. Umekage. We thank Professor stimulated by the CtrA. Akira Hiraishi of Toyohashi University of Technology for comments about the experiment. CtrA and extracellular DNA production 107

References Short hairpin RNAs of designed sequences can be extracellularly produced by the marine bacterium Rhodovulum sulfidophilum. J. Ando, T., Suzuki, H., Nishimura, S., Tanaka, T., Hiraishi, A. et al. (2006) Gen. Appl. Microbiol., 60, 222–226. Characterization of extracellular RNAs produced by the marine Nagao, N., Hirose, Y., Misawa, N., Ohtsubo, Y., Umekage, S. et al. photosynthetic bacterium Rhodovulum sulfidophilum. J. Biochem., (2015a) Complete genome sequence of Rhodovulum sulfidophilum 139, 805–811. DSM 2351, an extracellular nucleic acids producing bacterium. Ge- Antoine, R. and Locht, C. (1992) Isolation and molecular characteriza- nome Announcements, 3, e00388-15. tion of a novel broad-host-range plasmid from Bordetella Nagao, N., Yamamoto, J., Komatsu, H., Suzuki, H., Hirose, Y. et al. bronchiseptica with sequence similarities to plasmids from Gram- (2015b) The gene transfer agent-like particle of the marine pho- positive organisms. Mol. Microbiol., 6, 1785–1799. totrophic bacterium Rhodovulum sulfidophilum. Biochem. Biophys. Brimacombe, C. A., Stevens, A., Jun, D., Mercer, R., Lang, A. S. et al. Rep., 4, 369–374. (2013) Quorum-sensing regulation of a capsular polysaccharide Nagashima, K. V., Hiraishi, A., Shimada, K., and Matsuura, K. (1997) receptor for the Rhodobacter capsulatus gene transfer agent Horizontal transfer of genes coding for the photosynthetic reaction (RcGTA). Mol. Microbiol., 87, 802–817. centers of purple bacteria. J. Mol. Evol., 45, 131–136. Guzman, M. S., McGinley, B., Santiago-Merced, N., Gupta, D., and Quon, K. C., Marczynski, G. T., and Shapiro, L. (1996) Cell cycle con- Bose, A. (2017) Draft genome sequences of three closely related trol by an essential bacterial two-component signal transduction isolates of the purple nonsulfur bacterium Rhodovulum protein. Cell, 84, 83–93. sulfidophilum. Genome Announcements, 5, e00029-17. Schaefer, A. L., Taylor, T. A., Beatty, J. T., and Greenberg, E. P. (2002) Hiraishi, A. and Ueda, Y. (1994) Intrageneric structure of the genus Long-chain acyl-homoserine lactone quorum-sensing regulation of Rhodobacter: transfer of Rhodobacter sulfidophilus and related Rhodobacter capsulatus gene transfer agent production. J. marine species to the genus Rhodovulum gen. nov. Int. J. Syst. Bacteriol., 184, 6515–6521. Bacteriol., 44, 15–23. Skerker, J. M. and Laub, M. T. (2004) Cell-cycle progression and the Kikuchi, Y. (2010) Extracellular nucleic acids of the marine phototrophic generation of asymmetry in Caulobacter crescentus. Nature Rev. bacterium Rhodovulum sulfidophilum and related bacteria: Physi- Microbiol., 2, 325–337. ology and biotechnology. In Extracellular Nucleic Acids, ed. by Solioz, M. and Marrs, B. (1977) The gene transfer agent of Kikuchi, Y. and Rykova, E. Y., Springer-Verlag, Heidelberg, pp. Rhodopseudomonas capsulata. Purification and characterization of 55–67. its nucleic acid. Arch. Biochem. Biophys., 181, 300–307. Kikuchi, Y., Suzuki, H., and Umekage, S. (2010) Produktion definierter Suzuki, H., Daimon, M., Awano, T., Umekage, S., Tanaka, T. et al. RNAs im Kulturüberstand von Bakterien. LABORWELT, 11, 6–7. (2009a) Characterization of extracellular DNA production and Lang, A. S. and Beatty, J. T. (2000) Genetic analysis of a bacterial ge- flocculation of the marine photosynthetic bacterium Rhodovulum netic exchange element: the gene transfer agent of Rhodobacter sulfidophilum. Appl. Microbiol. Biotechnol., 84, 349–356. capsulatus. Proc. Natl. Acad. Sci. USA., 97, 859–864. Suzuki, H., Umekage, S., Tanaka, T., and Kikuchi, Y. (2009b) Extracel- Lang, A. S. and Beatty, J. T. (2007) Importance of widespread gene lular tRNAs of the marine photosynthetic bacterium Rhodovulum transfer agent genes in a-proteobacteria. Trends Microbiol., 15, 54– sulfidophilum are not aminoacylated. Biosci. Biotechnol. Biochem., 62. 73, 425–427. Lang, A. S. and Beatty, J. T. (2010) Gene transfer agents and defective Suzuki, H., Ando, T., Umekage, S., Tanaka, T., and Kikuchi, Y. (2010) bacteriophages as sources of extracellular prokaryotic DNA. In Extracellular production of an RNA aptamer by ribonuclease-free Extracellular Nucleic Acids, ed. by Kikuchi, Y. and Rykova, E. Y., marine bacteria harboring engineered plasmids: a proposal for in- Springer-Verlag, Heidelberg, pp. 15–24. dustrial RNA drug production. Appl. Environ. Microbiol., 76, 786– Lorenz, M. G. and Wackernagel, W. (1994) Bacterial gene transfer by 793. natural genetic transformation in the environment. Microbiol. Rev., Suzuki, H., Umekage, S., Tanaka, T., and Kikuchi, Y. (2011) Artificial 56, 563–602. RNA aptamer production by the marine bacterium Rhodovulum Marrs, B. (1974) Genetic recombination in Rhodopseudomonas sulfidophilum: Improvement of the aptamer yield using a mutated capsulata. Proc. Natl. Acad. Sci. USA, 71, 971–973. transcriptional promoter. J. Biosci. Bioeng., 112, 458–461. Masuda, S., Hori, K., Maruyama, F., Ren, S., Sugimoto, S. et al. (2013) Tani, K. and Nasu, M. (2010) Role of extracellular DNA in bacterial Whole-genome sequence of the purple photosynthetic bacterium ecosystem. In Extracellular Nucleic Acids, ed. by Kikuchi, Y. and Rhodovulum sulfidophilum strain W4. Genome Announcements, 1, Rykova, E. Y., Springer-Verlag, Heidelberg, pp. 25–37. e00577-13. Umekage, S., Uehara, T., Fujita, Y., Suzuki, H., and Kikuchi, Y. (2012) Mercer, R. G., Callister, S. J., Lipton, M. S., Pasa-Tolic, L., Strnad, H. In vivo circular RNA expression by the permuted Intron-Exon et al. (2010) Loss of the response regulator CtrA causes pleiotropic method. In Innovations in Biotechnology, ed. by Agbo, E. C., Intech, effects on gene expression but does not affect growth phase regula- pp. 75–90. tion in Rhodobacter capsulatus. J. Bacteriol., 192, 2701–2710. Vlassov, V. V., Laktionov, P. P., and Rykova, E. Y. (2007) Extracellular Nagao, N., Suzuki, H., Numano, R., Umekage, S., and Kikuchi, Y. (2014) nucleic acids. BioEssays, 29, 654–667.