PLASMID
27,246-250
(1992)
Transposition CHARLES I. HOOVER,’
of Tn435 7 in Porphyromonas
EMILIYA ABARBARCHUK,
gingivalis
CLIFFORD Y. NG, AND JEFFREY
Department of Stomatology, 604 Health Sciences West, University San Francisco, California 94143-0512
R. FELTON
of California,
Received August 1, 199 1; revised December 27, 199 1 Genetic analysis of Porphyromonas gingivalis, an obligately anaerobic gram-negative bacterium, has been hindered by the apparent lack of naturally occurring bacteriophages, transposable elements, and plasmids. Plasmid R75 1: : *Q4 has previously been used as a suicide vector to demonstrate transposition of Tn4351 in B. uniformis. The erythromycin resistance gene on Tn4351 functions in Bacteroides and Porphyromonas. Erythromycin-resistant transconjugants were obtained at a mean frequency of 1.6 x lo-’ from matings between Escherichia co/i HB 10 1 containing R751 ::‘a4 and P. gingivalis 33277. Southern blot hybridization analysis indicated that about half of the erythromycin-resistant P. gingivalis transconjugants contained simple insertions of Tn4351 and half contained both Tn4351 and R75 1 sequences. The presence of R75 1 sequences in some P. gingivalis transconjugants most likely occurred from Tn4351-mediated cointegration of R7S 1, since we were unable to detect autonomous plasmid in these P. gingivalis transconjugants. The P. gingivalis-Tn4351 DNA junction fragments from different transconjugants varied in size. These results are consistent with transposition of Tn4351 and with insertion at several different locations in the P. gingivalis chromosome. Tn4351 may be useful as a mutagen to isolate well-defined mutants of P. gingivaiis. 0 1992 Academic Press. Inc.
Porphyromonas gingivalis (Shah and Collins, 1988) is an obligately anaerobic gramnegative coccobacillus which has been implicated asa periodontal pathogen. In vitro studies have suggestedthat various P. gingivalis gene products, including proteases, may function as virulence factors. P. gingivalis proteases have been shown, in vitro, to degrade numerous host components (for review see Mayrand and Holt, 1988). However, the importance of these proteasesin pathogenicity has not been determined in vivo. A powerful approach to assessthe role of P. gingivalis proteases in vivo would be to isolate mutants deficient in specific proteases and compare their pathogenicity with that of the wild-type parent strain in the ligature-induced periodontitis model (Holt et al., 1988). To construct P. gingivalis strains having only a single known mutation in a chosen gene, genetic tools such as transposons and shuttle vectors which function in P. gingiva’ To whom correspondence 0147-619X/92
should be addressed.
$5.00
Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved
lis are needed. Since no naturally occurring bacteriophages, transposable elements, or plasmids have been described in P. gingivalis (Dickinson et al., 1987; Sako et al., 1988) we investigated the ability of conjugal and transpositional systems from colonic Bacteroides species to function in P. gingivalis. Colonic Bacteroides-based systems were chosen for the following reasons: (a) Porphyromonas is phylogenetically more closely related to Bacteroides than to Escherichia coli (Paster et al., 1991; Woese, 1987), (b) RNA polymerase(s) from P. gingivalis and E. coli appear to have different subunit structures and to exhibit different DNA template specificities (Klimpel and Clark, 1990), and (c) no antibiotic resistance genesknown to function in E. coli also function in Bacteroides (Odelson et al., 1987; Salyers et al., 1987). R75 1::*04 (Shoemaker et al., 1986) is a conjugal broad-host-range IncP plasmid that contains a partial tandem duplication of Tn4351, a transposon originally found on B. fragilis plasmid pBF4. Since R75 1: : *Q4 can transfer itself from E. coli to Bacteroides but
246
SHORT COMMUNICATIONS
.. ..-
IS4351L
9 0.5 2
a
‘Tc’
..
erm F I!X?~IR
B 3.8 kb - Ava I fragment
qsmm
b
FIG. 1. Restriction endonuclease map of integrated Tn4351. In Tn4351, insertion elements (IS4351, and IS4351R)are indicated by the hatched areas and the ap proximate location of the erythromycin (err&) and aerobic tetracycline resistance (*Tc’) genesare shown (Rasmussen et al., 1987). AvaI digestion of integrated Tn4351 results in a 3.8-kb fragment and two junction fragments (a and b). AvuI digestion leaves 0.5 kb of IS4351L attached to junction fragment a and 0.6 kb of IS4351a attached to junction fragment b.
does not replicate in Bacteroides it servesas a suicide vector for delivery of Tn4352 into Bacteroides species.Tn4351 contains two antibiotic resistance genes flanked by two directly repeated insertion sequences (Fig. 1); one gene codes for erythromycin resistance which is expressed in Bacteroides but not in E. coli and the other gene codes for tetracycline resistance which is expressedin aerobically grown E. coli but not in anaerobically grown E. coli or Bacteroides (Shoemaker et al., 1985). In this report we demonstrate that R75 1: : *a4 can be transferred from E. coli to P. gingivalis by conjugation and that Tn4351 can transpose from R75 1: : *Q4 to the P. gingivalis chromosome. E. coli HB 101 containing R75 1: : *Q4 was kindly provided by Dr. N. B. Shoemaker (University of Illinois, Urbana, IL) and was maintained on Luria-Bertani (LB) agar containing 10 &ml tetracycline. P. gingivalis 33277 was obtained from the American Type Culture Collection (Rockville, MD) and was maintained on LRBB agar [Brucella agar base(Difco Laboratories, Detroit, MI) supplemented with 5% laked rabbit blood, 2.5 pg/ ml hemin, 5.0 &ml menadione, and 0.0 1% dithiothreitol (DTT)]. For conjugation experiments, donor cells (E. coli HBlO 1 con-
247
taming R75 1::*Q4) were grown aerobically at 37°C for 16-18 h in 5 ml LB broth containing 10 &ml tetracycline, and recipient cells (P. gingivalis) were sequentially cultured two times in sTSB [Trypticase soy broth (Becton-Dickinson Microbiology Systems, Cockeysville, MD) supplemented with 0.25% yeast extract, 2.5 &ml hemin, 5.0 pg/ ml menadione, and 0.0 1%DTT]. In brief, several P. gingivalis colonies from an LRBB agar plate (incubated anaerobically at 37°C for 3-5 days) were collected on a cotton swab, resuspended in 5 ml sTSB and incubated anaerobically overnight at 37°C. Two milliliters of the overnight broth culture was then inoculated into 50 ml of sTSB and incubated anaerobically at 37°C for 18-36 h (A,, = 0.8 to 1.2). Bacterial matings were performed on LRBB agar. Two milliliters of donor cell culture and 2 ml of recipient cell culture were harvested by centrifugation in a single microcentrifuge tube (conjugation mixture). The mixed-cell pellet was then resuspended in 200 ~1of sTSB and 100-~1portions were spotted on LRBB agar. The spotted-LRBB agar plates were incubated aerobically at 37 “C for 2 to 4 h to allow mating to occur. Following this brief aerobic incubation the plates were incubated anaerobically at 37°C for 36-48 h. Each cultured bacterial spot was then harvested with a cotton swab and resuspended in 1 ml of sTSB. Portions ( 100 ~1)of this suspension and of lo-fold serial dilutions were plated on LRBB agar containing 20 pg/ml erythromycin and 100 &ml gentamicin to select for transconjugants, since P. gingivalis is naturally resistant to gentamicin while E. coli is naturally sensitive to it. Portions (100 ~1)were also plated on LRBB agar containing 100 pg/ml gentamicin alone and incubated anaerobically, for 5 to 7 days, to enumerate viable recipient cells. Donor cells were enumerated on LRBB agar after aerobic incubation for 1 to 2 days. Black-pigmented colonies which grew on LRBB agar containing erythromycin and gentamicin were considered to be transconjugants. The frequency of erythromycin-resistant (Em’) transconjugants varied from 8.2 X lop7 to 9.8 X lop9
248
SHORT COMMUNICATIONS
(mean 1.6 X lo-’ for 10 independent matings). The frequency of P. gingivalis transconjugants obtained was approximately 1Ofold lower than that reported for B. uniformis (Shoemaker et al., 1986). Most matings were done with a donor-to-recipient ratio ranging from 0.25 to 0.5. Spots ofthe cell suspensions initially contained lo* to lo9 recipient cells. Increasing the donor-to-recipient ratio to 1.O adversely affected the viability of recipient cells and therefore reduced the number of transconjugants, whereas increasing the length of aerobic incubation from 4 to 24 h slightly increased the number of transconjugants (data not shown). Southern blot hybridization analysis was used to determine the location of Tn4352 in the Em’ transconjugants. DNA was isolated from wild type and from Em’ transconjugants of P. gingivalis 33277 using a guanidine-isothiocyanate technique (Lippke et al., 1987). Plasmid DNA was extracted from E. coli HBlOl (R75 1::*R4) by the NaCl-SDS procedure of Godson and Vapnek (1973) and purified by CsCl-ethidium bromide equilibrium density gradient centrifugation. Two micrograms of high molecular weight DNA from each strain was digested with 10 units of AvaI for 1 h in buffer supplied by the manufacturer (Stratagene, La Jolla, CA). Restriction fragments were electrophoretically separated in 1% agarose gels submerged in TBE buffer (0.089 M Tris, 0.089 M boric acid, 0.002 M EDTA, pH 8.0) and blotted onto nylon membranes. Purified R75 1:: *94 was biotinylated using the BioNick Labeling System (Life Technologies, Gaithersburg, MD) according to the manufacturer’s instructions and usedto probe the Southern blots. Hybridized biotinylated R75 1: : *94 was detected with the PhotoGene Nucleic Acid Detection System (Life Technologies) according to the manufacturer’s instructions, except that the hybridization stringency was increased by raising the temperature of the 0.1 X SSC, 1% SDSwash from 50 to 65°C. Digestion of chromosomal DNA with AvaI allowed easier visualization of both junction fragments from the transconjugants than digestion with EcoRI,
12345678
23 1 kb 9.9 66 43
23 21
FIG. 2. Southern blot hybridization analysis of AvaIdigested chromosomal DNA from five independent Em’ transconjugants of P. gingivdis 33277 (lanes 3-7). HindIII-digested biotinylated X DNA (lane 1) was used for molecular sizing (numbers at left). The blot was probed with biotinylated R75 1:: *fl4. Lane 2 contains AvaI-digested R75 1: : *fl4 and lane 8 contains AvaI-digestedDNA from P. gingivulis 33277 (recipient control). No hybridization occurred between R75 1:: *fl4 and DNA from the recipient control (lane 8). The 3.8 at the right indicates the position of the internal 3.8-kb AvaI fragment of Tn4351. Junction fragments between Tn4351 and the P. gingivulis chromosome are indicated by arrows. Hybridization patterns in lanes 3, 6, and 7 represent simple insertions of Tn4351 (junction fragments in lane 6 are apparent with prolonged exposure). Hybridization patterns in lanes 4 and 5 are consistent with Tn4351-mediated cointegration of R75 1 (only one putative junction fragment is distinguishable in each).
which has been used in previous studies with colonic Bacteroides.AvuI digestion leaves0.5 kb of IS4351, and 0.6 kb of IS4351, attached to the chromosomal junction fragments (Fig. 1). In contrast, EcoRI digestion leaves less than 10 basesof IS4351R attached to one of the chromosomal junction fragments; therefore, usually only one junction fragment is detectable. Ten independent Em’ transconjugants were analyzed for evidence of transposition of Tn4351. Results for five representative transconjugants are shown in Fig. 2. Simple insertions of Tn4351 should yield three hybridizing AvaI DNA restriction fragments;
249
SHORT COMMUNICATIONS
one 3.8-kb fragment corresponding to the 3.8-kb internal fragment of Tn4351 and two AvuI fragments of varying sizescorresponding to junction fragments between Tn4351 and P. gingivalis chromosomal DNA (Fig. 1). Three of the Em’ transconjugants (Fig. 2; lanes 3, 6, and 7) exhibited a hybridization pattern consistent with simple insertion of Tn4351. The fact that the junction fragments from these three transconjugants differed in sizeindicatesinsertion of Tn4351 at different siteswithin the P. gingivalis chromosome.As reported with colonic Bucteroides, approximately half of the Em’ transconjugantsof P. gingivulis contained simple insertions of Tn4351 and half contained both Tn4351 and R75 1 DNA sequences.Southern blot hybridization patterns of two transconjugants which contain R75 1 sequencesare shown in Fig. 2 (lanes4 and 5). One putative junction fragment was visible with each of these transconjugantsin addition to the 3.8-kb internal fragment of Tn4351 and multiple AvuI R751 fragments. Analysis of DNA extracts from these transconjugants by agarose gel electrophoresisand CsCl banding failed to detect the presenceof autonomous plasmid (data not shown). The Southern blot hybridization patterns of thesetwo transconjugants most likely reflect Tn4351-mediated cointegration of R751 into the P. gingivalis chromosome. This conclusion is consistent with the known ability of Tn-4351to mediatecointegration of R751 in B. uniformis (Shoemaker et al., 1986). Only one previous study hasreported conjugal transfer of DNA into P. gingivalis. Progulske-Fox et al. (1989) found that R75 1 could mobilize the transfer of pE5-2 from E. coli to P. gingivalis. pE5-2 (Shoemakeret al., 1985) is an E. coli/Bacteroides shuttle vector constructed from pB8-51 (a cryptic colonic Bacteroides plasmid), the 3.8-kb EcoRI-D fragment of pBF4 (contains most of Tn4351), and RSFlO 10(an IncQ E. coli plasmid). Em’ transconjugantsoccurred at a frequency of 2 X 10m7.Dot blot hybridizations of DNA from 19 Em’ transconjugants with probes for Tn4351 and RSFlOlO demon-
strated that most of the transconjugants( 14/ 19) contained both Tn4351 and RSFl 0 10 sequences. Although agarose gels were not shown, the authors indicated that DNA extracts from these transconjugants contained pE5-2. SincepE5-2 containsan origin ofreplication from pB8-51 they concluded that these transconjugants resulted from autonomous replication of pE5-2 in P. gingivalis. A few Em’ transconjugants (2/l 9) contained Tn4351 sequences but did not contain RSFlOlO sequencesor autonomous pE5-2. Although these results are consistent with transposition of Tn4351, the authors suggestedthat these transconjugants may have occurredby homologousrecombination. Presumably transposition was not considered possible,since the 3.8-kb EcoRI-D fragment of pBF4, usedin constructing pE5-2, is missing all but 9 bp of one insertion sequence (IS4351J as well as the distal inverted repeat of the remaining insertion sequence (IS435JR)of Tn4352. In the present study, the use of Southern blot hybridization analysis demonstrated transposition of Tn4351 in P. gingivalis. The varying sizesofjunction fragmentsfrom independent Em’ transconjugants indicated insertion of Tn4351 at several different locations in the P. gingivalis chromosome. Tn4351 may thereforebe usefulasa mutagen to isolate mutants of P. gingivalis deficient in specificgenefunctions. Recently we haveisolated several trypsin-like proteasedeficient mutants of P. gingivulis from matingswith E. coli HB 10I (R75 1: : *n4). Phenotypic and genetic characterization of these mutants is in progress. ACKNOWLEDGMENTS This study was supported by Grant R29 DE08207 Institutes of Health, National Institute of Dental Research. C. I. Hoover was supported in part by an NIH-NIDR Institutional National Research Service Award (lT32 DEO7204).
fromthe National
REFERENCES DICKINSON, D. P., RIVOLI, P. S., BREWSTER,J. M., AND
GENCO, R. J. (1987). Plasmids in black-pigmented
SHORT COMMUNICATIONS members of the genus Bacteroides. J. Dent. Res. 66, 223. [Abstract 9331 GODSON,G. N., AND VAP~K, D. (1973). A simple method of preparing large amounts of&X 174 RFI supercoiled DNA. B&him. Biophys. Acta 299, 516520. HOLT, S. C., EBERSOLE, J., FELTON,J., BRUNSVOLD,M., AND KORNMAN,K. S. (1988). Implantation of Bacteroides gingivalis in nonhuman primates initiates progression of periodontitis. Science 239,55-57. KLIMPEL, K. W., AND CLARK, V. L. (1990). The RNA polymerases of Porphyromonas gingivalis and Fusobacterium nucieatum are unrelated to the RNA polymerase of Escherichia coli. J. Dent. Res. 69, 15671572. LIPPKE,J. A., STRZEMPKO,M. N., RAIA, F. F., SIMON, S. L., AND FRENCH,C. K. (1987). Isolation of intact high-molecular weight DNA by using guanidine isothiocyanate. Appl. Environ. Microbial. 53,2588-2589. MAYRAND, D., AND HOLT, S. C. ( 1988). Biology of asaccharolytic black-pigmented Bacteroides species. Microbiol. Rev. 52, 134-152. ODELSON,D. A., RASMUSSEN,J. L., SMITH, C. J., AND MACRINA, F. L. (1987). Review: Extrachromosomal systems and gene transmission in anaerobic bacteria. Plasmid 17,87-109. PASTER,B. J., FRASER,G. J., DE~HIRST, F. E., AND OLSEN,I. (199 1). Phylogeny of Bacteroides and related bacteria. J. Dent. Res. 70, 318. [Abstract 4251 PROGULSKE-Fox, A., OBERSTE,A., DRUMMOND, C., AND MCARTHUR, W. P. (1989). Transfer of plasmid pE5-2 from Escherichia coli to Bacteroides gingivalis
and B. intermedius. Oral Microbial. Immunol. 4,132134. RASMUSSEN,J. L., ODELSON,D. A., AND MACRINA, F. C. (1987). Complete nucleotide sequence of insertion element IS4351 from Bacteroides fragihs. J. Bacteriol. 169, 3573-3580. SAKO, K., TAKAZOE, I., AND OKUDA, K. (1988). Isolation and characterization of plasmid DNA from Bacteroides strains isolated from the human oral cavity. Oral Microbial. Immunol. 3,72-76. SALYERS,A. A., SHOEMAKER,N. B., ANDGUTHRIE,E. P. (1987). Recent advances in Bacteroides genetics. In “Critical Reviews in Microbiology” (W. M. O’Leary, Ed.), Vol. 14, pp. 49-71. CRC Press,Cleveland, OH. SHAH, H. N., AND COLLINS,M. D. (1988). Proposal for reclassification of Bacteroides asaccharolyticus, Bacteroides gingivalis, and Bacteroides endodontalis in a new genus, Porphyromonas. Int. J. Syst. Bacterial. 38, 128-131. SHOEMAKER,N. B., GET~Y, C., GARDNER,J. F., AND SALYERS,A. A. (1986). Tn4351 transposes in Bacteroides spp. and mediates integration of plasmid R75 1 into the Bacteroides chromosome. J. Bacterial. 165, 929-936. SHOEMAKER,N. B., GUTHRIE, E. P., SALYERS,A. A., AND GARDNER,J. F. (1985). Evidence that the clindamycinerythromycin resistance gene of Bacteroides plasmid pBF4 is on a transposable element. J. Bacteriol. 162, 626-632. WOESE,C. R. (1987). Bacterial evolution. Microbial. Rev. 51,221-271. Communicated by Stuart B. Levy
27,246-250
(1992)
Transposition CHARLES I. HOOVER,’
of Tn435 7 in Porphyromonas
EMILIYA ABARBARCHUK,
gingivalis
CLIFFORD Y. NG, AND JEFFREY
Department of Stomatology, 604 Health Sciences West, University San Francisco, California 94143-0512
R. FELTON
of California,
Received August 1, 199 1; revised December 27, 199 1 Genetic analysis of Porphyromonas gingivalis, an obligately anaerobic gram-negative bacterium, has been hindered by the apparent lack of naturally occurring bacteriophages, transposable elements, and plasmids. Plasmid R75 1: : *Q4 has previously been used as a suicide vector to demonstrate transposition of Tn4351 in B. uniformis. The erythromycin resistance gene on Tn4351 functions in Bacteroides and Porphyromonas. Erythromycin-resistant transconjugants were obtained at a mean frequency of 1.6 x lo-’ from matings between Escherichia co/i HB 10 1 containing R751 ::‘a4 and P. gingivalis 33277. Southern blot hybridization analysis indicated that about half of the erythromycin-resistant P. gingivalis transconjugants contained simple insertions of Tn4351 and half contained both Tn4351 and R75 1 sequences. The presence of R75 1 sequences in some P. gingivalis transconjugants most likely occurred from Tn4351-mediated cointegration of R7S 1, since we were unable to detect autonomous plasmid in these P. gingivalis transconjugants. The P. gingivalis-Tn4351 DNA junction fragments from different transconjugants varied in size. These results are consistent with transposition of Tn4351 and with insertion at several different locations in the P. gingivalis chromosome. Tn4351 may be useful as a mutagen to isolate well-defined mutants of P. gingivaiis. 0 1992 Academic Press. Inc.
Porphyromonas gingivalis (Shah and Collins, 1988) is an obligately anaerobic gramnegative coccobacillus which has been implicated asa periodontal pathogen. In vitro studies have suggestedthat various P. gingivalis gene products, including proteases, may function as virulence factors. P. gingivalis proteases have been shown, in vitro, to degrade numerous host components (for review see Mayrand and Holt, 1988). However, the importance of these proteasesin pathogenicity has not been determined in vivo. A powerful approach to assessthe role of P. gingivalis proteases in vivo would be to isolate mutants deficient in specific proteases and compare their pathogenicity with that of the wild-type parent strain in the ligature-induced periodontitis model (Holt et al., 1988). To construct P. gingivalis strains having only a single known mutation in a chosen gene, genetic tools such as transposons and shuttle vectors which function in P. gingiva’ To whom correspondence 0147-619X/92
should be addressed.
$5.00
Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved
lis are needed. Since no naturally occurring bacteriophages, transposable elements, or plasmids have been described in P. gingivalis (Dickinson et al., 1987; Sako et al., 1988) we investigated the ability of conjugal and transpositional systems from colonic Bacteroides species to function in P. gingivalis. Colonic Bacteroides-based systems were chosen for the following reasons: (a) Porphyromonas is phylogenetically more closely related to Bacteroides than to Escherichia coli (Paster et al., 1991; Woese, 1987), (b) RNA polymerase(s) from P. gingivalis and E. coli appear to have different subunit structures and to exhibit different DNA template specificities (Klimpel and Clark, 1990), and (c) no antibiotic resistance genesknown to function in E. coli also function in Bacteroides (Odelson et al., 1987; Salyers et al., 1987). R75 1::*04 (Shoemaker et al., 1986) is a conjugal broad-host-range IncP plasmid that contains a partial tandem duplication of Tn4351, a transposon originally found on B. fragilis plasmid pBF4. Since R75 1: : *Q4 can transfer itself from E. coli to Bacteroides but
246
SHORT COMMUNICATIONS
.. ..-
IS4351L
9 0.5 2
a
‘Tc’
..
erm F I!X?~IR
B 3.8 kb - Ava I fragment
qsmm
b
FIG. 1. Restriction endonuclease map of integrated Tn4351. In Tn4351, insertion elements (IS4351, and IS4351R)are indicated by the hatched areas and the ap proximate location of the erythromycin (err&) and aerobic tetracycline resistance (*Tc’) genesare shown (Rasmussen et al., 1987). AvaI digestion of integrated Tn4351 results in a 3.8-kb fragment and two junction fragments (a and b). AvuI digestion leaves 0.5 kb of IS4351L attached to junction fragment a and 0.6 kb of IS4351a attached to junction fragment b.
does not replicate in Bacteroides it servesas a suicide vector for delivery of Tn4352 into Bacteroides species.Tn4351 contains two antibiotic resistance genes flanked by two directly repeated insertion sequences (Fig. 1); one gene codes for erythromycin resistance which is expressed in Bacteroides but not in E. coli and the other gene codes for tetracycline resistance which is expressedin aerobically grown E. coli but not in anaerobically grown E. coli or Bacteroides (Shoemaker et al., 1985). In this report we demonstrate that R75 1: : *a4 can be transferred from E. coli to P. gingivalis by conjugation and that Tn4351 can transpose from R75 1: : *Q4 to the P. gingivalis chromosome. E. coli HB 101 containing R75 1: : *Q4 was kindly provided by Dr. N. B. Shoemaker (University of Illinois, Urbana, IL) and was maintained on Luria-Bertani (LB) agar containing 10 &ml tetracycline. P. gingivalis 33277 was obtained from the American Type Culture Collection (Rockville, MD) and was maintained on LRBB agar [Brucella agar base(Difco Laboratories, Detroit, MI) supplemented with 5% laked rabbit blood, 2.5 pg/ ml hemin, 5.0 &ml menadione, and 0.0 1% dithiothreitol (DTT)]. For conjugation experiments, donor cells (E. coli HBlO 1 con-
247
taming R75 1::*Q4) were grown aerobically at 37°C for 16-18 h in 5 ml LB broth containing 10 &ml tetracycline, and recipient cells (P. gingivalis) were sequentially cultured two times in sTSB [Trypticase soy broth (Becton-Dickinson Microbiology Systems, Cockeysville, MD) supplemented with 0.25% yeast extract, 2.5 &ml hemin, 5.0 pg/ ml menadione, and 0.0 1%DTT]. In brief, several P. gingivalis colonies from an LRBB agar plate (incubated anaerobically at 37°C for 3-5 days) were collected on a cotton swab, resuspended in 5 ml sTSB and incubated anaerobically overnight at 37°C. Two milliliters of the overnight broth culture was then inoculated into 50 ml of sTSB and incubated anaerobically at 37°C for 18-36 h (A,, = 0.8 to 1.2). Bacterial matings were performed on LRBB agar. Two milliliters of donor cell culture and 2 ml of recipient cell culture were harvested by centrifugation in a single microcentrifuge tube (conjugation mixture). The mixed-cell pellet was then resuspended in 200 ~1of sTSB and 100-~1portions were spotted on LRBB agar. The spotted-LRBB agar plates were incubated aerobically at 37 “C for 2 to 4 h to allow mating to occur. Following this brief aerobic incubation the plates were incubated anaerobically at 37°C for 36-48 h. Each cultured bacterial spot was then harvested with a cotton swab and resuspended in 1 ml of sTSB. Portions ( 100 ~1)of this suspension and of lo-fold serial dilutions were plated on LRBB agar containing 20 pg/ml erythromycin and 100 &ml gentamicin to select for transconjugants, since P. gingivalis is naturally resistant to gentamicin while E. coli is naturally sensitive to it. Portions (100 ~1)were also plated on LRBB agar containing 100 pg/ml gentamicin alone and incubated anaerobically, for 5 to 7 days, to enumerate viable recipient cells. Donor cells were enumerated on LRBB agar after aerobic incubation for 1 to 2 days. Black-pigmented colonies which grew on LRBB agar containing erythromycin and gentamicin were considered to be transconjugants. The frequency of erythromycin-resistant (Em’) transconjugants varied from 8.2 X lop7 to 9.8 X lop9
248
SHORT COMMUNICATIONS
(mean 1.6 X lo-’ for 10 independent matings). The frequency of P. gingivalis transconjugants obtained was approximately 1Ofold lower than that reported for B. uniformis (Shoemaker et al., 1986). Most matings were done with a donor-to-recipient ratio ranging from 0.25 to 0.5. Spots ofthe cell suspensions initially contained lo* to lo9 recipient cells. Increasing the donor-to-recipient ratio to 1.O adversely affected the viability of recipient cells and therefore reduced the number of transconjugants, whereas increasing the length of aerobic incubation from 4 to 24 h slightly increased the number of transconjugants (data not shown). Southern blot hybridization analysis was used to determine the location of Tn4352 in the Em’ transconjugants. DNA was isolated from wild type and from Em’ transconjugants of P. gingivalis 33277 using a guanidine-isothiocyanate technique (Lippke et al., 1987). Plasmid DNA was extracted from E. coli HBlOl (R75 1::*R4) by the NaCl-SDS procedure of Godson and Vapnek (1973) and purified by CsCl-ethidium bromide equilibrium density gradient centrifugation. Two micrograms of high molecular weight DNA from each strain was digested with 10 units of AvaI for 1 h in buffer supplied by the manufacturer (Stratagene, La Jolla, CA). Restriction fragments were electrophoretically separated in 1% agarose gels submerged in TBE buffer (0.089 M Tris, 0.089 M boric acid, 0.002 M EDTA, pH 8.0) and blotted onto nylon membranes. Purified R75 1:: *94 was biotinylated using the BioNick Labeling System (Life Technologies, Gaithersburg, MD) according to the manufacturer’s instructions and usedto probe the Southern blots. Hybridized biotinylated R75 1: : *94 was detected with the PhotoGene Nucleic Acid Detection System (Life Technologies) according to the manufacturer’s instructions, except that the hybridization stringency was increased by raising the temperature of the 0.1 X SSC, 1% SDSwash from 50 to 65°C. Digestion of chromosomal DNA with AvaI allowed easier visualization of both junction fragments from the transconjugants than digestion with EcoRI,
12345678
23 1 kb 9.9 66 43
23 21
FIG. 2. Southern blot hybridization analysis of AvaIdigested chromosomal DNA from five independent Em’ transconjugants of P. gingivdis 33277 (lanes 3-7). HindIII-digested biotinylated X DNA (lane 1) was used for molecular sizing (numbers at left). The blot was probed with biotinylated R75 1:: *fl4. Lane 2 contains AvaI-digested R75 1: : *fl4 and lane 8 contains AvaI-digestedDNA from P. gingivulis 33277 (recipient control). No hybridization occurred between R75 1:: *fl4 and DNA from the recipient control (lane 8). The 3.8 at the right indicates the position of the internal 3.8-kb AvaI fragment of Tn4351. Junction fragments between Tn4351 and the P. gingivulis chromosome are indicated by arrows. Hybridization patterns in lanes 3, 6, and 7 represent simple insertions of Tn4351 (junction fragments in lane 6 are apparent with prolonged exposure). Hybridization patterns in lanes 4 and 5 are consistent with Tn4351-mediated cointegration of R75 1 (only one putative junction fragment is distinguishable in each).
which has been used in previous studies with colonic Bacteroides.AvuI digestion leaves0.5 kb of IS4351, and 0.6 kb of IS4351, attached to the chromosomal junction fragments (Fig. 1). In contrast, EcoRI digestion leaves less than 10 basesof IS4351R attached to one of the chromosomal junction fragments; therefore, usually only one junction fragment is detectable. Ten independent Em’ transconjugants were analyzed for evidence of transposition of Tn4351. Results for five representative transconjugants are shown in Fig. 2. Simple insertions of Tn4351 should yield three hybridizing AvaI DNA restriction fragments;
249
SHORT COMMUNICATIONS
one 3.8-kb fragment corresponding to the 3.8-kb internal fragment of Tn4351 and two AvuI fragments of varying sizescorresponding to junction fragments between Tn4351 and P. gingivalis chromosomal DNA (Fig. 1). Three of the Em’ transconjugants (Fig. 2; lanes 3, 6, and 7) exhibited a hybridization pattern consistent with simple insertion of Tn4351. The fact that the junction fragments from these three transconjugants differed in sizeindicatesinsertion of Tn4351 at different siteswithin the P. gingivalis chromosome.As reported with colonic Bucteroides, approximately half of the Em’ transconjugantsof P. gingivulis contained simple insertions of Tn4351 and half contained both Tn4351 and R75 1 DNA sequences.Southern blot hybridization patterns of two transconjugants which contain R75 1 sequencesare shown in Fig. 2 (lanes4 and 5). One putative junction fragment was visible with each of these transconjugantsin addition to the 3.8-kb internal fragment of Tn4351 and multiple AvuI R751 fragments. Analysis of DNA extracts from these transconjugants by agarose gel electrophoresisand CsCl banding failed to detect the presenceof autonomous plasmid (data not shown). The Southern blot hybridization patterns of thesetwo transconjugants most likely reflect Tn4351-mediated cointegration of R751 into the P. gingivalis chromosome. This conclusion is consistent with the known ability of Tn-4351to mediatecointegration of R751 in B. uniformis (Shoemaker et al., 1986). Only one previous study hasreported conjugal transfer of DNA into P. gingivalis. Progulske-Fox et al. (1989) found that R75 1 could mobilize the transfer of pE5-2 from E. coli to P. gingivalis. pE5-2 (Shoemakeret al., 1985) is an E. coli/Bacteroides shuttle vector constructed from pB8-51 (a cryptic colonic Bacteroides plasmid), the 3.8-kb EcoRI-D fragment of pBF4 (contains most of Tn4351), and RSFlO 10(an IncQ E. coli plasmid). Em’ transconjugantsoccurred at a frequency of 2 X 10m7.Dot blot hybridizations of DNA from 19 Em’ transconjugants with probes for Tn4351 and RSFlOlO demon-
strated that most of the transconjugants( 14/ 19) contained both Tn4351 and RSFl 0 10 sequences. Although agarose gels were not shown, the authors indicated that DNA extracts from these transconjugants contained pE5-2. SincepE5-2 containsan origin ofreplication from pB8-51 they concluded that these transconjugants resulted from autonomous replication of pE5-2 in P. gingivalis. A few Em’ transconjugants (2/l 9) contained Tn4351 sequences but did not contain RSFlOlO sequencesor autonomous pE5-2. Although these results are consistent with transposition of Tn4351, the authors suggestedthat these transconjugants may have occurredby homologousrecombination. Presumably transposition was not considered possible,since the 3.8-kb EcoRI-D fragment of pBF4, usedin constructing pE5-2, is missing all but 9 bp of one insertion sequence (IS4351J as well as the distal inverted repeat of the remaining insertion sequence (IS435JR)of Tn4352. In the present study, the use of Southern blot hybridization analysis demonstrated transposition of Tn4351 in P. gingivalis. The varying sizesofjunction fragmentsfrom independent Em’ transconjugants indicated insertion of Tn4351 at several different locations in the P. gingivalis chromosome. Tn4351 may thereforebe usefulasa mutagen to isolate mutants of P. gingivalis deficient in specificgenefunctions. Recently we haveisolated several trypsin-like proteasedeficient mutants of P. gingivulis from matingswith E. coli HB 10I (R75 1: : *n4). Phenotypic and genetic characterization of these mutants is in progress. ACKNOWLEDGMENTS This study was supported by Grant R29 DE08207 Institutes of Health, National Institute of Dental Research. C. I. Hoover was supported in part by an NIH-NIDR Institutional National Research Service Award (lT32 DEO7204).
fromthe National
REFERENCES DICKINSON, D. P., RIVOLI, P. S., BREWSTER,J. M., AND
GENCO, R. J. (1987). Plasmids in black-pigmented
SHORT COMMUNICATIONS members of the genus Bacteroides. J. Dent. Res. 66, 223. [Abstract 9331 GODSON,G. N., AND VAP~K, D. (1973). A simple method of preparing large amounts of&X 174 RFI supercoiled DNA. B&him. Biophys. Acta 299, 516520. HOLT, S. C., EBERSOLE, J., FELTON,J., BRUNSVOLD,M., AND KORNMAN,K. S. (1988). Implantation of Bacteroides gingivalis in nonhuman primates initiates progression of periodontitis. Science 239,55-57. KLIMPEL, K. W., AND CLARK, V. L. (1990). The RNA polymerases of Porphyromonas gingivalis and Fusobacterium nucieatum are unrelated to the RNA polymerase of Escherichia coli. J. Dent. Res. 69, 15671572. LIPPKE,J. A., STRZEMPKO,M. N., RAIA, F. F., SIMON, S. L., AND FRENCH,C. K. (1987). Isolation of intact high-molecular weight DNA by using guanidine isothiocyanate. Appl. Environ. Microbial. 53,2588-2589. MAYRAND, D., AND HOLT, S. C. ( 1988). Biology of asaccharolytic black-pigmented Bacteroides species. Microbiol. Rev. 52, 134-152. ODELSON,D. A., RASMUSSEN,J. L., SMITH, C. J., AND MACRINA, F. L. (1987). Review: Extrachromosomal systems and gene transmission in anaerobic bacteria. Plasmid 17,87-109. PASTER,B. J., FRASER,G. J., DE~HIRST, F. E., AND OLSEN,I. (199 1). Phylogeny of Bacteroides and related bacteria. J. Dent. Res. 70, 318. [Abstract 4251 PROGULSKE-Fox, A., OBERSTE,A., DRUMMOND, C., AND MCARTHUR, W. P. (1989). Transfer of plasmid pE5-2 from Escherichia coli to Bacteroides gingivalis
and B. intermedius. Oral Microbial. Immunol. 4,132134. RASMUSSEN,J. L., ODELSON,D. A., AND MACRINA, F. C. (1987). Complete nucleotide sequence of insertion element IS4351 from Bacteroides fragihs. J. Bacteriol. 169, 3573-3580. SAKO, K., TAKAZOE, I., AND OKUDA, K. (1988). Isolation and characterization of plasmid DNA from Bacteroides strains isolated from the human oral cavity. Oral Microbial. Immunol. 3,72-76. SALYERS,A. A., SHOEMAKER,N. B., ANDGUTHRIE,E. P. (1987). Recent advances in Bacteroides genetics. In “Critical Reviews in Microbiology” (W. M. O’Leary, Ed.), Vol. 14, pp. 49-71. CRC Press,Cleveland, OH. SHAH, H. N., AND COLLINS,M. D. (1988). Proposal for reclassification of Bacteroides asaccharolyticus, Bacteroides gingivalis, and Bacteroides endodontalis in a new genus, Porphyromonas. Int. J. Syst. Bacterial. 38, 128-131. SHOEMAKER,N. B., GET~Y, C., GARDNER,J. F., AND SALYERS,A. A. (1986). Tn4351 transposes in Bacteroides spp. and mediates integration of plasmid R75 1 into the Bacteroides chromosome. J. Bacterial. 165, 929-936. SHOEMAKER,N. B., GUTHRIE, E. P., SALYERS,A. A., AND GARDNER,J. F. (1985). Evidence that the clindamycinerythromycin resistance gene of Bacteroides plasmid pBF4 is on a transposable element. J. Bacteriol. 162, 626-632. WOESE,C. R. (1987). Bacterial evolution. Microbial. Rev. 51,221-271. Communicated by Stuart B. Levy
