CHROMOSOMAL MULTIPLICITY IN BURKHOLDERIA CEPACIA

Thomas G. Lessiea and Brendan D. Manningb.

U.S. EPA Environmental Research Laboratory, Gulf Breeze, FL 32561; a Present address: Department of Microbiology, the University of Massachusetts, Amherst, MA 01003; and b Present address: Department of Biology, Yale University, New Haven, CT 06520

SUMMARY

We have used CHEF gel electrophoresis to screen preparations of large DNA from different Burkholderia cepacia isolates for the presence of DNA species corresponding to the linearized forms of the three chromosomes of 3.4, 2.5, and 0.9 Mb identified in B. cepacia strain 17616. DNA preparations from the different isolates contained between two and four linearized replicons in the size range of 1.1 to 5.8 Mb, confirming that chromosomal multiplicity is common among B. cepacia strains. Estimates of overall genome size varied between 4.6 and 8.5 Mb. We also compared the distribution of r-RNA (rrn) genes in selected isolates by determining the number and sizes of DNA fragments generated by treatment of their DNA with Ceu I, an enzyme which appears to cleave exclusively within 23S r-RNA genes. Ceu I fragment patterns of digests of DNA from strain 17616 were consistent with earlier results suggesting the presence of three rrn operons on the 3.4-Mb replicon, one on the 2.5-Mb replicon, and one on the 0.9-Mb replicon. Other B. cepacia isolates appeared to contain six sets of rrn genes.

INTRODUCTION

Burkholderia cepacia (formerly Pseudomonas cepacia) is a member of the -subclass of the proteobacteria (Palleroni, 1992; Yakabuuchi et al., 1992). This bacteriium has attracted attention because of its unusual degradative abilitities (Stanier et al., 1966; Lessie et al., 1986) and its potential as an agent of bioremediation (Shields et al., 1991; Daubaras et al., 1995). Macrorestriction fragment mapping of the genome of B. cepacia strain 249 (ATCC 17616) indicated that it contained three chromosomes of 3.4, 2.5, and 0.9 Mb (Cheng and Lessie, 1994). Different biosynthetic and degradative functions were associated with the 3.4- and 2.5-Mb replicons, and all three replicons were shown to contain r-RNA genes. Recently Rodley et al. (1995) reported a more detailed macrorestriction fragment map of the genome of another of B. cepacia strain, ATCC 25416. This strain also contained three chromosomes. Multiple chromosomes had been noted previously only among members of the ß-subclass of the proteobacteria (Suwanto and Kaplan, 1992; Michaux et al., 1993; Lessie and Cheng, 1994). We therefore undertook an analysis of the genomes of additional B. cepacia isolates to determine if chromosomal multiplicity was characteristic of this species. For these experiments we relied on the recent development of procedures for the manipulation of large DNA molecules (Cantor et al., 1988; Holloway, 1993). We also have compared the Ceu I fragment patterns of DNA from different B. cepacia isolates in an effort to gain further information about the genomic distribution of genes related to r-RNA synthesis. This enzyme cleaves at a conserved 19-bp sequence within 23S genes, and has been used to locate r-RNA genes on the chromosomes of other bacteria (Honeycutt et al., 1993; Lieu and Sanderson, 1995; Rodley et al., 1995).

MATERIALS AND METHODS

The bacteria examined in this study are listed in Table 1. All of the strains grew at 37C in inorganic salts medium (Cheng and Lessie, 1994) supplemented with 0.5% (w/v) trehalose, cellobiose, glucose, gluconate, mannitol, L-threonine, L-arginine, or citrate, or with 1% (w/v) penicillin G, as sole source of carbon and energy. Strains 249 (ATCC 17616), DB01 (ATCC 29424), G4, CRE-7, and 67-46 also grew well on potassium phthalate, but failed to utilize D-serine. Another group of strains including the isolates, 383 (ATCC 17760), SW3, 542, 382 (ATCC 17759), ATCC 25416, and NTC 10661, grew well on D-serine, but failed to utilize phthalate. Strain AC1100 (ATCC 53867) failed to utilize either of these compounds. All of the isolates excepting strains 542 and AC1100 exhibited a bright green sheen on EMB/glucose indicator plates (Sage and Lessie, 1990), indicative of the presence of high levels of glucosedehydrogenase.

Our strategy for evaluating the chromosomal content of the different strains was based upon observations indicating that preparations of unrestricted DNA obtained by lysis of bacteria in agarose plugs contain significant amounts of randomly linearized forms of any circular replicons present. Such DNA species are resolved readily by CHEF gel electrophoresis (Michaux et al., 1993; Cheng and Lessie, 1994; Rodley et al., 1995). For analysis of DNA species from different strains, bacteria were grown in inorganic salts medium supplemented with 1% (w/v) yeast extract. Preparations of large DNA were obtained by lysing cells in 1% Incert agarose (FMC Co, Rockland, ME) at cell densities equivalent to between 5 x 109 and 2 x 1010 bacteria per ml as described previously (Cheng and Lessie, 1994). The lysis buffer (pH 9.5) used in the present experiments contained 10 mM Tris, 450 mM EDTA, and 1% lauryl sarcosine, with 1 mg/ml of pronase E (Sigma Chem. Co., St. Louis, Mo) substituted for proteinase K. Ten ul agarose plugs containing ca 2 µg of unrestricted DNA were embedded in the wells of 13 x 13 x 0.5 cm 0.8% or 1% (w/v) Fastlane agarose gels (FMC Co., Rockland, ME) . Large linear DNA species were resolved electrophoretically in 0.5x TBE buffer in a contour-clamped homogeneous electric field (CHEF) apparatus (Owl Scientific Co., Boston, MA) operated at 50 V and 12C for 160 hr. The pulse time was ramped from 600 to 3400 sec in increments of 240 sec. The electrophoresis program used for resolution of DNA fragments generated by treatment of the DNA with Xba I was described previously (Cheng and Lessie, 1994). Macrorestriction fragments generated by treatment of the DNA with Ceu I (purchased from New England Biolabs, Beverley, MA) were resolved in 1% Fastlane Agarose, using a CHEF DR-II Pulsed-Field Electrophoresis system (Bio Rad. Co., Hercules, CA). Electophoresis was for 48 hr a 4V/cm with ramping between 60 and 600 sec. DNA size markers, including concatamers of coliphage lambda and chromosomal DNAs from Saccharomyces cerevisiae, Schizosaccharomyces pombe, and Hansulena wingeii purchased from Bio Rad Co.

RESULTS AND DISCUSSION

Genomic Complexity of B. cepacia Isolates. The numbers and sizes of large DNA species resolved from the B. cepacia strains characterized in this study are summarized in Table 1. Preparations of large DNA from each of the B. cepacia isolates contained at least two large DNA species. For example, strains DB01 and CRE-7 contained three DNA species similar in size to the linearized forms of the chromosomes detected in strain 17616. Strains 25416 and 382, as well as strain 10661, also appeared to contain three replicons, but the mid-size replicon was ca 1 Mb larger than the corresponding replicon of strain 17616. Strain G4 contained four replicons of 3.4, 2.4, 1.2, and 1.1 Mb. The linear 1.2- and 1.1-Mb species were resolved when the electrophoresis was carried out at 120 V for 48 hr over a range of pulse times from 10 to 300 s with ramping increments of 3 s. None of the other strains listed in Table 1 contained two replicons in this size range.

Six of the P. cepacia strains listed in Table 1, 383, 542, 67- 46, AC1100, SW3, and the strain designated PHK-GB, appeared to contain two large replicons. The difference in the number of DNA species from strains PHK-GB and DB01 was of particular interest since these represented different laboratory stocks of the same isolate. Strain PHK was isolated by P. Kaiser in the laboratory of D.W Ribbons on the basis of its ability to utilize phthalate as sole carbon source (Keyser et al., 1976) and deposited in the American Type Culture Collection, where it was assigned the ATCC number 29424. A stock of this strain was obtained by D.P. Ballou from the Ribbons laboratory and subsequently maintained at the University of Michigan as strain DB01 (Zylstra et al., 1989). PHK-GB represents another stock of strain PHK maintained at the Gulf Breeze Environmental Research Laboratory by Dr. R.W. Eaton. As noted above, strain DB01 contained three replicons of 1.3, 2.4, and 3.4 Mb. However Strain PHK- GB contained only two replicons of 1.3 Mb and 5.8 Mb. To resolve this discrepancy we obtained a stock of strain PHK from the American Type Culture Collection and examined its DNA. This strain contained three replicons of the same size as those present in strain DB01.

To confirm that strains SB01 and PHK-GB were variants of strain PHK, we compared the Xba I macrorestriction fragment patterns of DNA from these strains. The results indicated that Xba I digestion of DNA from the American type Culture collection strain of strain PHK (ATCC 29424) and from strain DB01 yielded the same pattern of macrorestriction fragments. The Macrorestriction fragment pattern of DNA from strain PHK-GB was similar but not identical to that of strains PHK and DB01. The simplest interpretation is that in the PHK-GB variant there had been a fusion of the 3.4- and 2.4-Mb replicons present in the original isolate. We recently have noted a similar replicon fusion involving the two smallest of the three chromosomes in a derivative of strain 17616. The B. cepacia genome is rich is IS elements which can mediate replicon fusions and activate gene expression (Lessie and Gaffney, 1986; Haugland et al., 1990; Lessie et al., 1990). It is possible that the types of chromosome fusions we have observed in strains 29424 and 17616 represent additional examples of IS-element dependent rearrangement of the B. cepacia genome.

The estimated genome sizes for the B. cepacia strains examined in this study ranged between 4.6 and 8.5 Mb of DNA. These values did not include replicons less than 0.9 Mb, such as the 170-kb cryptic plasmid of strain 17616 (Gaffney and Lessie, 1987). The differences in overall genome size and in sizes and number of replicons present in the different isolates we have characterized raise important questions about the origin of large elements of the P. cepacia genome. One is whether components of the genome might have been acquired by lateral transfer. Another is whether IS elements have played an important role in rearrangments of the genome of this bacterium. We are presently attempting to address these questions by examining the extent of homology among replicons from the different strains as well as the distribution of different genes and IS elements on the various replicons.

Chromosomal Distribution of rrn Genes in B. cepacia. The first example of multiple chromosomes in a bacterium was reported by Suwanto and Kaplan (1989), who established that Rhodobacter sphaeroides contained two distinct circular replicons containing t-RNA as well as r-RNA (rrn) genes. Multiple replicons were also defined in R. meliloti (Honeycutt et al., 1993) The genome of this bacterium was comprised of a 3.4-Mb chromosome containing three rrn loci and two megaplasmids of 1.7 and 1.4 Mb specifying functions related to nitrogen fixation and symbiosis, but lacking rrn genes. Southern hybridization experiments indicated that all three replicons in B. cepacia 17616 contained genes for 16S and 23S RNA (Cheng and Lessie, 1994). The 3.4-Mb replicon appeared to contain three sets of such genes. The 2.5- and 0.9-Mb replicons each appeared to contain one set of rrn genes. To obtain an independent estimate of the number of rrn genes on each replicon we determined the number of DNA fragments generated by digestion of genomic DNA from strain 17616 with Ceu I. We also examined Ceu I digests of DNA from strain 249-2(TGL6-), a derivative of strain 17616 deleted for portions of the 2.5 and 0.9-Mb replicons and cured of the cryptic plasmid present in its parent (Gaffney and Lessie, 1987; Cheng and Lessie, 1994). Ceu I digests of DNA from strain 17616 and 249-2(TGL6-) each contained five Ceu I fragments (see Table 2). The results were consistent with our earlier estimate of the number of rrn genes. Ceu I cleaved the 3.4-Mb chromosome of strains 17616 and 249-2(TGL6-) into three fragments of 1.2, 0.9 and 0.6-Mb. It cleaved the 2.5- and 0.9-Mb chromosomes of strain 17616 and the corresponding 1.8 and 0.65-Mb replicons of strain 249-2(TGL6-) only once. Thus, the deletion events that produced the 1.8- and 0.65-Mb replicons appear not to have affected the number of 23S RNA genes.

We also examined the number of Ceu I fragments generated by treatment of DNA from strains DB01, CRE-7, and 24516, isolates which like strain 17616 contained three chromosomes. The DNA digests from these strains contained six Ceu I fragments, suggesting the presence of an additional set of rrn genes. Table 2 compares the number and sizes of DNA fragments generated by Ceu I treatment of DNA from strains 17616, 249-2(TGL6-), and 24516 and those obtained by similar treatment of DNA from E. coli and S. typhimurium (Lieu and Sanderson, 1995). The tight clustering of rrn operons characteristic of E. coli and S. typhimurium (indicated by the preponderance of low molecular weight fragments in Ceu I digests of DNA from these bacteria) was not observed in the case of strain 17616. The number and sizes of fragments we detected in Ceu I digests of DNA from strain 25416 were in general agreement with those reported by Rodley et al. (1995). However both the number and sizes of fragments detected in Ceu I digests of DNA from strains 17616 and 25416 differed. These differences, as well as the aforementioned differences in number and sizes of replicons in different B. cepacia isolates, suggest a high degree of genomic plasticity among members of this genus.

ACKNOWLEDGEMENTS

This research was supported in part by grant DE-FG02-ER20051 from the Department of Energy. During part of the study B.D. Manning was an U.S. EPA Cooperative Education Student.

REFERENCES

Cantor CR, Smith CL and Mathew MK (1988) Pulsed-field gel electrophoresis of very large DNA molecules. Ann. Rev. Biophys. Biophys. Chem. 17:287-304.

Cheng HP and Lessie TG (1994) Multiple replicons constituting the genome of Pseudomonas cepacia 17616. J. Bacteriol. 176:4034-4042.

Daubaras DL, Hershberger CD, Kitano K, and Chakrabarty AM. 1995. Sequence analysis of a gene cluster involved in metabolism of 2,4,5-trichlorophenoxyacetic acid by Burkholder cepacia AC1100. Appl. Environ. Microbiol. 61:1279-1289.

Gaffney TG and Lessie TG (1987) Insertion sequence-dependent rearrangements of Pseudomonas cepacia plasmid pTGL1. J. Bacteriol. 169:224-230.

Haugland RA, Sangodkar UMX, and Chakrabarty AM (1990) Repeated sequences including RS1100 from Pseudomonas cepacia AC1100 function as IS elements. Mol. Gen. Genet. 220:222-228.

Honeycutt RJ, McClelland M, and Sobral, BWS (1993) Physical map of the genome of Rhizobium meliloti 1021. J. Bac. 175: 6945-6952.

Holloway B.W. (1993) Genetics for all bacteria. Ann. Rev. Micro. 47:659-684.

Keyser P, Pujar BF, Eaton RW and Ribbons DW (1976) Biodegradation of the phthalates and their esters by bacteria. Env. Health Perspect. 18:159-166.

Lessie, TG and Gaffney T (1986) Catabolic potential of Pseudomonas cepacia, p. 439-476. In: J.R. Sokatch, and L.N. Ornston. (ed.), The Bacteria, a Treatise on Structure and Function, vol. X. The Biology of Pseudomonas. Academic Press, New York.

Lessie TG, Wood MS, Byrne A, and Ferrante A. (1990) Transposable gene activating elements in Pseudomonas cepacia, p279-291. In: S. Silver, A.M. Chakrabartry, B. Iglewski, and S. Kaplan. (ed.), Pseudomonas: biotransformations, pathogenesis, and evolving biotechnology. ASM, Washington, D.C.

Lieu SL and Sanderson KE (1995) I-CeuI reveals conservation of the genome of independent strains of Salmonella typhimurium. J. Bacteriol. 177:3355-3357.

Michaux S, Paillison J, Carles-Nurit MJ, Bourg G, Allard-Servent A, and Ramuz, M (1993) Presence of two independent chromosomes in theBrucella melitensis 16M genome. J. Bacteriol. 175:701-705.

Palleroni NJ. (1992) Present situation in the taxonomy of aerobic pseudomonads, p. 105-115. In E. Galli, S. Silver, and B. Witholt (ed.), Pseudomonas: molecular biology and biotechnology. American Society for Microbiology, Washington, DC.

Rodley PD, Romling U, and Tummler B (1995) A physical genome map of the Burkholderia cepaciatype strain. Mol. Micro. 17:57-67.

Sage A, Linker A, Evans LR, and Lessie TG (1990) Hexose phosphate metabolism and exopolysaccharide formation in Pseudomonas cepacia.

Shields MS, Montgomery SO, Cuskey SM, Chapman PJ, and Pritchard, PH (1991) Mutants of Pseudomonas cepacia G4 defective in catabolism of aromatic compounds and trichloroethylene. Appl. Environ. Microbiol. 57:1935-1941.

Stanier RY, Palleroni, NJ and Doudoroff M (1966) The aerobic pseudomonads: a taxonomic study. J. Gen. Microbiol. 43:159-271.

Suwanto A and Kaplan S. 1992. Physical and genetic mapping of the Rhodobacter sphaeroides 2.4.1 genome: presence of two unique circular chromosomes. J. Bacteriol. 171: 5850-5859.

Yakabuuchi E, Kosako Y, Oyaizu H, Yano I, Hotta H, HashimotoY, Ezaki T, and Arakawa M (1992) Proposal of Burkholderia gen. nov. and transfer of seven species of the genus Pseudomonas homology group II to the new genus, with the type species Burkholderia cepacia (Palleroni and Holmes 1981) comb. nov. Microbiol. Immunol. 36:1251-1275.

Zylstra G, Olsen RH, and Ballou DP (1989) Cloning, expression, and regulation of the Pseudomonas cepacia protocatechate 3,4-dioxygenase genes. J. Bacteriol. 17:5907-5914.