|
|
||||||||
1Department of Medical Microbiology, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG, UK 2Laboratorium voor Mikrobiologie, Faculteit Wetenschappen, Universiteit Gent, K. L. Ledeganckstraat 35, B-9000 Gent, Belgium#dReceived 11 October 2002 Accepted 12 February 2003
Correspondence: John R. W. Govan (john.r.w.govan{at}ed.ac.uk)
| Abstract |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Most isolates of the B. cepacia complex exhibit high-level resistance to all major classes of antibiotics (Lewin et al., 1993; Pitt et al., 1996; Nzula et al., 2002). The B. cepacia complex is also one of the few groups of bacteria to exhibit intrinsic resistance to cationic antimicrobial peptides (Hancock, 1997). Some strains are susceptible in vitro to ceftazidime and meropenem, arguably the most potent anti-cepacia agents; however, the majority of strains, including the highly transmissible B. cenocepacia lineage ET12, are resistant to these agents (Lewin et al., 1993; Nzula et al., 2002).
All bacteria in the B. cepacia complex have large genomes (mean size approx. 8 Mbp), comprising multiple replicons that may contribute to genomic plasticity (Lessie et al., 1996; Wigley & Burton 2000; Parke & Gurian-Sherman, 2001; Mahenthiralingam et al., 2002). Ironically, the B. cepacia complex could be considered as both friend and foe, as some strains are highly effective as biopesticides in the control of plant fungal diseases and in bioremediation of contaminated soils (Holmes et al., 1998; Parke & Gurian-Sherman, 2001). These dual roles raise important medical, agricultural and ecological issues (Govan et al., 1996, 2000; Govan & Vandamme, 1998; Environmental Protection Agency, 2002), including the significance of horizontal gene transfer in assessment of the risk to humans of using these bacteria as biopesticides or in bioremediation (Holmes et al., 1998; LiPuma & Mahenthiralingam, 1999; Govan et al., 2000; Parke & Gurian-Sherman, 2001).
Attempts to explain the genomic plasticity, adaptability and virulence of the B. cepacia complex have paid little attention to the potential contribution of bacteriophages. In many pathogens, these bacterial viruses are recognized as important contributors to virulence, in the form of bacterial lysogens, or as vectors in horizontal gene transfer. Interest in the use of phage-induced bacterial lysis for therapeutic purposes was widespread in the 1920s, but declined with the arrival of the antibiotic era. However, as antibiotic resistance increasingly threatens standard therapies against bacterial infections, there is renewed interest in the antimicrobial properties of these highly specific agents (Pirisi, 2000; Sulakvelidze et al., 2001).
Little is known of the phages of the B. cepacia complex. Early reports on the B. cepacia phages CP1 (Cihlar et al., 1978) and CP 75 (Matsumoto et al., 1986) predate the revision of B. cepacia taxonomy. However, a recent report from our laboratories described two transducing bacteriophages, NS1 and NS2, whose host range included the five genomovars (IV) that were known at the time, and also extended to Pseudomonas aeruginosa (Nzula et al., 2000). Lytic phages with an interspecies host range within the B. cepacia complex have also been reported in association with soil-borne strains of B. cenocepacia (LiPuma et al., 2000). These results suggest that lysogenic conversion and transduction could play a role in the evolution of species of the B. cepacia complex as human pathogens, and indicate the need for further studies on the host range and properties of phages associated with the B. cepacia complex and related bacteria.
In this study, we investigated the prevalence of lysogeny within the nine current species of the B. cepacia complex and isolated lytic phages from natural habitats of these bacteria, including the plant rhizosphere. Our results show that the host range of the phage panel includes seven genomovars and, in the case of individual phages, is not genomovar-specific.
| METHODS |
|---|
|
|
|---|
|
Media.
Bacteria were grown in nutrient broth with 0.5 % yeast extract (NBYE) at 37 °C in a shaking incubator. Soft overlay agar for phage experiments comprised LuriaBertani (LB) broth with 0.3 % bacteriological agar (Difco). The nutrient agar used was Columbia agar base (39 g l-1; Oxoid).
Isolation of lysogenic phages.
Temperate phages were assayed and maintained as described previously (Nzula et al., 2000). Lysogeny was investigated using the following method: the 20 strains designated in Table 1 with the symbol
were prepared as saline suspensions (approx. 106 c.f.u. ml-1), inoculated onto tryptone soy agar (TSA; Oxoid) using a multipoint inoculator and incubated at 30 °C for 6 h. Bacterial growth was inverted over chloroform vapour for 15 min and then allowed to air-dry for 15 min. Soft agar overlays (2.5 ml), inoculated with 100 µl exponential-phase culture of each of the 20 strains, were layered over the original bacterial growth and allowed to set; the plates were incubated overnight at 37 °C. Phage plaques were identified in the overlay in the proximity of the original inoculum and used to prepare single-plaque preparations as follows: an agar plug containing a single phage plaque was removed using a sterile glass pipette, transferred to 10 ml phage buffer (10 mM Tris/HCl, pH 8.0; 10 mM MgCl2), vortexed for 30 s, centrifuged at 3000 g for 30 min and filtered (pore size 0.2 µm; Millipore).
Isolation of lytic environmental phage.
The natural habitats of the B. cepacia complex include soils, river sediments and plants, particularly the plant rhizosphere (Fisher et al., 1993; Butler et al., 1995; Parke & Gurian-Sherman, 2001). Therefore, 20 samples of soil, river sediment and rhizosphere (soil plus root material) were collected. The presence of phage was then investigated using a modification of the phage enrichment technique described by Weiss et al. (1994), as follows: approximately 10 g sample was suspended in 15 ml LB broth and dispersed by shaking in an orbital incubator for 30 min at 30 °C. After removal of soil particles by centrifugation (4000 g for 20 min), the supernatant was filter-sterilized (pore size 0.2 µm; Acrodisc) and 1 ml aliquots were added to 15 sterile tubes. To each extract was added 25 µl exponential-phase culture from one of 15 propagating strains chosen to represent the genomovars of the B. cepacia complex strain panel (Mahenthiralingam et al., 2000a), and the contents were incubated at 37 °C overnight. The bacterial cells were removed by centrifugation (4000 g for 30 min), the supernatant was membrane-filtered as before and 10 µl filtrate was spotted onto single-layer lawns of the propagating strain. Phage plaques were identified after overnight incubation at 37 °C, and single-plaque stocks were prepared as described in the previous section.
High-titre phage preparations.
High-titre phage preparations were prepared as follows: 100 µl single-plaque preparation, containing approximately 105 p.f.u. ml-1, was added to 2.5 ml soft nutrient agar, previously seeded with 100 µl exponential-phase culture of the propagating strain. The mixture was then overlaid on nutrient agar and allowed to set. After 18 h at 37 °C, overlays showing semi-confluent lysis were transferred into 10 ml phage buffer. The lysate was then vortexed and centrifuged at 3200 g for 30 min and the supernatant was membrane-filtered. Phage titres were determined, as p.f.u. ml-1, by incorporating 100 µl host bacteria (exponential-phase NBYE culture) and 100 µl phage stock in 2.5 ml soft agar overlay, and lytic plaques were enumerated after 18 h incubation at 37 °C. Stock preparations were maintained at 4 °C.
Host range of phages.
Stock phage preparations were diluted in phage buffer to approximately 108 p.f.u. ml-1 against the propagating strain, and 10 µl was spotted onto single-layer lawns (prepared from exponential-phase NBYE cultures) of potential host bacteria. Lytic activity was recorded after 24 h at 37 °C on a scale ranging from < 10 plaques (+) to confluent lysis (+++) (Table 1).
Electron microscopy.
Stock phage preparations (approx. 108 p.f.u. ml-1) were centrifuged at 100 000 g for 1 h. Phage pellets were resuspended in 1 M ammonium acetate, negatively stained with 2 % (w/v) potassium phosphotungstate solution (pH 7.0) and examined with a Hitachi model HU-12A transmission electron microscope.
Phage DNA extraction and RFLP profiling.
In preparation for DNA extraction, high-titre phage stocks (containing at least 1010 p.f.u. ml-1) were prepared using soft agar overlays, as described above. DNA was extracted from 10 ml phage stock using the Wizard Lambda preparation DNA purification system in conjunction with the Vac-Man laboratory vacuum manifold (both from Promega). Extracted DNA was eluted in sterile distilled water and stored at -20 °C. DNA quality was assessed on an E-gel pre-cast 0.8 % agarose gel (Invitrogen Life Technologies). In cases where DNA was not of sufficient quality for DNA restriction, purification was performed using the PCR protocol from the QIAquick gel extraction kit (Qiagen). Purified DNA was eluted in 30 µl elution buffer and stored at 4 °C. To determine genome size and to confirm that the phages were different from one another, approximately 1 µg DNA was restricted using 10 U HindIII (Promega), incubated for 3 h at 37 °C and visualized on 0.6 % 0.5x TBE agarose gel alongside 1 µl Ready-Load Lambda DNA/HindIII fragments (Invitrogen Life Technologies).
| RESULTS |
|---|
|
|
|---|
Isolation of environmental phages
Five virulent phages, JB1, JB3, JB5, RL1c and RL2, were isolated from soils and from the rhizosphere of various plants (Table 2). Most positive samples included decayed plant material collected from moist environments, but phages were also isolated from dry soils.
|
Host range of B. cepacia complex phages NS1, NS2 and newly isolated phages
To confirm that distinct phages were being accumulated and investigated, we determined the host range of the 19 phages against a preliminary bacterial panel comprising the 20 isolates of the B. cepacia complex that were used in the lysogeny screen, and also the HindIII RFLP profile (and hence an approximate genome size; Table 2). With the exception of phages JB3, DK2 and DK3, the genomes of the B. cepacia complex phages were within the range 4048 kbp (Table 2). If several phages shared the same host range and RFLP profile, only one phage was used for further study. An exception was made for phages RL1c and RL1t, which shared the same host range and RFLP profile but produced different plaque morphologies: clear plaques associated with virulent phage (RL1c) or turbid, temperate phage plaques (RL1t). Similar host ranges and RFLP profiles were observed with the temperate phages DK2 and DK3, which had respectively been isolated from B. cenocepacia C3166 and B. stabilis C3174. As previously observed for NS1 and NS2, none of the newly identified phages was inactivated by treatment with chloroform. In addition, no evidence of bacteriocin activity was found during the search for B. cepacia complex phages.
The host range of phages NS1, NS2 and eight novel phages (JB1, JB3, JB5, DK1, DK3, RL1c, RL1t and RL2) was then determined against an enlarged panel of 66 isolates of the B. cepacia complex and 55 isolates representing other pseudomonads. The host range of individual phages included multiple species of the B. cepacia complex. Collectively, the host range of the phage panel included seven of the presently recognized B. cepacia genomovar species; no phage activity was detected against the single representatives of B. cepacia genomovar VI or B. ambifaria (Table 1). However, within each B. cepacia species, there was wide variation in susceptibility to an individual phage. B. multivorans appeared to be least susceptible to phages investigated in this study: of nine B. multivorans strains examined as potential phage hosts, only strain C2775 showed susceptibility. As observed previously for NS1 and NS2, the host range of some of the phages was not restricted to the B. cepacia complex. P. aeruginosa strains C1546 and J2852 were susceptible to phage JB3, and B. gladioli strain C3654 was susceptible to phages NS2, DK1, RL1c, RL1t and JB5.
Electron microscopy
In accordance with previous studies on phages NS1 and NS2 (Nzula et al., 2000), electron microscopy revealed the novel phages JB1, JB5, DK2/DK3, RL1c/RL1t and RL2 to be morphologically T-even-like phages, with hexagonal heads and contractile tails of variable length. In contrast, phages JB3 and DK1 were lambda-like, with hexagonal heads and flexuous, non-contractile tails.
| DISCUSSION |
|---|
|
|
|---|
A broad host range for B. cepacia complex phages could contribute to the genomic plasticity of these bacteria, and their evolution from highly metabolically active soil saprophytes to plant and human pathogens and, recently, also animal pathogens (Berriatua et al., 2001). Lysogenic conversion and transduction are important processes by which chromosomal host genes can be acquired and exchanged between bacteria. We have previously demonstrated in vitro transfer of antibiotic-resistance genes between B. vietnamensis strains by phages NS1 and NS2 (Nzula et al., 2000).
As reported for Shigella flexneri (Allison & Verma, 2000), prophages may contribute to O-antigen modification in the B. cepacia complex (Kenna et al., 2003), and to the role of B. cepacia complex LPS as a potent virulence determinant (Shaw et al., 1995; Hughes et al., 1997). Opportunities for transduction and lysogenic conversion would exist not only in natural environments shared by various B. cepacia complex species and related bacteria, but also in CF airway secretions, where mixed infections are frequent and bacterial populations can reach densities in excess of 109 c.f.u. ml-1. Based on a close taxonomic relationship and shared insertion sequences and environmental habitats, we were particularly keen to test the phage panel against isolates of Burkholderia pseudomallei, the causative agent of melioidosis and a potential agent for bioterrorism (Mack & Titball, 1998). In collaboration with Dr Ty Pitt (PHLS, Colindale, London, UK), only phage NS2 was found to be active, lysing 13 of 40 B. pseudomallei strains tested (unpublished results). The potential importance of broad-host-range phages such as NS2 is also suggested by recent reports of B. pseudomallei infection in CF patients (including coinfection with B. cepacia) following travel to Thailand, where melioidosis is endemic (Schulin & Steinmetz, 2001; Visca et al., 2001).
Further studies are required to determine the transducing potential and other biological properties (e.g. nucleic acid content and bacterial receptors) of the B. cepacia complex phages that were accumulated in this study. Meanwhile, several preliminary observations merit comment. The shared host range and RFLP profiles of the temperate phages DK2 and DK3, respectively isolated from strains of B. cenocepacia and B. stabilis, suggest that integration of the same phage can occur in different species of the B. cepacia complex. While the presence of multiple prophages within a single strain is common in P. aeruginosa (Holloway et al., 1960), we found no evidence of polylysogeny during our investigation of B. cepacia complex phages. The variable plaque morphology exhibited by the environmental phages RL1c and RL1t is interesting and, to our knowledge, has not been reported previously for B. cepacia complex phages.
The relative lack of susceptibility of B. multivorans to the phage panel was interesting. Taken together, B. multivorans and B. cenocepacia account for almost 90 % of clinical isolates of the B. cepacia complex; however, in contrast to B. cenocepacia', B. multivorans is rarely isolated from natural environments (Bevivino et al., 2002; authors unpublished data). In our study, of nine B. multivorans strains tested, only one (C2775) exhibited phage susceptibility. However, this resistance may be misleading, as a recent study in our laboratories identified a novel B. cepacia complex phage (RU2) from soil, which plates on B. multivorans C3164 and also on three other B. cepacia complex isolates that are resistant to the primary phage panel (Table 1), namely J2552 and J2553 (both B. anthina) and E571 (B. ubonensis).
Whole-genome sequencing of bacteria provides increasing evidence for widespread exchange of chromosomal genes and other extrachromosomal elements, mediated by phages. Thus, analyses following the recent sequencing and annotation of B. cenocepacia J2315T (= LMG 16656T = NCTC 13227T; http://www.sanger.ac.uk/Projects/B_cepacia/) are keenly awaited. In relation to the issue of multiple lysogeny in the B. cepacia complex, we performed a BLAST search of the provisional J2315T genome sequence and found evidence of a single prophage. If confirmed, this would be an interesting result as, in this study and in a more extensive search for lysogeny in J2315T, we isolated only one temperate phage, DK4 (authors unpublished data). The availability of broad-host-range phages complements the panel of B. cepacia complex strains (Mahenthiralingam et al., 2000a) and should facilitate future research on these highly adaptable and increasingly important bacteria. Furthermore, in addition to the established therapeutic use of lytic phages, phage-encoded lytic enzymes may provide novel therapeutic agents against B. cepacia complex infections (Schuch et al., 2002), for which there are few antibiotic options at present (Nzula et al., 2002).
| Acknowledgments |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. T. Kenna, H. Yesilkaya, K. J. Forbes, V. A. Barcus, P. Vandamme, and J. R. W. Govan Distribution and genomic location of active insertion sequences in the Burkholderia cepacia complex J. Med. Microbiol., January 1, 2006; 55(1): 1 - 10. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. De Soyza, C. D. Ellis, C. M. A. Khan, P. A. Corris, and R. D. de Hormaeche Burkholderia cenocepacia Lipopolysaccharide, Lipid A, and Proinflammatory Activity Am. J. Respir. Crit. Care Med., July 1, 2004; 170(1): 70 - 77. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |