Purification and partial characterization of a ...

6 downloads 0 Views 839KB Size Report
isolated from Bacteroides ovatus H47. Can. J. Microbiol. 39: 169-174. A new intracellular bacteriocin isolated from a human fecal strain of Bacteroides ovatus ...
Purification and partial characterization of a bacteriocin isolated from Bacteroides ovatus H47

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13 For personal use only.

C. M. S. MIRANDA,L. M. FARIAS,M. A. R. CARVALHO,' AND C . A. V. DAMASCENO Laboratorio de Microbiologia Oral e Anaerobios, Departamento de Microbiologia, Znstituto de Cigncias Biologicas, Universidade Federal de Minas Gerais, Caixa Postal 2486, 30.161-960 Belo Horizonte, MG, Brazil A. H. TOTOLA AND C. A. P. TAVARES Laboratorio de Bioquimica e Zmunologia de Parasitos, Departamento de Bioquimica e Zmunologia, Znstituto de Cigncias Bioldgicas, Universidade Federal de Minas Gerais. Caixa Postal 2486, 30.161-960 Belo Horizonte, MG, Brazil

E. 0.CISALPINO Laboratorio de Microbiologia Oral e Anaerobios, Departamento de Microbiologia, Instituto de Cigncias Biologicas, Universidade Federal de Minas Gerais, Caixa Postal 2486, 30.161-960 Belo Horizonte, MG, Brazil AND

E. C. VIEIRA Laboratorio de Gnotobiologia, Departamento de Bioquimica e Zmunologia, Instituto de Cigncias Bioldgicas, Universidade Federal de Minas Gerais, Caixa Postal 2486, 30.161-960 Belo Horizonte, MG, Brazil Received March 19, 1992 Revision received June 30, 1992 Accepted August 14, 1992 MIRANDA,C. M. S., FARIAS,L. M., CARVALHO, M. A. R., DAMASCENO, C. A. V., TOTOLA,A. H., TAVARES, C. A. P., CISALPINO, E. O., and VIEIRA,E. C. 1993. Purification and partial characterization of a bacteriocin isolated from Bacteroides ovatus H47. Can. J. Microbiol. 39: 169-174. A new intracellular bacteriocin isolated from a human fecal strain of Bacteroides ovatus was partially purified through ammonium sulfate precipitation, ion exchange chromatography, gel filtration, and preparative polyacrylamide gel electrophoresis (PAGE). The bacteriocin is stable at a pH range of 3-10, at 60°C for 24 h, and at - 70°C for 6 months. It is inactivated by proteolytic enzymes. The molecular weight, estimated by sodium dodecyl sulfate - PAGE, is 78 kDa. Fifty strains of the Bacteroides fragilis group were isolated from fecal samples, and 41 of the isolates were shown to produce an antagonistic substance against at least 1 indicator strain. Iso-, auto-, and hetero-antagonisms were observed. Key words: Bacteroides ovatus, bacteriocin, human feces, bacterial growth inhibition. MIRANDA,C. M. S., FARIAS,L. M., CARVALHO, M. A . R., DAIVIASCENO,C. A. V., TOTOLA,A. H., TAVARES, E. O., et VIEIRA,E. C. 1993. Purification and partial characterization of a bacteriocin C. A. P., CISALPINO, isolated from Bacteroides ovatus H47. Can. J. Microbiol. 39 : 169-174. Une nouvelle bactkriocine intracellulaire d'une souche de Bacteroides ovatus isolCe de matibres fCcales humaines a CtC partiellement purifite par precipitation au sulfate d'ammonium, chromatographie par Cchange ionique, filtration sur gel et electrophorbse prkparative sur gel de polyacrylamide (PAGE). La bactkriocine est stable dans un Ccart de pH de 3 a 10, a 60°C pendant 24 h, et a - 70°C pendant 6 mois. Elle est inactivee par des enzymes protColytiques. Le poids molCculaire est de 78 kDa tel qu'estimt par le dodecyl sulfate de sodium - PAGE. Cinquante souches du groupe de Bacteroides fragilis ont CtC isolCes de matibres fCcales et 41 de ces isolements produisaient une substance antagoniste dirigte contre au moins 1 souche indicatrice. Des iso-, auto- et hCtCro-antagonismes ont ete obsewb. Mots cl6s : Bacteroides ovatus, bactkriocine, matieres fCcales humaines, inhibition de la croissance bacterienne.

Introduction Substances produced by microbes that colonize the human intestines have aroused the interest of many investigators, owing to the variety of species found. Since the first report on bacteriocin (Gratia 1925), very few papers on these antimicrobial substances produced by Bacteroides have appeared. These species have been frequently isolated from clinical samples and are of utmost importance in the ecological equilibrium of the human colon (Booth et al. 1977). In this paper, the isolation and partial purification of a bacteriocin from Bacteroides ovatus is reported. Materials and methods Bacterial strains Strains of Bacteroides eggertii (2), Bacteroides uniformis (3), Bacteroides ovatus (3), Bacteroides vulgatus (4), Bacteroides ' ~ u t h o rto whom all correspondence should be addressed. Printed

in

Canada / lrnprirne a u Canada

thetaiotaomicron (4), Bacteroides distasonis (7), and Bacteroides fragilis (27) were isolated from healthy human fecal samples in bile-esculin agar Bacteroides fragilis (BBE) selected medium (Livingston et al. 1978). Characterization and identification at the species level were carried out. These 50 strains, as well as B. fragilis ATCC 23745, were tested for inhibitory activity against themselves and against two reference strains: Streptococcus sanguis ATCC 10557 and Streptococcus uberis ATCC 9927. Over 1700 tests were performed. The isolated strains were maintained in skim milk (Difco) at -20°C and transferred to supplemented brain-heart infusion blood agar (BHIA-S) (Difco) according to Holdeman et al. (1977). Selection of bacteriocinogenic strains Bacteriocin activity was detected as described by Booth et al. (1977). After growth for 18-24 h in BHI-S broth (Holdeman et al. 1977), the strains (10~-10' cfu/mL) were inoculated in five equidistant points in 1.5% agar-BHIA-S (Booth et al. 1977). After incubation under anaerobiosis (atmosphere containing 90% N, and 10% COJ at 37°C for 48 h, the cells were killed by exposure

170

CAN. J . MICROBIOL. VOL. 39, 1993

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13 For personal use only.

to chloroform for 20 min (Tagg et al. 1976). After evaporation of the residual chloroform, the CHC1,-treated colonies were overlayed with 3.5 mL of BHIA-S (0.7% agar) (Booth et al. 1977). containing 0.2 mL of the indicator strain (lo6-10' cfu/mL) grown in BHI-S broth for 24 h. After incubation, the cultures were evaluated for the presence or absence of growth inhibition halos of the indicator bacteria. Cross-streak tests without the addition of chloroform were performed as controls (Pugsley and Oudega 1988). All experiments were carried out in duplicate.

Extraction of bacteriocin A strain of B. ovatus (H47) was selected for its wide spectrum of activity and stable production of the antagonistic substance. After growth in BHI-S broth under anaerobiosis at 37°C for 48 h, the cells were centrifuged at 23 400 x g at 4°C for 20 min. The supernatant was precipitated with 70010 ammonium sulfate, by adding the solid salt slowly with constant stirring at 4°C (Green and Hughes 1955). The precipitate was collected by centrifugation at 23 400 x g for 20 min, resuspended in 20 mL of 0.05 M phosphate buffer, pH 7.2. and exhaustively dialyzed against 0.01 M of the same buffer. The extracellular fraction (S-70) was thus obtained. The cellular pellet was washed with the same buffer, centrifuged, resuspended, and ultrasonicated (13 cycles for 2 min and at 50 W) in a model 450 Bronson sonicator. The resulting extract was submitted to centrifugation at 23 400 x g for 60 min. The supernatants were slowly concentrated through fractional precipitation with ammonium sulfate (0-30% and 30-70%) under constant stirring. Each precipitate was collected by centrifugation at 23 400 x g for 20 min and each fraction was dissolved in 20 mL of 0.05 M, phosphate buffer, pH 7.2, and exhaustively dialyzed against the same buffer. These preparations constituted the intracellular fractions (C-30 and C-70). Tests of activity and titration of crude bacteriocin Fractions S-70, C-30, and C-70 were tested for activity against Streptococcus sanguis ATCC 10557 after growth in BHI-S for 18-24 h at 37°C. A 0.1-mL aliquot was added to tubes containing molten semisolid BHIA-S at 45°C (0.7% agar). This inoculum was poured over solid agar plates. Aliquots of 50 pL of C-30, C-70, or S-70 containing 14.00, 5.39, and 14.67 mg/mL of protein, respectively, and previously treated with chloroform (50 pL/mL) for 30 min (Tagg et al. 1976), were placed directly onto the surface of the media containing the indicator strains. Evidence of activity was provided by the presence of halos of inhibition of growth of the indicator strains after 24 h of incubation. Aliquots of all active fractions were kept frozen at - 70°C. The C-70 fraction was further tested against the following indicator strains: Staphylococcus epidermidis ATCC 14990, Strep tococcus pyogenes ATCC10389, Staphylococcus aureus ATCC 25923, Proteus vulgaris ATCC 6380, Klebsiella pneumoniae ATCC 27736, Enterobacter cloacae ATCC 1333047, Escherichia coli ATCC 25922, Shigella sonnei ATCC 11060, Salmonella typhimurium ATCC 13311, Actinobacillus actinomycetemcomitans YA. Stre~tococcusmutans IM2PB14. Actinomvces viscosus T14V. and ~e$ostreptococcus a n a e r o b i u s . 27337. ~~~~ Titration of active fractions was performed by successive dilutions in 0.1 M Tris (Tris(hydroxymethy1)aminomethane) buffer, pH 7.2, and activity was tested as described above. The titer of the inhibitory activity was defined as the reciprocal of the highest dilution that showed a clearly visible zone of growth inhibition of the revealing strain and was expressed as arbitrary units (AU) of bacteriocin per millilitre. Protein determination Protein was determined according to Lowry et al. (1951). Bovine serum albumin was used as a standard. Partial purifcation The C-70 fraction was submitted to purification by fast protein, peptide, and polynucleotide liquid chromatography (FPLC)

(Pharmacia) in a 9.0 x 0.9 cm column of Q-Sepharose previously equilibrated with 0.05 M phosphate buffer, pH 7.2. A 10-mL aliquot of the fractions was applied and eluted with the same buffer with a gradient of 1.0 M NaCI, pH 7.2, at a rate of 2 mL/min. The peaks with activity were collected, dialyzed against 0.01 M ammonium bicarbonate buffer, pH 8.2, lyophilyzed, and resuspended in 0.01 M NaCl at pH 7.2 for protein determination. The peaks with bacteriocinogenic activity, obtained from Q-Sepharose chromatography, were submitted to a new fractionation in Sephacryl S-200 (Pharmacia) superfine column (1.6 x 82 cm) previously equilibrated with 0.05 M, ammonium bicarbonate, pH 8.4. The active fractions obtained were submitted to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) at 10%. Staining was done according to Tufion and Johansson (1980). The remaining gel was exhaustively washed for 18 h with 0.01 M. Tris-HC1 buffer at pH 7.2 to remove SDS. Then the gel was placed under sterile conditions on a Petri dish containing BHI agar (1.5% agar) and covered with the same agar (1%) containing the revealing strain. The lines of bacterial growth inhibition were determined after incubation at 37°C for 24 h. The molecular weights of the active fractions were determined through comparison of the rate of flux (Rf) of the active fractions with the R, of the molecular weight standard (SDSdH; Sigma Chemical Co.).

Stability of the bacteriocin The effect of pH on the stability of the bacteriocin was determined according to Gomori (1955) with the following buffers and pHs: KCI-HCI (2.0), glycine-HC1 (3.0), citrate (4.0), phosphatecitrate (6.0), Tris-HC1(KO), and carbonate-bicarbonate (10.0). Aliquots of 50 pL of the partially purified bacteriocin containing 1 mg/mL of protein in 450 pL of each filter-sterilized buffer were used. The mixtures were incubated for 24 h at 37°C and then 50 uL of each mixture was placed onto the indicator culture as described. The effect of temperature was evaluated by heating 100 pL of the partially purified bacteriocin previously treated with chloroform (50 pL of chloroform in 1 mL of the preparation) at time intervals of 1,2, and 24 h, at 80 and 100°C for 20 and 60 min. After cooling to 4"C, the inhibitory activity was tested and compared with the nontreated control. The effect of several enzymes on the partially purified bacteriocin was tested with buffers described by Gomori (1955). Aliquots of 50 pL of bacteriocin in 450 pL of enzyme solutions were used as follows: P-amylase (Sigma), 200 pg/mL in pH 5.0 citrate buffer; collagenase (Merck & Co.), 200 pg/mL; catalase (Merck). 300 pg/mL; lysozyme. 200 pg/mL; protease type VII (Sigma), 200 pg/mL; proteinase K (Sigma), 100 pg/mL, all in pH 7.2 borate buffer; pronase (Sigma), 250 pg/mL; trypsin (Sigma), 200 pg/mL, both in pH 8.0 Tris-HCI buffer; and pepsin, 200 pg/mL in pH 3.0 citrate buffer. Incubation was carried out for 2 h at 37°C. The effects of these enzymes were determined by their activities against Streptococcus sanguis ATCC 10557. Two controls were run, one without bacteriocin and with enzyme, the other without enzyme and with bacteriocin. Detection of interfering factors The search for phages was done by observation of fraction C-70, using an electron microscope. The C-70 fraction was mixed in a 1: 1 proportion with 4% phosphotungstic acid for 1 min and uranyl acetate for 30 s. One drop of the mixture was put in the screen previously covered with Formvar and charcoal. A Phillips 301 ultramicroscope was used with magnification from 25 000 to 45 000. To search for possible inhibition by fatty acids, tests were conducted in trypticase soy agar with and without the addition of 1% soluble starch (Walstad et al. 1974; Turner and Jordan 1981). The pH in the inhibition zone and in the noninoculated medium was measured with a microelectrode (Microelectrodes Inc., New Hampshire). The production of hydrogen peroxide was investigated with the enzymatic method described by Bergmeyer and Bernt (1974).

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13 For personal use only.

MIRANDA ET AL.

171

FIG. 1. Test for bacteriocin activity produced by strain H-47 of Bacteroides ovatus against sensitive strains of the B. fragilis group (A) and of Streptococcus sanguis (B).

Results Of the 1700 tests carried out, only 154 were positive for antagonism: 143, 10, and 1 for iso-, hetero-, and autoantagonism, respectively. Of the 50 tested strains, 9 did not produce antagonistic substances in relation to any indicator strain utilized. The bacteriocin-producing reference strain (B. fragilis ATCC 23745) exhibited isoantagonism against 10 indicator strains and heteroantagonism against Streptococcus sanguis. The highest level of bacteriocin production was reached after 48 h of cultivation. Figure 1 shows a typical experiment with bacteriocin preparation and an indicator strain. Only the intracellular fractions C-30 and C-70 obtained by precipitation with ammonium sulfate at 30 and 70% saturation, respectively, were active against Streptococcus sanguis ATCC 10557 used as an indicator strain. Fraction C-70, the most active of the two, was further tested against the remaining indicator strains and showed activity only against Streptococcus pyogenes ATCC 10389.

The C-30 fraction was active up to a 1:4 dilution and was inactivated after 15 days at -70°C. Unpurified fraction C-70 was active up to a 1: 16 dilution; after purification the titer was raised to 1:20, and the activity was unchanged after 6 months at - 70°C. Fractions C-30 and C-70, submitted to SDS-PAGE and activity tests, exhibited the same pattern of inhibition. On Q-Sepharose FPLC, only the first peak was active and was further submitted to chromatography on an S-200 Sephacryl column. Multiple protein peaks were obtained. The bacteriocinogenic activity was detected only in the second peak (Fig. 2A). On comparing the Rfvalues, determined from the activity tests and electrophoretic pattern of the several active fractions, the molecular weight of the bacteriocin was estimated as 78 kDa (Fig. 2B). In either reducing or nonreducing conditions, the bacteriocin was active in the monomeric form. It was stable in a pH range of 3-10 but was inactivated at pH 2. The activity was destroyed upon incubation at temperatures above

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13 For personal use only.

CAN. J.

MICROBIOL. VOL.

39, 1993

Fraction number

FIG.2. (A) S-200 Sephacryl column (82 x 1.6 cm) fractionation of the active material obtained from Q-Sepharose chromatography. Elution buffer: 0.05 M, ammonium bicarbonate, pH 8.0. Solid line, absorbance at 280 nm; broken line, bacteriocin titer. (B) Electrophoresis of the fractions from the second peak from S-200 Sephacryl column chromatography. Lane 1, molecular weight standard (SDSdH; Sigma); lane 2, crude bacteriocin; lanes 34, 42, and 43, fractions with negative bacteriocinogenic activity; lanes 35 to 41, fractions with positive bacteriocinogenic activity. 80°C for 20 min. Trypsin, pronase, proteinase K, and protease type VII inactivated the bacteriocin. No bacteriophagelike particles or bacteriophage subunit could be detected when negative staining of the active C-70 extract was examined with the electron microscope. No significant alterations in pH in the inhibition zone and in the noninoculated medium were detected when compared with the control experiment using the same indicator strain. Production of hydrogen peroxide, investigated by the enzymatic method (Bergmeyer and Bernt 1974), was not detected, as indicated by the absence of the characteristic

color in the medium inoculated with the test strain; a control of Streptococcus sanguis (Walstad et al. 1974) was used. The possibility of inhibition by fatty acids was discarded because of the lack of interference upon addition of soluble starch to the medium (Walstad et al. 1974).

Discussion Of the 50 strains of Bacteroides isolated from human feces, 41 produced bacteriocin against at least 1 indicator strain. Isoantagonism was detected. Ten strains were also active against a strain from another genus (Streptococcus

MIRANDA ET AL.

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13 For personal use only.

sanguis). These data represent a contribution to the knowledge of bacteriocin production, which is valuable given the ecological importance both of this phenomenon and of the bacterial group (Booth et al. 1977; Salyers 1984; Nakamura et al. 1984). Fifty isolated strains and 1 reference strain were tested against 52 indicator strains. Of the total of 1700 tests conducted, 1546 were negative. The negative results d o not necessarily mean that antagonistic substances were absent, since the indicator strain may not be sensitive t o the antagonistic substance because of a lack of specific receptors or because of resistance (Jakes and Zinder 1974; Graaf and Klaasen-Boor 1977; Reeves 1979; Harkness and Olschlager 1991). The autoantagonism that was detected is quite unusual, since this phenomenon is considered typical only among Gram-positive bacteria (Kutter 1966; Dajani et al. 1970; Gagliano and Hinsdill1970; Kramer and Brandis 1972). The sensitivity of a bacterium t o its own bacteriocin may be linked t o the lack of immunity that may occur owing t o a n imbalance in the concentrations of proteins that confer immunity to the cell itself (Ryan et al. 1955; Levinsohn et al. 1968; Vidaver et al. 1972). The data obtained with heteroantagonism are also quite peculiar. It has been reported that the bacteriocins of Gramnegative bacteria are considered t o have a more limited spectrum of action. Gram-negative bacteriocin activity is barely detected against Gram-positive microorganisms (Reeves 1979; Hammond et al. 1987). However, the antagonistic action of a species o n unrelated species of bacteria may occur through different mechanisms of action (Brock and Davie 1963; Vidaver et al. 1972). The results reported herein indicate that the strain isolated from B. ovatus produced a bacteriocin that had activity both against closely related species and against species from different genera; the latter finding has not been reported in the literature. The reported molecular weight (78 kDa) is different from the ones reported for human strains of the B. fragilis group: 13.5-18.7 kDa (Mossie et al. 1979), 300 kDa (Booth et al. 1977), and 7 kDa (Southern et al. 1984). The absence of activity of fraction S-70 and the gradual loss of activity of fraction C-30 o n storage at - 70°C suggest the presence of unspecific proteolytic enzymes. The results reported in the literature o n the stability of different bacteriocins towards factors such as pH, temperature, and proteolytic enzymes are very variable (Booth et al. 1977; Mossie et al. 1979; Southern et al. 1984; Hamrnond et al. 1987; Parrot et al. 1990; Lyon and Glatz 1991). This is probably due t o the different methods used by the authors. The protein nature of the bacteriocin from B. ovatus was evidenced by ( i ) the loss of activity when exposed t o proteolytic enzymes, (ii) the absorption at 280 nm, (iii) the increase in activity during the process of purification, ( i v ) the color reaction with Lowry's method, and (v) the stained bands o n SDS-PAGE (Fig. 2B). The experiments reported herein rule out the possibilities that the antagonistic action is due t o acids, hydrogen peroxide, bacteriophage, or fatty acids, the latter being discarded by the method of Walstad et al. (1974).

Acknowledgements The authors are grateful to Dr. Jacques R. Nicoli and Dr. Mauricio Rezende f o r suggestions; Dr. Maria

173

Evangelina F. Fonseca (Universidade Federal d o Rio de Janeiro) for the ultramicroscopic analysis; and Nicolau Santiago Pnmola, Nelder de Figueiredo Gontijo, Tulio Marcos Santos, Luzia Rosa Rezende, and Isabel de Paula Carvalho for technical help. This work was supported by Conselho Nacional de Desenvolvimento Cientifico e Tecnologico, Financiadora de Estudos e Projetos, ProReitoria de Pesquisas d a Universidade Federal de Minas Gerais, Coordenagao de Aperfeigoamento de Pessoal de Nivel Superior, Universidade Federal d o Piaui, and Curso de Pos-Graduagao d a Universidade Federal d o Rio de Janeiro. Bergmeyer, H.U., and Bernt, E. 1974. Determination with glucose oxidase and peroxidase. Methods Enzym. Anal. 3: 1205-15. Booth, S.J., Johnson, J.L., and Wilkins, T.D. 1977. Bacteriocin production by strains of Bacteroides isolated from human feces and the role of these strains in the bacterial ecology of the colon. Antimicrob. Agents Chemother. 11: 718-724. Brock, T.D., and Davie, J.M. 1963. Probable identity of a group D hemolysin with a bateriocin. J. Bacteriol. 86: 708-712. Dajani, A.S., Gray, E.D., and Wannamaker, L.W. 1970. Effect of bactericidal substance from Staphylococcus aureus on group A streptococci. I. Biochemical alterations. Infect. Immun. 1: 485-490. Gagliano, V.J., and Hinsdill, R.D. 1970. Characterization of a Streptococcus aureus bacteriocin. J. Bacteriol. 104: 117-125. Gomori, G. 1955. Preparation of buffers for use in enzyme studies. Methods Enzyrnol. 1: 138-146. Graaf, K., and Klaasen-Boor, P. 1977. Purification and characterization of a complex between cloacin and its immunity protein isolated from Enterobacter cloacae (Glo DF 13). J. Biochem. (Tokyo), 73: 107-114. Gratia, A. 1925. Sur un remarquable exemple d'antagonisme entre deux souches de colibacille. C.R. Hebd. Seances Acad. Sci. 140: 1032-1033. Green, A.A., and Hughes, W. 1955. Protein fractionation on the basis of solubility in aqueous solutions of salts and organic solvents. Methods Enzymol. 1: 67-90. Hammond, B.F., Lillard, S.E., and Stevens, R.H. 1987. A bacteriocin of Actinobacillus actinomycetemcomitans.Infect. Immun. 55: 686-691. Harkness, R.E., and Olschlager, T. 1991. The biology of colicin M. FEMS Microbiol. Rev. 88: 27-48. Holdeman, L.V., Catio, E.P., and Moore, W.E.C. (Editors).1977. Anaerobe laboratory manual. 4th ed. Virginia Polytechnic Institute and State University, Blacksburg. Jakes, K., and Zinder, N.D. 1974. Purification and properties of colicin E3 immunity protein. J. Biol. Chem. 249: 438-444. Kramer, J., and Brandis, H. 1972. Characterisierung eines Streptococcus agalactiae Bacteriocins. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1 Orig. Reihe A, 219: 290-301. Kutter, A.G. 1966. Production of bacteriocins by group A streptococci with special reference to the nephritogenic types. J. Exp. Med. 124: 279-291. Levinsohn R., Konisk, J., and Nomura, M. 1968. Interaction of colicins with bacterial cells. J. Bacteriol. 96: 811-821. Livingston, S.J., Komikos, S.D., and Yee, R.B. 1978. New medium for selection and presumptive identification of the Bacteroides fragilis group. J. Clin. Microbiol. 7: 448-453. Lowry, S.E., Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275. Lyon, W.J., and Glatz, B.A. 1991. Isolation and characterization of propiocinin PLG-1, bacteriocin produced by a strain of Propionibacterium thoenii. Appl. Environ. Microbiol. 57: 701-706.

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by Hunan Normal University on 06/05/13 For personal use only.

174

CAN. J . MICROBIOL. VOL. 39, 1993

Mossie, K.G., Jones, D.T., Robb, F.T., and Woods, D.R. 1979. Characterization and mode of action of a bacteriocin produced by a Bacteroides fragilis strain. Antimicrob. Agents Chemother. 16: 724-730. Nakamura, T., Kanagawa, N., Taniguchi, H., and Fujimura, S. 1984. Bacteriocin activity of the bacterium Bacterionema matruchotii isolated from dental plaque in man. Arch. Oral Biol. 29: 739-743. Parrot, M., Caufield, P.W., and Lavoie, M.C. 1990. Preliminary characterization of four bacteriocins from Streptococcus mutans. Can. J. Microbiol. 36: 123-130. Pugsley, A.P., and Oudega, B. 1988. Methods for studying colicins and their plasmids. In Plasmids: practical approach. Edited by K.G. Hardy. IRL Press, Oxford. pp. 106-107. Reeves, P. 1979. The concept of bacteriocins. Zentralbl. Bakteriol. Mikrobiol. Hyg. Abt. I Orig. A, 224: 78-89. Ryan, F.Q., Fried, P., and Mukai, F. 1955. A colicin produced by cells that are sensitive to it. Biochim. Biophys. Acta, 18: 131. Salyers, A.A. 1984. Bacteroides of human lower intestinal tract. Annu. Rev. Microbiol. 38: 293-313.

Southern, J.A., Katz, W., and Woods, D.R. 1984. Purification and properties of a cell bound bacteriocin from Bacteroides fragilis strain. Antimicrob. Agents Chemother. 25: 253-257. Tagg, J.R., Dajani, A.S., and Wannamaker, L.W. 1976. Bacteriocins of Gram-positive bacteria. Bacteriol. Rev. 40: 722-756. Tuiion, and Johansson, K.E. 1980. Yet another improved silver staining method for the detection of protein in PAGE. J. Biochem. Biophys. Methods, 9: 171-179. Turner, J.W., and Jordan, H.V. 1981. Bacteriocin-like activity within the genus Actinomyces. J. Dent. Res. 60: 1000-1007. Vidaver, A.K., Mathys, M.L., Thomas, M.E., and Schuster, M.L. 1972. Bacteriocins of the phytopathogens Pseudomonas syringae, P. glycinea, and P. phaseolicola. Can. J. Microbiol. 18: 705-713. Walstad, D.L., Reitz, R.C., and Sparling, P.F. 1974. Growth inhibition among strains of Neisseria gonorrheae due to production of inhibitory free fatty acids and lysophosphatidyl ethanolamine: absence of bacteriocins. Infect. Immun. 10: 481-488.