INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Linear DNA elements have been described in various gram-positive and gram-negative genera (9). Some of the linear elements, have inverted repeats at their ends and proteins bound to their 5' ends. Such structural characteristics have been found in the linear elements of actinomycetes, bacteriophages, and viruses. This class of elements has been termed invertrons (18). A second class of bacterial linear plasmids having covalently closed ends at the termini of their DNAs has been found in the genus Borrelia (8). Linear plasmids were first described in Streptomyces sp. (7). Among the mycolic acid-containing actinomycetes, linear plasmids have been described in Rhodococcus opacus (11), Mycobacterium sp. (17), Rhodococcus fascians ( 1), and Rhodococcus erythropolis (2, 14).

Rhodococcus sp. strain RHA1 is a gram-positive polychlorinated biphenyl (PCB) degrader that efficiently transforms a wide range of PCB congeners (20, 21). Diverse degradative genes have been isolated from this strain, as shown in Table 1. These include three -subunit genes (bphA1, ORF1, and ORF3), three -subunit genes (bphA2, ORF2, and ORF4), two ferredoxin component genes (bphA3 and ORF5), and one ferredoxin reductase component gene (bphA4) of the ring-hydroxylating dioxygenases, two dihydrodiol dehydrogenase genes (bphB and bphB2), seven extradiol ring cleavage dioxygenase genes (bphC, bphC2, bphC3, bphC4, bphC5, bphC6, and etbC), three ring cleavage compound hydrolase genes (bphD, etbD1, and etbD2), one hydroxypentadienoate hydrolase gene (bphE), and one hydroxyoxovalerate aldolase gene (bphF). RHA1 contains three linear plasmids, pRHL1 (1,100 kb), pRHL2 (390 kb), and pRHL3 (280 kb), and the bphA1A2A3A4CB and bphDEF gene clusters have been found to be localized on pRHL1 and pRHL2, respectively (16). Some deletions in these plasmids have resulted in the loss of some degradative genes and in growth deficiencies on biphenyl (3).

In this study, to obtain insight into the involvement of the RHA1 linear plasmids in propagation and assembly of degradative genes, localization of the degradative genes on plasmids and transfer of linear plasmids were examined with a particular focus on pRHL2. We then characterized the termini of pRHL2 to address the structural features of this linear element.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Bacterial strains, plasmids, and culture conditions. The strains and plasmids used in this study are shown in Table 2. Rhodococcus strains were grown at 30°C in Luria-Bertani (LB) medium, diluted LB medium, or W minimal salt medium supplemented with 0.2% biphenyl or ethylbenzene vapor (15). RCA1 is a bphA1A2A3A4CB deletion mutant of RHA1. RHA1 mutant RDO5 lacks the pRHL2 plasmid and carries a kanamycin resistance gene in its chromosome. This strain was grown in the presence of 100 µg of kanamycin per ml.

Preparation and detection of linear plasmid DNA. Rhodococcus strains were grown in 10 ml of diluted LB medium to an optical density at 600 nm of 0.8. Cells were harvested by centrifugation at 2,000 × g for 10 min, washed twice with 0.4 M sucrose, and recentrifuged. Then they were resuspended in 0.5 ml of TES (0.3 M sucrose, 0.25 M EDTA, 0.25 M Tris-HCl; pH 8) and mixed with 1.4% low-melting-point SeaPlaque agarose (FMC, Rockland, Maine) in TBE (0.49 M Tris, 0.49 M boric acid, 0.001 M EDTA; pH 8) (12). The resulting mixture was pipetted into plug molds (Bio-Rad Laboratories, Richmond, Calif.). After incubation at 4°C for 15 min, the agarose plugs were pushed out of the molds into 5 ml of a lysozyme solution (1 mg/ml) in TES. After incubation at 37°C for 2 h with swaying, the plugs were transferred to 5 ml of NDS (0.5 M EDTA, 0.01 M Tris [pH 9], 1% [wt/vol] lauroyl sarcosine) containing proteinase K (1 g/ml) and incubated at 50°C for 20 to 40 h. The proteinase K solution was predigested at 50°C for 1 h. For preparation of non-proteinase K-treated plasmid DNA, proteinase K was omitted. To digest DNA with restriction enzymes, an agarose plug containing DNA was soaked in 200 µl of restriction enzyme buffer for 30 min. The buffer was replaced with 200 µl of fresh buffer, and 9 to 36 U of restriction enzymes was added. The plugs were incubated at 37°C for at least 20 h. For pulsed-field gel electrophoresis (PFGE), the plugs were inserted into the wells of agarose gels consisting of 1% agarose in TBE. Electrophoresis was performed with TBE as the running buffer at 14°C by using a CHEF DRII PFGE system (Bio-Rad). The voltage, pulse time, and total running time were varied according to the size range of the fragments to be separated. Specific conditions are provided in the legends to the figures. Saccharomyces cerevisiae YNN295 chromosomes (Bio-Rad), a bacteriophage lambda ladder (Bio-Rad), and the Kb DNA ladder (Stratagene, La Jolla, Calif.) were used as size markers. Sodium dodecyl sulfate (SDS)-PFGE analysis was carried out by adding SDS (final concentration, 0.2%) to both the buffer and the agarose gel (13). SDS-PFGE was performed for 23 h in a 200-V electric field by using pulse times ranging from 60 to 90 s. Conventional agarose gel electrophoresis was carried out with 1.5% agarose in TAE buffer (0.04 M Tris, 0.04 M acetate, 0.001 M EDTA).

Southern hybridization analysis. For gene assignment, DNA fragments separated by PFGE or conventional agarose gel electrophoresis were transferred to Hybond N nylon membrane filters (Amersham International plc, Buckinghamshire, United Kingdom). Southern hybridization was carried out by using the protocols provided by Amersham and a probe labeled with the digoxygenin (DIG) system (Boehringer Manheim Biochemicals, Indianapolis, Ind.). Probes were labeled as described in the DIG system manual. pRHL2 DNAs used for probes were recovered from PFGE gels by electroelution. A block of gel containing pRHL2 DNA was cut out and was enclosed in a dialysis bag filled with TBE. The dialysis bag was placed in TBE in the PFGE apparatus, and the DNA was eluted from the gel block by PFGE. PFGE was performed for 6 h by using the conditions described above for SDS-PFGE. The DNA was recovered from TBE in the dialysis bag by butyl alcohol concentration and ethyl alcohol precipitation. The DNA recovered was digested with a restriction enzyme and labeled with the DIG system.

Construction of subclones. Subclones of pRHL2 were selected from an RHA1 total DNA cosmid library constructed by using pK4HKcos (24). Screening of pRHL2 subclones was performed by colony hybridization using restriction fragments of pRHL2 as the probes. Colony hybridization was carried out by using the protocols of Amersham. The degradative gene and terminal fragment probes were employed to obtain subclones containing the corresponding genes or fragments.

Mating experiment. Mating was performed by using RCA1 as the donor and RDO5 as the recipient. Three milliliters of RDO5 cells grown in 10 ml of LB medium containing kanamycin and 3 ml of RCA1 cells grown in 10 ml of LB medium were mixed and transferred onto a nitrocellulose filter with a diameter of 25 mm and a pore size of 0.45 µm (Advantec, Tokyo, Japan) by filtration. The filter was placed on the surface of an LB medium plate and incubated at 30°C for 40 h. Then the cells were suspended in 1 ml of 0.9% NaCl, and 100-µl portions of appropriate dilutions were spread onto a W minimal salt medium agar plate containing 100 mg of kanamycin per liter. The plates were supplemented with biphenyl and incubated at 30°C for 2 days. The colonies formed on the plates were isolated as candidates for transconjugants and were subjected to plasmid DNA analysis by PFGE.

Cloning and sequencing of the terminal fragments. pRHL2 DNA was separated by PFGE and recovered by electroelution. It was digested with PstI and was ligated with pUC18 linearized with PstI and SmaI. Escherichia coli JM109 was transformed by this ligation mixture, and transformants were selected on LB agar plates containing 50 mg of ampicillin per liter, 2 mM isopropyl--D-thiogalactopyranoside, and 0.04% 5-bromo-4-chloro-3-indolyl--D-galactoside. The terminal fragments of pRHL2 were subcloned in pUC18 and pUC19 and were sequenced by the dideoxy termination method (19) using an ALF Express DNA sequencer (Pharmacia). Sequence analysis was carried out by using the GeneWorks program (IntelliGenetics Inc., Mountain View, Calif.) and the FASTA program provided by DDBJ.

Nucleotide sequence accession number. The nucleotide sequences of the pRHL2 left and right ends determined in this study have been deposited in the DDBJ, EMBL, and GenBank nucleotide sequence detabases under accession numbers AB048369 and AB048370, respectively.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Localization of degradative genes on the linear plasmids in Rhodococcus sp. strain RHA1. We have previously reported that Rhodococcus sp. strain RHA1 has three linear plasmids, pRHL1 (1,100 kb), pRHL2 (390 kb), and pRHL3 (280 kb) (16). In this study, the sizes of these linear plasmids were reevaluated by PFGE with improved conditions suitable for separating DNA fragments ranging from 200 to 500 kb long. The sizes of pRHL2 and pRHL3 were estimated to be 450 and 330 kb, respectively (Fig. 1). To localize the variety of degradative genes shown in Table 1, Southern hybridization analysis of the linear plasmids separated by PFGE was carried out using each gene probe. As a result, etbD1 was localized on pRHL1, which contained the bphA1A2A3A4CB genes, and bphC2, bphC4, and ORF3·4·5 etbD2 were localized on pRHL2, which included the etbC bphDEF genes (Table 1). It is assumed that the etb genes are involved in ethylbenzene degradation based on the profile of induction by ethylbenzene and the substrate specificity of their products (5, 6, 24). In contrast, bphC3, bphC5, and bphC6 hybridized at the position in the wells in which the chromosomal DNA should remain. These results indicated that degradative genes are widely distributed in the RHA1 genome, except for pRHL3.

View larger version (33K): [in this window] [in a new window]   FIG. 1.   PFGE of RHA1 linear plasmids performed with long (A) and short (B) pulse times. The positions of the chromosome and linear plasmids are indicated. (A) The pulse time was increased from 60 to 90 s during 23 h of electrophoresis. The voltage was adjusted to 6 V/cm. (B) The pulse time was increased from 20 to 30 s during 23 h of electrophoresis. The voltage was adjusted to 6 V/cm. Lane 1, S. cerevisiae chromosome marker, lane 2, total DNA of RHA1; lane 3,  ladder marker.

Physical map of pRHL2. We constructed restriction maps of pRHL2 using physical methods and hybridization analysis. Because the G+C content of Rhodococcus DNA is high, restriction enzymes AflII, AseI, and HpaI with AT-rich recognition sequences were employed to generate a manageable number of fragments from pRHL2. All the DNA fragments generated by restriction enzyme digestion were separated by PFGE (data not shown). Five HpaI fragments smaller than 6 kb were separated by conventional agarose gel electrophoresis (data not shown). The estimated sizes of the restriction fragments are shown in Table 3. The total size of pRHL2 was estimated to be approximately 450 kb.

View larger version (30K): [in this window] [in a new window]   FIG. 2.   Physical map of pRHL2 and localization of degradative genes. The locations of degradative genes are indicated by bars below the map. The insert region of each pRHL2 subclone is also indicated below the map.

Conjugal transfer of pRHL2. To investigate whether pRHL2 is self-transmissible, mating experiments were carried out using RHA1 mutant strains RCA1 and RDO5 as the donor and the recipient, respectively. RCA1 contained a derivative of pRHL1, pRHL1-1, which lacks the bphA1A2A3A4CB region. pRHL1-1 is 100 kb larger than pRHL1, suggesting that pRHL1-1 was generated by not only deletion of the bphACB region but also an unknown insertion. Strain RDO5 lacked all of pRHL2 and contained an insertion of the kanamycin resistance (Km^r) gene in its chromosome. RCA1 and RDO5 were not able to grow on biphenyl because of deficiencies in bphA1A2A3A4CB and bphD, respectively. Thus, only the derivative of RDO5 that received pRHL2 from RCA1 could grow on biphenyl in the minimum medium containing kanamycin. BP⁺ and Km^r colonies were obtained at a frequency of 7.5 × 10⁵ colony per recipient. Seven independently isolated transconjugants were examined in the presence of pRHL2 and the Km^r gene by PFGE (Fig. 3) and Southern hybridization analysis (data not shown), respectively. All the transconjugants tested possessed both pRHL2 and the Km^r gene, indicating that pRHL2 was transmitted from RCA1 to RDO5.

View larger version (69K): [in this window] [in a new window]   FIG. 3.   Linear plasmids in transconjugants. Linear plasmids of a donor (RCA1), a recipient (RDO5), and transconjugants were separated by PFGE. The pulse time was increased from 60 to 90 s during 23 h of electrophoresis. The voltage was adjusted to 6 V/cm. Lane 1, S. cerevisiae chromosome marker; lane 2, RCA1; lane 3, RDO5; lanes 4 to 10, RDO5 transconjugants.

Terminal structure of pRHL2. Blunt-ended terminal structures have been reported for linear plasmids in Rhodococcus (11) and actinomycetes (10). On the basis of the assumption that the ends of pRHL2 are blunt, we cloned the terminal fragments of pRHL2. pRHL2 DNA was isolated following separation by PFGE and was digested with PstI. The resulting fragments were ligated with pUC18 linearized with PstI and SmaI and used to transform E. coli JM109 cells. The transformants obtained harbored 4.5 and 4.6-kb plasmids that were designated pTPS3 and pTPS4, respectively. Southern hybridization analysis of pRHL2 fragments generated with AseI or HpaI was performed by using the inserts of pTPS3 and pTPS4 as probes. The insert of pTPS3 hybridized to A7 and H4, and the insert of pTPS4 hybridized to A5 and H8, indicating that pTPS3 and pTPS4 carry the right and left ends of pRHL2 shown in Fig. 2, respectively.

View larger version (32K): [in this window] [in a new window]   FIG. 4.   Alignment of the terminal sequences of pRHL2 and pHG201. The dashes indicate gaps; the asterisks indicate identical nucleotide residues in pRHL2 and pHG201. The 3-bp perfect terminal inverted repeats are indicated by boldface type. The inverted repeats of pHG201 containing the control motif GCTXCGC which were identified by Kalkus et al. (11) are indicated by converging arrows.

View larger version (33K): [in this window] [in a new window]   FIG. 5.   Southern hybridization of linear plasmids of RHA1 with the terminus probes of pRHL2. (A) Results of PFGE. The sizes of marker fragments and the positions of the chromosome and linear plasmids are indicated on the left and right, respectively. The pulse time was increased from 60 to 90 s during 23 h of electrophoresis. The voltage was adjusted to 6 V/cm. Lane 1, S. cerevisiae chromosome marker; lane 2, linear plasmids of RHA1. (B and C) Results of hybridization performed with the right (B) and left (C) end probes. The 1.1-kb EcoRI and 0.45-kb EcoRI-SalI fragments of pRHL2 were used as the right and left end probes, respectively. The lanes contained linear plasmids of RHA1. The bands of linear plasmids are indicated by arrowheads.

Association of proteins with the plasmid termini. Linear plasmids in actinomyces have been well-characterized. A general feature of linear replicons of actinomycetes is the presence of terminal proteins that covalently bind to the 5' end of the plasmid DNA (18). Kinashi and Shimaji-Murayama (13) have reported that an intact protein-bound linear plasmid of Streptomyces can move during PFGE in the presence of SDS but not in the absence of SDS. To prepare intact protein-bound plasmids, proteinase K was omitted from the lysis treatment used for the cells during plasmid DNA preparation in a gel plug. The resultant DNA was analyzed by PFGE in the absence or presence of SDS. Plasmid DNAs that were not treated with proteinase K remained at the origin of electrophoresis (Fig. 6A). In the presence of 0.2% SDS (Fig. 6B), the plasmid DNAs moved as far as the proteinase K-treated plasmid DNAs. These results suggest that the proteins did bind to these linear plasmids. To determine whether proteins were bound to the termini of pRHL2, proteinase K-treated and non-proteinase K-treated total DNAs were digested with AseI or HpaI in gel plugs. The resultant DNA fragments were separated by PFGE and were subjected to Southern hybridization analysis by using the terminal fragments of pRHL2 as probes. When the cells were lysed without proteinase K, the terminal fragment probes hybridized to the DNAs in the wells (Fig. 6C to E). When proteinase K was employed in cell lysis, the probes hybridized to the positions appropriate for the sizes of the A5, A7, H4, and H8 fragments. The right terminus probe also produced minor bands in the presence of SDS that seemed to originate from the termini of pRHL1 and pRHL3. These results suggest that each terminus of pRHL2 is linked to a protein. Based on the presence of blunt-ended termini and perfect terminal inverted repeats and the association between proteins and termini, we concluded that pRHL2 has invertron termini, as has been reported for linear replicons in Streptomyces (18).

View larger version (38K): [in this window] [in a new window]   FIG. 6.   (A and B) Normal PFGE (A) and SDS-PFGE (B) of linear plasmid DNAs obtained with and without proteinase K treatment. The pulse time was increased from 60 to 90 s during 23 h of electrophoresis. The voltage was adjusted to 6 V/cm. Lane 1, S. cerevisiae chromosome marker; lane 2, proteinase K-treated plasmid DNA; lane 3, non-proteinase K-treated plasmid DNA. The positions of the chromosome and plasmids are indicated on the left and right. (C to E) Results of PFGE (C) and hybridization (D and E) of total restriction fragments obtained with and without proteinase K treatment. Hybridization was performed with the right (D) and left (E) end probes. The 1.1-kb EcoRI and 0.45-kb EcoRI-SalI fragments of pRHL2 were used as the right and left end probes, respectively. The pulse time was increased from 3 to 12 s during 21 h of electrophoresis. The voltage was adjusted to 5.1 V/cm. The sizes of marker fragments in panel C are indicated on the left. The bands of interest in panels D and E are indicated by solid arrowheads. Minor bands in panel D which seemed to be derived from the termini of pRHL1 and pRHL3 are indicated by open arrowheads. Lane 1, non-proteinase K-treated AseI digests; lane 2, proteinase K-treated AseI digests; lane 3,  ladder plus HindIII-digested  DNA markers; lane 4, non-proteinase K-treated HpaI digests; lane 5, proteinase K-treated HpaI digests.