Lack of cable pili expression by cblA-containing Burkholderia cepacia complexa

Umadevi Sajjan1, Lixia Liu2, Annie Lu1, Theodore Spilker2, Janet Forstner1 and John J. LiPuma2

Departments of Structural Biology and Biochemistry, Hospital for Sick Children, Toronto, Ontario, Canada1
Department of Pediatrics and Communicable Diseases, University of Michigan, 8323 MSRB III, Box 0646, 1150 W. Medical Center Drive, Ann Arbor, MI 48109, USA2

Author for correspondence: John J. LiPuma. Tel: +1 734 936 9767. Fax: +1 734 764 4279. e-mail: jlipuma{at}umich.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The Burkholderia cepacia complex consists of several closely related bacterial species (or genomovars) which although generally not pathogenic for healthy individuals, contribute significantly to morbidity and mortality among persons with cystic fibrosis (CF). Certain B. cepacia complex strains are more frequently recovered from CF sputum cultures than are others, and these typically reside in genomovar III. The ET12 clone is a genomovar III strain that predominates among CF patients in Canada and the United Kingdom and is characterized by distinctive cblA-encoded pili that have a cable-like morphology. In a previous survey of B. cepacia complex isolates recovered from 606 CF patients in the US, a single genomovar III ET12 isolate (isolate AU0007) was identified; several cblA-containing genomovar I isolates, however, were also detected. In the study reported here, analysis by PFGE revealed several distinct strain types among these genomovar I isolates, and sequence analysis of their cblA genes demonstrated 87·8–88·4% identity to the ET12 cblA sequence. Southern analysis indicated that the cblA variant from each genomovar I isolate resides on a 4 kbp EcoRI fragment, in contrast to ET12 isolates, in which cblA localizes to a 5 kbp EcoRI fragment. Western blot assay indicated expression of the 16 kDa major pilin subunit by ET12 isolates, including AU0007, but neither whole-cell nor surface-protein extracts of the genomovar I reacted. Electron microscopy revealed the complete absence of pili expression by the genomovar I isolates. In contrast to typical ET12 isolates, AU0007 appeared to be hyperpiliated with rigid pili that lacked the cable morphology and did not bind cytokeratin 13, which has been previously identified as the epithelial cell receptor for the ET12 cable-pili-associated adhesin.

Keywords: cblA gene, bacterial adherence, cystic fibrosis

Abbreviations: CF, cystic fibrosis; CK13, cytokeratin 13

a The GenBank accession numbers for the complete cblA nucleotide sequences for the isolates listed in Table 1 are AF455151AF455162.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The Burkholderia cepacia complex consists of several phylogenetically closely related yet distinct species of soil commensal and phytopathogenic bacteria (Coenye et al., 2001 ). New binomial designations have been assigned to some species within this group (i.e. B. multivorans, B. stabilis, B. vietnamiensis, B. ambifaria, B. anthina and B. pyrrocinia), while the remaining species continue to be referred to as genomovars (i.e. B. cepacia genomovars I, III and VI) pending the identification of species-specific phenotypic characteristics. Ultimately, the name B. cepacia will be restricted to genomovar I, the species that contains the original species type strain.

Certain B. cepacia complex strains have attracted considerable attention because of their potential commercial use as biopesticidal and bioremedial agents (LiPuma & Mahenthiralingam, 1999 ; Parke & Gurian-Sherman, 2001 ). At the same time, there has been a growing appreciation of the role B. cepacia complex species play as opportunistic human pathogens. Although generally not pathogenic for healthy humans, these species are capable of causing severe infection in certain vulnerable populations, particularly persons with cystic fibrosis (CF), the most common lethal genetic disease among whites (LiPuma, 1998 ). In CF, respiratory tract infection with B. cepacia complex is associated with increased rates of morbidity and mortality. In fact, following infection, a significant proportion of CF patients will die, with rapidly progressive necrotizing pneumonia and sepsis.

The bacterial virulence factors and pathogenic mechanisms involved in human infection due to B. cepacia complex remain to be elucidated. Nevertheless, it is clear that some specific strains are more frequently recovered from CF patients than are others. One such ‘epidemic’ strain, the ET12 lineage, is a genomovar III strain that predominates among patients in Ontario, Canada, and is found in about one-third of B. cepacia-infected patients in the United Kingdom (Pitt et al., 1996 ; Sun et al., 1995 ). This strain expresses a distinctive pilus (the cblA-encoded cable pilus) and an associated adhesin that mediate bacterial binding to CF respiratory epithelium (Sajjan et al., 1995 , 2000a , b ). The cognate receptor for cable-pili expressing B. cepacia, cytokeratin 13 (CK13), is a 55 kDa protein that is enriched in CF epithelia (Sajjan et al., 2000a ). Because the expression of cable pili may provide an important pathogenic mechanism contributing to inter-patient spread of B. cepacia complex, differential infection control measures based on the presence (or absence) of cblA have been proposed (Clode et al., 2000 ).

In a recent study only one of 606 B. cepacia complex-infected US CF patients harboured a cblA-containing ET12 strain (LiPuma et al., 2001 ). Analyses employing cblA-specific dot-blot and PCR assays, and genotyping by PFGE indicated, however, that nine additional patients were infected with non-ET12 strains also positive for cblA sequences. An additional cblA-positive isolate was identified from stream sediment. Interestingly, all 10 of these cblA-positive, non-ET12 isolates were B. cepacia genomovar I, a species not frequently found in CF sputum culture (LiPuma et al., 2001 ). In this study we investigated the expression of cable pili by these strains and characterized the cblA gene variants found therein.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains with cblA sequences.
Bacterial strains were obtained during a study of B. cepacia complex recovered from persons with CF as previously reported (LiPuma et al., 2001 ). In that study, 10 of 606 isolates (from 606 CF patients receiving care in 132 treatment centres in 105 US cities) were identified by dot-blot hybridization as containing genomic DNA sequences homologous to cblA. By using 16S rDNA- and recA-based species-specific PCR assays (LiPuma et al., 1999 ; Mahenthiralingam et al., 2000a ) one of these isolates (AU0007) was found to be genomovar III whereas the remaining nine were identified as genomovar I. For the present study, an additional isolate with cblA sequence (BC20) was identified from among B. cepacia complex isolates recovered from stream sediment (Wise et al., 1995 ); this too was identified as genomovar I. BC7 (Sajjan et al., 1995 ; Mahenthiralingam et al., 2000b ) and J2315 (Govan et al., 1993 ; Mahenthiralingam et al., 2000b ) are representatives of the genomovar III ET12 lineage. Although both isolates have been shown in previous work to contain cblA genes, J2315 lacks the 22 kDa adhesin and does not bind to CK13 (Sajjan et al., 2000b ). In contrast, BC7 both expresses this adhesin and binds CK13.

ATCC 25416 (American Type Culture Collection, Manassas, VA, USA) is a cblA-negative genomovar I strain and was used as a negative control.

Bacterial growth conditions.
For DNA sequence determination and PFGE analyses, bacteria from frozen stock were recovered on Mueller–Hinton agar (Difco) after incubation at 35 °C for 24–48 h. For Southern and Western blot analyses, and for electron microscopy, bacteria from frozen stock were grown on brain-heart infusion agar (Becton Dickinson). For CK13-binding assay, bacteria were grown in tryptic soy broth (Difco), as described by Sajjan et al. (2000b ).

cblA and upstream DNA sequence analysis.
By using cblA-specific PCR primers and reaction conditions previously described (Sajjan et al., 1995 ) DNA fragments of the predicted size were amplified from the 11 study isolates (Table 1) as well as from the positive control strains BC7 and J2315. A second PCR assay employing forward primer 5'-GAGCTCGAATTCGATATCGAGTGG-3' and reverse primer 5'-CTTGTCGTTCGTGAAGATCTTCGTG-3' was designed to amplify DNA sequences immediately upstream of the cblA start codon. This assay was carried out in a 50 µl final volume that included 1·5 mM MgCl2, 0·8 mM dNTPs, 0·4 mM of each primer, 100 ng template DNA and 1 U Taq DNA polymerase. PCR conditions were the same as for the cblA PCR except that an annealing temperature of 62 °C was used. Nucleotide sequences of cblA and the upstream segments were determined by using an ABI PRISM model 3700 DNA sequencer (Perkin-Elmer Applied Biosystems). Sequence analyses were performed using EditSeq and MegAlign software (DNAStar). The GenBank accession numbers for the complete cblA nucleotide sequences for the isolates listed in Table 1 are AF455151AF455162.


View this table:
[in this window]
[in a new window]
 
Table 1. B. cepacia complex isolates with cblA sequences

 
PFGE.
Isolate genotyping by using macrorestriction digest and PFGE was performed as previously described (Chen et al., 2001 ). Strain identity was determined by using published criteria (Tenover et al. 1995 ).

Southern blot analysis.
Chromosomal DNA (10 µg) from each isolate was digested with EcoRI and subjected to electrophoresis in 0·8% agarose. The resulting DNA fragments were transferred to a nylon membrane, fixed by UV cross-linking and hybridized with a digoxigenin-labelled cblA probe as described by Sajjan et al. (1995) . The bound probe was detected by using anti-digoxigenin antibody and a chemiluminescent substrate (CPD*, Roche Molecular Biochemicals).

Production of antibody to cable pilin protein.
Cable pilin subunit protein was prepared from B. cepacia strain BC7 as described previously (Sajjan et al., 1995 ). Two New Zealand White rabbits were injected subcutaneously with this preparation (10 µg), mixed with TiterMax classic adjuvant (Sigma-Aldrich), on days 1, 15, 30 and 45. Polyclonal antiserum was obtained 2 weeks after the last injection.

Reactivity to anti-cable pilus antibodies.
Whole-cell extracts of B. cepacia isolates were prepared as described previously (Sajjan & Forstner, 1992 ). Bacterial surface proteins were isolated after incubating bacteria at 60 °C for 15 min as described by Cravioto et al. (1982) . Whole-cell extracts or surface proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked with 3% BSA for 2 h at room temperature, incubated with anti-cable pilin antiserum (1:1000 diluted) overnight at 4 °C and washed. Bound antibody was detected using anti-rabbit IgG conjugated with alkaline phosphatase (Bio-Rad), and a colour substrate (NBT-BCIP, Roche Molecular Biochemicals).

CK13 binding assay.
Binding of B. cepacia to CK13 was determined as described previously (Sajjan et al., 2000a ). In brief, a cytokeratin-rich fraction, isolated from buccal epithelial cells as described by Franke et al. (1981) , was subjected to electrophoresis and the proteins were transferred to a nitrocellulose membrane. Membranes were blocked with 1% gelatin for 1 h at room temperature, incubated with 35S-labelled B. cepacia (1x109 c.f.u. ml-1) for 90 min at 37 °C, washed to remove unbound bacteria and exposed to X-ray film.

Electron microscopy.
Bacteria were grown on brain-heart infusion agar, lifted onto a Formvar-coated grid, and stained with 1% phosphotungstic acid for 30 s. Excess stain was removed by blotting with moist filter paper and the grid was allowed to air dry. Samples were examined by using a JEOL 1230 electron microscope equipped with a CCD camera.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and cblA sequence analyses
The 11 isolates identified as containing cblA sequences are listed in Table 1. The 10 CF sputum isolates were recovered from 10 patients receiving care in six US states. The environmental isolate (BC20) was obtained from blackwater stream sediment from another US state (Wise et al., 1995 ). Nine of the 10 CF isolates as well as the environmental isolate BC20 were confirmed as belonging to B. cepacia complex genomovar I based on recA PCR and RFLP analyses (Mahenthiralingam et al., 2000a ). The remaining CF isolate (AU0007) belonged to genomovar III.

Although DNA fragments of the predicted size were amplified from all 11 isolates by cblA-specific PCR, only the product from the genomovar III isolate (AU0007) had 100% nucleotide sequence identity with the cblA amplified from the ET12 strains J2315 and BC7. The cblA nucleotide sequences from the 10 genomovar I isolates showed high identity (range 98·2–100%) to each other, but lower identity (range 87·8–88·4%) to the ET12 cblA sequence. cblA sequences from three genomovar I isolates (AU1544, AU1555 and AU1693) shared 100% identity with each other.

To investigate whether differences in cable pili expression (see below) may result from upstream nucleotide sequence variations we amplified a 145 bp segment immediately upstream of the cblA start codon from all isolates. Again, the sequences from the ET12 isolates AU0007, J2315 and BC7 were identical to each other. The sequences from AU1544, AU1555 and AU1693 were also identical to each other and this sequence differed from the remaining seven genomovar I strains by a single nucleotide. The identity between the ET12 sequence and that of isolates AU1544, AU1555 and AU1693 was 96·6%; identity between the ET12 sequence and that of the remaining seven isolates was 95·9%. Although the putative -10 and -35 sequences are identical among all ET12 and genomovar I isolates, the 16 bp intervening region contains 3 bp that differ between the ET12 and the genomovar I isolates (Fig. 1).



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 1. Sequences of the 70 bp segment upstream of the cblA start codon (asterisks) from ET12 isolates (upper line) and genomovar I isolates (lower line). All 10 genomovar I isolates had identical DNA sequence in the 70 bp segment shown. Putative -10 and -35 sequences are underlined. Three nucleotide differences in the intervening segment are indicated by white letters.

 
Predicted CblA amino acid sequence
Alignment of the deduced amino acid sequences is depicted in Fig. 2. The genomovar I strains showed 99% sequence identity to each other; they had between 85% and 86·2% sequence identity to the ET12 sequence. Differences in amino acid sequence between the genomovar I isolates and the ET12 sequence were detected mainly in the middle to the C-terminus of the predicted protein.



View larger version (47K):
[in this window]
[in a new window]
 
Fig. 2. Alignment of deduced amino acid sequences. Row A, major pilin subunit protein from ET12 isolates (J2315, BC7 and AU0007). Row B, cblA-encoded peptide from isolate AU0113. Row C, cblA-encoded peptides from isolates AU0143, AU0666, AU0717, AU1540 and BC20. Row D, cblA-encoded peptide from isolates AU1544, AU1555 and AU1693. Row E, cblA-encoded peptide from isolate AU1550. Amino acids differing from the ET12 pilin subunit protein are shaded.

 
PFGE
Genotyping by PFGE demonstrated that the genomovar III isolate (AU0007) had a profile identical to strain J2315 and was thus indeed of the ET12 lineage (Fig. 3, lanes 1 and 2). The DNA profiles of BC7 and J2315 have been shown in previous work to be the same as well (Mahenthiralingam et al., 2000b ). Consistent with the cblA sequence analyses (above), genomovar I isolates AU1544, AU1555 and AU1693 had the same profile (Fig. 3, lanes 8, 10 and 11) and are thus considered clonal; the remaining seven isolates had different genomic macrorestriction profiles, indicating that they represent different strain types.



View larger version (131K):
[in this window]
[in a new window]
 
Fig. 3. PFGE genotyping analysis of cblA-positive isolates. Lanes 1 through 12, J2315, AU0007, AU0113, AU0143, AU0666, AU0717, AU1540, AU1544, AU1550, AU1555, AU1693, BC20, respectively. Profiles of J2315 and AU0007 are identical, as are those of AU1544, AU1555 and AU1693.

 
Southern blot analysis
The mobility of the cblA-containing EcoRI fragments from the 10 genomovar I strains differed from that of the ET12 strains (Fig. 4). The cblA-containing fragments from all 10 genomovar I isolates were approximately 4 kbp in size, whereas that of the ET12 isolates was approximately 5 kbp.



View larger version (72K):
[in this window]
[in a new window]
 
Fig. 4. Detection of cblA-containing EcoRI fragments by Southern blot analysis. ATCC, ATCC 25416 (cblA-negative). Only the relevant part of the blot is shown.

 
Reactivity of anti-cable pilin antibody with B. cepacia isolates
Western blot analysis of bacterial whole-cell extracts and surface proteins using anti-cable pilin antibody was used to examine expression of the major pilin subunit. Cable pilin antibody reacted strongly with a 16 kDa protein in both whole-cell extracts (Fig. 5) and surface-protein preparations (data not shown) from the genomovar III ET12 isolates BC7, J2315 and AU0007. In contrast, neither whole-cell nor surface-protein extracts from ATCC 25416, which lacks cblA, or any of the genomovar I strains reacted with this antibody under the conditions examined. This indicates either lack of expression of major cable pilin subunit protein or lack of antibody reactivity with an aberrant pilin subunit that may be expressed by these isolates.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 5. Detection of major cable pilin protein by Western blot analysis. Whole-cell extracts from each B. cepacia isolate indicated were subjected to SDS-PAGE, and proteins were transferred to a nitrocellulose membrane. The membrane was incubated with anti-pilin antibody and the antibody was detected with a colorimetric reaction as detailed in the text. The first lane contains molecular mass standards. The 16 kDa cable pilin protein was detected only in the three genomovar III ET12 isolates, BC7, J2315 and AU0007. All genomovar I isolates are negative. ATCC, ATCC 25416.

 
Binding of B. cepacia to CK13
As previously demonstrated (Sajjan et al., 2000a ) cable-pili-expressing isolate BC7 exhibited strong binding to CK13. In contrast, neither AU0007 nor any of the genomovar I isolates bound to the membrane-immobilized CK13-enriched preparation (Fig. 6). Also as previously shown (Sajjan et al., 2000b ), J2315 did not bind to CK13.



View larger version (10K):
[in this window]
[in a new window]
 
Fig. 6. Binding of B. cepacia isolates to CK13. A cytokeratin-rich fraction was isolated from buccal epithelial cells, transferred to a membrane, and the membrane was incubated with 35S-labelled bacteria as described in the text. Only isolate BC7 showed binding to CK13, which has a molecular mass of 55 kDa. ATCC, ATCC 25416.

 
Electron microscopy
Electron microscopy was used to investigate the surface expression of cable pili (Fig. 7). The positive control strain BC7 showed peritrichous pili tethered together to give the cable morphology described previously (Sajjan et al., 1995 ) (Fig. 7a). None of several genomovar I strains (AU0143, AU1540, AU1544 and AU1555) demonstrated cable pili on their surfaces (Fig. 7b). Although isolate AU0007 expressed pili, it was hyperpiliated compared to BC7 and demonstrated numerous rigid pili that failed to entangle each other (Fig. 7c). Replicate electron microscopic preparations gave reproducible results.



View larger version (58K):
[in this window]
[in a new window]
 
Fig. 7. Electron microscopy of negatively stained bacteria. Bacteria grown on brain-heart infusion agar were adsorbed on Formvar coated grids, stained with 1% phosphotungstic acid for 1 min and air-dried. The grids were observed under a JEOL 1200 EXII transmission electron microscope at 80 kV. Panels (a), (b) and (c) represent BC7, AU1544 and AU0007 respectively.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Although all species within the B. cepacia complex have been recovered from CF sputum culture, their distribution in CF patients is quite disproportionate (LiPuma et al., 2001 ). In the US, genomovar III is recovered from approximately half of B. cepacia-complex-infected patients. In Canada, this species accounts for an even greater proportion of such patients (Speert et al., 2002 ). This is due largely to the prevalence of the ET12 clonal lineage, a genomovar III strain that predominates among CF patients in Ontario and is believed to have spread between Canadian and UK patients during CF summer camps (Govan et al., 1993 ; Johnson et al., 1994 ).

The factors that account for the frequency of ET12 and other so-called ‘epidemic’ strains among CF patients are not clear. Few such strains have been characterized in detail (Mahenthiralingam et al., 1997 ) and factors critical to inter-patient spread have yet to be defined. The B. cepacia epidemic strain marker (BCESM), a 1·4 kb sequence encoding an ORF with homology to transcriptional regulatory genes, is found in ET12 and other strains that infect multiple CF patients (Mahenthiralingam et al., 1997 ). This sequence is generally absent from strains found only in single patients (i.e. for which there is no evidence of patient-to-patient spread). The role this element may have in contributing to inter-patient transmission is unknown.

Cable pili, another feature of the ET12 lineage, have been better characterized (Sajjan et al., 1995 ; Sun et al., 1995 ). These large intertwined peritrichous fibres, together with an associated adhesin, mediate bacterial adherence to respiratory epithelia via binding to CK13, a 55 kDa protein that has increased expression in CF (Sajjan et al., 2000a ). Recent work indicates that adherence mediated by cable pili also contributes to epithelial cell invasion and cytotoxicity during B. cepacia infection in CF (Sajjan et al., 2002 ). Thus, expression of cable pili is of critical importance in the pathology of ET12, a strain accounting for significant infection in CF.

The identification of cable pili in ET12 initially raised hopes that screening of B. cepacia isolates for this phenotype (or, more specifically, for the presence of cblA) may allow identification of ‘highly transmissible’ lineages (Hearst & Elliott, 1995 ). Such screening could, in turn, enable CF centres to selectively apply stringent infection control measures, isolating only those patients harbouring an epidemic strain. Similar recommendations for stratification of infection control based on the presence or absence of cblA have been made more recently (Clode et al., 2000 ).

However, such strategies are limited. Although cable pili mediate events important in the pathogenesis of infection by ET12, it is not yet clear whether expression of cable pili confers an enhanced capacity for inter-patient transmission, per se. Other B. cepacia complex lineages implicated in inter-patient spread do not contain cblA sequences (Mahenthiralingam et al., 1997 ). Indeed, PHDC, the genomovar III strain that predominates among CF patients in the mid-Atlantic region of the US, contains neither BCESM nor cblA (Chen et al., 2001 ). The B. multivorans strain responsible for a hospital-associated outbreak among CF patients in the UK (Whiteford et al. 1995 ) similarly lacks BCESM and cblA (Mahenthiralingam, 2000b ). Thus, neither cblA nor cable pili expression is a sensitive marker of B. cepacia strains that seem to possess an enhanced ability for spread in CF.

The results of the present study also indicate that cblA sequences are not specific for ET12. We identified several non-ET12 isolates positive for cblA by both dot-blot and PCR assays. Variant cblA genes with varying degrees of identity to the ET12 cblA gene were obtained from these isolates by PCR amplification using cblA-directed primers. Although we demonstrated pili expression by the ET12 isolates included in the study, none of the 10 cblA-containing non-ET12 isolates expressed pili under the conditions examined based on functional assays; this was confirmed by electron microscopy. The non-ET12 isolates represented several distinct strains by PFGE analysis, but interestingly all belonged to genomovar I, a species not commonly found in CF (LiPuma et al., 2001 ). Furthermore, in all of these isolates the cblA sequences resided on an approximately 4 kbp EcoRI fragment, suggesting a common chromosomal location.

Others have similarly found cblA sequences among non-ET12 isolates, albeit at a low frequency. Sun et al. (1995) identified a single cblA-positive strain from among non-ET12 isolates recovered from over 100 CF patients. Similar to the isolates described in the present study, the cblA gene nucleotide sequence from this isolate had 88% identity with the ET12 cblA sequence. Although neither microscopic studies nor epithelial cell binding assays were performed, antibody against purified ET12 cable pili did not react with this isolate. Because description of this isolate preceded current knowledge of the taxonomy of the B. cepacia complex, the genomovar of this isolate is not known.

In a larger survey of 627 B. cepacia isolates representing 132 distinct strain types (defined by using randomly amplified polymorphic DNA typing) recovered from both CF and non-CF sources, Mahenthiralingam et al. (1997) detected DNA homologous to cblA in five non-ET12 isolates. Only one of these was from a CF patient; the remaining four were recovered from the environment. cblA sequence analysis and cable pili expression studies were not performed and again, genomovar analysis was not yet available.

More recently, Richardson et al. (2001) identified two non-ET12 strains, of unclear genomovar, containing cblA sequences from among 75 isolates of Burkholderia spp. Although functional assays of pili expression were not performed, the predicted amino acid sequences of these two strains had only 68% and 78% identity with that of the ET12 pilin subunit protein. In contrast, a survey of 117 isolates from CF patients receiving care in 40 hospitals in the UK by Clode et al. (2000) demonstrated cblA sequences exclusively among ET12 lineage isolates; none of 76 non-ET12 isolates was cblA positive by PCR assay.

The frequency of cblA sequences in B. cepacia residing in the natural environment (or more specifically, among environmental genomovar I strains) is unknown. A more comprehensive survey of such isolates is needed to address this. Nevertheless, our findings suggest the possibility of inter-species transfer of potential virulence elements among B. cepacia complex species. Work is currently under way to assess this.

The reasons for the lack of cable pili expression in the cblA-positive genomovar I isolates identified in this study are not entirely clear. We found minor differences in the DNA upstream of the cblA start codon (specifically in the spacer region between the putative -10 and -35 sequences), and it is possible that this variation may affect transcription initiation. For example, substitution of AT by GC in this spacer region significantly increases the Ka of E. coli RNA polymerase (Auble et al., 1987 ). Perhaps the replacement of C and G at positions -18 and -25, respectively, in the ET12 sequence by T in genomovar I isolates decreases the Ka of RNA polymerase, resulting in a reduced rate of transcription. It is also entirely possible that the differences noted in this promoter region do not account for the lack of pili expression by these isolates. Several other genes are probably involved in cable pili expression. Perhaps nucleotide differences in one or more of these, particularly those residing upstream of cblA, are responsible for the lack of pili expression in these isolates. Ongoing studies are assessing this possibility.

The reasons for the failure of AU0007 to express typical cable pili and the associated adhesin also require elucidation. This genomovar III CF isolate is clearly of the ET12 clonal lineage based on PFGE analysis. It also shares 100% nucleotide identity with the other ET12 isolates analysed (BC7 and J2315) in cblA as well as in the 145 bp segment immediately upstream of the cblA start codon. Nevertheless, it is hyperpiliated with rigid-appearing pili and fails to express cable-pili-associated adhesin based on its lack of binding to CK13 (Sajjan et al., 2000b ); this is markedly different from other ET12 isolates examined to date. It is possible that the reasons for this lie elsewhere in the B. cepacia cbl pil operon; this is the subject of ongoing investigation.

In summary, although efforts to lessen the burden of strict infection control in CF are urgently needed, data from this study and those of others indicate that cblA lacks both the sensitivity and specificity to identify ‘transmissible’ strains. Until the factors contributing to the apparent proclivity of select B. cepacia complex strains to cause infection in CF are better defined, stratification of infection control measures will be problematic. In the interim, assays to assess cable pili expression (rather than the presence of cblA sequences) may provide better specificity for identification of ET12 isolates. This will be particularly helpful in regions where tracking of ET12 is important in infection control surveillance.


   ACKNOWLEDGEMENTS
 
Supported by a grant from the Cystic Fibrosis Foundation (United States) (to J.J.L.). U.S. and J.F. acknowledge support provided by the Canadian Cystic Fibrosis Foundation. The authors gratefully acknowledge the generosity and cooperation of participating CF centres and microbiology laboratories for submission of clinical isolates.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Auble, D. T., Allen, T. L. & DeHaseth, P. I. (1987). Promoter recognition by Escherichia coli RNA polymerase. Effects of substitutions in the spacer DNA separating the -10 and -35 regions. J Biol Chem 261, 11202-11206.[Abstract/Free Full Text]

Chen, J. S., Witzmann, K., Spilker, T., Fink, R. & LiPuma, J. J. (2001). Endemicity and inter-city spread of Burkholderia cepacia genomovar III in cystic fibrosis. J Pediatr 139, 643-649.[Medline]

Clode, F. E., Kaufmann, M. E., Malnick, H. & Pitt, T. L. (2000). Distribution of genes encoding putative transmissibility factors among epidemic and non-epidemic strains of Burkholderia cepacia from cystic fibrosis patients in the United Kingdom. J Clin Microbiol 38, 1763-1766.[Abstract/Free Full Text]

Coenye, T., Vandamme, P., Govan, J. R. W. & LiPuma, J. J. (2001). Taxonomy and identification of the Burkholderia cepacia complex. J Clin Microbiol 39, 3427-3436.[Free Full Text]

Cravioto, A., Scotland, S. M. & Rowe, B. (1982). Hemagglutination activity and colonization factor antigens I and II in enterotoxigenic and non-enterotoxigenic strains of E. coli isolated from humans. Infect Immun 43, 508-514.

Franke, W. W., Schiller, D. L., Moll, R. , Winter, E., Schmid, E., Engelbrecht, L., Denk, H., Kreplear, R. & Platzer, B. (1981). Diversity of cytokeratins. Differentiation specific expression of cytokeratin polypeptides in epithelial cells and tissues. J Mol Biol 153, 933-955.[Medline]

Govan, J. R. W., Brown, P. H., Maddison, J., Doherty, C. J., Nelson, J. W., Dodd, Greening, A. P. & Webb, A. K. (1993). Evidence for transmission of Pseudomonas cepacia by social contact in cystic fibrosis. Lancet 342, 15–19.[Medline]

Hearst, J. E. & Elliott, K. E. (1995). Identifying the killer in cystic fibrosis. Nat Med 1, 626-627.[Medline]

Johnson, W. M., Tyler, S. D. & Rozee, K. R. (1994). Linkage analysis of geographical and clinical clusters in Pseudomonas cepacia infections by multilocus enzyme electrophoresis and ribotyping. J Clin Microbiol 32, 924-930.[Abstract]

LiPuma, J. J. (1998). Burkholderia cepacia: management issues and new insights. Clin Chest Med 19, 473-486.[Medline]

LiPuma, J. J. & Mahenthiralingam, E. (1999). Commercial use of Burkholderia cepacia. Emerg Infect Dis 5, 305-306.[Medline]

LiPuma, J. J., Dulaney, B. J., McMenamin, J. D., Whitby, P. W., Stull, T. L., Coenye, T. & Vandamme, P. (1999). Development of rRNA-based PCR assays for identification of Burkholderia cepacia complex isolates recovered from cystic fibrosis patients. J Clin Microbiol 37, 3167-3170.[Abstract/Free Full Text]

LiPuma, J. J., Spilker, T., Gill, L., Campbell, P. W., Liu, L. & Mahenthiralingam, E. (2001). Disproportionate distribution of Burkholderia cepacia complex species and transmissibility factors in cystic fibrosis. Am J Respir Crit Care Med 164, 92-96.[Abstract/Free Full Text]

Mahenthiralingam, E., Simpson, D. A. & Speert, D. P. (1997). Identification and characterization of a novel DNA marker associated with epidemic Burkholderia cepacia strains recovered from patients with cystic fibrosis. J Clin Microbiol 35, 808-816.[Abstract]

Mahenthiralingam, E., Bischof, J., Byrne, S. K., Radomski, C., Davies, J. E., Av-Gay, Y. & Vandamme, P. (2000a). DNA-based 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.[Abstract/Free Full Text]

Mahenthiralingam, E., Coenye, T, Chung, J. W., Speert, D. P., Govan, J. R., Taylor, P. & Vandamme, P. (2000b). Diagnostically and experimentally useful panel of strains from the Burkholderia cepacia complex. J Clin Microbiol 38, 910-913.[Abstract/Free Full Text]

Parke, J. L. & 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.[Medline]

Pitt, T. L., Kaufmann, M. E., Patel, P. S., Benge, L. C., Gaskin, S. & Livermore, D. M. (1996). Type characterisation and antibiotic susceptibility of Burkholderia (Pseudomonas) cepacia isolates from patients with cystic fibrosis in the United Kingdom and the Republic of Ireland. J Med Microbiol 44, 203-210.[Abstract]

Richardson, J., Stead, D. E. & Coutts, R. H. A. (2001). Incidence of the cblA major subunit pilin gene amongst Burkholderia species. FEMS Microbiol Lett 196, 61-66.[Medline]

Sajjan, U. S. & Forstner, J. F. (1992). Identification of the mucin-binding adhesin of Pseudomonas cepacia isolated from patients with cystic fibrosis. Infect Immun 60, 1434-1440.[Abstract]

Sajjan, U. S., Sun, L., Goldstein, R. & Forstner, J. F. (1995). Cable (cbl) type II pili of cystic fibrosis-associated Burkholderia (Pseudomonas) cepacia: nucleotide sequence of the cblA major subunit pilin gene and novel morphology of the assembled appendage fibers. J Bacteriol 177, 1030-1038.[Abstract]

Sajjan, U. S., Sylvester, F. A. & Forstner, J. (2000a). Cable-piliated Burkholderia cepacia binds to cytokeratin 13 of epithelial cells. Infect Immun 68, 1787-1795.[Abstract/Free Full Text]

Sajjan, U., Wu, Y., Kent, G. & Forstner, J. (2000b). Preferential adherence of cable-piliated Burkholderia cepacia to respiratory epithelia of CF knockout mice and human cystic fibrosis lung explants. J Med Microbiol 49, 875-885.[Abstract/Free Full Text]

Sajjan, U. S., Ackerly, C. & Forstner, J. (2002). Interaction of cblA adhesin positive Burkholderia cepacia with squamous epithelia. Cell Microbiol (in press).

Speert, D. P., Henry, D., Vandamme, P., Corey, M. & Mahenthiralingam, E. (2002). Epidemiology of Burkholderia cepacia complex inpatients with cystic fibrosis in Canada: geographical distribution and clustering of strains. Emerg Infect Dis 8, 181-187.[Medline]

Sun, L., Jiang, R.-Z., Steinbach, S. & 7 other authors (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.[Medline]

Tenover, F. C., Arbeit, R. D., Goering, R. V., Mickelsen, P. A., Murray, B. E., Persing, D. H. & Swaminathan, B. (1995). Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 33, 2233-2239.[Free Full Text]

Whiteford, M. L., Wilkenson, J. D., McColl, J. H., Conoln, F. M., Michie, J. R., Evans, T. J. & Paton, J. Y. (1995). Outcome of Burkholderia (Pseudomonas) cepacia colonisation in children with cystic fibrosis following a hospital outbreak. Thorax 50, 1194-1198.[Abstract]

Wise, M. G., Shimkets, L. J. & McArthur, J. V. (1995). Genetic structure of a lotic population of Burkholderia (Pseudomonas) cepacia. Appl Environ Microbiol 61, 1791-1798.[Abstract]

Received 18 March 2002; revised 19 June 2002; accepted 11 July 2002.