Burkholderia cepacia complex bacteria - Wiley Online Library

Journal of Applied Microbiology ISSN 1364-5072

REVIEW ARTICLE

Burkholderia cepacia complex bacteria: opportunistic pathogens with important natural biology E. Mahenthiralingam1, A. Baldwin2 and C.G. Dowson2 1 Cardiff School of Biosciences, Cardiff University, Main Building, Cardiff, UK 2 Department of Biological Sciences, Warwick University, Coventry, UK

Keywords Burkholderia, cystic fibrosis, environmental microbiology, molecular epidemiology, pathogenesis. Correspondence Eshwar Mahenthiralingam, Cardiff School of Biosciences, Cardiff University, Main Building, Museum Avenue, Cardiff, Wales, UK, CF10 3TL. E-mail: [email protected]

2007 ⁄ 1411: received 1 September 2007, revised 1 December 2007 and accepted 2 December 2007 doi:10.1111/j.1365-2672.2007.03706.x

Abstract Interaction with plants around their roots and foliage forms the natural habitat for a wide range of gram-negative bacteria such as Burkholderia, Pseudomonas and Ralstonia. During these interactions many of these bacteria facilitate highly beneficial processes such as the breakdown of pollutants or enhancement of crop growth. All these bacterial species are also capable of causing opportunistic infections in vulnerable individuals, especially people with cystic fibrosis (CF). Here we will review the current understanding of the Burkholderia cepacia complex (Bcc) as a group of model opportunistic pathogens, contrasting their clinical epidemiology with their ecological importance. Currently, the B. cepacia complex is composed of nine formally named species groups which are all difficult to identify using phenotypic methods. Genetic methods such as 16S rRNA and recA gene sequence analysis have proven useful for Bcc species identification. Multilocus sequence typing (MLST) is also emerging as a very useful tool for both Bcc strain and species identification. Historically, Burkholderia cenocepacia was the most dominant Bcc pathogen in CF, however, probably as a result of strict infection control practices introduced to control the spread of this species, its prevalence has been reduced. Burkholderia multivorans is the now the most dominant Bcc infection encountered in the UK CF population, a changing epidemiology that also appears to be occurring in the US CF population. The distribution of Bcc species residing in the natural environment may vary considerably with the type of environment examined. Clonally identical Bcc strains have been found to occur in the natural environment and cause infection. The contamination of medical devices, disinfectants and pharmaceutical formulations has also been directly linked to several outbreaks of infection. In the last 10 years considerable progress has been made in understanding the natural biology and clinical infections caused by this fascinating group of bacteria.

Introduction In 1950, William Burkholder first described Pseudomonas cepacia as a plant pathogen capable of causing onion rot (Burkholder 1950). As additional P. cepacia isolates were characterized it became clear that this group of bacteria were highly versatile and capable of a variety of complex interactions such as the degradation of complex aromatic pollutants (O’Sullivan and Mahenthiralingam 2005), the

protection and growth promotion of plants (Parke and Gurian-Sherman 2001) and the infection of humans and animals in addition to plant disease (Mahenthiralingam et al. 2005). The phenotypic diversity observed within P. cepacia bacteria led to the initial classification of these isolates as closely related pseudomonads. However, when molecular taxonomic analysis was performed on these Pseudomonas species, several, including P. cepacia, were transferred to a new genus, Burkholderia (Yabuuchi et al.

ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 1539–1551

1539

The B. cepacia complex

E. Mahenthiralingam et al.

1992). At the time Burkholderia cepacia was designated as the type species for this new genus (Yabuuchi et al. 1992), which in hindsight was an interesting choice, as 5 years later Peter Vandamme et al. (1997) showed that isolates originally classified as this single species were composed of at least five distinct genetic species designated as genomovars; the collective of the genomovars was named as the B. cepacia complex (Bcc). During the last 10 years four further genomovars have been described, and with the discovery of tests that allow the discrimination of each, all have now been formally named as species. The diversity and identification, role in cystic fibrosis (CF) and other diseases and environmental functions of the Bcc are described. The current diversity Since 1997, the taxonomy of Bcc species has changed considerably, with nine formally named species defined up to 2004; in genomovar order these are as follows (Table 1): B. cepacia (Vandamme et al. 1997); Burkholderia multivorans (Vandamme et al. 1997); Burkholderia cenocepacia (Vandamme et al. 2003); Burkholderia stabilis (Vandamme et al. 2000); Burkholderia vietnamiensis

(Gillis et al. 1995; Vandamme et al. 1997); Burkholderia dolosa (Vermis et al. 2004); Burkholderia ambifaria (Coenye et al. 2001a); Burkholderia anthina (Vandamme et al. 2002) and Burkholderia pyrrocinia (Vandamme et al. 2002). Vermis et al. (2002) proposed Burkholderia ubonensis as the tenth genomovar within the Bcc and while this has been confirmed via phlogenetic examination of the 16S rRNA gene sequence of this species (Coenye and Vandamme 2003), a formal polyphasic study of this species has yet to be published. All Bcc species demonstrate considerable phenotypic variability (Vandamme et al. 1997), even within sequential clinical isolates of the same strain (Larsen et al. 1993). Simple alterations in carbon source can for example change the colonial morphology of Bcc species grown on agar (Fig. 1). The ability to alter their metabolism and biochemical profile in this way can make their identification by phenotypic methods alone very difficult; however, it must be noted that problems of identification of Bcc based on phenotypic methods are multifactorial, and not necessarily just due to growth differences on different media. Since 2004, there have been no further formal species descriptions for bacteria in the Bcc, however, sequence analysis of the recA gene (Payne et al. 2005) demonstrated

Table 1 Current Burkholderia cepacia complex (Bcc) diversity within the Cardiff collection Number of isolates identified by MLST Species ⁄ or MLST subgroup

Genomovar

Total no.

CF (%)

CLIN (%)

Burkholderia cepacia Burkholderia multivorans Burkholderia cenocepacia IIIA B. cenocepacia IIIB B. cenocepacia IIIC B. cenocepacia IIID Burkholderia stabilis Burkholderia vietnamiensis

I II III III III III IV V

45 93 148 123 16 14 25 41

14 77 123 84 0 14 9 16

12 9 21 6 0 0 9 5

Burkholderia Burkholderia Burkholderia Burkholderia Burkholderia Group K

VI VII VIII IX X –

7 113 10 17 2 59

6 6 2 1 0 20

– – – – – – Total

4 8 13 3 18 39 798

4 4 4 1 1 2 388

BCC1 BCC2 BCC3 BCC4 BCC5 BCC6

dolosa ambifaria anthina pyrrocinia ubonensis

(31) (83) (83) (68) (100) (36) (39) (85) (5) (20) (6) (34) (100) (50) (31) (33) (5) (5)

(27) (10) (14) (5)

ENV (%)

IND (%)

(38) (6) (1) (20) (100)

(36) (12)

17 6 2 25 16 0 3 17

2 1 2 8

(12) (41)

(4) (1) (1) (6) 0 0 4 (16) 3 (7)

0 0 0 0 1 (50) 5 (8)

1 106 8 16 1 10

(14) (94) (80) (94) (50) (17)

0 1 (1) 0 0 0 24 (41)

0 1 (12) 2 (15) 0 1 (5) 0 72

0 3 1 2 16 37 287

(37) (7) (67) (89) (95)

0 0 6 (46) 0 0 0 51

References (Vandamme et al. 1997) (Vandamme et al. 1997) (Vandamme et al. 2003) (Vandamme et al. 2003) (Vandamme et al. 2003) (Vandamme et al. 2003) (Vandamme et al. 2000) (Gillis et al. 1995; Vandamme et al. 1997) (Vermis et al. 2004) (Coenye et al. 2001a) (Vandamme et al. 2002) (Vandamme et al. 2002) (Vermis et al. 2002) (Vermis et al. 2002; Mahenthiralingam et al. 2006) (Baldwin et al. 2005) (Baldwin et al. 2005) (Baldwin et al. 2005) (Baldwin et al. 2005) (Dalmastri et al. 2007) (Dalmastri et al. 2007)

CF, isolate from cystic fibrosis; CLIN, isolate from a non-CF clinical infection; ENV, isolate from the natural environment; IND, isolate from an industrial source; MLST, multilocus sequence typing.

1540




(a)

(b) 1

3

4 6

9

8

8 9

11

11

10 12

13

13

14

14 15

16

16

17 18

5 7

6

7

12

2 4

3

5

10

15

1

2

17

18

Figure 1 Morphological differences of Burkholderia cepacia complex (Bcc) bacteria are grown on different carbon sources. Eighteen strains were grown on basal salts medium containing 4 g l)1 glucose (panel a) or 4 g l)1 glycerol (panel b) as the sole carbon source. Strains are as follows: 1, Burkholderia cepacia ATCC 25416T; 2, Bcc group K 383*; 3, Bcc group K strain LMG 23255*; 4, Burkholderia multivorans ATCC 17616*; 5, Burkholderia ambifaria AMMDT*; 6, B. ambifaria MC40-6*; 7, Burkholderia dolosa LMG 18943; 8, Burkholderia vietnamiensis G4*; 9, Burkholderia cenocepacia IIIA J2315*; 10, B. cenocepacia IIIA BCC0524; 11, B. cenocepacia IIIA BCC0223; 12, B. cenocepacia IIIA BCC0048; 13, B. cenocepacia IIIB HI2424*; 14, B. cenocepacia IIIB MC0-3*; 15, Burkholderia stablis LMG 18888; 16, B. cenocepacia IIIA CZ1238; 17, B. cenocepacia IIIA BCC1434; 18, B. cenocepacia BCC1436. All strains marked with an asterisk have genome sequences available (see http://img.jgi.doe.gov).

the presence of four new recA phylotypes designated B. cepacia group K, B. cepacia group AW, BCC1 and BCC2. Subsequent multilocus sequence typing (MLST, see later) studies (Baldwin et al. 2005, 2007; Mahenthiralingam et al. 2006; Dalmastri et al. 2007) have fully resolved these groups and also defined further novel phylogenetic clusters. The ability of MLST to differentiate the existing Bcc species is greater than the analysis of the recA gene alone and also shows an excellent correlation to the multiple polyphasic taxonomic methods used to fully characterize these bacteria (Vandamme et al. 1997); hence MLST will probably become the gold standard for identification of these bacteria in future (see later). Currently seven distinct MLST groups have been designated in addition to the formally named species: Bcc group K, BCC1, BCC2, BCC3, BCC4, BCC5 and BCC6 (Table 1); with additional data from polyphasic taxonomic techniques several of these novel MLST species group are currently in the process of being formally designated as new species (E. Vanlaere and P. Vandamme, personal communication). Identification approaches for the Bcc Differentiation of Bcc species by biochemical analysis alone is not straightforward and only B. multivorans and B. stabilis can be distinguished from the rest of the complex by

relatively simple tests (Henry et al. 2001). Coenye et al. (2001b) review the best methods for phenotypic identification, and recommend that these include the use of selective media such as B. cepacia selective agar (BCSA; Hardy Diagnostics, Santa Maria, CA, USA) or B. cepacia medium (Mast Diagnostics, Merseyside, UK) for clinical isolates and respiratory specimens; and less selective media such as ‘‘P. cepacia’’ polycyclic hydrocarbon medium (PCAT) and Trypan Blue Tetracycline (TB-T) for environmental isolates (Ramette et al. 2005). Other growth agar that can assist with isolation and identification of Bcc bacteria include the use of Stewart’s medium (Vanlaere et al. 2006). Many commercial biochemical analysis kits also fail to provide accurate identification of Bcc bacteria. However, if commercial kits are combined with positive growth on BCSA and specific tests, such as decarboxylation of lysine and ornithine, oxidation of sucrose and adonitol, oxidase positivity, haemolysis and growth at 42C, then generally, a correct phenotypic identification can be made. However, misidentification or lack of identification of Bcc bacteria is still a problem that faces many diagnostic microbiology laboratories (Coenye et al. 2001b). To confirm a presumptive phenotypic identification of Bcc bacteria, genetic methods are the most accurate and reliable. Phylogenetic comparison of the full-length 16S rRNA gene sequence of Bcc bacteria can distinguish all the formally named species (Coenye and Vandamme 2003). In addition, fluorescent in situ hybridization probes based on the 16S rRNA gene have shown promise for the direct microscopic identification of Bcc bacteria sputum smears (Brown and Govan 2007). However, partial 16S rRNA gene sequencing or restriction fragment length polymorphism (RFLP) analysis of this gene is not sufficiently discriminatory to resolve all the species as the gene is >98% identical for members of the Bcc. Sequence polymorphism within the protein-coding gene, recA, has proven much more discriminatory, enabling species-specific polymerase chain reaction (PCR) to be designed (Mahenthiralingam et al. 2000) and contributing to the formal naming of several species (Table 1). Positive amplification of a 1040-bp recA product using primers BCR1 and BCR2 (Mahenthiralingam et al. 2000) has remained 100% predictive that an isolate is a member of the Bcc, but recA RFLP or recA-based species-specific PCR are not accurate for all current groups within the complex. The B. multivorans and B. ambifaria recA-specific primers give good sensitivity and specificity for these species (Vermis et al. 2002), however, given the considerable amount of new Bcc diversity defined since the design of several of the original PCR primers (Mahenthiralingam et al. 2000), the remaining species PCR should no longer be considered absolutely specific (E. Mahenthiralingam, unpublished data). Other genes such as opcL gene encoding the


1541



peptidoglycan-associated outer-membrane lipoprotein of Bcc bacteria have also been validated as useful for species identification (Plesa et al. 2004), however, sequence datasets and use of this marker by the research community remains limited. If possible, identification should always be confirmed by additional testing as follows. Phylogenetic analysis of the recA gene sequence is discriminatory (Mahenthiralingam et al. 2000; Payne et al. 2005) and can place an isolate within a named or novel Bcc group. In addition, recA analysis also subdivided B. cenocepacia into four distinct subgroups referred to as B. cenocepacia IIIA, IIIB, IIIC and IIID (Table 1; Vandamme et al. 2003). However, as noted previously, this single gene sequence-based method has now been surpassed by MLST. The Bcc MLST scheme examines nucleotide polymorphisms in seven genomically disparate housekeeping genes located on the first and second chromosomes (ATP synthase b-chain, atpD; glutamate synthase large subunit, gltB; DNA gyrase B, gyrB; recombinase A, recA; GTP-binding protein, lepA; acetoacetyl-CoA reductase, phaC and tryptophan synthase, trpB), and uses the resulting allelic profiles to assign a clonal sequence type (ST) to each unique strain. MLST is uniquely able to provide unequivocal strain and Bcc species identification within a single approach (Baldwin et al. 2005); a public database of MLST sequence data (http://pubmlst.org/bcc/) has also been established to enable characterization of isolates by other researchers. Does the use of MLST for Bcc species identification and strain typing have limitations? To date we have found no instances where MLST has failed to provide an experimentally useful identification for a Bcc isolate. However, as noted earlier, the Bcc and other Burholderia are highly diverse organisms, with genomes that may recombine at a high rate. We have studied mutation and recombination rates in the MLST target genes and found that they occur at very low rates which do not impact the efficacy of the technique (Waine et al. 2007). Failure to amplify one of the MLST loci using the current PCR primers (Baldwin et al. 2005) may occur in less than 1% of the isolates examined, however, the sequence from the other six genes is normally sufficient to assign a workable strain and species identity for the isolate. In addition, such variation in single MLST loci has only been observed for isolates which appear to be very novel species groups and hence would fail to be accurately classified by other Bcc species identification methods (Mahenthiralingam et al. 2000; Plesa et al. 2004). Therefore, application of MLST to these novel groups is still driving forward their classification, although it cannot be applied with 100% efficiency in these rare occurrences of novel isolate types. Finally, the expense of DNA sequence analysis is often cited as a limitation or reason not to use MLST. 1542

However, this argument no longer stands given the highthroughput achievable with modern PCR techniques and DNA sequence analysis. Typing techniques such as pulsed field gel electrophoresis (PFGE) requires 3–5 days to process limited numbers of isolates, and may also require repeated lane-to-lane comparison against closely related strains to accurately assign a strain type. The longer technical time costs required for conventional PFGE typing are more than the costs of MLST analysis of a similar number of isolates. PFGE typing also only derives a strain type for a Bcc isolate, with the application of other diagnostic approaches (Mahenthiralingam et al. 2000; Plesa et al. 2004) being needed for Bcc species identification. In addition, although PFGE has been historically applied to epidemiological surveillance of CF pathogens by national reference centres, the data these centres generate cannot be accessed in a transferable manner by the wider research or public health community. In comparison, if laboratory anywhere in the world generates DNA sequence for just a single MLST gene, this can be searched against the public MLST database and will often achieve an experimentally useful strain typing and species identification outcome. Hence, MLST is ideally suited to an easily accessible epidemiological surveillance role. Epidemiology in CF Bcc bacteria emerged as problematic CF pathogens 30 years ago. The first significant report of P. cepacia infection in people with CF was made by Isles et al. (1984) and within a year a second report from Tablan et al. (1985) had confirmed that infections with these bacteria were emerging as a problem in CF. These early reports described the virulent nature of B. cepacia infection, which in contrast to infection with other CF pathogens such as Pseudomonas aeruginosa, led to a rapid, uncontrollable and fatal clinical decline in about 10% of the individuals (a clinical course that became known as cepacia syndrome). In addition to the virulent nature of these infections, subsequent molecular epidemiological studies demonstrated that Bcc bacteria were also capable of transmission through social contact (Lipuma et al. 1990; Govan et al. 1993), an alarming problem for the CF community. As a result, and following the publication of national guidelines, Bcc-infected patients were invariably cohorted and treated in clinics which were separate from CF individuals with P. aeruginosa infection or no infection. These initial infection control measures and education of patients and care staff took time to implement. By the mid-1990s, there had been several reports of a major epidemic strain that was prevalent in Canadian and United Kingdom CF populations; this strain was referred to by several names: using multilocus enzyme electrophoresis, Johnson et al. (1994)



initially designated it as electrophoretic type 12 or ET12; Sun et al. (1995) defined it as the cable pilus strain and Mahenthiralingam et al. (1996b) referred to random amplified polymorphic DNA (RAPD) analysis type 2. Whether the ET12 strain emerged in Canada and then spread to the United Kingdom is the subject of much debate. In 1996, Govan et al. (1996) highlighted the movement between European and North American CF populations and suggested that ‘Edinburgh ⁄ Toronto ⁄ ET12 epidemic B. cepacia’ lineage as it was designated may be clonally related to the first documented outbreaks in Ontario, Canada (Isles et al. 1984). The seminal study by Isles et al. (1984) reported the outcome of the surveillance of P. cepacia CF infection from 1971 to 1981 in patients attending the Hospital for Sick Children in Toronto, Canada. Subsequent molecular epidemiological studies demonstrated that Bcc CF infection within this region of Canada was almost entirely due to just the ET12 strain (Mahenthiralingam et al. 1996a; Speert et al. 2002). Hence it is probable that the CF isolates described in Toronto as early as 1971 (Isles et al. 1984) were actually ET12. By 1981, Isles et al. (1984) described that the prevalence of infection in the Toronto treatment centre had risen to 18% of the 500 patient population. Hence almost a decade prior to the identification of ET12 among the UK CF community (Govan et al. 1993, 1996), the strain appeared to be dominant in eastern Canada, suggesting that it had probably also emerged in Canada. MLST analysis has also demonstrated that ET12 isolates (designated by conserved PFGE fingerprint and presence of the cable pilus gene) fall into at least four very closely related ST which only differ at one of the seven sequence loci (Baldwin et al. 2005); ST29, 30 and 31 were recovered from Canadian CF patients, while ST28 is specific to UK CF patients. Hence, tracing the heritage of the ET12 strain is worthy of further systematic study, especially in terms of understanding the subtle strain differences in the North American and European CF populations and why this lineage rapidly became such a dominant and highly virulent CF pathogen. Several other epidemic Bcc CF strains were subsequently described, and initially the majority of these problematic strains were found to belong to the species B. cenocepacia. RAPD types 1, 2, 4 and 6, included epidemic B. cenocepacia strains that predominated in the Canadian CF population (Mahenthiralingam et al. 1996a; Speert et al. 2002). Sequence analysis of the recA gene showed that these epidemic strains resided in phylogenetic subgroup IIIA of B. cenocepacia (Mahenthiralingam et al. 2000, 2001; Speert et al. 2002). Patient-to-patient transmission (Govan et al. 1993), the ability to super-infect and replace infection with B. multivorans and high mortality postinfection (Mahenthiralingam et al. 2001) were traits associated with several


of these B. cenocepacia IIIA strains. Recent studies have also shown that CF infection with B. cenocepacia and in particular the ET12 strain is associated with significantly reduced survival, while outcome from infection with B. multivorans was similar to that of P. aeruginosa-infected case controls (Jones et al. 2004). Several virulence factors have been described for B. cenocepacia IIIA strains, and in particular for ET12 strains the combination of two factors, the cable pilus and B. cenocepacia pathogenicity island, appears to be unique (reviewed by Mahenthiralingam et al. 2005). The lipopolysaccharide (LPS) of Bcc bacteria is also highly inflammatory (Govan 2003). Indeed aminoarabinose biosynthesis enzymes which modify the LPS lipid A core have been shown to be essential for viability in B. cenocepacia (Ortega et al. 2007). In the United States, other epidemic strains such as the Philadelphia-District Columbia (PHDC) and Mid-West strains were also identified and classified as the B. cenocepacia recA IIIB subgroup (Chen et al. 2001). In addition to its prevalence within the US CF population, the PHDC strain has also been found to infect European CF patients (Coenye et al. 2004). Although many of these initial epidemiological surveys showed that B. cenocepacia frequently caused highly virulent and transmissible infections in large numbers of individuals with CF, this is not the only Bcc species associated with these pathogenic traits. Epidemic outbreaks of B. multivorans CF infection were described in the United Kingdom (Whiteford et al. 1995) and France (Segonds et al. 1999). Clusters of multiple CF individuals infected with B. multivorans, B. cepacia and B. dolosa, respectively, have also been reported in the United States (Biddick et al. 2003). A study in Portugal had noted multiple CF patients infected with a single strain of B. cepacia (Cunha et al. 2003), however, subsequently the same group demonstrated that this outbreak of infection was because of a contaminated nonsterile saline solution that was used in various nasally applied therapies (Cunha et al. 2007). Significantly reduced survival has been reported for B. dolosa CF infection (Kalish et al. 2006). Overall, in common with many other opportunistic pathogens, it appears that severe disease and death may result after infection with all Bcc species dependent on the clinical state and predisposition of CF individuals at the time of infection. Clinical outcomes may vary greatly among CF patients infected with the same strain (Govan et al. 1993; Frangolias et al. 1999). Once chronically infected, treatment is also very difficult as all the different Bcc species are highly resistant to antibiotics (Nzula et al. 2002). The problems associated with patient-to-patient transmission and poor clinical outcome has led to the development of stringent infection control procedures for managing CF patients with Bcc infection (Saiman and Siegel 2004). Have these measures made a difference to


1543



the rates of Bcc infection? A review of the Bcc epidemiology observed in three CF populations (United States, Canada and Italy) prior to 2002 demonstrated that B. cenocepacia was the most prevalent species encountered in CF (mean prevalence = 67%), with B. multivorans as the next most dominant Bcc species (mean prevalence = 17%). In terms of specific populations, the largest surveys of Bcc CF infection have been performed in the United States and in 2001, Lipuma et al. (2001) demonstrated that 50% of the 606 patients studied were infected with B. cenocepacia (referred to as genomovar III at the time); B. multivorans accounted for 37Æ8% of the cases of infection in this study. By 2005, the same group reported that for 1218 individuals examined, B. cenocepacia infected 45Æ6% (a 4Æ4% reduction in prevalence since 2001), while B. multivorans infections had increased slightly to 38Æ7% (Reik et al. 2005). Outside of B. cenocepacia and B. multivorans, the remaining formally named species account for less than 10% of all CF infections caused by the complex. Reik et al. (2005) showed that for these remaining Bcc species, B. cepacia, B. stabilis, B. vietnamiensis and B. dolosa, were more likely to be encountered as CF pathogens in the United States than B. ambifaria, B. anthina and B. pyrrocinia. The general dominance of B. cenocepacia in CF is also reflected in epidemiological reports emerging from other countries. Kidd et al. (2003) showed that B. cenocepacia and B. multivorans constitute 39% and 21%, respectively of the Bcc bacteria infecting CF patients at a large Australian centre. In Brazilian CF survey (Carvalho et al. 2007), although B. cenocepacia was dominant (67% of isolates), B. vietnamiensis was the next most commonly encountered species (15% of isolates) indicating that regional differences may be present in the epidemiological distribution of Bcc in CF. Overall, these figures suggest that the spread of transmissible B. cenocepacia strains and other epidemic strains has been reduced via the implementation of strict infection control practices (Saiman and Siegel 2004). Indeed, a recent review describing the current status of Bcc CF infection in the United Kingdom clearly demonstrates that the epidemiology has changed (Govan et al. 2007). Since 2000, B. multivorans infection has been more prevalent than B. cenocepacia infection among the UK CF population (Govan et al. 2007). The majority of new cases of infection with B. multivorans are also because of genetically unique strains suggesting the sporadic acquisition of infection from environmental sources. Other infections and the capacity to cause disease Understanding the clinical epidemiology of the Bcc outside CF is more difficult, as although these bacteria are 1544

encountered as opportunistic pathogens outside this genetic disorder, they cause smaller numbers of infections. In addition, for the novel MLST-defined Bcc groups (Table 1) this problem is further confounded by the difficulty associated with their identification. Bcc infection may also occur in individuals with chronic granulomatous disease (CGD; Winkelstein et al. 2000), however, given the rarity of this disorder few isolates are available for study. In total, four CGD isolates (two B. multivorans, and one each of B. vietnamiensis and B. cenocepacia IIIB) are present in the Cardiff collection (see later). Reik et al. (2005) also examined 90 non-CF patients who had Bcc infections; a wide variety of clinical specimens were examined and included tracheal aspirates, sputum, blood, urine, bronchial lavage, sinus, bone marrow, wounds, faeces, pancreatic aspirates, cornea, pleural biopsy and spinal fluid. They found that B. cenocepacia (25Æ6%), B. cepacia (18Æ9%) and B. multivorans (15Æ6%) were the most commonly encountered species in non-CF cases of infection. Over 20% of these non-CF patients had Bcc infections which were of indeterminate taxonomic status at the time (Reik et al. 2005). General associations with clinical infection may also be drawn from examining the statistics of Bcc culture collections such as the one held at Cardiff University (Baldwin et al. 2005, 2007; Mahenthiralingam et al. 2006). The collection currently comprises over 1500 isolates of which 798 isolates have been accurately identified and typed by MLST (Table 1). The Cardiff collection is based on isolates that have been drawn from other recognized Bcc collections from all over the world (Vandamme et al. 1997; Agodi et al. 2001; Lipuma et al. 2001; Speert et al. 2002; Cunha et al. 2003), published taxonomic studies of the complex (reviewed by Mahenthiralingam et al. 2005), systematic surveys of the natural environment (Ramette et al. 2005; Dalmastri et al. 2007), as well as the prospective collection of all Bcc isolates encountered at the CF treatment centres in Vancouver, Canada, between 1981 and 2000 (Mahenthiralingam et al. 1996b, 2001; Speert et al. 2002). Hence, it can be examined as a collection that is at least representative of the currently cultivable Bcc diversity. When the distribution of Bcc species is examined in this way several interesting trends are apparent, however, this analysis cannot be taken as completely accurate as it depends solely on the number of isolates of a particular type that have been contributed to the collection to date. From the collection, all Bcc groups, except B. cenocepacia IIIC, appear to have the capacity to cause CF infection with at least one isolate from each group deriving from this source (Table 1). The largest numbers of CF isolates in the collection reside in the following Bcc species (in rank order): B. cenocepacia IIIA, B. multivorans,



B. cenocepacia IIIB, Bcc group K, B. vietnamiensis and B. cenocepacia IIID and B. cepacia, with the remaining groups composed of 10 or less CF isolates (Table 1); for these species with few CF isolates the natural environment accounts for the majority of the isolates (Table 1). In terms of the percentage of isolates within each species group that were derived from CF infection, the rankings are slightly different, with B. cenocepacia IIID being 100% CF-associated, and B. cenocepacia IIIA, B. mutlivorans and B. dolosa all being greater than 85% linked to lung infection in this disease (Table 1). For isolates recovered from non-CF opportunistic infections, B. cenocepacia IIIA is again the most dominant in terms of total numbers, with B. cepacia, B. multivorans, B. stablis, B. cenocepacia IIIB, B. vietnamiensis and Bcc group K ranking behind it. It is interesting that despite having over 100 isolates of B. ambifaria, not one has been recovered from a source of infection outside of CF, suggesting that this species may have very limited virulence as an invasive human opportunistic pathogen. Of the novel MLST-defined species, all except BCC1, BCC4 and BCC6 have been recovered from non-CF infections. One aspect of Bcc biology that may play a significant role in allowing these organisms to cause infection is their ability to survive within a number of man-made products such as pharmaceuticals (Mahenthiralingam et al. 2006), cosmetics and toiletries (Perry 2001), disinfectants and preservatives (Geftic et al. 1979). Bcc bacteria have a remarkable ability to survive for very long periods in solutions containing high concentrations of antimicrobials (Geftic et al. 1979). Even solutions with minimal nutrients such as saline used in hospitals may become contaminated and cause large outbreaks, such as reported for B. cepacia among Portugese CF patients (Cunha et al. 2007). The species most commonly encountered as industrial contaminants in our collection are B. cepacia group K, B. cenocepacia (mainly IIIB, but also IIIA), BCC3, B. stabilis, B. vietnamiensis and B. cepacia (Table 1). Interestingly all the latter species have also been recovered from CF, non-CF infections and the natural environment and demonstrate a wide distribution in comparison with several of the other Bcc species (Table 1). Recently, using MLST we were able to identify a contamination-associated strain of Bcc group K that was globally distributed (Mahenthiralingam et al. 2006); the story behind its identification was fascinating. In 2004, Craig Venter’s group published a pioneering study exploring the microbial diversity of the Sargasso Sea by using high-throughput shotgun sequencing of extracted DNA to define the genetic variation in the sample and assemble metagenomes where possible (Venter et al. 2004). One of the bacterial metagenomes identified, designated Burkholderia SAR-1, was highly controversial as


Burkholderia are not known as salt water micro-organisms, leading many researchers to suggest that its presence in the metagenomic dataset was because of contamination (Mahenthiralingam et al. 2006). Using the SAR-1 metagenome, we obtained the sequence data for the seven MLST target genes and defined the sequence type for the ‘hypothetical’ strain from which the metagenome ST102 was derived. Subsequent examination of cultivable strains in our collection (Table 1) identified 11 isolates with the identical sequence type. These cultivable strains came from a variety of sources from several continents, and were all linked to human or animal infection. Sequence analysis of over 3 kb of random DNA from two of these matching strains demonstrated identical genomic sequence to SAR-1, corroborating the exact match seen at the seven MLST genes. Except for potential variation in elements of accessory genome, these cultivable strains were therefore absolutely identical genetic ‘clones’ of the strain from which the SAR-1 metagenome originated. However, they were not able to grow well in sea water suggesting that this was not their natural habitat. Several of the cultivable clones were recovered as contaminants of medical devices and products, adding weight to the argument that the SAR-1 metagenome sequence was also derived from sample contamination (Mahenthiralingam et al. 2006). The ability of MLST to work from a starting point of DNA sequence and trace back to cultivable bacteria which are identical genetic matches is testament to its resolving power as a typing tool. Environmental diversity Characterization of Bcc species in the environment has been more limited than investigation of their clinical epidemiology. The original description of P. cepacia (Burkholder 1950) and many of the initial studies on this group of Pseudomonads focussed on their environmental microbiology (reviewed by Parke and Gurian-Sherman 2001). Plants of the Gramineae group appear to be particularly important rhizospheric hosts for Bcc bacteria. Studies reclassifying species historically associated with certain plant species have shown the following: several Bcc species are associated with grasses and maize (Parke and Gurian-Sherman 2001); B. vietnamiensis occurs in high numbers on the rice rhizosphere; B. ambifaria is present on pea roots (Coenye et al. 2001a) and B. cepacia ⁄ B. cenocepacia species are associated with wheat rhizosphere (Parke and Gurian-Sherman 2001; Vandamme et al. 2003). Systematic studies examining the environmental distribution of Bcc species have only recently been published and to date these have focussed on the populations interacting with crop plants as one of the most important


1545



habitats for these bacteria (Parke and Gurian-Sherman 2001). In 2005, Ramette et al. (2005) examined the diversity of Bcc species present in the rhizosphere of maize plants grown in Michigan, United States. Burkholderia ambifaria and B. cepacia (64% and 20% of the total number of isolates, respectively) were the most commonly cultivated species; B. multivorans, B. cenocepacia, B. stabilis, B. dolosa and B. pyrrocinia were also found suggesting that the maize rhizosphere is a very rich environmental source of these bacteria. Payne et al. (2006) expanded these studies on the maize rhizosphere and applied a cultivation-independent approach based on the amplification and sequence analysis of the recA gene to total DNA extracted from the United States maize rhizosphere soil samples (Ramette et al. 2005). In addition to detecting recA genes associated with the cultivated species, B. ambifaria and B. pyrrocinia, 90% of the Burkholderia recA genes identified in the study were novel phylotypes suggesting that a wealth of novel species are present in the maize rhizosphere. An MLST-based study of Bcc populations associated with the Italian maize rhizosphere also showed that B. ambifaria was the most commonly encountered species associated with this crop (Dalmastri et al. 2007). However, in contrast to the United States-based study (Ramette et al. 2005), the novel groups BCC5 and BCC6 (Table 1) were the next most dominant groups cultivated from these maize samples suggesting that different global geographic locations lead to different Bcc populations being present on maize (Dalmastri et al. 2007). Is the natural environment a potential source of infection with Bcc species? This has been a major research question and an important issue in the risk assessment of bacteria related to opportunistic pathogens that are used as biopesticides (Parke and Gurian-Sherman 2001). In the past, researchers have often referred to environmental and clinical strains of B. cepacia as if they were in some way distinct from one another. Unlike P. aeruginosa which may be carried by around 10% of humans, e.g. as a gut colonizer, Bcc bacteria have not to date been recovered from human sources other than the sites of infection. Therefore in the absence of patient-to-patient transmission in CF infection, the natural environment must be the source of Bcc infection. Genetic typing studies have now shown that many instances of identical strains of Bcc bacteria residing in the environment are being encountered as infections. In a short letter in 2000, Govan et al. (2000) reported that a CF isolate of B. cepacia was identical by PFGE and several other characteristics to the type strain for this species, ATCC 25416 that was originally isolated from onion rot. More substantial evidence of identical strains occurring in both CF infection and the natural environment was subsequently described by 1546

Lipuma et al. (2002). They found by using PFGE fingerprinting and other genetic typing methods that the epidemic CF strain prevalent in the Mid-Atlantic region of the United States and Europe, PHDC (Chen et al. 2001; Coenye et al. 2004), was also abundant in onion field soils in New York State. Further instances of genetic identity between environmental and clinical isolates have now been described. PFGE fingerprinting was used to show genomic identity within pairs of B. cepacia, B. ambifaria and B. stabilis isolates recovered from the natural environment and clinical infections (Payne et al. 2005). Baldwin et al. (2007) recently presented a highly comprehensive evidence by MLST that greater that 20% of the clinical isolates were indistinguishable from environmental isolates. Multiple clinical isolates of B. cepacia, B. multivorans, B. cenocepacia IIIA, B. cenocepacia IIIB, B. stabilis, B. vietnamiensis and B. ambifaria were identified that each showed nucleotide identity at all seven MLST loci with environmental isolates recovered from diverse environments such as river water, onion, radish, maize rhizosphere, pharmaceutical solutions, hospital equipment, shampoo and industrial settings. It is interesting that man-made industrial settings also appear to be prone to contamination with Bcc bacteria (Table 1) and represent an environment that is worth further systematic study. Contamination with Bcc bacteria may also occur in the oil and fuel industry, which is not surprising given that these bacteria are able to grow on a diverse range of hydrocarbon substrates (reviewed in O’Sullivan and Mahenthiralingam 2005; O’Sullivan et al. 2007). Using MLST we have found several Bcc sequence types that overlap clinical, industrial and natural sources (Fig. 2), which clearly demonstrates the remarkably versatility of this group of bacteria to survive and grow in highly diverse environments. Environmental isolates within the Cardiff collection show an interesting distribution (Table 1). We have no environmental isolates for B. cenocepacia IIID and very few for B. multivorans, B. cenocepacia IIIA, B. stabilis and B. dolosa, despite the fact that all these species have been frequently encountered as CF and other clinical infections (Table 1); this odd distribution suggests that as yet we have not fully defined a natural habitat where these species are prominent. It is also striking that all 16 isolates of B. cenocepacia IIIC are environmental and to date this species has not been recovered from CF infection; all other B. cenocepacia subgroups are all highly associated with this disease. Similarly for B. ambifaria, B. pyrrocinia, BCC5 and BCC6, over 89% of the isolates from each species are from the natural environment (Table 1). Further systematic studies of the natural environment will be required to show if these collection-based statistics are truly reflective of the natural distribution of these species.




Natural environment

Clinical infection

30

15

206

3 4

6

16

Industrial processes Figure 2 Overlap of Burkholderia cepacia complex (Bcc) sequence types from different sources. The 798 isolates in the Cardiff collection are discriminated into 376 sequence types (ST). The total number of ST and the occurrence of identical clones (overlapping ST) recovered from clinical, industrial and natural sources are shown.

Natural functional diversity The ability for Burkholderia species to thrive in the diverse range of environments is testament to the fact they can be considered as one of the most versatile groups of gram-negative bacteria. Recently, endosymbiotic Burkholderia species have also been characterized (Van Oevelen et al. 2002; Partida-Martinez et al. 2007), adding a further very exciting dimension to their natural biology. All Burkholderia species possess very large genomes ranging between 6 and 9 Mb in size, and it is this huge genetic capacity which underpins their versatility in disease and natural biology (Mahenthiralingam and Drevinek 2007). All species also separate this DNA into two or more chromosomal replicons (Mahenthiralingam et al. 2005) which may add greater flexibility in the acquisition, loss and expression of genes. Over 30 Burkholderia genome-sequencing projects have been initiated, and draft ⁄ finished genomes are available for B. cepacia group K, B. multivorans, B. cenocepacia, B. vietnamiensis, B. ambifaria and B. dolosa strains. All Bcc species characterized have a minimum of three large chromosomal replicons and an average genome size of 7Æ5 Mb (Mahenthiralingam and Drevinek 2007), and hence encode over 7000 genes. It is also clear from their genomes that 10% or more of these 7000 genes appear to have been acquired through horizontal gene transfer and reside as elements of foreign DNA such as genomic islands, integrated phages or plasmids. The extensive genomic dataset has been very useful

in allowing the genetic basis for the pathogenesis and environmental versatility of Bcc bacteria to be studied in more detail. Bcc bacteria play very important roles in the production of many important crops such as peas, maize, rice and wheat, protecting seedling plants from attack by fungal and nematode pathogens and leading to general growth promotion (Parke and Gurian-Sherman 2001). These beneficial traits have led to the commercial use of several strains as biological control agents in the United States (Parke and Gurian-Sherman 2001). However, after a risk assessment in 1999 of Bcc bacteria as model opportunistic pathogens with biopesticidal uses, the United States Environmental Protection Agency placed a moratorium on the new registrations of products containing these bacteria (see Federal Register; http://www.epa.gov /fedrgstr/EPA-PEST/2004/September/Day-29/p21695.htm). Several Bcc bacteria also have great biotechnological potential as bioremediation agents and are capable of degrading a number of man-made pollutants (O’Sullivan and Mahenthiralingam 2005). For example, B. vietnamiensis strain G4 is one of the most efficient degraders of trichloroethylene and has been used in field trial to clean up aquifers contaminated with this pollutant (O’Sullivan and Mahenthiralingam 2005). Possession of considerable biotechnological potential and the contrasting capacity to cause opportunistic infection represent a dilemma that has been the subject of considerable debate (Parke and Gurian-Sherman 2001; Chiarini et al. 2006). It is clear that certain species within the Bcc like B. ambifaria cause very low levels of the disease, but does this make them safe for use as a biotechnological agent that will be environmentally dispersed? A potential way around this dilemma is to engineer strains of Bcc bacteria that are attenuated for virulence functions yet retain their biotechnologically useful traits. However, as Bcc virulence is known to be multifactorial, for attenuation to be effective, multiple pathogenic traits will need to be knocked out in potential biotechnological strains. Using a powerful form of transposon mutagenesis in combination with genome sequence data, O’Sullivan et al. (2007) were able to identify multiple genes that were important for both pollutant degradation and rhizosphere colonization in B. vietnamiensis strain G4. Mapping these environmentally important genes may allow the transfer of beneficial genetic pathways to biotechnological host strains which do not have the capacity to cause disease. Alternatively this molecular knowledge may enable the creation of Bcc strains that are attenuated in multiple virulence gene functions yet retain their ecologically useful genetic content. Indeed, one of the mutations identified in B. vietnamiensis G4 was the inactivation of a large surface adhesin gene, a YadA homologue, which


1547



encodes an invasin protein that is an important virulence factor in many bacteria (O’Sullivan et al. 2007). The ecological phenotype expressed by the YadA mutant was one of enhanced rhizosphere colonization and hence it potentially makes an ideal platform to begin to engineer safe biotechnological agents that are better able to interact with plants, yet attenuated in virulence functions. Conclusions In the years since the discovery that B. cepacia isolates were composed of multiple genetically distinct species (Vandamme et al. 1997), considerable advances in our understanding of the Bcc have been made. We have very good molecular tests to identify each formally named Bcc species and MLST is proving very valuable in global strain tracking and in the identification of new potentially complex species. While the species identity of the major Bcc CF pathogens has been largely worked out, the epidemiology of infection in this population is clearly changing and will need on-going surveillance. Improving treatment and developing therapeutic agents that can eradicate chronic lung infection with these antibiotic-resistant bacteria should also be a major priority for future CF microbiology research. Understanding the risks posed by the environment and exact routes of acquisition of CF infection from this Bcc reservoir also needs to be studied further. The debate over what constitutes an environmental or clinical strain has also been answered. Genetic studies have now clearly shown that at the genomic and nucleotide levels, identity between environmental and clinical isolates exists, demonstrating that the capacity to grow freely in natural habitats and cause opportunistic infections can be encoded within any given strain of the Bcc. Future studies on the biotechnological use of Bcc should now focus on safe ways to harness their great potential to improve agriculture and reduce global pollution. Acknowledgements The authors thank the following funding agencies for their support to this research on the B. cepacia complex: the Wellcome Trust (grants 072853 and 075586), the UK Cystic Fibrosis Trust (grant PJ535) and the Natural Environment Research Council (grant NER ⁄ T ⁄ S ⁄ 2001 ⁄ 00299). References Agodi, A., Mahenthiralingam, E., Barchitta, M., Giannino, V., Sciacca, A. and Stefani, S. (2001) Burkholderia cepacia complex infection in Italian patients with cystic fibrosis: prevalence, epidemiology, and genomovar status. J Clin Microbiol 39, 2891–2896.

1548

Baldwin, A., Mahenthiralingam, E., Thickett, K.M., Honeybourne, D., Maiden, M.C., Govan, J.R., Speert, D.P., Lipuma, J.J. et al. (2005) Multilocus sequence typing scheme that provides both species and strain differentiation for the Burkholderia cepacia complex. J Clin Microbiol 43, 4665–4673. Baldwin, A., Mahenthiralingam, E., Drevinek, P., Vandamme, P., Govan, J.R.W., Waine, D.J., Lipuma, J.J., Chiarini, L. et al. (2007) Environmental Burkholderia cepacia complex isolates in human infection. Emerg Infect Dis 13, 458–461. Biddick, R., Spilker, T., Martin, A. and Lipuma, J.J. (2003) Evidence of transmission of Burkholderia cepacia, Burkholderia multivorans and Burkholderia dolosa among persons with cystic fibrosis. FEMS Microbiol Lett 228, 57–62. Brown, A.R. and Govan, J.R. (2007) Assessment of fluorescent in situ hybridization and PCR-based methods for rapid identification of Burkholderia cepacia complex organisms directly from sputum samples. J Clin Microbiol 45, 1920– 1926. Burkholder, W. (1950) Sour skin, a bacterial rot of onion bulbs. Phytopathology 64, 468–475. Carvalho, A.P., Ventura, G.M., Pereira, C.B., Leao, R.S., Folescu, T.W., Higa, L., Teixeira, L.M., Plotkowski, M.C. et al. (2007) Burkholderia cenocepacia, B. multivorans, B. ambifaria and B. vietnamiensis isolates from cystic fibrosis patients have different profiles of exoenzyme production. Apmis 115, 311–318. Chen, J.S., Witzmann, K.A., Spilker, T., Fink, R.J. and Lipuma, J.J. (2001) Endemicity and inter-city spread of Burkholderia cepacia genomovar III in cystic fibrosis. J Pediatr 139, 643–649. Chiarini, L., Bevivino, A., Dalmastri, C., Tabacchioni, S. and Visca, P. (2006) Burkholderia cepacia complex species: health hazards and biotechnological potential. Trends Microbiol 14, 277–286. Coenye, T. and Vandamme, P. (2003) Diversity and significance of Burkholderia species occupying diverse ecological niches. Environ Microbiol 5, 719–729. Coenye, T., Mahenthiralingam, E., Henry, D., Lipuma, J.J., Laevens, S., Gillis, M., Speert, D.P. and Vandamme, P. (2001a) Burkholderia ambifaria sp. nov., a novel member of the Burkholderia cepacia complex including biocontrol and cystic fibrosis-related isolates. Int J Syst Evol Microbiol 51, 1481–1490. Coenye, T., Vandamme, P., Govan, J.R. and Lipuma, J.J. (2001b) Taxonomy and identification of the Burkholderia cepacia complex. J Clin Microbiol 39, 3427–3436. Coenye, T., Spilker, T., Van Schoor, A., Lipuma, J.J. and Vandamme, P. (2004) Recovery of Burkholderia cenocepacia strain PHDC from cystic fibrosis patients in Europe. Thorax 59, 952–954. Cunha, M.V., Leitao, J.H., Mahenthiralingam, E., Vandamme, P., Lito, L., Barreto, C., Salgado, M.J. and SA-Correia, I. (2003) Molecular analysis of Burkholderia cepacia complex



isolates from a Portuguese cystic fibrosis center: a 7-year study. J Clin Microbiol 41, 4113–4120. Cunha, M.V., Pinto-De-oliveira, A., Meirinhos-Soares, L., Salgado, M.J., Melo-Cristino, J., Correia, S., Barreto, C. and Sa-correia, I. (2007) Exceptionally high representation of Burkholderia cepacia among B. cepacia complex isolates recovered from the major Portuguese cystic fibrosis center. J Clin Microbiol 45, 1628–1633. Dalmastri, C., Baldwin, A., Tabacchioni, S., Bevivino, A., Mahenthiralingam, E., Chiarini, L. and Dowson, C. (2007) Investigating Burkholderia cepacia complex populations recovered from Italian maize rhizosphere by multilocus sequence typing. Environ Microbiol 9, 1632–1639. Frangolias, D.D., Mahenthiralingam, E., Rae, S., Raboud, J.M., Davidson, A.G., Wittmann, R. and Wilcox, P.G. (1999) Burkholderia cepacia in cystic fibrosis. Variable disease course. Am J Respir Crit Care Med 160, 1572–1577. Geftic, S.G., Heymann, H. and Adair, F.W. (1979) Fourteenyear survival of Pseudomonas cepacia in a salts solution preserved with benzalkonium chloride. Appl Environ Microbiol 37, 505–510. Gillis, M., Vanvan, T., Bardin, R., Goor, M., Hebbar, P., Willems, A., Segers, P., Kersters, K. et al. (1995) Polyphasic taxonomy in the genus Burkholderia leading to an emended description of the genus and proposition of Burkholderia vietnamiensis sp-nov for N-2-fixing isolates from rice in Vietnam. Int J Syst Bacteriol 45, 274–289. Govan, J.R. (2003) The Burkholderia cepacia complex and cytokine induction: an inflammatory tale. Pediatr Res 54, 294– 296. Govan, J.R., Brown, P.H., Maddison, J., Doherty, C.J., Nelson, J.W., Dodd, M., Greening, A.P. and Webb, A.K. (1993) Evidence for transmission of Pseudomonas cepacia by social contact in cystic fibrosis. Lancet 342, 15–19. Govan, J.R., Hughes, J.E. and Vandamme, P. (1996) Burkholderia cepacia: medical, taxonomic and ecological issues. J Med Microbiol 45, 395–407. Govan, J.R.W., Balendreau, J. and Vandamme, P. (2000) Burkholderia cepacia – friend and foe. Asm News 66, 124–125. Govan, J.R., Brown, A.R. and Jones, A.M. (2007) Evolving epidemiology of Pseudomonas aeruginosa and the Burkholderia cepacia complex in cystic fibrosis lung infection. Future Microbiol 2, 153–164. Henry, D.A., Mahenthiralingam, E., Vandamme, P., Coenye, T. and Speert, D.P. (2001) Phenotypic methods for determining genomovar status of the Burkholderia cepacia complex. J Clin Microbiol 39, 1073–1078. Isles, A., Maclusky, I., Corey, M., Gold, R., Prober, C., Fleming, P. and Levison, H. (1984) Pseudomonas cepacia infection in cystic fibrosis: an emerging problem. J Pediatr 104, 206–210. Johnson, W.M., Tyler, S.D. and Rozee, K.R. (1994) Linkage analysis of geographic and clinical clusters in Pseudomonas cepacia infections by multilocus enzyme electrophoresis and ribotyping. J Clin Microbiol 32, 924–930.


Jones, A.M., Dodd, M.E., Govan, J.R., Barcus, V., Doherty, C.J., Morris, J. and Webb, A.K. (2004) Burkholderia cenocepacia and Burkholderia multivorans: influence on survival in cystic fibrosis. Thorax 59, 948–951. Kalish, L.A., Waltz, D.A., Dovey, M., Potter-Bynoe, G., Mcadam, A.J., Lipuma, J.J., Gerard, C. and Goldmann, D. (2006) Impact of Burkholderia dolosa on lung function and survival in cystic fibrosis. Am J Respir Crit Care Med 173, 421–425. Kidd, T.J., Bell, S.C. and Coulter, C. (2003) Genomovar diversity amongst Burkholderia cepacia complex isolates from an Australian adult cystic fibrosis unit. Eur J Clin Microbiol Infect Dis 22, 434–437. Larsen, G.Y., Stull, T.L. and Burns, J.L. (1993) Marked phenotypic variability in Pseudomonas cepacia isolated from a patient with cystic fibrosis. J Clin Microbiol 31, 788–792. Lipuma, J.J., Dasen, S.E., Nielson, D.W., Stern, R.C. and Stull, T.L. (1990) Person-to-person transmission of Pseudomonas cepacia between patients with cystic fibrosis. Lancet 336, 1094–1096. Lipuma, J.J., Spilker, T., Gill, L.H., Campbell, P.W. III, Liu, L. and Mahenthiralingam, E. (2001) Disproportionate distribution of Burkholderia cepacia complex species and transmissibility markers in cystic fibrosis. Am J Respir Crit Care Med 164, 92–96. Lipuma, J.J., Spilker, T., Coenye, T. and Gonzalez, C.F. (2002) An epidemic Burkholderia cepacia complex strain identified in soil. Lancet 359, 2002–2003. Mahenthiralingam, E. and Drevinek, P. (2007) Comparative genomics of Burkholderia species. In Burkholderia: Molecular Biology and Genomics ed. Coenye, T. and Vandamme, P. Norwich, UK: Horizon Scientific Press. Mahenthiralingam, E., Campbell, M.E., Foster, J., Lam, J.S. and Speert, D.P. (1996a) Random amplified polymorphic DNA typing of Pseudomonas aeruginosa isolates recovered from patients with cystic fibrosis. J Clin Microbiol 34, 1129–1135. Mahenthiralingam, E., Campbell, M.E., Henry, D.A. and Speert, D.P. (1996b) Epidemiology of Burkholderia cepacia infection in patients with cystic fibrosis: analysis by randomly amplified polymorphic DNA fingerprinting. J Clin Microbiol 34, 2914–2920. Mahenthiralingam, E., Bischof, J., Byrne, S.K., Radomski, C., Davies, J.E., Av-Gay, Y. and Vandamme, P. (2000) DNAbased diagnostic approaches for identification of Burkholderia cepacia complex, Burkholderia vietnamiensis, Burkholderia multivorans, Burkholderia stabilis, and Burkholderia cepacia genomovars I and III. J Clin Microbiol 38, 3165– 3173. Mahenthiralingam, E., Vandamme, P., Campbell, M.E., Henry, D.A., Gravelle, A.M., Wong, L.T., Davidson, A.G., Wilcox, P.G. et al. (2001) Infection with Burkholderia cepacia complex genomovars in patients with cystic fibrosis: virulent transmissible strains of genomovar III can replace Burkholderia multivorans. Clin Infect Dis, 33, 1469–75.


1549



Mahenthiralingam, E., Urban, T.A. and Goldberg, J.B. (2005) The multifarious, multireplicon Burkholderia cepacia complex. Nat Rev Microbiol 3, 144–156. Mahenthiralingam, E., Baldwin, A., Drevinek, P., Vanlaere, E., Vandamme, P., Lipuma, J.J. and Dowson, C.G. (2006) Multilocus sequence typing breathes life into a microbial metagenome. PLoS ONE 1, e17. Nzula, S., Vandamme, P. and Govan, J.R. (2002) Influence of taxonomic status on the in vitro antimicrobial susceptibility of the Burkholderia cepacia complex. J Antimicrob Chemother 50, 265–269. O’Sullivan, L.A. and Mahenthiralingam, E. (2005) Biotechnological potential within the genus Burkholderia. Lett Appl Microbiol 41, 8–11. O’Sullivan, L.A., Weightman, A.J., Jones, T.H., Marchbank, A.M., Tiedje, J.M. and Mahenthiralingam, E. (2007) Identifying the genetic basis of ecologically and biotechnologically useful functions of the bacterium Burkholderia vietnamiensis. Environ Microbiol 9, 1017–1034. Ortega, X.P., Cardona, S.T., Brown, A.R., Loutet, S.A., Flannagan, R.S., Campopiano, D.J., Govan, J.R. and Valvano, M.A. (2007) A putative gene cluster for aminoarabinose biosynthesis is essential for Burkholderia cenocepacia viability. J Bacteriol 189, 3639–3644. Parke, J.L. and Gurian-Sherman, D. (2001) Diversity of the Burkholderia cepacia complex and implications for risk assessment of biological control strains. Annu Rev Phytopathol 39, 225–258. Partida-Martinez, L.P., Monajembashi, S., Greulich, K.O. and Hertweck, C. (2007) Endosymbiont-dependent host reproduction maintains bacterial–fungal mutualism. Curr Biol 17, 773–777. Payne, G.W., Vandamme, P., Morgan, S.H., Lipuma, J.J., Coenye, T., Weightman, A.J., Jones, T.H. and Mahenthiralingam, E. (2005) Development of a recA gene-based identification approach for the entire Burkholderia genus. Appl Environ Microbiol 71, 3917–3927. Payne, G.W., Ramette, A., Rose, H.L., Weightman, A.J., Jones, T.H., Tiedje, J.M. and Mahenthiralingam, E. (2006) Application of a recA gene-based identification approach to the maize rhizosphere reveals novel diversity in Burkholderia species. FEMS Microbiol Lett 259, 126–132. Perry, B. (2001) Cosmetic microbiology. Microbiol Today 28, 185–187. Plesa, M., Kholti, A., Vermis, K., Vandamme, P., Panagea, S., Winstanley, C. and Cornelis, P. (2004) Conservation of the opcL gene encoding the peptidoglycan-associated outermembrane lipoprotein among representatives of the Burkholderia cepacia complex. J Med Microbiol 53, 389–398. Ramette, A., Lipuma, J.J. and Tiedje, J.M. (2005) Species abundance and diversity of Burkholderia cepacia complex in the environment. Appl Environ Microbiol 71, 1193–1201. Reik, R., Spilker, T. and Lipuma, J.J. (2005) Distribution of Burkholderia cepacia complex species among isolates

1550

recovered from persons with or without cystic fibrosis. J Clin Microbiol 43, 2926–2928. Saiman, L. and Siegel, J. (2004) Infection control in cystic fibrosis. Clin Microbiol Rev 17, 57–71. Segonds, C., Heulin, T., Marty, N. and Chabanon, G. (1999) Differentiation of Burkholderia species by PCR-restriction fragment length polymorphism analysis of the 16S rRNA gene and application to cystic fibrosis isolates. J Clin Microbiol 37, 2201–2208. Speert, D.P., Henry, D., Vandamme, P., Corey, M. and Mahenthiralingam, E. (2002) Epidemiology of Burkholderia cepacia complex in patients with cystic fibrosis, Canada. Emerg Infect Dis 8, 181–187. Sun, L., Jiang, R.Z., Steinbach, S., Holmes, A., Campanelli, C., Forstner, J., Sajjan, U., Tan, Y. et al. (1995) The emergence of a highly transmissible lineage of cbl+ Pseudomonas (Burkholderia) cepacia causing CF centre epidemics in North America and Britain. Nat Med 1, 661–666. Tablan, O.C., Chorba, T.L., Schidlow, D.V., White, J.W., Hardy, K.A., Gilligan, P.H., Morgan, W.M., Carson, L.A. et al. (1985) Pseudomonas cepacia colonization in patients with cystic fibrosis: risk factors and clinical outcome. J Pediatr 107, 382–387. Van Oevelen, S., De Wachter, R., Vandamme, P., Robbrecht, E. and Prinsen, E. (2002) Identification of the bacterial endosymbionts in leaf galls of Psychotria (Rubiaceae, angiosperms) and proposal of ‘Candidatus Burkholderia kirkii’ sp. nov. Int J Syst Evol Microbiol 52, 2023–2027. Vandamme, P., Holmes, B., Vancanneyt, M., Coenye, T., Hoste, B., Coopman, R., Revets, H., Lauwers, S. et al. (1997) Occurrence of multiple genomovars of Burkholderia cepacia in cystic fibrosis patients and proposal of Burkholderia multivorans sp. nov. Int J Syst Bacteriol 47, 1188–1200. Vandamme, P., Mahenthiralingam, E., Holmes, B., Coenye, T., Hoste, B., De Vos, P., Henry, D. and Speert, D.P. (2000) Identification and population structure of Burkholderia stabilis sp. nov. (formerly Burkholderia cepacia genomovar IV). J Clin Microbiol 38, 1042–1047. Vandamme, P., Henry, D., Coenye, T., Nzula, S., Vancanneyt, M., Lipuma, J.J., Speert, D.P., Govan, J.R. et al. (2002) Burkholderia anthina sp. nov. and Burkholderia pyrrocinia, two additional Burkholderia cepacia complex bacteria, may confound results of new molecular diagnostic tools. FEMS Immunol Med Microbiol 33, 143–149. Vandamme, P., Holmes, B., Coenye, T., Goris, J., Mahenthiralingam, E., Lipuma, J.J. and Govan, J.R. (2003) Burkholderia cenocepacia sp. nov– a new twist to an old story. Res Microbiol 154, 91–96. Vanlaere, E., Hansraj, F., Vandamme, P.A. and Govan, J.R. (2006) Growth in Stewart’s medium is a simple, rapid and inexpensive screening tool for the identification of Burkholderia cepacia complex. J Cyst Fibros 5, 137–139. Venter, J.C., Remington, K., Heidelberg, J.F., Halpern, A.L., Rusch, D., Eisen, J.A., Wu, D., Paulsen, I. et al. (2004)



Environmental genome shotgun sequencing of the Sargasso Sea. Science 304, 66–74. Vermis, K., Coenye, T., Mahenthiralingam, E., Nelis, H.J. and Vandamme, P. (2002) Evaluation of species-specific recAbased PCR tests for genomovar level identification within the Burkholderia cepacia complex. J Med Microbiol 51, 937–940. Vermis, K., Coenye, T., Lipuma, J.J., Mahenthiralingam, E., Nelis, H.J. and Vandamme, P. (2004) Proposal to accommodate Burkholderia cepacia genomovar VI as Burkholderia dolosa sp. nov. Int J Syst Evol Microbiol 54, 689–691. Waine, D.J., Henry, D.A., Baldwin, A., Speert, D.P., Honeybourne, D., Mahenthiralingam, E. and Dowson, C.G. (2007) Reliability of multilocus sequence typing of the Burkholderia cepacia complex in cystic fibrosis. J Cyst Fibros 6, 215–219.


Whiteford, M.L., Wilkinson, J.D., Mccoll, J.H., Conlon, F.M., Michie, J.R., Evans, T.J. and Paton, J.Y. (1995) Outcome of Burkholderia (Pseudomonas) cepacia colonisation in children with cystic fibrosis following a hospital outbreak. Thorax 50, 1194–1198. Winkelstein, J.A., Marino, M.C., Johnston, R.B. Jr., Boyle, J., Curnutte, J., Gallin, J.I., Malech, H.L., Holland, S.M. et al. (2000) Chronic granulomatous disease. Report on a national registry of 368 patients. Medicine (Baltimore) 79, 155–169. Yabuuchi, E., Kosako, Y., Oyaizu, H., Yano, I., Hotta, H., Hashimoto, Y., 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.


1551