Differentiation of plasmids in marine diazotroph assemblages determined by randomly amplified polymorphic DNA analysis

Keri E. Beeson1, Deana L. Erdner1, Christopher E. Bagwell2, Charles R. Lovell2 and Patricia A. Sobecky1

School of Biology, Georgia Institute of Technology, 310 Ferst Drive, Atlanta, GA 30332-0230, USA1
Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, USA2

Author for correspondence: Patricia A. Sobecky. Tel: +1 404 894 5819. Fax: +1 404 894 0519. e-mail: patricia.sobecky{at}biology.gatech.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Nitrogen fixation by diazotrophic bacteria is a significant source of new nitrogen in salt marsh ecosystems. Recent studies have characterized the physiological and phylogenetic diversity of oxygen-utilizing diazotrophs isolated from the rhizoplanes of spatially separated intertidal macrophyte habitats. However, there is a paucity of information regarding the traits encoded by and the diversity of plasmids occurring in this key ecological functional group. Five-hundred and twenty-one isolates cultivated from the rhizoplanes of Juncus roemarianus, Spartina patens and different growth forms (short-form and tall-form) of Spartina alterniflora were screened for the presence of plasmids. One-hundred and thirty-four diazotrophs carrying plasmids that ranged in size from 2 to >100 kbp were identified. The majority of the marine bacteria contained one plasmid. Diazotrophs from the short-form S. alterniflora rhizoplane contained significantly fewer plasmids relative to isolates from tall-form S. alterniflora, J. roemarianus and S. patens. Although some plasmids exhibited homology to a nifH gene probe, the majority of the plasmids were classified as cryptic. Two oligonucleotide primers were developed to facilitate genotypic typing of the endogenously isolated marine plasmids by the randomly amplified polymorphic DNA (RAPD)-PCR technique. These primers proved to be more effective than 21 commercially available primers tested to generate RAPD-PCR patterns. Analysis of the RAPD-PCR patterns indicated as many as 71 different plasmid genotypes occurring in diazotroph bacterial assemblages within and between the four different salt marsh grass rhizoplane habitats investigated in this study.

Keywords: extrachromosomal element, rhizoplane, RAPD-PCR, endogenous isolation

Abbreviations: RAPD, randomly amplified polymorphic DNA


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
To date, there have been a limited number of studies attempting to characterize the molecular diversity of plasmid populations encountered in naturally occurring marine bacterial assemblages (Dahlberg et al., 1997 ; Sobecky et al., 1997 ). Plasmids isolated from marine bacterial communities, either by the exogenous ‘plasmid capture’ method (Bale et al., 1987 , 1988 ; Dahlberg et al., 1997 ) or obtained from culturable marine bacterial hosts (endogenous isolation), have been shown to exhibit broad host-ranges (Sobecky et al., 1998 ), to be mobilizable (Powers et al., 2000 ) and to have self-transfer capabilities (Sandaa & Enger, 1994 ; Dahlberg et al., 1998 ). However, plasmids in marine microbial communities are unrelated to the well characterized broad and narrow host-range incompatibility and replication (inc/rep) groups derived from plasmids found in the Enterobacteriaceae (Dahlberg et al., 1997 ; Sobecky et al., 1997 ). The lack of homology to the known inc/rep groups also extends to plasmids occurring in bacterial populations from freshwater (Osborn et al., 2000 ) and soils (Kobayashi & Bailey, 1994 ; Brønstad et al., 1996 ; van Elsas et al., 1998 ).

To gain a better understanding of bacterial gene flux mediated by plasmids in marine microbial communities, knowledge of the distribution, diversity and genes encoded by naturally occurring marine plasmids is necessary. The development of inc/rep probes and primers specific for plasmid replication origins (Sobecky et al., 1998 ; Cook et al., 2001 ) and RFLP comparisons (Dahlberg et al., 1998 ) are two approaches that have been used to examine the distribution, relationships and genetic diversity of marine plasmids. These methods can be labour-intensive, time-consuming and often require considerable amounts of purified plasmid DNA. Such factors can limit the applicability of these methods, making it difficult to design studies to address spatial and/or temporal changes in plasmid populations.

An objective of this study was to develop a method that would facilitate the comparison and differentiation of marine plasmids based on the randomly amplified polymorphic DNA (RAPD) method, a PCR-based procedure developed for identifying chromosomal polymorphisms (Welsh & McClelland, 1990 ; Williams et al., 1990 ). The RAPD-PCR-based technique was subsequently used to determine the molecular diversity of plasmid populations within marine diazotroph (nitrogen-fixing) bacterial assemblages from the rhizoplanes of salt marsh macrophytes Juncus roemarianus, Spartina alterniflora and Spartina patens (Bagwell et al., 1998 ; Bagwell & Lovell, 2000 ; Bergholz et al., 2001 ). These salt marsh macrophytes exist in discrete spatial and tidal zones in a coastal Atlantic salt marsh long-term ecological study site. In addition, unique polymorphic RAPD-PCR amplicons were investigated for their potential use as plasmid-specific DNA probes.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Site description and diazotroph characterization.
Juncus roemarianus (hereafter referred to as Juncus), Spartina patens and short-form (<=30 cm) and tall-form (>=1 m) Spartina alterniflora macrophytes were collected with root zone sediments intact from a salt marsh study site on Goat Island in the North Inlet estuary, near Georgetown, SC, USA (79°12'W, 33°20'N) as previously described (Bagwell et al., 1998 , 2001 ). At this low-elevation location (70 cm above mean sea level; Morris & Haskin, 1990 ) a band of tall-form S. alterniflora surrounded a small tidal creek, and short-form S. alterniflora extended inland for several metres from the edge of the tall-form Spartina zone. Mixed and pure stands of S. patens and Juncus occurred along the high marsh-terrestrial fringe border (Bergholz et al., 2001 ). The isolation, physiological characterization and nitrogen fixation activity of oxygen-tolerant heterotrophic diazotroph bacterial assemblages obtained from Juncus, S. alterniflora and S. patens roots and rhizomes have been described previously (Bagwell et al., 1998 , 2001 ; Bergholz et al., 2001 ).

Identification and phenotypic analysis of plasmid-containing isolates.
The alkaline lysis method used to determine the presence of plasmids in marine bacterial isolates has been described previously (Sobecky et al., 1997 ). Minor modifications to this method included a twofold increase in cell culture volume (10 ml) grown in Bacto Marine Broth 2216 (Difco) and a twofold decrease in lysis (Solution A) and denaturation (Solution B) buffer volumes. The supernatant was immediately electrophoresed in 0·65% agarose. Gels were run at 5 V cm-1 and stained with ethidium bromide. Plasmid incidence data were analysed for statistical significance using Statview (release 4.1). Data were subjected to the G-squared test for significance (P<=0·01). Escherichia coli strains containing plasmids pRR10 (5·0 kb), pRR54 (8·3 kb) (Roberts et al., 1990 ), pRK290 (20 kb) (Ditta et al., 1980 ), R6K (38 kb) (Stalker et al., 1979 ) and RK2 (60 kb) (Pansegrau et al., 1994 ), and Sinorhizobium meliloti 102F34 containing three plasmids of 100, 150 and 220 kb, respectively, were used as plasmid size standards.

Antibiotic resistance and heavy metal tolerance phenotypes were determined as described by Reyes et al. (1999) . Briefly, antibiotic-containing disks (Difco) were applied to lawns of marine bacterial isolates on half-strength YTSS agar (Sobecky et al., 1997 ). The antibiotics used were as follows: kanamycin (10 µg), neomycin (30 µg), tetracycline (30 µg) and trimethoprim (5 µg). Plates were scored after incubation for 18–24 h at 30 °C and resistance was determined according to the method of Bauer et al. (1966) . The heavy metal salts and their concentrations were as follows: 100 µg mercuric chloride ml-1, 570 µg cadmium chloride ml-1 and 1360 µg zinc chloride ml-1. Resistance to the metal salts was assayed according to Reyes et al. (1999) .

RAPD primers.
The 21 different PCR primers (Operon Technologies) synthesized for RAPD procedures and sequences of those primers generating RAPD profiles are given in Table 1. Two additional primers, designated DIAZO1 and MPRP2, were generated. The sequence of DIAZO1 was based on analysis of the sequence of the 536 kbp plasmid pNGR234a from Rhizobium sp. (Freiberg et al., 1997 ). The sequence of MPRP2 was selected based upon analysis of the complete 6·9 kbp pPS41 sequence (Powers et al., 2000 ) and partial sequences of four additional marine plasmids: p09022, p0329, p23023 and p0908 (P. Sobecky & J. Eisen, unpublished). The plasmid sequences were analysed using OligoCalc (Fislage et al., 1997 ) to calculate the most frequently occurring oligomers of a given length (6-mer and 8-mer).


View this table:
[in this window]
[in a new window]
 
Table 1. Sequences and amplification efficiencies of primers used in this study

 
RAPD-PCR amplifications.
PCR amplifications were performed in a 25 µl reaction volume containing 3·0 mM MgCl2, 250 µM each dNTP, 25 pmol primer, 1·25 U TaKaRa Taq (TaKaRa), 1x PCR amplification buffer (10 mM Tris/HCl, pH 8·3; 50 mM KCl) and 20–100 ng template DNA. Template DNA was obtained by excision of the plasmid DNA band directly from the agarose gel (SeaKem LE agarose; ISC BioExpress) using GeneCapsule (Geno Technology). The agarose plug containing plasmid DNA was melted at 92 °C for 20 min and 10 µl of the molten DNA/agarose mixture was used in PCR reactions. The reaction mixtures were incubated in a thermocycler (model 2400; Perkin Elmer) at 94 °C for 1·5 min and 55 °C for 3·5 min. Samples were then subjected to 30 cycles of 95 °C for 30 s, 35 °C for 3 min and 72 °C for 1 min, followed by a final extension for 10 min at 72 °C. Samples were mixed with 5 µl 6x loading buffer [0·25% (w/v) xylene cyanol FF, 40% (w/v) sucrose] (Maniatis et al., 1982 ) and heated at 55 °C for 15 min to facilitate loading and electrophoresis. PCR products were electrophoresed in 1·5% agarose. Gels were run at 5 V cm-1 for 2 h and stained with ethidium bromide. The mobilities of mass standards and RAPD amplicon bands were measured (EagleSight, Stratagene) and the sizes of all bands were also confirmed visually. To standardize between gel comparisons, relative mobilities of known mass standards were compared and normalized for each gel.

Plasmid RAPD-PCR data analysis.
Plasmid RAPD profiles were analysed by treating each RAPD-PCR product band on the agarose gel as an individual unit. The individual units were scored with either a 1, indicating the presence of the band in the profile, or a 0, indicating the absence of the band in the profile. A minimum of two to three independent RAPD-PCR reactions were performed on each individual plasmid excised from independent gel runs. RAPD amplicon bands were compared visually to confirm reproducibility of plasmid fingerprint patterns. Those amplicons that were either faint or absent from replicate reactions were omitted from subsequent analyses. The presence/absence dataset was analysed with RAPDistance, version 1.04 (Armstrong et al., 1996 ). The similarity between all plasmids was determined using the Jaccard coefficient, which calculates the proportion of positive bands shared by each sample pair (Sneath & Sokol, 1973 ). The Jaccard coefficient was applied to all combinations of plasmid pairs to generate a similarity matrix of pairwise comparisons. The RAPD-PCR profiles of those plasmid pairs with similarity coefficients between 0·1 and 0·9 were also confirmed by visual examination.

Southern hybridizations.
Specific RAPD-PCR amplicon bands were excised from the agarose gels, melted (92 °C, 20 min) and 5 µl used as PCR template. PCR amplification was performed in a 50 µl volume containing 1x PCR buffer (10 mM Tris/HCl, pH 8·3; 50 mM KCl), 1·5 mM MgCl2, 250 µM each dNTP, 25 pmol DIAZO1 or MPRP2 primer and 1·25 U TaKaRa Taq. Amplification was performed for 35 cycles under the following conditions: 95 °C for 1 min, 55 °C for 1 min, 72 °C for 1 min, with initial incubation at 95 °C and 55 °C for 2 min and final extension at 72 °C for 1 min. To verify that only one PCR product of the expected size was obtained, the PCR reaction was electrophoresed in 1% agarose. Purified products were used as template for PCR labelling with digoxigenen-11-dUTP according to the manufacturer’s instructions (Roche). Unincorporated nucleotides were removed with the QIAquick nucleotide removal kit (Qiagen). An 880 bp EcoRI fragment digested from pYMP1, containing the nifH nitrogenase gene from Klebsiella pneumoniae, was used as the source for the nifH DNA probe (Bagwell et al., 1998 ). Briefly, 350 ng pYMP1 DNA was used as template for a PCR labelling reaction (50 µl reaction volume) with digoxigenin-11-dUTP, according to the manufacturer’s instructions (Roche). PCR amplification was carried out with 25 pmol vector primer sequences (5'-GTTTTCCCAGTCACGAC-3') and (5'-AACAGCTATGACCATG-3') (Vieira & Messing, 1982 ). Digoxigenin-labelled PCR products were purified by the QIAquick PCR purification kit (Qiagen).

Following electrophoresis to screen for plasmid DNA, gels were blotted onto nylon membranes (Schleicher & Schuell), as described by Sobecky et al. (1997) . Southern hybridizations conducted with the DIG-labelled RAPD-PCR amplicon and nifH probes were essentially done as described in Reyes et al. (1999) . Pre-hybridization and hybridization were carried out at 65 °C. Lower stringency hybridization was also carried out at 40 °C with the nifH probe. DIG-labelled probes were added at 5–10 ng ml-1 for 16 h. Membranes were exposed to BioMax X-ray film (Kodak) at 37 °C. When necessary, bound DIG-labelled probe was removed from the membrane, according to the manufacturer’s recommendations (Roche).

16S rRNA gene amplification and RFLP analysis.
16S rRNA gene analysis by restriction digestion was performed on selected plasmid-containing isolates. Genomic DNA was obtained from isolates by lysis of one to three colonies suspended in 100 µl sterile dH2O (100 °C, 10 min). Cell debris was removed by centrifugation (8000 g, 15 min). An aliquot of the supernatant (1 µl) was used as template for 16S rRNA gene amplification. PCR amplification was performed as described by Sobecky et al. (1998) . Aliquots (1 µl) of PCR products were digested in separate restriction digestion reactions with 10 U MspI (Promega) and 20 U EcoRI (NEB) at 37 °C for 2 h. The digested products were separated by electrophoresis in 1·5% agarose containing ethidium bromide for 2·5 h at 5 V cm-1.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Incidence of plasmids in rhizoplane diazotroph assemblages
The endogenous plasmid isolation method was used to determine plasmid incidence in oxygen-utilizing heterotrophic diazotroph assemblages previously cultivated from the rhizoplanes of Juncus, S. patens and the short and tall growth forms of S. alterniflora (Bagwell et al., 1998 ; Bagwell & Lovell, 2000 ; Bergholz et al., 2001 ). A total of 521 diazotrophs were screened for plasmids (Table 2). There was no significant difference in the percentage of plasmid-containing bacterial isolates obtained from Juncus, tall-form S. alterniflora and S. patens (27, 30 and 28% respectively; Table 2). In contrast, diazotrophs from the rhizoplane of short-form S. alterniflora had a significantly lower plasmid percentage (12%; Table 2) relative to isolates from the other plant types (G-squared values of 7·71, 8·53 and 7·04 for Juncus, S. alterniflora and S. patens, respectively, and P values<0·01 for all three plant types).


View this table:
[in this window]
[in a new window]
 
Table 2. Plasmid incidence in marine diazotrophs isolated from salt marsh grass rhizoplanes

 
The 200 plasmids detected in the 134 plasmid-containing diazotrophs (Table 2) were compared to known plasmid size standards to obtain an estimation of plasmid mass. A broad range of plasmid sizes, 2 to >100 kbp, were observed (data not shown). Although the majority of the isolates contained single plasmid bands, numerous isolates contained multiple plasmid bands (Fig. 1). The method used for identifying a plasmid is based on its presence as a covalently closed circular DNA molecule. Linear plasmid DNA molecules would not be likely to be detected by the procedure used in this study. The diazotroph isolates have been previously identified by physiological and 16S rRNA phylogenetic analyses as belonging to the genera Aeromonas, Azotobacter, Enterobacter, Pseudomonas, Rhizobium and Vibrio, or to undescribed genera (Bagwell et al., 1998 ; Bagwell & Lovell, 2000 ).



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 1. Incidence of one, two and three or more plasmids detected in plasmid-containing diazotroph isolates from the rhizoplane of Juncus ({blacksquare}), short-form () and tall-form ({square}) S. alterniflora and S. patens ().

 
The plasmid-bearing and putative plasmid-free diazotroph isolates were analysed for antibiotic and heavy-metal-resistance phenotypes. While it is possible that the alkaline lysis procedure used to identify plasmids from marine diazotrophs may not have detected plasmids considerably greater in size than 400 kbp or of linear form, there does not appear to be any obvious correlation between plasmid content and antibiotic and heavy metal resistance (data not shown). Previously, Piceno et al. (1999) analysed diazotroph assemblage composition of nifH amplified from rhizoplane/rhizosphere sediment DNA. To determine if any of the plasmids encoded genes for nitrogen fixation activity, total DNA (chromosome and plasmid) of the 134 plasmid-bearing isolates (Table 2) was subjected to Southern hybridization with a nifH probe (Bagwell et al., 1998 ). Interestingly, only 8 of the 200 plasmids exhibited homology to nifH (data not shown). Lowering the stringency of hybridization (i.e. <75% homology) did not reveal any additional marine plasmids that had homology to the nifH probe used in this study nor to a negative control containing vector plasmid only (data not shown).

RAPD analysis of plasmids from rhizoplane diazotrophs
Plasmid DNA recovered from agarose gels served as a DNA template for the RAPD-PCR method used to characterize the genetic diversity of the endogenous plasmid populations (Fig. 2a). Short oligonucleotides have been shown to anneal to plasmid DNA from E. coli (Elaichouni et al., 1994 ). However, initial attempts to generate RAPD fingerprint profiles of plasmids from marine diazotrophs with 21 commercially available 10-mer primers resulted in either a low percentage of amplification (Table 1) or a complete lack of amplicons. Two additional 10-mer primers, designated MPRP2 and DIAZO1, were subsequently designed (Table 1). The majority of the 200 plasmids obtained from Juncus, short- and tall-form S. alterniflora and S. patens (66, 9, 37 and 39 plasmids, respectively) were subjected to PCR amplification with MPRP2 and DIAZO1. Amplification with primer MPRP2 was achieved for 57 out of 128 plasmids tested (45%; Table 1). Amplification with DIAZO1 resulted in discernable fingerprint profiles for 99 of the 151 plasmids tested (47, 7, 18 and 27 plasmids from Juncus, short- and tall-form S. alterniflora and S. patens, respectively; Table 1 and Fig. 2b). Forty-five of the 99 plasmids amplified with both DIAZO1 and MPRP2 primers (data not shown).



View larger version (56K):
[in this window]
[in a new window]
 
Fig. 2. Plasmid profiles of representative diazotroph isolates and corresponding RAPD-PCR fingerprint patterns. (a) Ethidium-bromide-stained 0·65% agarose gel of representative plasmids isolated from diazotrophs associated with the rhizoplane of short-form (lane 1) and tall-form (lane 2) S. alterniflora, S. patens (lanes 3, 4 and 6) and Juncus (lane 5). Multiple plasmid bands detected in diazotrophs from short-form S. alterniflora (1a, 1b) and S. patens (6a, 6b) were excised and amplified individually. The expected position of the chromosomal DNA band (chr) is indicated. (b) Corresponding RAPD-PCR fingerprint profiles of plasmid bands excised from the gel shown in (a) and amplified with the DIAZO1 primer. Lanes: 1, pSSC2-8a and pSSC2-8b (approx. 120 and 85 kbp, respectively, from short-form S. alterniflora); 2, pTM2-3 (approx. 85 kbp from tall-form S. alterniflora); 3 and 4, pPC2-2 and pPC2-3, respectively (approx. 60 kbp from S. patens); 5, pJC2-29 (approx. 5 kbp from Juncus); 6, pPG2-22a and pPG2-22b (approx. 65 and 35 kbp, respectively, from S. patens); 7, 200 bp molecular mass markers.

 
Amplification of plasmids with DIAZO1 and MPRP2 yielded consistently reproducible amplification patterns (Fig. 3). The reproducibility of the RAPD-PCR patterns produced by individual plasmids was confirmed by conducting two or more independent PCR assays from plasmid DNA excised from different gel runs (Fig. 3b and c, lanes 1 and 2). RAPD fingerprint profiles were obtained for plasmids ranging in size from approximately 3 to 120 kb from rhizoplane isolates cultivated from each of the four plant types (Fig. 2). We did not detect a correlation between plasmid size and amplification with DIAZO1. The number of amplicons in each RAPD profile (Fig. 2b) varied from 2 to 15 and ranged in size from 100 to 1500 bp, irrespective of plasmid size (i.e. small plasmids could have as many amplicons as large plasmids). These ranges have also been reported for chromosomal DNA (Harrison et al., 1992 ; Roberts & Crawford, 2000 ; Moschetti et al., 2001 ). In addition, it is possible to distinguish between individual plasmids by noting polymorphisms, apparent among plasmids obtained from different diazotroph strains (Fig. 2b). Plasmids co-occurring within the same isolate can also be amplified independently (Fig. 2b, lanes 1a, 1b and 6a, 6b). To verify that chromosomal DNA was not contributing to amplification products obtained from excised plasmid DNA, chromosomal DNA (Fig. 2a) was excised separately and amplified with DIAZO1 and MPRP2. When amplicons were obtained, chromosomal RAPD profiles were markedly different and distinguishable from plasmid DNA RAPD patterns (data not shown).



View larger version (47K):
[in this window]
[in a new window]
 
Fig. 3. Representative diazotroph plasmid from a S. patens isolate and corresponding RAPD fingerprint patterns demonstrating reproducibility of the RAPD-PCR method used in this study. (a) Ethidium-bromide-stained 0·65% agarose gel of an approximately 40 kbp plasmid designated pPC1-10. The expected position of the chromosomal DNA band (chr) is indicated. (b) Corresponding RAPD-PCR fingerprint patterns of plasmid band excised from the gel shown in (a) and amplified with primers DIAZO1 (lane 1) and MPRP2 (lane 2). Lane 3, 200 bp molecular mass markers. (c) RAPD-PCR was performed independently on a replicate alkaline lysis preparation of plasmid pPC1-10 excised from separate gel runs. RAPD-PCR patterns obtained with primers DIAZO1 (lane 1) and MPRP2 (lane 2). Lane 3, 200 bp molecular mass markers. The bracket indicates a representative example of amplicon bands that were omitted from subsequent Jaccard similarity calculations because of inconsistent or weak amplification.

 
DNA probes from RAPD amplicons
Genus- and strain-specific DNA probes have been derived from amplicons obtained from RAPD fingerprints of Streptomyces lydicus (Roberts & Crawford, 2000 ), Aeromonas hydrophila (Oakey et al., 1999 ) as well as virioplankton populations from marine water columns (Wommack et al., 1999 ). To determine if RAPD-PCR amplicons from marine plasmids could be used as a means of developing plasmid-specific DNA probes, polymorphic RAPD-PCR bands were selected for individual plasmids (data not shown). A DNA probe was generated from a 400 bp RAPD amplicon obtained with primer MPRP2 from an approximately 60 kbp plasmid. The plasmid was isolated from a diazotroph cultivated from S. patens (pPC2-3; Fig. 4a, lane 3). The probe hybridized only to the plasmid from which it was derived and to a plasmid isolated from another S. patens diazotroph with an identical RAPD profile (Fig. 4b, lanes 2 and 3). No cross-reactivity was observed between the pPC2-3 RAPD-derived probe and either chromosomal DNA or any marine plasmids (Fig. 4b, lanes 4–9) that had dissimilar RAPD profiles. Southern hybridization using polymorphic RAPD amplicons as DNA probes from additional marine plasmids tested confirmed the specificity of each individual probe for the plasmid from which it was derived and only those plasmids having identical RAPD patterns (data not shown).



View larger version (39K):
[in this window]
[in a new window]
 
Fig. 4. Plasmid profiles and Southern blot analysis test of homology with RAPD-PCR product. The DNA probe is a gel-purified 400 bp RAPD fragment obtained with primer MPRP2 amplified from the approximately 60 kbp plasmid pPC2-3 isolated from an S. patens rhizoplane diazotroph. (a) Ethidium-bromide-stained 0·65 % agarose gel. (b) Corresponding autoradiograph of Southern analysis of gel. Lanes: 1, E. coli TG1 containing 60 kbp RK2; 2 and 3, pPC2-2 (from an S. patens diazotroph) and pPC2-3, respectively; 4, pPG2-22 (from an S. patens diazotroph); 5 and 6, pTM1-13 and pTM2-3, respectively (from tall-form S. alterniflora diazotrophs); 7 and 8, pSSC2-7 and pSSC2-8, respectively (from short-form S. alterniflora diazotrophs); 9, pJC1-8 (from a Juncus diazotroph). Representative plasmids lack homology to the RAPD DNA probe generated with primer MPRP2 from pPC2-3. The expected position of the chromosomal DNA band (chr) is indicated.

 
Grouping of plasmids from diazotroph assemblages
The genetic similarity of the 99 plasmids amplified with the DIAZO1 primer (Table 1) was determined using the Jaccard coefficient. The Jaccard coefficient calculates the proportion of positive bands shared and unshared (i.e. polymorphic bands) by each sample pair (Sneath & Sokal, 1973 ). The Jaccard coefficient was applied to all combinations of plasmid pairs to generate a similarity (S) matrix of pairwise comparisons. There was an uneven distribution of S values, with the greatest number of plasmid pairs occurring at the high (S>0·9) and low (S<0·2) ends of the range and the least number of plasmid pairs at S values between 0·5 and 0·6 (Fig. 5). The plasmid pairs thus formed two distinct groups, designated for purposes of this study as either belonging to high (S>0·5) or low (S<0·5) similarity groups. These groupings were also confirmed by visual analyses of their RAPD profiles. The low-similarity group contained 4756 plasmid pairs (Fig. 5). Of these, 4596 pairs had S values of 0, i.e. the two plasmids in each pair had no RAPD amplicons in common. The remaining 160 plasmid pairs shared only one or two RAPD amplicons. The high-similarity group initially contained only 46 plasmid pairs (Fig. 5). Of these, 26 pairs had an S value of 1, i.e. the two plasmids in each pair had identical RAPD banding patterns. Of the other 20 pairs, three were removed from the high-similarity group because both plasmids in the pair produced too few amplicons (<=2 bands). The remaining 43 plasmid pairs in the high-similarity group shared between 3 and 11 RAPD amplicons each and contained between 0 and 4 polymorphic bands. The high-similarity group was further subdivided into groups of identical (S=1) and similar (0·5<S<0·9) plasmids.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 5. Distribution of pairwise similarity values between plasmids. The Jaccard coefficient of similarity was calculated using the RAPD-PCR banding patterns generated for the 99 plasmids. Pairwise comparison of the 99 plasmids (992) resulted in 9801 similarity coefficients; only the 4802 unique values are shown.

 
Molecular diversity of plasmids in diazotroph assemblages
Plasmids having identical RAPD profiles were detected within the same diazotroph host (Fig. 2b, lanes 1a and 1b) as well as in different hosts (Fig. 2b, lanes 3 and 4). A small number of diazotroph hosts (8, 1 and 1 isolates from Juncus, short- and tall-form S. alterniflora, respectively; data not shown) each contained two or three plasmids having identical RAPD profiles (e.g. Fig. 2b, lanes 1a and 1b). These plasmids are likely to be a nicked form of the same plasmid or multimers. Thus, they were considered to be one plasmid for subsequent analyses. However, in most cases RAPD profiles indicated the presence of different plasmids in those hosts containing multiple plasmid bands (e.g. Fig. 2b, lanes 6a and 6b). Specifically, 12 plasmids (6 pairs) obtained from different hosts isolated from S. patens, 6 plasmids (3 pairs) from tall-form S. alterniflora and 2 plasmids (1 pair) from short-form S. alterniflora had an S value of 1 (data not shown). Each plasmid was identical only to the other plasmid in the pair and not to any other plasmid. Those diazotroph hosts containing genetically identical plasmids were considered to be phylogenetically the same as determined by 16S rRNA gene RFLP analysis (data not shown).

Similar plasmids (0·5<S<0·9) were detected within the same bacterial host as well as in different hosts. In six instances, similar plasmids were found within the same host. Three diazotrophs from S. patens, two from Juncus and one from tall-form S. alterniflora each contained two plasmids with similar RAPD profiles. Two of these plasmid pairs, one from Juncus and one from S. patens, also showed similar RAPD patterns when tested with the MPRP2 primer (data not shown). Genetically similar plasmids in different bacterial hosts were only detected in isolates cultivated from Juncus and short-form S. alterniflora. Four plasmids from each plant type grouped into two distinct groups (data not shown).

Overall, plasmids belonging to the high-similarity group (0·5<S<1) were observed only within the same plant type. Specifically, plasmids isolated from Juncus (n=6 plasmids), S. patens (n=15 plasmids) and tall-form S. alterniflora (n=6 plasmids) isolates were similar to either one or two other plasmids within the same plant type (Fig. 6). However, one plasmid isolated from a Juncus diazotroph and all of the plasmids from short-form S. alterniflora diazotrophs were similar to three other plasmids (Fig. 6). Numerous plasmids from isolates cultivated from Juncus (n=30 plasmids), S. patens (n=14 plasmids) and tall- and short-form S. alterniflora (n=11 and 2 plasmids, respectively) rhizoplanes were similar to no other plasmids based on RAPD profiles (0<S<0·49; Figs 4 and 6). In most cases, dissimilar (genetically diverse) plasmids were isolated from phylogenetically different hosts. Analyses of all possible pairwise comparisons indicated that the 99 plasmid RAPD profiles divided into 71 groups.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 6. Number of plasmids isolated from diazotrophs associated with the rhizoplanes of Juncus, short- and tall-form S. alterniflora and S. patens that have similarity to either no other plasmid ({blacksquare}), one plasmid (), two plasmids ({square}) or three plasmids () based on RAPD-PCR fingerprint patterns.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The presence of plasmids in marine sediment and water-column bacterial populations is well documented (Kobori et al., 1984 ; Hermansson et al., 1987 ; Belliveau et al., 1991 ; Aviles et al., 1993 ; Dahlberg et al., 1997 ). In contrast to efforts to characterize plasmid populations from the rhizoplane of terrestrial plants (Lilley et al., 1996 ; van Elsas et al., 1998 ), few studies have characterized plasmid distribution and diversity in bacterial assemblages associated with the roots and rhizomes of salt marsh macrophytes. Plasmid incidence, with the exception of isolates from short-form S. alterniflora, was comparable for diazotrophs associated with Juncus, S. patens and tall-form S. alterniflora rhizoplane microenvironments. The growth zone of short-form Spartina is subject to interstitial porewater stagnation, low oxygen availability and above average hydrogen sulfide concentrations and salinity. These environmental stresses can reduce the efficiency of nitrogen uptake by Spartina (Bradley & Morris, 1990 ) and may act as selective forces on the incidence and composition of plasmids occurring in associated diazotrophs.

As the method of plasmid isolation used in this study is not likely to facilitate isolation of plasmids larger than 400 kbp, linear plasmids and plasmids resident in non-culturable diazotroph bacterial assemblages, plasmid incidence, as well as the occurrence of plasmid-encoded nitrogen fixation genes, may be underestimated. Attempts to assign particular phenotypic traits to plasmids from aquatic and terrestrial habitats (Zawadzki et al., 1996 ; Sobecky et al., 1997 ; van Elsas et al., 1998 ), as well as results from this study, suggest that many naturally occurring plasmids are cryptic. Some of these plasmids may in fact be advantageous to hosts occurring in either ‘localized’ niches or specific environmental conditions that vary temporally and spatially (Eberhard, 1990 ). For example, plasmid-encoded traits conferring niche specificity to salt marsh diazotrophs may include the ability to tolerate fluctuations in oxygen and salinity (Takeyama et al., 1991 ), oxidation or reduction of sulfur compounds (Zillig et al., 1985 ), enhanced growth capabilities in rhizoplane microenvironments (Moenne-Loccoz & Weaver, 1995 ) and catabolism of plant sugars and organic acids (Coplin, 1989 ).

Methods such as RFLP profiling and inc/rep typing have been used to classify the molecular diversity of plasmids isolated from marine (Dahlberg et al., 1997 ; Sobecky et al., 1997 ), soil (Drønen et al., 1999 ) and terrestrial habitats (Lilley et al., 1996 ). The genotypic typing of genomes by RAPD-PCR can reveal intra- and interspecific differences (Williams et al., 1990 ; Welsh & McClelland, 1990 ). RAPD analyses are increasingly being used to identify micro-organisms (Harrison et al., 1992 ; Moschetti et al., 2001 ) as well as determine clonal diversity between strains of the same bacterial species (Dye et al., 1995 ; Hyytia et al., 1999 ; Roberts & Crawford, 2000 ). In addition, RAPD-PCR has been used to compare microbial communities within aquatic (Franklin et al., 1999 ) and terrestrial (Picard et al., 2000 ) environments and to generate probes for the spatial and temporal comparison of marine virioplankton (Wommack et al., 1999 ). RAPD-PCR has been used on bacterial isolates with and without plasmids (Son et al., 1998 ). However, it has been noted that plasmid DNA needs to be extensively purified as the presence of contaminating chromosomal DNA appears to inhibit plasmid amplification (Elaichouni et al., 1994 ). We have found that chromosomal DNA does not contribute to amplification products obtained with plasmid DNA excised directly from agarose gels and further purification is not necessary. We did detect slight differences in the RAPD patterns obtained under the standard conditions used in this study (Fig. 3b and c). However, these differences were likely due to either the amount of PCR product loaded or to the amount of plasmid template excised from independent gel runs. In those infrequent instances (i.e. less than 5% of all plasmids tested), bands were omitted from subsequent Jaccard coefficient calculations (Fig. 3c).

Our results indicate that while a subset of commercially available oligomers produced either too few or no amplicons, it was feasible to design oligomers to generate RAPD DNA fingerprinting profiles for the majority of the plasmids isolated from marine diazotrophs tested in this study. The 10-mer primers DIAZO1 and MPRP2 have a G+C content of 70 mol%, which is expected to generate reproducible patterns and more amplification products (Caetano-Anollés et al., 1991 ) relative to primers with a lower G+C content (Williams et al., 1990 ; Harrison et al., 1992 ). The 3' ends of the RAPD primers were composed of hexamers that repeatedly occurred in the sequences of plasmids from marine bacteria and a Rhizobium sp. isolate (Freiberg et al., 1997 ). The four bases at the 5' end of the primers were chosen arbitrarily as it has been proposed that these bases are more tolerant of mismatch (Mori et al., 1999 ). We postulate that the 52 plasmids that did not yield an RAPD profile with either DIAZO1 or MPRR2 may lack these particular repetitive sequences. In this study, a limited number of sequences were used in RAPD primer design. Increasing the number of plasmid sequences analysed or obtaining additional DNA sequences of those plasmids from marine diazotrophs that did not amplify may identify other repetitive hexamers for future primer development.

Remarkably, all possible pairwise comparisons of the plasmids indicated that rather than clustering into a few or a small number of groups as we initially hypothesized, the 99 RAPD profiles divided into nearly as many structural groups as endogenous plasmids analysed (71 groups). These findings are in contrast to the 5–16 structural groups, as determined by RFLP, of exogenous HgR plasmids isolated from soils (Drønen et al., 1999 ), sugar beet phyllosphere/rhizoplane (Lilley et al., 1996 ) and marine (Dahlberg et al., 1997 ) habitats. Fifty-seven of the groups contained only one plasmid based on their unique RAPD patterns. The patterns of the remaining 14 groups indicated possible additions, deletions or structural variants of genetically similar plasmids (Lilley et al., 1996 ; Zhang et al., 2001 ). While it has been suggested that fingerprinting by RFLP may overestimate plasmid diversity (Dahlberg et al., 1997 ), we believe that the use of RAPD-PCR coupled with subsequent DNA probe generation provides a robust and rapid approach for differentiating plasmid variation and diversity in natural microbial communities.

To the best of our knowledge, this study represents the first report of RAPD-PCR genotypic typing of plasmids from marine environments. Studies are under way to determine the applicability of RAPD-PCR in monitoring the evolution of plasmids due to recombination and rearrangement events (Arber, 2000 ). Future studies will also address physiochemical environmental changes associated with seasonality and other factors related to geographic locations that may be important in the determination of plasmid composition and structure in marine microbial assemblages.


   ACKNOWLEDGEMENTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
This work was supported by the Office of Naval Research Grant N00014-98-1-0078 and NOAA Office of Sea Grant RR100-289/2000607 to P.A.S. We thank J. Felsenstein for helpful suggestions on data analysis, J. Eisen and The Institute for Genomic Research for providing plasmid sequences.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Arber, W. (2000). Genetic variation: molecular mechanisms and impact on microbial evolution. FEMS Microbiol Rev 24, 1-7.[Medline]

Armstrong, J., Gibbs, A., Peakall, R. & Weiller, G. (1996). RAPDistance programs; version 1.04 (http://life.anu.edu.au/molecular/software/rapd.html).

Aviles, M., Codina, J. C., Perez-Garcia, A., Cazorla, F., Romero, P. & de Vicente, A. (1993). Occurrence of resistance to antibiotics and metals and of plasmids in bacterial strains isolated from marine environments. Water Sci Technol 27, 475-478.

Bagwell, C. E. & Lovell, C. R. (2000). Microdiversity of culturable diazotrophs from the rhizoplanes of the salt marsh grasses Spartina alterniflora and Juncus roemerianus. Microb Ecol 39, 128-136.[Medline]

Bagwell, C. E., Piceno, Y. M., Ashburne-Lucas, A. & Lovell, C. R. (1998). Physiological diversity of the rhizosphere diazotroph assemblages of selected salt marsh grasses. Appl Environ Microbiol 64, 4276-4282.[Abstract/Free Full Text]

Bagwell, C. E., Dantzler, M., Bergholz, P. W. & Lovell, C. R. (2001). Host-specific ecotype diversity of rhizoplane diazotrophs of the perennial glasswort Salicornia virginica and selected salt marsh grasses. Aquat Micro Ecol 23, 293-300.

Bale, M. J., Fry, J. C. & Day, M. J. (1987). Plasmid transfer between strains of Pseudomonas aeruginosa on membrane filters attached to river stones. J Gen Microbiol 133, 3099-3107.[Medline]

Bale, M. J., Fry, J. C. & Day, M. J. (1988). Transfer and occurrence of large mercury resistance plasmids in river epilithon. Appl Environ Microbiol 54, 972-978.

Bauer, A. W., Kirby, M. M., Sherris, J. C. & Turck, M. (1966). Antibiotic susceptibility testing by a standard single disk method. Am J Clin Pathol 45, 493-496.[Medline]

Belliveau, B. H., Starodub, M. E. & Trevors, J. T. (1991). Occurrence of antibiotic and metal resistance and plasmids in Bacillus strains isolated from marine sediment. Can J Microbiol 37, 513-520.[Medline]

Bergholz, P. W., Bagwell, C. E. & Lovell, C. R. (2001). Physiological diversity of rhizoplane diazotrophs of the saltmeadow cordgrass, Spartina patens: implications for host specific ecotypes. Microb Ecol 42 (in press).

Bradley, P. M. & Morris, J. T. (1990). Influence of oxygen and sulfide concentration on nitrogen uptake kinetics in Spartina alterniflora. Ecology 71, 282-287.

Brønstad, K., Drønen, K., Ovreaas, L. & Torsvik, V. L. (1996). Phenotypic diversity and antibiotic resistance in soil bacterial communities. J Ind Microbiol Technol 17, 253-259.

Caetano-Anollés, G., Bassam, B. J. & Gresshoff, P. M. (1991). DNA amplification fingerprinting using very short arbitrary oligonucleotide primers. Bio-Technology 9, 553-557.[Medline]

Cook, M. A., Osborn, A. M., Bettandorff, J. & Sobecky, P. A. (2001). Endogenous isolation of replicon probes for assessing plasmid ecology of marine sediment microbial communities. Microbiology 147, 2089-2101.[Abstract/Free Full Text]

Coplin, D. L. (1989). Plasmids and their role in the evolution of plant pathogenic bacteria. Annu Rev Phytopathol 27, 187-212.

Dahlberg, C., Linberg, C., Torsvik, V. L. & Hermansson, M. (1997). Conjugative plasmids isolated from bacteria in marine environments show various degrees of homology to each other and are not closely related to well-characterized plasmids. Appl Environ Microbiol 63, 4692-4697.[Abstract]

Dahlberg, C., Bergstrom, M. & Hermansson, M. (1998). In situ detection of high levels of horizontal plasmid transfer in marine bacterial communities. Appl Environ Microbiol 64, 2670-2675.[Abstract/Free Full Text]

Ditta, G., Stanfield, S., Corbin, D. & Helinski, D. R. (1980). Broad host range DNA cloning system for gram-negative bacteria: construction of a gene bank of Rhizobium meliloti. Proc Natl Acad Sci USA 77, 7347-7351.[Abstract]

Drønen, A. K., Torsvik, V. L. & Top, E. M. (1999). Comparison of the plasmid types obtained by two distantly related recipients in bi-parental exogenous plasmid isolations from soil. FEMS Microbiol Lett 176, 105-110.

Dye, M., Skøt, L., Mytton, L. R., Harrison, S. P., Dooley, J. J. & Cresswell, A. (1995). A study of Rhizobium leguminosarum biovar trifolii populations from soil extracts using randomly amplified polymorphic DNA profiles. Can J Microbiol 41, 336-344.

Eberhard, W. G. (1990). Evolution in bacterial plasmids and levels of selection. Q Rev Biol 65, 3-22.[Medline]

Elaichouni, A., van Emmelo, J., Claeys, G., Vershraegen, G., Verhelst, R. & Vaneechoutte, M. (1994). Study of the influence of plasmids on the arbitrary primer polymerase chain reaction fingerprint of Escherichia coli strains. FEMS Microbiol Lett 115, 335-340.[Medline]

van Elsas, J. D., Gardener, B. B. M., Wolters, A. C. & Smit, E. (1998). Isolation, characterization, and transfer of cryptic gene-mobilizing plasmids in the wheat rhizosphere. Appl Environ Microbiol 64, 880-889.[Abstract/Free Full Text]

Fislage, R., Berceanu, M., Humboldt, Y., Wendt, M. & Oberender, H. (1997). Primer design for a prokaryotic differential display RT-PCR. Nucleic Acids Res 25, 1830-1835.[Abstract/Free Full Text]

Franklin, R. B., Taylor, D. R. & Mills, A. L. (1999). Characterization of microbial communities using randomly amplified polymorphic DNA (RAPD). J Microbiol Methods 35, 225-235.[Medline]

Freiberg, C., Fellay, R., Bairoch, A., Broughton, W. J., Rosenthal, A. & Perret, X. (1997). Molecular basis of symbiosis between Rhizobium and legumes. Nature 387, 394-401.[Medline]

Harrison, S. P., Mytton, L. R., Skøt, L., Dye, M. & Cresswell, A. (1992). Characterization of Rhizobium isolates by amplification of DNA polymorphisms using random primers. Can J Microbiol 38, 1009-1015.[Medline]

Hermansson, M., Jones, G. W. & Kjelleberg, S. (1987). Frequency of antibiotic and heavy metal resistance, pigmentation, and plasmids in bacteria of the marine air–water interface. Appl Environ Microbiol 53, 2338-2342.[Medline]

Hyytia, E., Hielm, S., Bjorkroth, J. & Korkeala, H. (1999). Biodiversity of Clostridium botulinum type E strains isolated from fish and fishery products. Appl Environ Microbiol 65, 2057-2064.[Abstract/Free Full Text]

Kobayashi, N. & Bailey, M. J. (1994). Plasmids isolated from the sugar beet phyllosphere show little or no homology to molecular probes currently available for plasmid typing. Microbiology 140, 289-296.[Abstract]

Kobori, H., Sullivan, C. W. & Shizuya, H. (1984). Bacterial plasmids in Antarctic natural assemblages. Appl Environ Microbiol 48, 515-518.

Lilley, A. K., Bailey, M. J., Day, M. J. & Fry, J. C. (1996). Diversity of mercury resistance plasmids obtained by exogenous isolation from the bacteria of sugar beet in three successive years. FEMS Microbiol Ecol 20, 211-227.

Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982). Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Moenne-Loccoz, Y. & Weaver, R. W. (1995). Plasmids influence growth of rhizobia in the rhizosphere of clover. Soil Biol Biochem 27, 1001-1004.

Mori, E., Lio, P., Daly, S., Damiani, G., Perito, B. & Fani, R. (1999). Molecular nature of RAPD markers from Haemophilus influenzae Rd genome. Res Microbiol 150, 83-93.[Medline]

Morris, J. T. & Haskin, B. (1990). A 5-yr record of aerial primary production and stand characteristics of Spartina alterniflora. Ecology 71, 2209-2217.

Moschetti, G., Blaiotta, G., Villani, F., Coppola, S. & Parente, E. (2001). Comparison of statistical methods for identification of Streptococcus thermophilus, Enterococcus faecalis, and Entercoccus faecium from randomly amplified polymorphic DNA patterns. Appl Environ Microbiol 67, 2156-2166.[Abstract/Free Full Text]

Oakey, H. J., Gibson, L. F. & George, A. M. (1999). DNA probes specific for Aeromonas hydrophila (HG1). J Appl Microbiol 86, 187-193.[Medline]

Osborn, A. M., Pickup, R. W. & Saunders, J. R. (2000). Development and application of molecular tools in the study of IncN-related plasmids from lakewater sediments. FEMS Microbiol Lett 186, 203-208.[Medline]

Pansegrau, W., Lanka, E., Barth, P. T. & 7 other authors (1994). Complete nucleotide-sequence of Birmingham IncP-{alpha} plasmids – compilation and comparative analysis. J Mol Biol 239, 623–663.[Medline]

Picard, C., Di Cello, F., Ventura, M., Fani, R. & Guckert, A. (2000). Frequency and biodiversity of 2,4-diacetylphloroglucinol-producing bacteria isolated from the maize rhizosphere at different stages of plant growth. Appl Environ Microbiol 66, 948-955.[Abstract/Free Full Text]

Piceno, Y. M., Noble, P. A. & Lovell, C. R. (1999). Spatial and temporal assessment of diazotroph assemblage composition in vegetated salt marsh sediments using denaturing gradient gel electrophoresis analysis. Microb Ecol 38, 157-167.[Medline]

Powers, L. G., Mallonee, J. T. & Sobecky, P. A. (2000). Complete nucleotide sequence of a cryptic plasmid from the marine bacterium Vibrio splendidus and identification of open reading frames. Plasmid 43, 99-102.[Medline]

Reyes, N. S., Frischer, M. E. & Sobecky, P. A. (1999). Characterization of mercury resistance mechanisms in marine sediment microbial communities. FEMS Microbiol Ecol 30, 273-284.[Medline]

Roberts, M. A. & Crawford, D. L. (2000). Use of randomly amplified polymorphic DNA as a means of developing genus- and strain-specific Streptomyces DNA probes. Appl Environ Microbiol 66, 2555-2564.[Abstract/Free Full Text]

Roberts, R. C., Burioni, R. & Helinski, D. R. (1990). Genetic characterization of the stabilizing functions of a region of broad-host-range plasmid RK2. J Bacteriol 172, 6204-6216.[Medline]

Sandaa, R. A. & Enger, O. (1994). Transfer in marine sediments of the naturally occurring plasmid pRAS1 encoding multiple antibiotic resistance. Appl Environ Microbiol 60, 4234-4238.[Abstract]

Sneath, P. H. & Sokal, R. R. (1973). Numerical Taxonomy. San Francisco: W. H. Freeman.

Sobecky, P. A., Mincer, T. J., Chang, M. C. & Helinski, D. R. (1997). Plasmids isolated from marine sediment microbial communities contain replication and incompatibility regions unrelated to those of known plasmid groups. Appl Environ Microbiol 63, 888-895.[Abstract]

Sobecky, P. A., Mincer, T. J., Chang, M. C., Toukdarian, A. & Helinski, D. R. (1998). Isolation of broad-host-range replicons from marine sediment bacteria. Appl Environ Microbiol 64, 2822-2830.[Abstract/Free Full Text]

Son, R., Nasreldin, E. H., Zaiton, H., Samuel, L., Rusul, G. & Nimita, F. (1998). Use of randomly amplified polymorphic DNA analysis to differentiate isolates of Vibrio parahaemolyticus from cockles (Anadara granosa). World J Microbiol Biotechnol 14, 895-901.

Stalker, D. M., Kolter, R. & Helinski, D. R. (1979). Nucleotide sequence of the region of an origin of replication of the antibiotic resistance plasmid R6K. Proc Natl Acad Sci USA 76, 1150-1154.[Abstract]

Takeyama, H., Burgess, J. G., Sudo, H., Sode, K. & Matsunaga, T. (1991). Salinity-dependent copy number increase of a marine cyanobacterial endogenous plasmid. FEMS Microbiol Lett 90, 95-98.

Vieira, J. & Messing, J. (1982). The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19, 259-268.[Medline]

Welsh, J. & McClelland, M. (1990). Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res 18, 7213-7218.[Abstract]

Williams, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A. & Tingey, S. V. (1990). DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18, 6531-6535.[Abstract]

Wommack, K. E., Ravel, J., Hill, R. T. & Colwell, R. R. (1999). Hybridization analysis of Chesapeake Bay virioplankton. Appl Environ Microbiol 65, 241-250.[Abstract/Free Full Text]

Zawadzki, P., Riley, M. A. & Cohan, F. M. (1996). Homology among nearly all plasmids infecting three Bacillus species. J Bacteriol 178, 191-198.[Abstract]

Zhang, X., Kosier, B. & Priefer, U. B. (2001). Symbiotic plasmid rearrangement in Rhizobium leguminosarum bv. viciae VF39SM. J Bacteriol 183, 2141-2144.[Abstract/Free Full Text]

Zillig, W., Yeats, S., Holz, I., Bock, A., Gropp, E., Rettenberger, M. & Lutz, S. (1985). Plasmid-related anaerobic autotrophy of the novel archaebacterium Sulfolobus ambivalens. Nature 313, 789-791.[Medline]

Received 6 August 2001; revised 3 September 2001; accepted 7 September 2001.