A Defective Bacteriophage Produced by Bacillus subtilis MAFF 118147 and a Mutant Producing No Normal Particles of the Defective Bacteriophage

Abstract

Bacillus subtilis (natto) MAFF 118147 produced a defective bacteriophage, different in DNA size from PBND8, a previously reported defective phage of B. subtilis (natto) IAM 1207. The phage, named DF118147, had a small head (diameter 40 nm) and contractile tail (17 × 264 nm). Its DNA was 13.5 kb in size and yielded a smear pattern in gel electrophoresis after digestion with Hin dIII. Following acridine orange treatment, a mutant of B. subtilis (natto) MAFF 118147, in whose culture the 13.5-kb DNA fragment was not detected, was obtained. Electron microscopy observation showed that the head and tail of the defective phage produced by the mutant were not joined. A mutant yielding no normal defective phage particles is useful as a host in which other types of bacteriophage are amplified, especially for isolating their DNA.

Introduction

After the addition of mitomycin C to the culture, Bacillus subtilis 168 produces the defective bacteriophage, PBSX, whose genes are coded on a 28-kb region in the chromosomal DNA of the 168 strain (Seaman et al., 1964; Kunst et al., 1997; Moszer, 1998). PBSX has a small head and a contractile tail. Instead of its own genes, the defective phage particle packages a DNA fragment 13 kb in length, which is randomly cut out of the host Bacillus genomic DNA (Zahler, 1993). Although the defective phage kills sensitive B. subtilis strains, producing a clear zone on a lawn of such strains when a suspension of the defective phage is spotted, unlike common phages (e.g., Escherichia coli lambda phage) it is unable to propagate in the sensitive cells. The existence of a defective phage from B. natto was reported as a PBSX-like particle when PBSX was discovered (Seaman et al., 1964). However, the authors did not describe further details along with its morphology.

B. subtilis (natto) IAM 1207, which had been isolated and identified as B. natto, produced a defective bacteriophage, PBND8, after induction with mitomycin C (Tsutsumi et al., 1990). PBND8 has similar morphology to PBSX, but a smaller head. PBND8 has an 8-kb DNA fragment generated randomly by cutting the genome DNA of IAM 1207, and is about 5 kb smaller than that of PBSX (Tsutsumi et al., 1990). A DNA region highly homologous to the gene-coding region of PBSX was found on the chromosomal DNA of B. subtilis (natto) BEST195 (Nishito et al., 2010).

As defective phage particles are also produced constitutively from Bacillus cells under normal cultural conditions, i.e., without induction by mitomycin C, it is likely that defective phages could contaminate the purification process of virulent phages. A host strain that does not produce any defective phages is desired to avoid contamination of the defective phage. In this article, a defective phage of B. subtilis (natto) MAFF 118147 and the construction of a mutant producing no normal particles of the defective phage are described. Interestingly, the defective phage of B. subtilis (natto) MAFF 118147 contains a DNA fragment that is longer than that reported for PBND8 and equal to that of PBSX.

Materials and Methods

Microorganisms and culture media    The strains used in this study are listed in Table 1. B. subtilis MAFF 118147 and MAFF 302078 were obtained from the NIAS Genebank (Tsukuba, Japan). Construction of the mutants in Table 1 is described in detail below. The mutants were deposited at the NIAS Genebank. B. subtilis 1A1 and IAM 1207 were obtained from the Bacillus Genetic Stock Center (Columbus, OH, USA) and the IAM Culture Collection (Tokyo, Japan), respectively.

Table 1. Bacillus subtilis strains used in this study

Strain	Relevant characteristicsa	Origina	NIAS Genebank No.b
MAFF 118147	PGA+, DP118147+	Fermented food, Miyagi Pref., Japan
MAFF 302078	Rifs	Soil, Saitama Pref., Japan
NA1	PGA−, DP118147+	Spontaneous mutant from MAFF 118147	MAFF 211989
NA2	PGA−, DP118147−	NA1 mutated with AO	MAFF 211990
NA3	Rifr	Spontaneous mutant from MAFF 302078	MAFF 211991
1A1	trpC2	Marburg strain
IAM 1207 (=JCM 20105)c		From Sawamura as B. natto

a  Abbreviations: PGA, polyglutamate production; Rif, rifampicin; AO, acridine orange.

b  Deposited at the NIAS Genebank (Tsukuba, Japan) with the indicated accession numbers.

c  The IAM Culture Collection was dissolved in 2006 and its collections were transferred to JCM (Riken BioResource Center, Tsukuba, Japan).

Bacillus subtilis strains were grown in Luria-Bertani medium [LB; 1% tryptone (BD, Franklin Lakes, NJ, USA), 0.5% yeast extract (BD), 1% NaCl] supplemented with 10 mM MgSO₄ (LBMg) or nutrient broth (Nissui Pharmaceutical, Tokyo, Japan), or on agar plates containing LBMg at 37°C for 16 – 24 h or at 30°C for 24 – 48 h. Induction by mitomycin C was not carried out in this study.

For PGA production, B. subtilis was grown on GSP plates [1% glucose, 1.5% monosodium glutamate, 1% phytone (BD), and 1.5 – 1.7% agar] at 37°C overnight (Nagai et al., 1997).

Plaque formation of B. subtilis (natto) phage JNDMP was described previously (Nagai and Yamasaki, 2009).

Test for biotin requirement    Minimal medium, consisting of 0.2% NH₄SO₄, 1.4% K₂HPO₄, 0.6% KH₂PO₄, 50 µg/mL each of tryptophan and monosodium glutamate, 0.1% sodium citrate·2H₂O, 0.02% MgSO₄·7H₂O and 0.5% glucose, was prepared according to Anagnostopoulos and Spizizen (1961). Sterile biotin solution was added at a final concentration of 0.1 µg/mL. Twenty microliters of a 1/100 dilution in 0.85% NaCl of an overnight culture grown in LBMg was inoculated into 2 mL of minimal medium in the presence or absence of biotin. After 24-h incubation, optical density at 660 nm was measured. In the second experiment, 20 µL of the culture without biotin (or a 1/100 dilution of the culture for B. subtilis Marburg) was inoculated into 2 mL of minimal medium in the presence or absence of biotin.

Isolation of defective phage particles and their DNA    Defective phage particles and their DNA were isolated according to Nagai and Yamasaki (2009) with minor modifications. Briefly, Bacillus stains were grown in LBMg or nutrient broth overnight at 37°C. The culture supernatant was recovered by centrifugation. Defective phage particles in the supernatant were precipitated by adding NaCl and polyethylene glycol 6000 (PEG), and suspended in SM buffer. After digestion with DNase and RNase, and then with proteinase K in 1% SDS, DNA was extracted in the aqueous phase using phenol-chloroform (1:1). The DNA was dissolved in TE after precipitation with ethanol. Digestion with proteinase K was often skipped.

For isolation of defective phages in cells, cells were collected from the overnight culture and suspended in SM. After cell lysis with 1 – 2 mg/mL lysozyme for 50 min at 37°C, NaCl and PEG were added, and phage particles were precipitated and suspended in SM buffer.

The QIAEX II Gel Extraction Kit (Qiagen, Venlo, the Netherlands) was used to purify DNA samples for digestion with Hin dIII.

Electron microscopy observation    The phage suspension was negatively stained with 2% uranyl acetate and observed under a JEM 2000EX transmission electron microscope (Jeol, Tokyo, Japan). Observations were carried out at the Hanaichi Ultrastructure Research Institute in Aichi, Japan.

Growth inhibition test    After centrifugation of the B. subtilis MAFF 118147 culture grown in LBMg for 24 h, the supernatant was filtered through a 0.2-µm membrane filter (Advantec, Tokyo, Japan). Three µL of the filtrate were spotted onto 4 mL of 0.6% agarose containing 20 µL of a 24-h culture of a tested strain, which was solidified on LBMg agar. After incubation at 37°C for 24 h, the production of a growth inhibition zone on the plate was determined.

Fractionation of culture filtrate    The culture, filtered through a 0.2-µm membrane filter as described above, was then filtered through a USY-5 or USY-20 ultrafiltration membrane (Advantec) with a molecular weight cut-off of 50 kDa or 200 kDa, respectively.

Mutation with acridine orange    Mutation with acridine orange was carried out mainly in accordance with a method described previously (Hara et al., 1982). B. subtilis (natto) MAFF 118147 was grown in 2 mL of LBMg with acridine orange in a test tube wrapped with aluminum foil and shaken. After incubation overnight at 37°C, appropriate dilutions of the culture were plated on LBMg agar and grown overnight at 30°C. Colonies on the plates were picked with sterile toothpicks and inoculated into 200 µL of LBMg in a 96-well microtiter plate. The plates were sealed with vinyl tape and then incubated at 37°C overnight with rotary shaking. After centrifugation, the supernatant was spotted onto a lawn of B. subtilis NA3 (suspended in 0.6% agarose gel) on LBMg agar supplemented with 2.5 µg/mL rifampicin. After overnight incubation at 37°C, colonies without a growth inhibition zone were isolated.

Results and Discussion

Biotin requirement of B. subtilis (natto) MAFF 118147    From a food production standpoint, especially in Japan, it is important to determine whether B. subtilis strains are applicable to natto fermentation. The ability to ferment natto is tightly linked to biotin auxotrophy (Kiuchi et al., 1987; Ueda, 1989; Nagai and Tamang, 2010; Kubo et al., 2011). Natto-fermenting strains were formerly classified as B. natto, and were re-classified as B. subtilis (or B. subtilis subsp. subtilis). The classification was supported by the results of phylogenetic studies, in which natto-fermenting strains formed a cluster within the B. subtilis cluster (Gibson and Gordon, 1974; Tamang et al., 2002; Kubo et al., 2011). However, for the sake of convenience, natto-fermenting strains are often referred to as “B. subtilis (natto)” in food microbiology.

B. subtilis 1A1 (natto non-fermenting laboratory strain) grew well in the minimal medium without biotin, but both B. subtilis (natto) IAM 1207 and MAFF 118147 showed poor growth (Table 2). The second cultures, which were inoculated with 20 µL of the first culture (or a dilution for B. subtilis 1A1) in the minimal medium, gave similar results, indicating no influence of contaminated biotin from the LBMg cultures used for inoculating the minimal medium. In terms of biotin requirement, B. subtilis MAFF 118147 belongs to the B. subtilis (natto) group. The phenotype with the biotin requirement of both of B. subtilis (natto) IAM1207 and MAFF118147 was somewhat leaky.

Table 2. Growth of Bacillus subtilis strains in minimal media with or without biotin

Strain		1A1	IAM 1207	MAFF 118147
Without biotin	1st	0.623	0.111	0.091
	2nd	0.636	0.157	0.102
With biotin	1st	0.671	0.689	0.802
	2nd	0.679	0.873	0.788

Twenty microliters of the first culture (or a 1/100 dilution of the first culture for B. subtilis 1A1) without biotin was inoculated into the second minimal medium. The values represent optical densities at 660 nm. The minimal medium without biotin showed a value of 0.001.

Defective bacteriophage of B. subtilis (natto) MAFF 118147    Defective phage particles were observed by electron microscopy in the culture of B. subtilis (natto) MAFF 118147 without induction by mitomycin C (Fig. 1) and the defective phage was named DP118147 for “defective phage of MAFF 118147.” DP118147 had a small head (diameter 40 nm) and contractile tail (17 × 264 nm).

Fig. 1.

Defective phage of B. subtilis (natto) MAFF 118147(A) and its contractile particle (B)

Bar = 100 nm.

The DNA contained in DP118147 was 13.5 kb in size, equal to that of PBSX and PBND8 DNA (Fig. 2A). Although it was reported that PBND8 contained DNA 8 kb in size (Tsutsumi et al., 1990), the defective phage produced by IAM 1207 contained 13.5-kb DNA in this study. Morphological differences were also observed (Nagai, unpublished results). The DNA of DP118147 gave a smear pattern below the 13.5-kb fragment by 0.8% agarose gel electrophoresis after digestion with Hin dIII (Fig. 2B). The pattern indicated that DP118147 was composed of highly heterogeneous fragments similar to PBDN8 DNA (Tsutsumi et al., 1990).

Fig. 2.

DNA isolated from defective phages

Panel A, Defective phage DNAs. Panel B, Digestion of defective phage DNAs with Hin dIII. M, marker, 2.5-kb DNA ladder (Takara Bio, Shiga, Japan); A, PBSX; B, defective phage produced from B. subtilis (natto) IAM1207; C, DP118147. H; digested with Hin dIII. Electrophoresis was carried out on a 0.8% agarose gel.

Strains sensitive to the culture filtrate of B. subtilis (natto) MAFF 118147    Among the 140 B. subtilis strains obtained from the NIAS Genebank, 73 (52%) were sensitive to the culture filtrate of B. subtilis (natto) MAFF 118147. B. subtilis MAFF 302078 was chosen as an indicator strain for DP118147 because of its apparent strong sensitivity to the culture filtrate of B. subtilis (natto) MAFF 118147 and the formation of a uniform lawn in the upper layer of the agarose gel. A rifampicin-resistant mutant of MAFF 302078, named NA3, was used for the experiments described below.

B. subtilis is known to produce a wide variety of antibiotics (Stein, 2005), and, as the antibiotics can also produce growth inhibitory zones similar to defective phages, this similarity can interfere with the screening of mutants for the production of defective phages. The presence of antibiotics, which have much lower molecular weights than defective phage particles, in the culture filtrate of B. subtilis (natto) MAFF 118147 was examined by a simple fractionation technique. When the culture of B. subtilis (natto) MAFF 118147, after being filtered through a 0.2-µm membrane filter, was fractionated through an ultrafiltration membrane with a molecular weight cut-off of 50 kDa, defective phage DNA and strong growth inhibitory activity was detected in the retentate fraction. The filtrate had weak growth inhibitory activity and no DP118147 DNA (data not shown). Even though some phage particles passed through the membrane with a molecular weight cut-off of 200 kDa, strong growth inhibitory activity remained in the retained fraction. It was concluded that although weak inhibitory activity with low molecular weight for NA3 growth was detected in the MAFF118147 culture, it would not interfere with screening for a mutant producing defective phage.

Screening of a mutant producing no defective phage    B. subtilis (natto) produces the highly viscous polymer, poly-γ-glutamic acid (PGA), which is not easily separated from DNA during conventional DNA preparation because of its physicochemical resemblance to DNA. It was reported that PGA protects cells from infection by phages (Kimura and Itoh, 2003). To overcome these problems, a colony that produced no PGA on GSP plates was selected and named NA1. It is known that B. subtilis (natto) often loses its PGA productivity by high-frequency transposition of the insertion sequence IS4Bsu1 to comP, which controls PGA synthesis (Nagai et al., 2000). However, whether the defect in PGA production of NA1 was caused by transposition of IS4Bsu1 was not confirmed.

In order to construct a mutant producing no DP118147, NA1 was mutated with acridine orange (AO). Acridine dyes are DNA-intercalating agents and cause frameshift mutations by nucleotide insertion or deletion (Nasim and Brychcy, 1979). One among 96 colonies isolated after the drug treatment showed no clear zone on the lawn of NA3 and was named NA2. Though the exact frequency remains to be determined, this apparent high mutation frequency might be simply due to the length of the region or genes to be mutated. Mutation of PGA synthesis directed by 3 genes on the chromosomal DNA occurred at frequencies of 10 – 60% (Hara et al., 1982; Ashiuchi et al., 1999). On the other hand, reversion from auxotroph to prototroph in B. subtilis with acridine yellow occurred at low frequencies of 0 to 10⁻⁵, depending on the auxotroph type (Stewart, 1968). The high mutation frequency with AO might imply the existence of genes involved in the production of DP1181147 on plasmid, as AO is known to eliminate plasmids (e.g., F plasmid of Escherichia coli) at a high frequency of 100% (Hirota, 1960). However, this is not the case with NA2, as no plasmid could be detected in the parent strain, MAFF 118147, judging from the agarose gel electrophoresis banding patterns of DNA prepared by an alkaline lysis method (data not shown). In the experiment, B. subtilis (natto) MAFF 118100 harboring 2 types of plasmids was used as a positive plasmid control.

The 13.5-kb DNA fragment packaged in defective phage particles was not detected in either the culture broth or the cell lysate of NA2 (Fig. 3). It remains to be determined whether the DNA fragment was excised normally from the NA2 chromosomal DNA and then destroyed by DNase or not excised at all. Electron microscopy observation of the NA2 culture showed that imperfect particles of DP118147 were produced; moreover, the head and tail failed to join (Fig. 4). The contracted tail was also observed (Fig. 4C), indicating that the tail could contract despite the loss of morphological integrity as a phage particle.

Fig. 3.

Agarose gel electrophoresis of DNA from cultures of B. subtilis (natto) MAFF 118147 and its mutant NA2, and from the cell lysate of NA2

DNA was purified from cultures and cell lysates. DNA samples from the culture and cell lysate of the mutant NA2 were concentrated three times more than that from the culture of B. subtilis (natto) MAFF 118147. Electrophoresis was carried out on a 0.8% agarose gel. M, marker, 2.5-kb DNA ladder (Takara Bio, Shiga, Japan); A, DNA of DP118147; B, DNA from the culture of NA2 (not visible); C, DNA from the lysate of NA2 (not visible).

Fig. 4.

Imperfect DP118147 particles produced by NA2

A, tails; B, head; C, contractile tail. Bar = 200 nm. Enlarged images of a head (B) and a contractile tail (C) are shown under the main panel.

The growth curves of mutant NA2 and its parent strain MAFF 118147 showed that they grew at the same rate (data not shown). The efficiency of plaque formation of JNDMP on NA1 (PGA⁻ mutant) was 51% higher than that on its parent strain MAFF 118147, and the efficiency on NA2 was 99% of that on NA1. NA2 was not killed by DP118147. These results indicate that the mutation caused no critical damage to cell growth, susceptibility to phage JNDMP, or immunity against DP118147. The increase in susceptibility to a phage after mutation of PGA production might support the conclusion that PGA is a defense system against attacks by bacteriophages (Kimura and Itoh, 2003).

The strategy employed in this study to mutate defective phage production might be applicable to all B. subtilis wild-type strains producing extracellular defective bacteriophage particles, unless the genes associated with defective phage production have been sequenced as B. subtilis 168 and B. subtilis (natto) BEST195, which could be easily mutated by double crossover homologous recombination.

References

•  Anagnostopoulos,  C. and  Spizizen,  J. (1961). Requirements for transformation in Bacillus subtilis. J. Bacteriol., 81, 741-746.
•  Ashiuchi,  M.,  Soda,  K., and  Misono,  H. (1999). A poly-γ-glutamate synthetic system of Bacillus subtilis IFO 3336: Gene cloning and biochemical analysis of poly-γ-glutamate produced by Escherichia coli clone cells. Biochem. Biophys. Res. Commun., 263, 6-12.
•  Gibson,  T. and  Gordon,  R. (1974). Genus I. Bacillus Cohn 1872. In “Bergey's Manual of Determinative Bacteriology”, 8th edition,  R.  Buchanan and  N.  Gibbons, eds. N. Williams & Wilkins, Baltimore, Md, pp. 529-550.
•  Hara,  T.,  Aumayr,  A.,  Fujio,  Y., and  Ueda,  S. (1982). Elimination of plasmid-linked polyglutamate production by Bacillus subtilis (natto) with acridine orange. Appl. Environ. Microbiol., 44, 1456-1458.
•  Hirota,  Y. (1960). The effect of acridine dyes on mating type factors in Escherichia coli. Proc. Natl. Acad. Sci. U.S.A., 46, 57-64.
•  Kiuchi,  K.,  Taya,  N.,  Sulistyo,  J., and  Funane,  K. (1987). Isolation and identification of natto bacteria from market-sold natto starters. Rep. Natl. Food Res. Inst., 50, 18-21.
•  Kimura,  K. and  Itoh,  Y. (2003). Characterization of poly-γ-glutamate hydrolase encoded by a bacteriophage genome: Possible role in phage infection of Bacillus subtilis encapsulated with poly-γ-glutamate. Appl. Environ. Microbiol., 69, 2491-2497.
•  Kubo,  Y.,  Rooney,  A.P.,  Tsukakoshi,  Y.,  Nakagawa,  R.,  Hasegawa,  H., and  Kimura,  K. (2011). Phylogenetic analysis of Bacillus subtilis strains applicable to natto (fermented soybean) production. Appl. Environ. Microbiol., 77, 6463-6469.
•  Kunst,  F.,  Ogasawara,  N.,  Moszer,  I.,  Albertini,  A.M.,  Alloni,  G.,  Azevedo,  V.,  Bertero,  M.G.,  Bessières,  P.,  Bolotin,  A.,  Borchert,  S.,  Borriss,  R.,  Boursier,  L.,  Brans,  A.,  Braun,  M.,  Brignell,  S.C.,  Bron,  S.,  Brouillet,  S.,  Bruschi,  C.V.,  Caldwell,  B.,  Capuano,  V.,  Carter,  N.M.,  Choi,  S.-K.,  Codani,  J.-J.,  Connerton,  I.F.,  Cummings,  N.J.,  Daniel,  R.A.,  Denizot,  F.,  Devine,  K.M.,  Düsterhöft,  A.,  Ehrlich,  S.D.,  Emmerson,  P.T.,  Entian,  K.D.,  Errington,  J.,  Fabret,  C.,  Ferrari,  E.,  Foulger,  D.,  Fritz,  C.,  Fujita,  M.,  Fujita,  Y.,  Fuma,  S.,  Galizzi,  A.,  Galleron,  N.,  Ghim,  S.-Y.,  Glaser,  P.,  Goffeau,  A.,  Golightly,  E.J.,  Grandi,  G.,  Guiseppi,  G.,  Guy,  B.J.,  Haga,  K.,  Haiech,  J.,  Harwood,  C.R.,  Hénaut,  A.,  Hilbert,  H.,  Holsappel,  S.,  Hosono,  S.,  Hullo,  M.-F.,  Itaya,  M.,  Jones,  L.,  Joris,  B.,  Karamata,  D.,  Kasahara,  Y.,  Klaerr-Blanchard,  M.,  Klein,  C.,  Kobayashi,  Y.,  Koetter,  P.,  Koningstein,  G.,  Krogh,  S.,  Kumano,  M.,  Kurita,  K.,  Lapidus,  A.,  Lardinois,  S.,  Lauber,  J.,  Lazarevic,  V.,  Lee,  S.-M.,  Levine,  A.,  Liu,  H.,  Masuda,  S.,  Mauël,  C.,  Médigue,  C.,  Medina,  N.,  Mellado,  R.P.,  Mizuno,  M.,  Moestl,  D.,  Nakai,  S.,  Noback,  M.,  Noone,  D.,  O'Reilly,  M.,  Ogawa,  K.,  Ogiwara,  A.,  Oudega,  B.,  Park,  S.-H.,  Parro,  V.,  Pohl,  T.M.,  Portetelle,  D.,  Porwollik,  S.,  Prescott,  A.M.,  Presecan,  E.,  Pujic,  P.,  Purnelle,  B.,  Rapoport,  G.,  Rey,  M.,  Reynolds,  S.,  Rieger,  M.,  Rivolta,  C.,  Rocha,  E.,  Roche,  B.,  Rose,  M.,  Sadaie,  Y.,  Sato,  T.,  Scanlan,  E.,  Schleich,  S.,  Schroeter,  R.,  Scoffone,  F.,  Sekiguchi,  J.,  Sekowska,  A.,  Seror,  S.J.,  Serror,  P.,  Shin,  B.-S.,  Soldo,  B.,  Sorokin,  A.,  Tacconi,  E.,  Takagi,  T.,  Takahashi,  H.,  Takemaru,  K.,  Takeuchi,  M.,  Tamakoshi,  A.,  Tanaka,  T.,  Terpstra,  P.,  Tognoni,  A.,  Tosato,  V.,  Uchiyama,  S.,  Vandenbol,  M.,  Vannier,  F.,  Vassarotti,  A.,  Viari,  A.,  Wambutt,  R.,  Wedler,  E.,  Wedler,  H.,  Weitzenegger,  T.,  Winters,  P.,  Wipat,  A.,  Yamamoto,  H.,  Yamane,  K.,  Yasumoto,  K.,  Yata,  K.,  Yoshida,  K.,  Yoshikawa,  H.-F.,  Zumstein,  E.,  Yoshikawa,  H., and  Danchin,  A. (1997). The complete genome sequence of the Gram-positive bacterium Bacillus subtilis. Nature, 390, 249-256.
•  Moszer,  I. (1998). The complete genome of Bacillus subtilis: from sequence annotation to data management and analysis. FEBS Lett., 430, 28-36.
•  Nagai,  T.,  Koguchi,  K., and  Itoh,  Y. (1997). Chemical analysis of poly-γ-glutamic acid produced by plasmid-free Bacillus subtilis (natto): evidence that plasmids are not involved in poly-γ-glutamic acid production. J. Gen. Appl. Microbiol., 43, 139-143.
•  Nagai,  T. and  Tamang,  J. (2010). Fermented legumes: Soybean and non-soybean products. In “Fermented Foods and Beverages of the World.”  J.P.  Tamang and  K.  Kailasapathy, eds. CRC Press, Boca Raton, FL, pp. 191-224.
•  Nagai,  T.,  Tran,  L.S.P.,  Inatsu,  Y., and  Itoh,  Y. (2000). A new IS4 family insertion sequence, IS4Bsu1, responsible for genetic instability of poly-γ-glutamic acid production in Bacillus subtilis. J. Bacteriol., 182, 2387-2392.
•  Nagai,  T. and  Yamasaki,  F. (2009). Bacillus subtilis (natto) bacteriophages isolated in Japan. Food Sci. Technol. Res., 15, 293-298.
•  Nasim,  A. and  Brychcy,  T. (1979). Genetic effects of acridine compounds. Mutation Res., 65, 261-288.
•  Nishito,  Y.,  Osana,  Y.,  Hachiya,  T.,  Popendorf,  K.,  Toyoda,  A.,  Fujiyama,  A.,  Itaya,  M., and  Sakakibara,  Y. (2010). Whole genome assembly of a natto production strain Bacillus subtilis natto from very short read data. BMC Genomics, 11, 243. doi.10.1186/1471-2164-11-243
•  Seaman,  E.,  Tarmy,  E., and  Marmur,  J. (1964). Inducible phages of Bacillus subtilis. Biochemistry, 3, 607-613.
•  Stein,  T. (2005). Bacillus subtilis antibiotics: structures, syntheses and specific functions. Mol. Microbiol., 56, 845-857.
•  Stewart,  C.R. (1968). Mutagenesis by acridine yellow in Bacillus subtilis. Genetics, 59, 23-31.
•  Tamang,  J.P.,  Thapa,  S.,  Dewan,  S.,  Jojima,  Y.,  Fudou,  R., and  Yamanaka,  S. (2002). Phylogenetic analysis of Bacillus strains isolated from fermented soybean foods of Asia: kinema, chungkokjang and natto. J. Hill Res., 15, 56-62.
•  Tsutsumi,  T.,  Hirokawa,  H., and  Shishido,  K. (1990). A new defective phage containing a randomly selected 8 kilobase-pairs fragment of host chromosomal DNA inducible in a strain of Bacillus natto. FEMS Microbiol. Lett., 72, 41-46.
•  Ueda,  S. (1989). Industrial application of B. subtilis. In “Bacillus subtilis: Molecular Biology and Industrial Application”.  B.  Maruo and  H.  Yoshikawa, eds. Kodansha, Tokyo, Japan, pp. 143-161.
•  Zahler,  S.A. (1993). Temperate bacteriophages. In “Bacillus subtilis and Other Gram-Positive Bacteria: Biochemistry, Physiology, and Molecular Genetics”.  A.L.  Sonenshein,  J.A.  Hoch, and  R.  Losick, eds. American Society for Microbiology, Washington, D.C., pp. 831-842.