Open Access

Tandem repeat regions within the Burkholderia pseudomallei genome and their application for high resolution genotyping

  • Jana M U'Ren1,
  • James M Schupp1,
  • Talima Pearson1,
  • Heidie Hornstra1,
  • Christine L Clark Friedman1,
  • Kimothy L Smith1,
  • Rebecca R Leadem Daugherty1,
  • Shane D Rhoton1,
  • Ben Leadem1,
  • Shalamar Georgia1,
  • Michelle Cardon1,
  • Lynn Y Huynh1,
  • David DeShazer2,
  • Steven P Harvey3,
  • Richard Robison4,
  • Daniel Gal5,
  • Mark J Mayo5,
  • David Wagner1,
  • Bart J Currie5 and
  • Paul Keim1Email author
Contributed equally
BMC Microbiology20077:23

DOI: 10.1186/1471-2180-7-23

Received: 10 November 2006

Accepted: 30 March 2007

Published: 30 March 2007

Abstract

Background

The facultative, intracellular bacterium Burkholderia pseudomallei is the causative agent of melioidosis, a serious infectious disease of humans and animals. We identified and categorized tandem repeat arrays and their distribution throughout the genome of B. pseudomallei strain K96243 in order to develop a genetic typing method for B. pseudomallei. We then screened 104 of the potentially polymorphic loci across a diverse panel of 31 isolates including B. pseudomallei, B. mallei and B. thailandensis in order to identify loci with varying degrees of polymorphism. A subset of these tandem repeat arrays were subsequently developed into a multiple-locus VNTR analysis to examine 66 B. pseudomallei and 21 B. mallei isolates from around the world, as well as 95 lineages from a serial transfer experiment encompassing ~18,000 generations.

Results

B. pseudomallei contains a preponderance of tandem repeat loci throughout its genome, many of which are duplicated elsewhere in the genome. The majority of these loci are composed of repeat motif lengths of 6 to 9 bp with 4 to 10 repeat units and are predominately located in intergenic regions of the genome. Across geographically diverse B. pseudomallei and B.mallei isolates, the 32 VNTR loci displayed between 7 and 28 alleles, with Nei's diversity values ranging from 0.47 and 0.94. Mutation rates for these loci are comparable (>10-5 per locus per generation) to that of the most diverse tandemly repeated regions found in other less diverse bacteria.

Conclusion

The frequency, location and duplicate nature of tandemly repeated regions within the B. pseudomallei genome indicate that these tandem repeat regions may play a role in generating and maintaining adaptive genomic variation. Multiple-locus VNTR analysis revealed extensive diversity within the global isolate set containing B. pseudomallei and B. mallei, and it detected genotypic differences within clonal lineages of both species that were identical using previous typing methods. Given the health threat to humans and livestock and the potential for B. pseudomallei to be released intentionally, MLVA could prove to be an important tool for fine-scale epidemiological or forensic tracking of this increasingly important environmental pathogen.

Background

The environmental saprophyte Burkholderia pseudomallei is the causative agent of melioidosis, a disease endemic to tropical regions of Southeast Asia and northern Australia. Symptoms range in severity from fatal sepsis and acute community-acquired pneumonia to benign and localized abscesses. Infection in humans and animals generally occurs through direct contact of open wounds or abrasions with contaminated water and soil, by ingestion of contaminated drinking water, or inhalation of infectious aerosols. Melioidosis is a serious public health threat in Thailand and northern Australia, where it is associated with a case fatality rate of approximately 50 and 20%, respectively [1]. In addition, B. pseudomallei has recently attracted attention as a potential biological weapon, and is listed as a Category B biothreat agent by the U.S. Centers for Disease Control and Prevention (CDC) [2].

The close genetic relationship of B. pseudomallei to B. mallei has previously been demonstrated by DNA hybridization studies [3]. More recently, studies have revealed that B. mallei is a clonal lineage of B. pseudomallei, and its recent evolutionary divergence is marked by gene deletions and intra-chromosomal rearrangements [47]. B. mallei, the etiologic agent of glanders, is an obligate parasite of the family Equidae, but can also infect humans through direct contact with infected animals [8] or occupational exposure [9]. Glanders was once a globally distributed disease, but is currently predominant only in the Middle East, Africa, Asia and Central and South America. Due to its highly infectious nature and ability to infect via aerosol, it was used as a biological weapon during World War I and World War II [10, 11]. It is also listed as a Category B biothreat agent by the CDC [2].

Due to the severe nature of melioidosis, the molecular epidemiology of B. pseudomallei has been investigated using various DNA restriction-based methods, including Pulse Field Gel Electorphoresis (PFGE) [12, 13] and ribotyping [14, 15]. PFGE has the ability to resolve potentially polymorphic, large DNA restriction fragments, while ribotyping uses restriction fragment length polymorphisms associated with rRNA genes [16]. Although both of these methods have been successful in the epidemiological tracking of pathogens [17], their technical nature can make large datasets more difficult to handle. Also, neither method is easily standardized for transfer throughout the scientific and public health community, and can often lack discriminatory power among closely related isolates within a species or between closely related species [18].

Other procedures that have been used for molecular typing of B. pseudomallei involve PCR, such as random amplified polymorphic DNA (RAPD) [19, 20] and multilocus sequence typing (MLST) [6]. RAPD detects differences in genomes by amplifying segments of unknown DNA. Drawbacks to this technique include the presence/absence binary nature of the data and the difficulty in reproducing banding patterns between reactions (attributed to PCR artifacts). MLST uses concatenated nucleotide sequences from seven housekeeping genes, that are assumed to be selectively neutral or under purifying selection [21]. This method provides nucleotide data for multiple haplotypes, is easily amenable to phylogenetic analyses and can be standardized across laboratories. The MLST scheme developed for B. pseudomallei is also applicable to B. mallei and B. thailandensis. However, MLST can be time consuming and expensive, and most importantly lacks discriminatory power within closely related B. pseudomallei isolates and among the vast majority of B. mallei isolates, which are all close genetic relatives [6].

Recently, a reliable PCR-based method using variable-number tandem repeat (VNTR) loci has become a popular tool for the molecular typing of pathogens [18, 2225]. A VNTR locus consists of tandemly repeated sequences of DNA that vary in copy number, creating PCR amplicon size polymorphisms that are easily detected with gel electrophoresis. Due to increased mutation rates when compared to other regions of DNA and their multi-allelic nature, VNTRs allow superior discrimination between closely related isolates. These loci have been successfully implemented for forensic, epidemiological and phylogenetic analyses of bacterial pathogens with low genetic diversity, such as Bacillus anthracis, F. tularensis, and Y. pestis [23, 2630].

Due to the success of VNTR typing in other pathogens, the primary objective of this study was to develop a high-resolution VNTR typing system for B. pseudomallei that is suitable for epidemiological, forensic, phylogenetic and population genetic studies. Thus the first task for this study was to characterize tandem repeat loci, including their distribution and frequency within the B. pseudomallei genome. Additionally, in order to develop a comprehensive multiple-locus VNTR typing system that utilizes loci with varying degrees of polymorphism, the second task was to screen loci that were characteristic of the tandem repeat loci throughout the genome and examine levels of polymorphism. Finally, in order to understand the effects that mechanisms such as recombination and mutation have on generating the high levels of diversity observed in this pathogen, it was essential to examine the mutation rates for the non-duplicated VNTR loci chosen for the typing system, as well as a representative sample of the duplicated tandem repeat regions. Furthermore, the estimation of mutation rates will allow for future epidemiological studies that model the transmission of melioidosis in natural populations, similar to published studies on plague [26].

In this manuscript we describe a multiple-locus VNTR analysis (MLVA) genotyping system in which 32 independent, tandemly inserted repeated motifs identified in the B. pseudomallei K96243 genome are amplified using fluorescently labeled primers in multiplexed PCRs and separated using capillary electrophoresis. These loci were highly polymorphic across a globally distributed set of 66 B. pseudomallei and 21 B. mallei isolates, as well as a few very closely related B. pseudomallei isolates from an outbreak event and two individual patients.

Results

Tandem repeats within the Burkholderia pseudomallei genome

We observed that in comparison to other bacterial pathogens with similarly sized genomes, such as Bacillus anthracis Ames and Yersinia pestis CO92, the Burkholderia pseudomallei K96243 genome harbors a relatively large number of tandem repeat arrays (Figure 1). The large (4,074,542 bp) chromosome of B. pseudomallei contains 285 (69.9 arrays/Mbp) while the small (3,173,005 bp) chromosome contains 324 (102.1 arrays/Mbp) tandem repeat arrays (Table 1). In contrast, the Y. pestis genome contains only 174 arrays and B. anthracis contains just 66 arrays, at densities of 37.4 arrays/Mb interval and 12.6 arrays/Mb, respectively. In B. pseudomallei, tandem repeat motif sizes on both chromosomes ranged from 3 to 16 bp with copy numbers ranging from 4 to 21 units (Figure 2, A1 and A2). Non-triplet repeat motifs were more common in intragenic regions than inside genes (Figure 2, B1 and B2).
Table 1

Summary of B. pseudomallei chromosomal repeat region frequency, duplication and location in coding regions

Chromosome

Size (bp)

GC%

Tandem repeat (TR) regions‡

TR/10 kb

#TR's in Sanger CDS

#TR's within 40 bp 5' of Sanger CDS

#TR's within 40 bp 3' of Sanger CDS

cis only duplicated TRs (in CDS)*

trans only duplicated TRs (in CDS)*

Both cis and trans duplicated TRs (in CDS)*

total dup arrays (in CDS)*

Large

4,074,542

67.7

285

0.699

72**

8

43

22 (4)

56 (12)

30 (2)

108 (18)***

Small

3,173,005

68.5

324

1.021

103**

11

42

25 (8)

48 (14)

41 (13)

114 (35)***

Total

7,247,547

68.1

609

0.860

175**

19

85

47(12)

104(26)

71(15)

222(53)***

‡ Regions with repeats ≥ 2 bp, ≥ 4 repeat units and array sizes ≥ 30 bp

† Average number

*duplications ≥ 20 bp and 80% similarity

**all but 4 and 8 (X1 and X2, resp.) of the non degenerate arrays had RU sizes of 3 bp multiples

***Average duplication size of 50 bp

Figure 1

Linear repeat array distribution of B. anthracis , Y. pestis and B. pseudomallei chromosomes. Nucleic acid repeat region "icicle" plots were generated with DNAStar GeneQuest software (Madison, WI). The horizontal scale indicates the linear position in base pairs along the respective chromosomes from the start position of the GenBank FASTA file sequence. The scale bar to the right of each icicle plot indicates 10 possible repeat sequence combinations as found by the GeneQuest software. The overall length, or number of possible repeat combinations of each icicle, is a measure of the size of the repeated sequence array found at that position. In general, the longer the icicle, the larger the repeat array. Note that both perfect and degenerate repeat arrays are found and displayed by GeneQuest, as indicated by the arrows and notes in panel C. The number of arrays/Mbp and total arrays are all repeat regions found by the software package Tandem Repeats Finder larger than 30 bp and with an internal similarity greater than or equal to 80%.

Figure 2

Repeat region motif size and total array size distribution. A) Frequencies of arrays consisting of different size repeat motifs in inter-, intragenic and duplicated locations. Degenerate repeats were determined as described in the Materials and Methods Section. B) Frequencies of arrays consisting of different total size classes, again in inter-, intragenic and duplicated locations, based upon triplet and non-triplet repeat motif copy number. Degenerate arrays are not included as consensus repeat motifs were not determined.

Distribution and location of tandem repeats

A χ2 goodness-of-fit test of the "observed" B. pseudomallei tandem repeat distribution to an "expected" Poisson distribution was significant for both the large (p < 0.001) and small chromosomes (p < 0.001) using 10 Kb intervals (Figure 3). The non-random observed distributions for both chromosomes are consistent with a clustered arrangement of arrays throughout both chromosomes. Additionally, the majority of the tandem repeats were found in intergenic regions of the chromosomes: 74.7% (n = 213) tandem repeats on the large chromosome and 68.2% (n = 221) on the small chromosome. However, a portion of these arrays (28.1% on the large chromosome and 35.2% on the small chromosome) were found inside or within 40 base pairs upstream of predicted ORFs (Table 1). Longer arrays (≥ 11 repeat units), including even those with triplet motifs, tended not to be found inside predicted protein coding regions on the large chromosome (Figure 2A1). Conversely, on the small chromosome, longer arrays with triplet repeat motifs were found in both inter- and intragenic locations in almost equal numbers (Figure 2A2). It was also observed that four-fold more degenerate arrays were found on the small chromosome than on the large, and the majority of these degenerate arrays were located inside coding regions (Figure 2A1, and 2A2).
Figure 3

Repeat array distribution Goodness-of-fit test against a Poisson distribution. The bar graphs in each of the panels indicate the observed and expected number of 10 Kbp intervals containing zero, one, two, three and four or more repeat arrays for the B. pseudomallei large (A) and small (B) chromosomes. For each chromosome, the total number of arrays, average arrays/interval used to generate the Poisson expected frequencies, and calculated p values are shown. Values above each bar indicate the observed or expected frequencies in each category.

We found that 36.3% of the total number of tandem repeat arrays on both chromosomes of B. pseudomallei are duplicated, at least partially (≥ 20 bp and ≥ 80% similarity), in other locations on either chromosome (Table 1). Most of these duplications were found in intergenic regions of the chromosomes and involved the repeat motif only and not the flanking sequences. The majority of duplicated tandem repeats on the large chromosome were, in fact, duplicated on the small chromosome, rather than on the large chromosome. In contrast, arrays duplicated on the small chromosome were found in equal numbers on both chromosomes (Table 1). Additionally, total array lengths were typically longer for duplicated tandem arrays. For example, 104 of the 108 duplicated arrays on the large chromosome, and 112 of the 114 duplicated arrays on the small chromosomes are larger than 200 bp, with the largest almost 6000 bp in size. It was observed that repeat regions that contained more than 20 repeat copies were found to be duplicated in some fashion, and repeat motifs of six and seven bp were more often duplicated than not (Figure 2).

MLVA development

In order to develop a MLVA system for B. pseudomallei, a variety of array sizes were screened, from 2 bp repeat motif by 7 repeat copy unit (i.e. 2 × 7) to degenerate repeat arrays greater than 500 bp but less than 1000 bp, for a total of 104 VNTR loci. We also screened both intra- and intergenically located arrays. Criteria used for including loci in the MLVA system were 1) variation within the screening panel (see Methods), either within the globally distributed or locally distributed outbreak sets, 2) robust (> 80% success) PCR amplification, and 3) highly discrete PCR amplicon sizes (minimal partial repeat differences), based upon locus repeat unit motif. Thirty-two loci met the above three criteria and were chosen for MLVA development (Tables 2 and 3).
Table 2

VNTR primer sequences and concentrations

Locus Name

Primer Sequence

PCR Mix

Final [Primer] uM

Dye

933 k

F: atggtggcggccgtcggcgaaaacc

1.1

0.20 *

Fam

 

R: gctcgaatgggtgtacgaagggccacgctgattc

 

0.2

 

2065 k

F: gggggacccggcgcacgacagg

1.1

0.20**

Vic

 

R: cggcgcgttgggacgatcggcttgat

 

0.2

 

2971 k

F: gcgcaagcgcgactcggccactcg

1.2

0.1

Pet

 

R: gtcgccgggcgcggggctacatcttctta

 

0.1

 

3145 k

F: ggcaggcaccgccggcatggaagc

1.2

0.2

Ned

 

R: gcgtcgcgcgtatcgatccgactgattgtacc

 

0.2

 

2666 k b

F: gctgcaagtccgccttcacgcgcatcag

2

0.13

Ned

 

R: gcggcggccggctcgagttggact

 

0.13

 

3671 k a

F: gcagcggctttggatcgcccgggttct

2

0.10*

Pet

 

R: gggccggggcgcggaagtcgaaagtt

 

0.1

 

2115 k a

F: ggtgcgtgctggtgtcgctgctgtgctatctgt

2

0.1

Vic

 

R: ggggaaggcgccggattgcccgagtt

 

0.1

 

2341 k

F: ggcttcgcacccgccccatttcagc

2

0.10**

Fam

 

R: gcaccgggcgcggcgcactcg

 

0.1

 

1500 k

F: cagagcgcggcgaggacgatcaaaaggag

2

0.10**

Fam

 

R: gccgcggctactggcgccaccattg

 

0.1

 

3091 k

F: aattcgtcggcagcgggcacggaagatg

3

0.20*

Vic

 

R: agcgggcacgcagcttgacggaacc

 

0.2

 

3152 k c

F: cggcgcggcgttcgtccggctactc

3

0.2

Pet

 

R: acgaatgcggggcccgaggttgacgatagg

 

0.2

 

3652 k

F: gattcggacggtcggccccgggtatcaa

3

0.25

Ned

 

R: gctggacgaaatccggggcgggacaaag

 

0.25

 

3564 k

F: ggccatgccgctgccgggttgagc

3

0.20*

Fam

 

R: cgcgggaagcgggttttgacgaagggtgtagttt

 

0.2

 

20 k

F: gcaccgcgagcgccgagcccgaac

4

0.20 *

Ned

 

R: gcgcccggcggccaaccctttgtcg

 

0.2

 

857 k

F: cgcgccggatacgccgtccaccag

4

0.2

Fam

 

R: acgccggcgccgcaatggctgtc

 

0.2

 

1690 k

F: cgtttcccgtttgatgcatttgcgttccctttgaa

4

2

Pet

 

R: catcgcggccgtcagaaaagttgagaaacctcgtc

 

2

 

2445 k

F: caggccgggccgtcgacgtgttcg

4

0.1

Vic

 

R: atcggggagggagggcgacgaggtgaagg

 

0.1

 

1367 k a

F: ggcgctgccgtggccggacgac

5

0.3

Ned

 

R: gccggcgaagcatcgaggcggtatg

 

0.3

 

1764 k

F: acccggtcggcacgctacggaactggttgtt

5

2

Pet

 

R: cggcggtgaactggcttggcggacctc

 

2

 

2815 k

F: cgaggacgcggctcaggtcgatgattttcagg

5

0.1

Fam

 

R: cggcgggcgggctttgcatgtcgt

 

0.1

 

2170 k

F: cgcatcggcgcaacgtcgtcatctcgt

6.1

0.10*

Fam

 

R: cggcgaccgcgcagggcagttga

 

0.1

 

389 k

F: gttacaagcgcgggtcggcaagaggctgaaa

6.1

0.10*

Vic

 

R: gccggtgttgaacgagtgggtggcgtaagc

 

0.1

 

1788 k

F: gcgcggcgagaacggcaagaacgaa

6.2

0.10*

Pet

 

R: gagcatcgggtgggcggcgcgtattgat

 

0.1

 

1217 k a

F: gcgagatgcgggcgtgtgcggtgtg

6.2

0.2**

Ned

 

R: gcggcggccgtgagcctgctgagaatc

 

0.2

 

397 k

F: cgcacgcgggcaggccgagacg

7

0.20**

Fam

 

R: gcggtcgcgcccttccacgcttcatc

 

0.2

 

2050 k

F: ccggcggccgcttcgtcgtctcg

7

0.2

Pet

 

R: cgcgaagtcgatccgcaactgcctgctcac

 

0.2

 

2862 k a

F: gattcggcgcggtccgtaccagcttgttgc

7

0.3

Vic

 

R: gcgcggggtatgtgacggggcagagc

 

0.3

 

140 k a

F: gcgcgcaccggccgcttcgactgacga

8

0.3

Fam

 

R: gcatacggtcgcgccgggcgggtggtaggaag

 

0.3

 

2356 k

F: ccgctgatcggcgtgctgacggtgtt

8

0.2

Ned

 

R: gctcggggcgctcggcgttctctg

 

0.2

 

2518 k a

F: caggcgcagttgtcgattgacgggtgtggac

8

0.2

Vic

 

R: acggcgggatgtgcgcggtctgacg

 

0.2

 

2124 k a

F: ctgcgcgtgctgcccggcgtcac

9

0.2

Vic

 

R: cgcgtggcggaatgcgcatgatagg

 

0.2

 

1934 k c

F: cgacgtgatccgcggctatctcgaagacg

9

0.2

Pet

 

R: ccgacgcggcttgccagcttggatcgttag

 

0.2

 

* 50% unlabeled Forward primer

** 75% unlabeled Forward primer

a Not recommended for globally diverse isolates

b Not used in phylogenetic analysis due to < 80% amplification

c Locus reported in Liu et al. 2006 [22]

Table 3

MLVA loci characteristics

Chromosome

VNTR Locus Name

Array start position in K96243

Consensus Repeat sequence

In CDS

Array Size in K96243 (bpxcopy#)

Amplicon Size Range (bp)

Number of Alleles

Nei's Diversity

Large

933 k

933861

CGGCGAGGGAAA

no

12 × 10

160–365

16

0.89

Small

2065 k

2064726

TCGAGTCA

no

8 × 8

238–370

21

0.9

Large

2971 k

2971247

CGTGCTT

no

7 × 9

201–314

22

0.92

Large

3145 k

3144932

CCTTCCTCG

no

9 × 8

220–345

14

0.86

Large

2666 k

2666129

CTTTCGCTA

yes

9 × 7

268–332

8

0.79

Large

3671 k

3671327

CTTGGAC

no

7 × 21

205–364

23

0.93

Small

2115 k

2115424

CGCCGGTT

no

8 × 15d

290–399

15

0.83

Large

2341 k

2340566

TTCGTGCGC

no

9 × 7

122–219

10

0.8

Large

1500 k

1500968

GGGAAAGTGCG

no

11 × 6

312–379

7

0.55

Small

3091 k

3091444

TCACGGC

no

7 × 12

202–287

11

0.86

Large

3152 k

3152382

GACTCG

no

6 × 17

160–371

26

0.94

Large

3652 k

3651903

CCGTAGTC

no

8 × 8

320–408

13

0.87

Large

3564 k

3563188

GCAGCCTTCTTCGCG

yes

15 × 30d

295–692

10

0.63

Large

20 k

20292

CGCCTCA

no

7 × 10

245–435

22

0.92

Small

857 k

857207

CGAAYGAGC

no

9 × 11

209–300

12

0.81

Small

1690 k

1689945

CGTCGATA

no

8 × 13

252–405

13

0.78

Small

2445 k

2444540

GGCACTTC

no

8 × 19

205–391

19

0.89

Small

1367 k

1366924

CGCRTCGAA

yes

9 × 24

454–686

26

0.92

Small

1764 k

1764166

GCCGCTGAAGTT

no

12 × 20

233–466

12

0.47

Large

2815 k

2815153

TGGCGTCTT

yes

9 × 7

223–439

19

0.86

Large

2170 k

2171435

ATGCCGTGG

no

9 × 24

229–513

25

0.93

Small

389 k

388768

GACGAACC

no

8 × 6

224–313

12

0.87

Small

1788 k

1788368

GTCGTGCGATCCTGCT

no

16 × 8

203–367

11

0.86

Large

1217 k

1217379

CGGACCTAGG

no

10 × 15

357–480

14

0.85

Small

397 k

397146

GCCCGAGA

no

8 × 12

226–401

17

0.88

Small

2050 k

2049749

CGATGCGGT/GCACCCAAC

yes/yes

9 × 8/9 × 8

377–549

18

0.92

Small

2862 k

2861834

CTCGCCTTTG

no

10 × 8

273–422

15

0.88

Large

140 k

139952

GCGCCGAA

no

8 × 15

367–675

28

0.93

Large

2356 k

2356018

CTTGGCGA

no

8 × 13

236–425

16

0.9

Small

2518 k

2517929

CCGCGAT

no

7 × 31

294–394

17

0.92

Small

2124 k

2123866

CCTTCGCG

no

8 × 23

332–490

14

0.88

Small

1934 k

1933513

CGAGTCGGCGGTT

no

13 × 16

224–645

21

0.91

B. pseudomallei and B. mallei genetic relationships

The 32-locus MLVA system was used to characterize 66 B. pseudomallei and 21 B. mallei isolates from diverse geographic locations (Table 4). These loci provide high levels of discrimination among different isolates of B. pseudomallei, with the number of alleles ranging between 7 to 28, and Nei's diversity values between 0.47 and 0.94 across all B. pseudomallei and B. mallei isolates (Table 3). Furthermore, the MLVA loci amplified equally well in both B. pseudomallei and closely related B. mallei strains, and showed variation between and among the two closely related species. MLVA loci did not PCR amplify in the more genetically distant B. thailandensis and B. cepacia.
Table 4

B. pseudomallei and B. mallei isolates

Species

Strain Name

Other Identifier

Country of Origin

Source

Site

Date

Tree Code

B. pseudomallei (66)

PHLS 5*

2002721617, NCTC 8016

Australia

Sheep

 

1949

Bp_Aust_Sheep_49

 

PHLS 91

2002721622, 84–1097

Australia

Sheep

Lung

1984

Bp_Aust_Sheep_84

 

PHLS 92

2002721623, 85–1097

Australia

Cow

Spleen

1985

Bp_AustCow_85

 

PHLS 83

 

Australia

Environment

Soil

 

Bp_Aust1_Env

 

PHLS 84*

 

Australia

Environment

Soil

 

Bp_Aust2_Env

 

PHLS 85

 

Australia

Environment

Soil

 

Bp_Aust3_Env

 

PHLS 104

 

Australia

Goat

Lymph node

 

Bp_Aust_Goat

 

146*

 

Australia

Animal

Right Udder

1992

Bp_Aust_NT_Animal_1_92

 

147

 

Australia

Animal

Med Lymph Node

1992

Bp_Aust_NT_Animal_2_92

 

213

 

Australia

Environmental

Soil

1993

Bp_Aust_NT_Envl_93

 

214

 

Australia

Environmental

Soil

1993

Bp_Aust_NT_Env2_93

 

465a

 

Australia

Human

Blood

1997

Bp_Aust_NT_Human_1_97

 

465e

 

Australia

Human

Sputum

1997

Bp_Aust_NT_Human_2_97

 

1459

 

Australia

Human

Sputum

2002

Bp_Aust_NT_Human_02

 

1627

 

Australia

Human

Sputum

2003

Bp_Aust_NT_Human_1_03

 

1628

 

Australia

Human

Throat

2003

Bp_Aust_NT_Human_2_03

 

PHLS 6

 

Bangledesh

Human

 

1960

Bp_Bangledesh_Human_60

 

PHLS 208

 

Ecuador

Human

  

Bp_Equador_Human

 

PHLS 68

 

Fiji

Human

Blood

1992

Bp_Fiji_Human_92

 

PHLS 33

2002721630, 7605

France

Environment

Manure

1976

Bp_France_Env_76

 

PHLS 24

2002721620, 7641

France

Horse

Stool

1976

Bp_France_Horse_76

 

PHLS 4075

 

Holland (tourist)

Human

Sputum

1999

Bp_tourist_2_99

 

PHLS 4152

 

Holland (tourist)

Human

Cervix

1999

Bp_tourist_3_99

 

PHLS 17

 

Indonesia

Monkey

Spleen

1990

Bp_Indo1_Monkey_90

 

PHLS 18*

 

Indonesia

Monkey

Pus

1990

Bp_Indo2_Monkey_90

 

PHLS 3477

 

Italy (Tourist SE Asia)

Human

Sputum

1998

Bp_Tourist1_98

 

PHLS 31*

 

Kenya

Environment

Water drain

1992

Bp_Kenya_Env_92

 

PHLS 25*

 

Madagascar

Environment

Soil

1977

Bp_Madagascar_Env_77

 

PHLS 71

 

Malaysia

Human

  

Bp_Malaysia1_Human

 

PHLS 72*

 

Malaysia

Human

  

Bp_Malaysia2_Human

 

PHLS 73

 

Malaysia

Human

  

Bp_Malaysia3_Human

 

PHLS 79

 

Malaysia

Human

  

Bp_Malaysia4_Human

 

PHLS 75*

 

Malaysia

Human

  

Bp_Malaysia5_Human

 

PHLS 9

2002721637, 521

Pakistan

Human

 

1988

Bp_Pakistan_Human_88

 

PHLS 16

 

Phillipines

Monkey

 

1990

Bp_Phillipines1_Monkey_90

 

PHLS 14

 

Phillipines

Monkey

Liver

1990

Bp_Phillipines2_Monkey_90

 

PHLS 39*

 

Singapore

Human

Blood

1988

Bp_Sing1_Human_88

 

PHLS 36

2002721635

Singapore

Human

 

1988

Bp_Sing2_Human_88

 

PHLS 38

 

Singapore

Human

 

1988

Bp_Sing3_Human_88

 

PHLS 40

 

Singapore

Human

 

1988

Bp_Sing4_Human_88

 

PHLS 19

 

Singapore

Environment

 

1991

Bp_Sing_Env_91

 

PHLS 3584

 

Sweden (Tourist SE Asia)

Human

Blood

1998

Bp_Tourist2_98

 

PHLS 8*

 

Thailand

Human

 

1988

Bp_Thai_Human_88

 

PHLS 20

 

Thailand

Human

Blood

1990

Bp_Thai_Human_90

 

PHLS 53

2002721633, 307a

Thailand

Human

Urine

1987

Bp_Thail_NE_Human_87

 

PHLS 43

 

Thailand

Human

 

1988

Bp_Thai1_NE_Human_88

 

PHLS 45

 

Thailand

Human

 

1988

Bp_Thai2_NE_Human_88

 

PHLS 47

 

Thailand

Human

 

1988

Bp_Thai4_NE_Human_88

 

PHLS 44*

 

Thailand

Human

 

1988

Bp_Thai5_NE_Human_88

 

PHLS 392

 

Thailand

Human

 

1989

Bp_Thai_NE_Human_89

 

PHLS 216

2002721626

Thailand

Environment

 

1990

Bp_Thai_NE_Env_90

 

PHLS 110

 

Thailand

Human

Urine

1992

Bp_Thai1_NE_Human_92

 

PHLS 111

 

Thailand

Human

Blood

1992

Bp_Thai2_NE_Human_92

 

PHLS 112*

 

Thailand

Human

 

1992

Bp_Thai3_NE_Human_92

 

PHLS 98/SID 2953*

 

United Kingdom

Human

 

1998

Bp_UK_Human1_98

 

PHLS 98/SID 3292*

 

United Kingdom

Human

 

1998

Bp_UK_Human2_98

 

99/SID 4349

 

United Kingdom

Human

 

1999

Bp_UK_Human_99

 

PHLS 2889

 

United Kingdom (Bangledesh national)

Human

Sputum

1998

Bp_Bangledesh_National_human_98

 

PHLS 3811

 

United Kingdom (Bangledesh national)

Human

Abscess

1999

Bp_Bangledesh_National_1_human_99

 

PHLS 3871

 

United Kingdom (Bangledesh national)

Human

Abscess

1999

Bp_Bangledesh_National_2_human_99

 

PHLS 3783*

 

United Kingdom (Tourist SE Asia)

Human

Sputum

1999

Bp_Tourist1_99

 

PHLS 35

2002721638, Ducrete

Vietnam

Human

 

1963

Bp_Vietnam_Human_63

 

PHLS 126

     

Bp1

 

ACTC 11668

     

Bp2

 

ACTC 15682

     

Bm_Hungary1_61

 

ACTC 23343

  

Type strain

  

Bp_TypeStrain

B. mallei (21)

ACTC 10399*

2002721275, GB11, NCTC 10245

China

Horse

Lung

1956

Bm_China_Horse_56

 

ACTC 15310

 

Hungary

  

1961

Bp3

 

NCTC 10229

GB5

Hungary

  

1961

Bm_Hungary2_61

 

NCTC 3708

GB9

India

Mule

Lung

1932

Bm_India_Mule_32

 

NCTC 3709

GB10

India

Horse

 

1932

Bm_India_Horse_32

 

NCTC 10260

GB6

Turkey

Human

 

1949

Bm_Turkey_Human_49

 

NCTC 10248

GB4

Turkey

Human

 

1950

Bm_Turkey_Human_50

 

NCTC 10247

GB7

Turkey

  

1960

Bm_Turkey_60

 

NCTC 120

GB3

United Kingdom

  

1920

Bm_UK_1920

 

85_503

  

Equine

  

Bm_equine

 

86_567

 

East India

Mule

  

Bm1

 

ISU

     

Bm2

 

Turkey_1

 

Turkey

   

Bm_Turkey1

 

Turkey_2

 

Turkey

   

Bm_Turkey2

 

Turkey_3

 

Turkey

   

Bm_Turkey3

 

Turkey_4

 

Turkey

   

Bm_Turkey4

 

Turkey_5

 

Turkey

   

Bm_Turkey5

 

Turkey_6

 

Turkey

   

Bm_Turkey6

 

Turkey_7

 

Turkey

   

Bm_Turkey7

 

Turkey_8

 

Turkey

   

Bm_Turkey8

 

Turkey_9

 

Turkey

   

Bm_Turkey9

* Isolates used in the screening panel.

Analysis of allelic variation at 23 loci using a Neighbor Joining distance algorithm revealed 62 genotypes among the 66 B. pseudomallei isolates and 19 genotypes among the 21 B. mallei isolates. Phylogenetic analysis of these VNTR data provided an extremely high level of strain discrimination even within B. pseudomallei isolates from single melioidosis patients (Patient 465 and chronic lung patient) and within isolates from a single B. pseudomallei outbreak focus in Australia (Goat Farms 1 and 2) (Figure 4). The average pairwise genetic distance was 0.86 for B. pseudomallei, and 0.61 for B. mallei.
Figure 4

Arbitrarily rooted phylogram of 66 B. pseudomallei and 21 B. mallei isolates. Colors indicate the geographic area from which the isolates were collected. Arrows indicate isolates from patients or from a specific outbreak event. Isolates that had identical MLST genotypes are bracketed and the sequence type is given. * indicates which B. mallei isolates were available on the MLST database.

A phylogram depicting this analysis indicates four highly diverse major clusters among the two Burkholderia sp., although there is less than 50% bootstrap support for these branches (Figure 4). These major clusters did not reveal any noticeable geographic or temporal relationships, with isolates from the same country or the same time period occurring in all groups. However, there are many instances in which the relationships between closely related isolates demonstrate clear geographic correlations with solid statistical support (Figure 4). Additionally, the tree indicates that overall, B. pseudomallei is much more diverse than B. mallei, although this could be due to the less geographically diverse nature of the B. mallei isolates. The tree clearly shows that the B. mallei isolates form a monophyletic group derived from a B. pseudomallei ancestor. The split between B. mallei and B. pseudomallei is supported by two MLVA loci (3564 k and 2445 k) that contain multiple alleles specific to B. mallei.

A comparison of a subset of isolates to other typing methods revealed that MLVA is much more discriminating between closely related isolates. MLST data for 37 of the 66 B. pseudomallei and four of the 21 B. mallei isolates used in this study were obtained from the online database [31]. A comparison of MLST and MLVA for these 37 B. pseudomallei isolates revealed seven instances where MLST sequence types were identical between isolates, while MLVA genotypes were different in all but two of these instances (Figure 4). Of particular note was the single MLST genotype for B. mallei and the multiple MLVA genotypes for the same isolates (n = 4). Additionally, a ribotyping study revealed three genotypes for seven of the B. mallei isolates (T2, T4, T5, T7, T9, GB5, GB6), while MLVA identified unique genotypes for every isolate [32].

Mutation rates of tandem repeats

Parallel serial passages experiments (PSPE) from a single B. pseudomallei isolate resulted in estimated ~18,000 generations of growth from which lineages were analyzed for variation in all MLVA loci. Mutational events were observed in 12 VNTR loci; the number and type of mutations observed are shown in Table 5. We observed comparable numbers of mutations for loci on each chromosome. There was a noticeable trend towards single repeat mutations (p = 0.0001) as well as bias towards insertion mutations (p = 0.0736) (Table 5). No discernable pattern was observed between loci that had mutations and those without mutations with respect to array size, repeat motif GC %, and/or amplification characteristics. The number of successful lineage PCR amplifications for the mutating MLVA loci ranged from 75–95 (out of 95 possible), averaging 90.25 ± 5.7; while those from the non-mutating loci ranged from 82–95, averaging 92.25 ± 3.1 (data not shown). (The basis of these failures is under investigation, but all mutation rates were corrected appropriately for these missing data.) We observed an average of 1.67 mutations per locus, and mutation rates for individual loci ranged from 5.3 × 10-5 to l.7 × 10-4. The combined mutation rate across all 32 loci was 1.113 × 10-3, which represents a discrimination power estimator for this MLVA typing system (Table 5). It is similar to the Y. pestis MLVA system rate and greater than the E. coli rate.
Table 5

B. pseudomallei1 MLVA loci mutation rate

Locus Name

Chromosome

Inside CDS

Array size*

Total Number of Mutations

Insertions

Deletions

Single Repeat Changes

Multiple Repeat Changes

Lineages**

Mutation Rate

1788 k

Small

no

16 × 9

1

-

1

1

-

94

5.3 × 10-5

2862 k

Small

no

10 × 11

3

3

-

2

1

90

1.7 × 10-4

1367 k

Small

yes

9 × 22

1

-

1

1

-

75

6.7 × 10-5

3145 k

Large

no

9 × 13

1

1

-

1

-

84

6.0 × 10-5

2170 k

Large

no

9 × 12

4

4

-

4

-

89

2.3 × 10-4

1690 k

Small

no

8 × 13

2

-

2

2

-

93

1.1 × 10-4

933 k

Large

no

12 × 14

1

-

1

1

-

91

5.5 × 10-5

2065 k

Small

no

8 × 14

1

1

-

1

-

95

5.3 × 10-5

2050 k

Small

yes

9 × 16

1

-

1

1

-

93

5.4 × 10-5

2518 k

Small

no

7 × 22

2

2

-

2

-

90

1.1 × 10-4

3152 k

Large

no

6 × 25

2

2

-

2

-

95

1.1 × 10-4

2815 k

Large

yes

9 × 19

1

1

-

1

-

94

5.3 × 10-5

Total

   

20

14

6

19

1

† 92.50

1.113 × 10 -3

1An isolate from the Arizona department of health (Bp9905-1902) was used for this study.

**Number of lineages successfully amplified out of 95 total

† Average number of lineages to amplify

We also examined mutation rates for 17 tandem repeat loci, not included in the final MLVA system, containing arrays found to be duplicated in up to four different locations within and/or between chromosomes (Table 6). In contrast to the MLVA loci, all duplicated loci screened consisted of either six or seven bp repeat motifs, as these were most commonly found with larger duplicated regions in the K96243 strain. Also, while the number of mutations for the duplicated arrays was equal to the MLVA loci, there were more mutations observed for large chromosome loci than small chromosome loci. Additionally, there was a nonsignificant trend towards multiple repeat mutations (p = 0.5127), as well as, a nominally significant trend towards deletion mutations (p = 0.0495) (Table 6). The multiple repeat mutations ranged from 2 to 6 repeat units. Two of the duplicated loci (1558 k and 3851 k), had less than 50% PCR amplification. Highly unpredictable PCR amplification was seen with three loci (3166 k, 1343 k and 2646 k). These PCR failures could be due to the difficult nature of PCR in a high GC organism such as B. psuedomallei, or could be indicative of loss of priming sites due to recombination. The PCR amplification success rates for the remaining loci were comparable to the MLVA loci. The duplicated loci averaged 2.6 mutations/locus, and combined mutation rate for 15 duplicated tandem repeat loci was also comparable to the non-duplicated MLVA loci, at 1.23 × 10-3 for ~18,000 generations.
Table 6

B. pseudomallei1 duplicated loci mutation rates

Locus Set

Locus Name

Array start position in K96243

Chromo-some

Inside CDS

Array Size *

Total Number of Mutations

Insertions

Deletions

Single Repeat Changes

Multiple Repeat Changes

Lineages **

Mutation Rate

1

1839 k

1839378

Large

no

7 × 2

1

-

1

1

-

82

6.12 × 10-5

1

3166 k

3166431

Small

no

-

-

-

-

-

-

-

 

2

1853 k

1853384

Large

no

7 × 27

3

-

3

2

1

94

1.71 × 10-4

2

2523 k

2523234

Small

no

7 × 5

-

-

-

-

-

88

 

2

817 k

817412

Small

no

7 × 17

3

-

3

-

3

89

1.69 × 10-4

3

1546 k

1546409

Large

no

6 × 20

-

-

-

-

-

83

 

3

2620 k

2620013

Large

no

6 × 20

3

1

2

3

-

93

1.64 × 10-4

3

3451 k

3451829

Large

no

6 × 27

1

1

-

1

-

94

5.34 × 10-5

3

3103 k

3103500

Small

no

6 × 27

2

1

1

2

-

94

1.07 × 10-4

4

200 k

199721

Small

no

7 × 12

-

-

-

-

-

94

 

4

735 k

734579

Small

no

6 × 9

-

-

-

-

-

92

 

5

1880 k

1879903

Small

no

7 × 6

-

 

-

-

-

92

 

5

3984 k

3983644

Large

no

7 × 43

7

3

4

3

4

94

3.82 × 10-4

6

1558 k

1558336

Large

no

6 × 11

-

-

-

-

-

37

 

6

1343 K

1343285

Small

yes

-

-

-

-

-

-

-

 

7

3851 k

3851246

Large

no

7 × 17

1

-

1

-

1

41

1.25 × 10-4

7

2646 K

2646281

Small

no

-

-

-

-

-

-

-

 

Total

     

21

6

15

12

9

† 83.36

1.23 × 10 -3

1An isolate from the Arizona department of health (Bp9905-1902) was used for this study.

*Estimated array size in the B. pseudomallei strain used in the mutation rate study

**Number of lineages successfully amplified out of 95 total

† Average number of lineages to amplify

Discussion

Burkholderia pseudomallei is a distinctive microbial pathogen due to its ability to survive and exploit a wide variety of environmental conditions, as well as, the opportunistic infection of animals. It can cause mild, chronic, or rapidly progressing and potentially fatal disease states in a range of animal hosts [33], and it has a demonstrated ability invade the cells of other eukaryotic organisms such as fungi and amoeba [34, 35]. It has been known to survive extreme environmental conditions for long periods of time, including nutrient starvation [36], and chlorine concentrations generally recognized as sufficient for potable water treatment [37]. This level of environmental flexibility and pathogenic potential may require the B. pseudomallei genome to be highly plastic in order to quickly adapt to different environments. Indeed, while the large chromosome primarily harbors genes essential for growth, the small chromosome contains more diverse genes that are primarily involved in survival and/or exploiting variable or contingent environmental conditions. Consequently, it is not biologically surprising that numerous genetic typing methodologies [6, 14, 15, 22], including the MLVA system reported here, find very high levels of genetic diversity within this organism. The high level of genetic diversity and host flexibility of the organism suggest enhanced mechanisms for generating and maintaining adaptive variation through processes such as selection, recombination and mutation.

The unusually high number of tandem repeats in B. pseudomallei (compared to other pathogenic bacteria with similarly sized genomes such as B. anthracis and Y. pestis, and other bacteria of similar GC content [5]) is indicative of potentially high genomic diversity which, in turn, may facilitate rapid genomic adaptation to a variable environment. While the majority of large VNTRs in B. pseudomallei are located intergenically and thus may have no direct phenotypic effect, it has been observed in other bacteria that such loci, when upstream of genes, can alter important biological functions through mechanisms such as transcriptional regulation and amino acid changes [3841]. Within coding regions we observed fewer tandem repeat arrays. The majority of these tandem arrays contain repeat units in multiples of three, which indicates the potential for adaptive variation. For example, Nierman et al. [5] observed variation in triplet repeat unit simple sequence repeat (SSR) loci that are located inside four genes coding for surface or putative virulence proteins in B. mallei and B. pseudomallei. A subsequent serial passage experiment of B. mallei through several mammalian hosts revealed indels in seven intragenic SSR loci, five of which caused frameshift mutations, while the other two were triplet repeats that only added or removed amino acids from the encoded protein [42]. This variation is consistent with the potential for phase variation during the infection cycle and may be a mechanism to avoid host defenses [5, 42]. Thus, given the similarity of B. mallei and B. pseudomallei, the unusually high number of tandem repeat loci in B. pseudomallei, as well as their non-random arrangement, as indicated by a deviation from the expected Poisson distribution (Figure 3), may indicate that coding and non-coding genomic regions use different molecular mechanisms to adapt to different selective pressures.

In addition to the large number of tandem repeats in B. pseudomallei, there was a prevalence of duplicated tandem repeats throughout the genome. In B. pseudomallei, 37.9% of tandem repeats in the large chromosome and 35.2% of tandem repeats in the small chromosome are found to be duplicated, at least in part, at other intra- and inter-chromosomal locations. Moreover, a serial passage experiment revealed that the duplicated loci show a contrasting trend towards deletions, as well as an increased frequency of multiple repeat changes in comparably sized repeat arrays, while displaying comparable mutation rates to non-duplicated loci; which is in contrast to the lack of bias in Y. pestis [43]. This suggests that the repeat regions within B. pseudomallei may facilitate large scale genomic rearrangements through recombination rather than slip-strand mispairing [44]. Although this has not been specifically studied in B. pseudomallei, it has been suggested that SSRs in Mycoplasma genomes may in fact facilitate genomic rearrangements via recombination [45], and that long tracts of tandem repeats may facilitate gene transfer [46]. Conversely, tandem repeats may not directly cause recombination, but rather be associated with regions that are prone to recombination for other reasons. Since recombination frequency is affected by the length of the homology between two loci [47] which in turn is controlled by slip strand repair, the observed tandem repeat patterns could represent an interesting interaction between slip strand expansion and recombination.

During in vitro passage, mutation events were observed in multiple B. pseudomallei VNTR loci suggesting similar mutation rates at many loci. The MLVA combined mutation rate reported in this study is 1.113 × 10-3 mutations/generation, compared to combined MLVA rates in E. coli and Y. pestis rates of 6.4 × 10-4 and 1.1 × 10-3 mutations/generation (respectively) [26, 43, 48]. The combined rate is, hence, comparable to those previously observed in E. coli and Y. pestis and offers similar subtyping discriminatory power. These rate calculations are dependent upon accurate estimation of the population growth parameters during serial passage and this may be particularly problematic for B. pseudomallei, which forms highly mucoid colonies. Experimental serial passage studies in E. coli and Y. pestis have previously identified a positive correlation between the in vitro mutation rate and natural locus diversity. This correlation was not detected in B. pseudomallei (analysis not shown) and it is not immediately obvious what differs between these pathogens. Perhaps due the much larger number of VNTR loci in B. pseudomallei, the current study was based upon an overwhelming number of equally and highly mutable loci, which are not commonly present in other genomes. In other words, the marker loci in E. coli and Y. pestis MLVA systems are stratified by their mutability but in the Burkholderia MLVA we may examining a number loci that are equally mutable. Thus, there is no correlation with array size. Another interesting difference is in the mutation products, where the majority (19:1) were single repeat changes. This bias was greater than observed in the E. coli and Y. pestis studies where the single-repeat mutational products were about 80% of the total observed. The lack of more two and three repeat changes needs to be explored in a larger in vitro population to see if this trend repeats reality in this particular genome.

Here we present a rapid PCR-based MLVA typing system using 32 independent VNTR loci. Although the initial development of a MLVA system in this organism was complicated by the quantity and duplicated nature of repeated regions found in B. pseudomallei and inconsistencies of the allelic size variation in comparison to the repeat unit size, we found 23 markers that were useful for phylogenetic analysis due to high diversity levels, minimal partial repeat differences and amplification success. An additional nine loci, while demonstrating some partial repeat sizes, are very useful for even finer scale resolution of closely related B. pseudomallei and B. mallei isolates from outbreak situations [49]. While no specific effort was made to design the MLVA primers specific to B. mallei, all B. mallei isolates tested amplified well at every locus, as expected given the phylogenetic relationship of the two species [6]. Conversely, B. thailandensis and B. cepacia did not amplify well in any of the loci, indicating that the MLVA loci primers will not support amplification in more distantly related bacterial species, although this has not been explicitly tested. Thus, this MLVA system represents a reliable method of identifying B. pseudomallei as well as B. mallei strains. Furthermore, this typing method is an easily transferable approach to high-resolution molecular typing analysis using low levels of crudely isolated DNA. The unique size and fluorescent label of each allele, as well as automated sizing software, allows for easy classification of each VNTR allele, and capillary electrophoresis significantly reduces run time.

Due to the relative effects of convergent evolution, reversal mutations, recombination, gene duplications and suggested horizontal gene transfer within Burkholderia pseudomallei, phylogenetic hypotheses have been difficult to establish. For example, neither MLST [6] nor MLVA are able to resolve the deeper relationships among distantly related B. pseudomallei isolates, as illustrated by the poor bootstrap support for deeper branches (Figure 4) and similar levels of consistency for a subset of the same isolates (~0.63) (data not shown). This lack of resolution results in the absence of a geographic correlation within basal clades, although more derived clades do demonstrate geographic associations between isolates (Figure 4). In comparison, an analysis of Thai and Australian isolates using MLST exhibited no overlap between sequence types for the two countries [50]. However, phylogenetic analysis of these data lacks strong bootstrap values to support this geographic differentiation. Also, the analysis of historical isolates of B. pseudomallei using MLST reveals an overlapping sequence type between Australia and Thailand environmental isolates, and does not support the genetic distinction of isolates from Australia [51]. Thus, phylogenetic hypotheses using both MLVA and MLST data are difficult to establish with isolates that are geographically and temporally distant.

The present typing system targets VNTR loci over a wide range of diversity levels and consequently provides resolution between B. pseudomallei and B. mallei, while still providing high levels of discrimination between closely related isolates due to the high variability of tandem repeat loci in these bacterial pathogens. Whereas a number of typing methodologies such as PFGE, ribotyping, RAPDs and MLST have detected differences between isolates, their resolving power among very closely related isolates is less than MLVA [6, 14, 15, 19]. For example, while MLST analysis provided only a single unique genotype for the B. mallei cluster, MLVA further resolved the B. mallei group into individual genotypes, even among very closely related isolates from Turkey with the same ribotype [32]. Additionally, B. pseudomallei isolates with the same sequence type often had different MLVA genotypes (Figure 4). This type of high resolution genotyping can define patterns of mutation within very closely related isolates from an outbreak, which can then be used for generating phylogenetic hypotheses [49].

A recent study by Liu et al. (2006) used six VNTR loci to differentiate B. pseudomallei isolates from an outbreak in Singapore [22]. Four of the six loci used were characterized in the present MLVA study. Two of these loci are included in this MLVA (Table 2), but the other two loci were found to be duplicated within the genome, and consequently were not included in MLVA development. This six-locus MLVA offered insight into the epidemiology of B. pseudomallei in Singapore, but presented limitations due to the lack of resolution inherent in agarose gel electrophoresis. Given the partial repeat sizes (as small as 3 bp) seen with capillary electrophoresis, it is doubtful that all alleles for these loci were detectable using agarose gels, and thus levels of diversity were underestimated. Additionally, because two of the VNTR loci that were used are duplicated within the genome, they are not recommended for phylogenetic analysis due to the confounding phylogenetic effects of gene duplication and associated possibilities for independent evolutionary trajectories.

Conclusion

In summary, the findings of this study suggest that the prevalence and location of tandemly repeated regions within the B. pseudomallei genome may generate and maintain adaptive variation in this bacterial pathogen. The intragenically located repeat regions, found twice as frequently on the "contingency-oriented" small chromosome [4], may provide for rapid changes in gene function. Duplicated repeat regions may facilitate genomic rearrangements which can lead to altered gene regulation. While the mutation rates of individual repeat regions do not appear to be enhanced over those in other organisms, the sheer number of these regions, some of which are quite large, provides great potential for genetic variation within this species.

Epidemiological characterization is important in any pathogen, but most especially for those that are emerging as global pathogens that may be exploited for biological terrorism, such as B. pseudomallei. While no typing system for B. pseudomallei can currently be used to reliably establish deep phylogeneic relationships, the B. pseudomallei-B. mallei multiplex MLVA typing system presented here provides unsurpassed ability to resolve very closely related isolates, even those from the same patient. Efficient and sensitive genetic typing tools, such as the MLVA system presented here, are important for facilitating the increasingly important epidemiological and phylogenetic characterization of emerging pathogens.

Methods

DNA preparation

DNA for 66 B. pseudomallei and 21 B. mallei isolates was obtained from different institutions which used different extraction methods such as Dneasy (Qiagen, Valencia, CA) [52] and phenol/chloroform extraction [32] and quantified using a Pico Green quantification kit (Molecular Probes, Eugene, OR) and a minifluorometer (Turner Biosystems, Sunnyvale, CA). DNA was then normalized to100 pg/μL for VNTR screening. Isolates for the global panel were selected to represent a wide variety of isolates in terms of geographic distribution, host source and date of isolation (Table 4).

VNTR identification

The complete genome sequence of Burkholderia pseudomallei strain K96243 was obtained from the National Center for Biotechnology Information [GenBank: NC_006350, NC_006351] and screened for potentially polymorphic repetitive sequences that were comprised of ≥ dinucleotide repeats, 4 copies and a total array size of 30 bp using GeneQuest (Lasergene, Inc., Madison, WI) and Tandem Repeats Finder [53]. Primers flanking repeat sequences were designed using Primer Express (Lasergene, Inc., Madison, Wis.).

To assess the variability of repeated regions among a globally distributed set of isolates and to develop a comprehensive typing system for this organism, 104 repeated regions (48 from the large chromosome, 56 from the small) were targeted for analysis and subsequent incorporation into a multiple-locus VNTR analysis (MLVA) system. These VNTR loci were selected based upon PCR amplicon size, array size, locus duplication, and proximity to other arrays. Loci resulting in small PCR fragment sizes (<1000 bp) were favored since such loci amplified better than larger regions, and are best suited for analytical platforms. Arrays with fewer than five copies of a motif were not selected for screening. Loci that were duplicated, either within or between chromosomes were also eliminated since multiple alleles would confuse a typing system. Lastly, repeat regions in close proximity (<1000 bp) to other repeat regions were avoided to preserve locus independence. Loci were not excluded based on their intra or intergenic location. The 104 candidate loci were examined for robust amplification and polymorphism across a screening panel which was comprised of 29 B. pseudomallei isolates, one B. mallei isolate (ATCC 10399), and one B. thailandensis isolate (ATCC 700388). B. pseudomallei stains in the screening panel included 15 closely related isolates from two different outbreaks in northern Australia [49], and 14 geographically diverse isolates from seven different countries (Table 4). This tiered screening panel allowed us to identify loci with varying degrees of polymorphism.

VNTR screening using universal tail PCR and genotype analysis

A high throughput five dye Universal Tail amplification and labeling methodology, developed for use in the low GC (x = 35%) bacterium B. anthracis [54], was used to screen the chosen repeat region loci for variation among a combination of 29 diverse and closely related B. pseudomallei isolates. The optimal Tm for labeling sequences in B. anthracis is 55°C, however due to the high G-C (x = 68.12%) content of the B. pseudomallei genome, all PCR reactions were performed at a Tm of 72°C.

The UT PCR labeling protocol provides for fluorescent labeling of any PCR amplicon with only four universal fluorescently labeled oligonucleotiodes. The fluorescently labeled universal primer is complimentary to a universal tailed primer sequence on the 5' end of the target specific forward primer (FAM = ACCCAACTGAATAGAGAGC, NED = ATCGACTGTGTTAGGTCAC, PET = CTGTCCTTACCTCAATCTC and VIC = ACGCACTTGACTTGTCTTC). This method significantly reduces the cost of initial screening by not having to order labeled primers for each locus.

PCR amplifications were performed using MJ Research 96-well DNA engines equipped with hot bonnets (Bio-Rad, Waltham, MA). Reaction volumes equaled 10 μL and contained the following: 10× Hot Master Taq buffer with Mg2+ (Brinkmann-Eppendorf, Westbury, New York), 200 μM deoxynucleoside triphosphates (Invitrogen, Carlsbad, CA), 5 μM tailed primer, 50 μM untailed primer, 50 μM fluorescently labeled universal primer (Applied Biosystems, Foster City, CA), 1U Hot Master Taq DNA Polymerase (Brinkmann-Eppendorf, Westbury, New York) and double-distilled H20. After an initial denaturation step at 94°C for 2 min, 30 cycles of touchdown PCR were performed (denaturation at 94°C for 30 sec, annealing for 30 sec with an 0.5°C/cycle decrement at 72°C, and an extension at 72°C for 30 sec) followed by 20 cycles of regular PCR (94°C for 30 sec, 55°C for 30 sec, 72°C for 30 sec), followed by a final extension step for 5 min at 72°C. Negative controls, containing all the components except DNA templates, were included in parallel. PCR samples were stored at -20°C until genotyped.

PCR amplicons were diluted with double-distilled H20 based upon their universal tail sequence (FAM and NED 1:50, PET 1:10 and VIC 1:5) and mixed in equal amounts to provide relatively equal fluorescent signals from each locus during subsequent electrophoresis on an Applied Biosystems 3100 DNA sequencer (Applied Biosystems, Foster City, CA). Size polymorphisms were subsequently analyzed and scored using GeneScan and Genotyper software (Applied Biosystems, Foster City, CA).

MLVA PCR and genotype analysis

Primers for 32 polymorphic VNTR loci were redesigned with fluorescently labeled forward primers, and optimized for 11 multiplex PCR reactions across B. mallei ATCC 10399 and the 14 globally diverse B. pseudomallei isolates used in the initial screening panel (Table 4). These isolates were chosen to increase future amplification success across an array of genetically diverse isolates. MLVA reaction primers (Table 2) were designed to provide uniquely labeled or sized amplicons for every allele at all 32 VNTR loci. PCR amplification of all loci was routinely accomplished using 11 reactions, which were pooled into nine electrophoretic runs.

All reactions contained a final concentration of 1× PCR buffer, 2 mM MgCl2, 200 μM of deoxynucleoside triphosphates, 0.08 units Taq DNA Polymerase (Invitrogen, Carlsbad, CA), 1.2 M Betaine (Sigma-Aldrich Co., St. Louis, MO), double-distilled H20, 1 μL of template DNA (~100 pg/μL) and the appropriate primer concentrations for each multiplex PCR (Table 1) for a total volume of 10 μL. Thirteen VNTR loci required the inclusion of a non-fluorescently labeled forward primer in order to decrease the amount of fluorescent amplicon, and thus obtain relatively equal fluorescent signals from each amplicon in the multiplex mix (Table 2). In cases where low DNA quantity affected multiplex PCR results, loci were amplified individually using the same concentrations above and 0.2 μM of both forward and reverse primers.

All PCR reactions were performed in MJ Research 96-well DNA engines equipped with hot bonnets (Bio-Rad, Waltham, MA). PCR reactions underwent an initial denaturation at 94°C for 5 min, 35 cycles of PCR were performed (denaturation at 94°C for 30 sec, annealing for 30 sec at 68°C, and an extension at 72°C for 30 sec) followed by a final extension step for 5 min at 72°C. Negative controls, containing all the components except DNA templates, were included in parallel. PCR samples were stored at -20°C until genotyped.

PCR products for all multiplex mixes were diluted 1:100 with double-distilled H20 and then mixed 1:1 with a 3:1 ratio of formamide to NAU Liz 1007 fluorescently labeled size standard. The PCR products were electrophorectically analyzed with an Applied Biosystems 3730xl DNA sequencer (Foster City, CA). Amplicons were scored using the ABI software program GeneMapper and genotyped according to predetermined allele sizes. An independent party has verified all sizes presented.

Mutation rate determination

A parallel serial passage experiment used to determine VNTR mutation rates began with a single isolated colony of the Bp9905-1902 strain (T = 0). Bp9905-1902 was a human clinical isolate obtained from the Arizona Department of Health. This colony was dispersed in nutrient broth and then used to start 95 independent clonal lineages by streaking for single colonies on 24 quartered plates. Each lineage was then serially passed 10 times over a 10 day period by streaking a single colony from the previous passage. DNA was extracted from all 95 T = 10 lineages by using an in-house phenol chloroform extraction protocol. PCR for each locus was performed using the universal tail VNTR screening method described above. Mutational events were then visualized using GeneMapper software (Applied Biosystems, Foster City, CA). Using viable plate counts, the number of generations (doublings) per colony was determined to be ~19.93 (log2 of the average colony size in cells), which corresponded to a total of 1.81 × 104 generations in the entire experiment (19.93 generations/colony × 10 passages × 91.03 average analyzed lineages/marker), allowing the detection of mutation rates of 10 -4 or greater (Table 1). For estimation of cell doubling see discussion and supplemental information in Girard and Wagner et al. [26, 24].

Statistical analyses

Data from 23 loci that displayed greater than 85% amplification success were used to generate an arbitrarily rooted distance-based phylogenetic tree using the Neighbor Joining algorithm in PAUP 4.0 b10 [55]. In order to estimate confidence levels for the analysis, a full heuristic bootstrapping analysis was conducted using a random generator seed for 2000 replicates. Individual marker diversity (D) was calculated as equal to 1-∑ (allele frequency)2 and based solely upon allele frequencies in the 87 isolates shown here (Table 1). A χ2 goodness-of-fit test was performed in 10 Kb intervals in order to examine the observed distribution against an expected Poisson distribution for both the large and small chromosomes.

Notes

Declarations

Acknowledgements

We would like to acknowledge Dr. Amy Vogler for her assistance with the statistical analyses, as well as, Dr. Matthew Van Ert and Mr. Eric Greene for their help in the editing this manuscript. Funding for this work was provided by the U.S. Departments of Energy and Homeland Security.

Authors’ Affiliations

(1)
Center for Microbial Genetics and Genomics, Northern Arizona University
(2)
U.S. Army Medical Research Institute for Infectious Diseases
(3)
U.S. Army Edgewood Chemical Biological Center, Aberdeen Proving Ground
(4)
Brigham Young University
(5)
Menzies School of Health Research, Charles Darwin University

References

  1. White NJ: Melioidosis. Lancet. 2003, 361 (9370): 1715-1722. 10.1016/S0140-6736(03)13374-0.View ArticlePubMedGoogle Scholar
  2. Rotz LD, Khan AS, Lillibridge SR, Ostroff SM, Hughes JM: Public health assessment of potential biological terrorism agents. Emerg Infect Dis. 2002, 8 (2): 225-230.PubMed CentralView ArticlePubMedGoogle Scholar
  3. Rogul M, Brendle JJ, Haapala DK, Alexander AD: Nucleic acid similarities among Pseudomonas pseudomallei, Pseudomonas multivorans, and Actinobacillus mallei. J Bacteriol. 1970, 101 (3): 827-835.PubMed CentralPubMedGoogle Scholar
  4. Holden MT, Titball RW, Peacock SJ, Cerdeno-Tarraga AM, Atkins T, Crossman LC, Pitt T, Churcher C, Mungall K, Bentley SD, Sebaihia M, Thomson NR, Bason N, Beacham IR, Brooks K, Brown KA, Brown NF, Challis GL, Cherevach I, Chillingworth T, Cronin A, Crossett B, Davis P, DeShazer D, Feltwell T, Fraser A, Hance Z, Hauser H, Holroyd S, Jagels K, Keith KE, Maddison M, Moule S, Price C, Quail MA, Rabbinowitsch E, Rutherford K, Sanders M, Simmonds M, Songsivilai S, Stevens K, Tumapa S, Vesaratchavest M, Whitehead S, Yeats C, Barrell BG, Oyston PC, Parkhill J: Genomic plasticity of the causative agent of melioidosis, Burkholderia pseudomallei. Proc Natl Acad Sci U S A. 2004, 101 (39): 14240-14245. 10.1073/pnas.0403302101.PubMed CentralView ArticlePubMedGoogle Scholar
  5. Nierman WC, DeShazer D, Kim HS, Tettelin H, Nelson KE, Feldblyum T, Ulrich RL, Ronning CM, Brinkac LM, Daugherty SC, Davidsen TD, Deboy RT, Dimitrov G, Dodson RJ, Durkin AS, Gwinn ML, Haft DH, Khouri H, Kolonay JF, Madupu R, Mohammoud Y, Nelson WC, Radune D, Romero CM, Sarria S, Selengut J, Shamblin C, Sullivan SA, White O, Yu Y, Zafar N, Zhou L, Fraser CM: Structural flexibility in the Burkholderia mallei genome. Proc Natl Acad Sci U S A. 2004, 101 (39): 14246-14251. 10.1073/pnas.0403306101.PubMed CentralView ArticlePubMedGoogle Scholar
  6. Godoy D, Randle G, Simpson AJ, Aanensen DM, Pitt TL, Kinoshita R, Spratt BG: Multilocus sequence typing and evolutionary relationships among the causative agents of melioidosis and glanders, Burkholderia pseudomallei and Burkholderia mallei. J Clin Microbiol. 2003, 41 (5): 2068-2079. 10.1128/JCM.41.5.2068-2079.2003.PubMed CentralView ArticlePubMedGoogle Scholar
  7. Kim HS, Schell MA, Yu Y, Ulrich RL, Sarria SH, Nierman WC, DeShazer D: Bacterial genome adaptation to niches: divergence of the potential virulence genes in three Burkholderia species of different survival strategies. BMC Genomics. 2005, 6: 174-10.1186/1471-2164-6-174.PubMed CentralView ArticlePubMedGoogle Scholar
  8. Howe C: Glanders. The Oxford Medicine. 1950, New York , Oxford Univ. Press, 185–202-Google Scholar
  9. Srinivasan A, Kraus CN, DeShazer D, Becker PM, Dick JD, Spacek L, Bartlett JG, Byrne WR, Thomas DL: Glanders in a military research microbiologist. N Engl J Med. 2001, 345 (4): 256-258. 10.1056/NEJM200107263450404.View ArticlePubMedGoogle Scholar
  10. Wheelis M: First shots fired in biological warfare. Nature. 1998, 395 (6699): 213-10.1038/26089.View ArticlePubMedGoogle Scholar
  11. United States. Army Medical Dept.: Military Medical Ethics Volume 2... U.S. Department Of Defense... 2003. 2004, [S.l. , s.n.Google Scholar
  12. Gal D, Mayo M, Smith-Vaughan H, Dasari P, McKinnon M, Jacups SP, Urquhart AI, Hassell M, Currie BJ: Contamination of hand wash detergent linked to occupationally acquired melioidosis. Am J Trop Med Hyg. 2004, 71 (3): 360-362.PubMedGoogle Scholar
  13. Cheng AC, Jacups SP, Gal D, Mayo M, Currie BJ: Extreme weather events and environmental contamination are associated with case-clusters of melioidosis in the Northern Territory of Australia. Int J Epidemiol. 2006, 35 (2): 323-329. 10.1093/ije/dyi271.View ArticlePubMedGoogle Scholar
  14. Sermswan RW, Wongratanacheewin S, Trakulsomboon S, Thamlikitkul V: Ribotyping of Burkholderia pseudomallei from clinical and soil isolates in Thailand. Acta Trop. 2001, 80 (3): 237-244. 10.1016/S0001-706X(01)00176-0.View ArticlePubMedGoogle Scholar
  15. Currie B, Smith-Vaughan H, Golledge C, Buller N, Sriprakash KS, Kemp DJ: Pseudomonas pseudomallei isolates collected over 25 years from a non-tropical endemic focus show clonality on the basis of ribotyping. Epidemiol Infect. 1994, 113 (2): 307-312.PubMed CentralView ArticlePubMedGoogle Scholar
  16. Popovic T, Kim C, Reiss J, Reeves M, Nakao H, Golaz A: Use of molecular subtyping to document long-term persistence of Corynebacterium diphtheriae in South Dakota. J Clin Microbiol. 1999, 37 (4): 1092-1099.PubMed CentralPubMedGoogle Scholar
  17. Miettinen MK, Siitonen A, Heiskanen P, Haajanen H, Bjorkroth KJ, Korkeala HJ: Molecular epidemiology of an outbreak of febrile gastroenteritis caused by Listeria monocytogenes in cold-smoked rainbow trout. J Clin Microbiol. 1999, 37 (7): 2358-2360.PubMed CentralPubMedGoogle Scholar
  18. Keim P, Price LB, Klevytska AM, Smith KL, Schupp JM, Okinaka R, Jackson PJ, Hugh-Jones ME: Multiple-locus variable-number tandem repeat analysis reveals genetic relationships within Bacillus anthracis. J Bacteriol. 2000, 182 (10): 2928-2936. 10.1128/JB.182.10.2928-2936.2000.PubMed CentralView ArticlePubMedGoogle Scholar
  19. Haase A, Melder A, Smith-Vaughan H, Kemp D, Currie B: RAPD analysis of isolates of Burkholderia pseudomallei from patients with recurrent melioidosis. Epidemiol Infect. 1995, 115 (1): 115-121.PubMed CentralView ArticlePubMedGoogle Scholar
  20. Haase A, Smith-Vaughan H, Melder A, Wood Y, Janmaat A, Gilfedder J, Kemp D, Currie B: Subdivision of Burkholderia pseudomallei ribotypes into multiple types by random amplified polymorphic DNA analysis provides new insights into epidemiology. J Clin Microbiol. 1995, 33 (7): 1687-1690.PubMed CentralPubMedGoogle Scholar
  21. Maiden MC, Bygraves JA, Feil E, Morelli G, Russell JE, Urwin R, Zhang Q, Zhou J, Zurth K, Caugant DA, Feavers IM, Achtman M, Spratt BG: Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc Natl Acad Sci U S A. 1998, 95 (6): 3140-3145. 10.1073/pnas.95.6.3140.PubMed CentralView ArticlePubMedGoogle Scholar
  22. Liu Y, Loh JP, Aw LT, Yap EP, Lee MA, Ooi EE: Rapid molecular typing of Burkholderia pseudomallei, isolated in an outbreak of melioidosis in Singapore in 2004, based on variable-number tandem repeats. Trans R Soc Trop Med Hyg. 2006, 100 (7): 687-692. 10.1016/j.trstmh.2005.08.017.View ArticlePubMedGoogle Scholar
  23. Johansson A, Farlow J, Larsson P, Dukerich M, Chambers E, Bystrom M, Fox J, Chu M, Forsman M, Sjostedt A, Keim P: Worldwide genetic relationships among Francisella tularensis isolates determined by multiple-locus variable-number tandem repeat analysis. J Bacteriol. 2004, 186 (17): 5808-5818. 10.1128/JB.186.17.5808-5818.2004.PubMed CentralView ArticlePubMedGoogle Scholar
  24. Le Flèche P, Hauck Y, Onteniente L, Prieur A, Denoeud F, Ramisse V, Sylvestre P, Benson G, Ramisse F, Vergnaud G: A tandem repeats database for bacterial genomes: application to the genotyping of Yersinia pestis and Bacillus anthracis. BMC Microbiology. 2001, 1 (1): 2-10.1186/1471-2180-1-2.PubMed CentralView ArticlePubMedGoogle Scholar
  25. Svraka S, Toman R, Skultety L, Slaba K, Homan WL: Establishment of a genotyping scheme for Coxiella burnetii. FEMS Microbiol Lett. 2006, 254 (2): 268-274.View ArticlePubMedGoogle Scholar
  26. Girard JM, Wagner DM, Vogler AJ, Keys C, Allender CJ, Drickamer LC, Keim P: Differential plague-transmission dynamics determine Yersinia pestis population genetic structure on local, regional, and global scales. Proc Natl Acad Sci U S A. 2004, 101 (22): 8408-8413. 10.1073/pnas.0401561101.PubMed CentralView ArticlePubMedGoogle Scholar
  27. Hoffmaster AR, Fitzgerald CC, Ribot E, Mayer LW, Popovic T: Molecular subtyping of Bacillus anthracis and the 2001 bioterrorism-associated anthrax outbreak, United States. Emerg Infect Dis. 2002, 8 (10): 1111-1116.PubMed CentralView ArticlePubMedGoogle Scholar
  28. Budowle B, Johnson MD, Fraser CM, Leighton TJ, Murch RS, Chakraborty R: Genetic analysis and attribution of microbial forensics evidence. Crit Rev Microbiol. 2005, 31 (4): 233-254. 10.1080/10408410500304082.View ArticlePubMedGoogle Scholar
  29. Johansson A, Forsman M, Sjostedt A: The development of tools for diagnosis of tularemia and typing of Francisella tularensis. Apmis. 2004, 112 (11-12): 898-907. 10.1111/j.1600-0463.2004.apm11211-1212.x.View ArticlePubMedGoogle Scholar
  30. Lindstedt BA: Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis. 2005, 26 (13): 2567-2582. 10.1002/elps.200500096.View ArticlePubMedGoogle Scholar
  31. Burkholderia pseudomallei Multi Locus Sequencing Typing Database.http://bpseudomallei.mlst.net/
  32. Harvey SP, Minter JM: Ribotyping of Burkholderia mallei isolates. FEMS Immunol Med Microbiol. 2005, 44 (1): 91-97. 10.1016/j.femsim.2004.12.002.View ArticlePubMedGoogle Scholar
  33. Cheng AC, Currie BJ: Melioidosis: epidemiology, pathophysiology, and management. Clin Microbiol Rev. 2005, 18 (2): 383-416. 10.1128/CMR.18.2.383-416.2005.PubMed CentralView ArticlePubMedGoogle Scholar
  34. Levy A, Chang BJ, Abbott LK, Kuo J, Harnett G, Inglis TJ: Invasion of spores of the arbuscular mycorrhizal fungus Gigaspora decipiens by Burkholderia spp. Appl Environ Microbiol. 2003, 69 (10): 6250-6256. 10.1128/AEM.69.10.6250-6256.2003.PubMed CentralView ArticlePubMedGoogle Scholar
  35. Inglis TJ, Rigby P, Robertson TA, Dutton NS, Henderson M, Chang BJ: Interaction between Burkholderia pseudomallei and Acanthamoeba species results in coiling phagocytosis, endamebic bacterial survival, and escape. Infect Immun. 2000, 68 (3): 1681-1686. 10.1128/IAI.68.3.1681-1686.2000.PubMed CentralView ArticlePubMedGoogle Scholar
  36. Wuthiekanun V, Smith MD, Dance DA, White NJ: Isolation of Pseudomonas pseudomallei from soil in north-eastern Thailand. Trans R Soc Trop Med Hyg. 1995, 89 (1): 41-43. 10.1016/0035-9203(95)90651-7.View ArticlePubMedGoogle Scholar
  37. Howard K, Inglis TJ: Novel selective medium for isolation of Burkholderia pseudomallei. J Clin Microbiol. 2003, 41 (7): 3312-3316. 10.1128/JCM.41.7.3312-3316.2003.PubMed CentralView ArticlePubMedGoogle Scholar
  38. Dawid S, Barenkamp SJ, St Geme JW: Variation in expression of the Haemophilus influenzae HMW adhesins: a prokaryotic system reminiscent of eukaryotes. Proc Natl Acad Sci U S A. 1999, 96 (3): 1077-1082. 10.1073/pnas.96.3.1077.PubMed CentralView ArticlePubMedGoogle Scholar
  39. van Ham SM, van Alphen L, Mooi FR, van Putten JP: Phase variation of H. influenzae fimbriae: transcriptional control of two divergent genes through a variable combined promoter region. Cell. 1993, 73 (6): 1187-1196. 10.1016/0092-8674(93)90647-9.View ArticlePubMedGoogle Scholar
  40. van Belkum A, van Leeuwen W, Scherer S, Verbrugh H: Occurrence and structure-function relationship of pentameric short sequence repeats in microbial genomes. Res Microbiol. 1999, 150 (9-10): 617-626. 10.1016/S0923-2508(99)00129-1.View ArticlePubMedGoogle Scholar
  41. van Belkum A: Short sequence repeats in microbial pathogenesis and evolution. Cell Mol Life Sci. 1999, 56 (9-10): 729-734. 10.1007/s000180050019.View ArticlePubMedGoogle Scholar
  42. Romero CM, DeShazer D, Feldblyum T, Ravel J, Woods D, Kim HS, Yu Y, Ronning CM, Nierman WC: Genome sequence alterations detected upon passage of Burkholderia mallei ATCC 23344 in culture and in mammalian hosts. BMC Genomics. 2006, 7: 228-10.1186/1471-2164-7-228.PubMed CentralView ArticlePubMedGoogle Scholar
  43. Vogler AJ, Keys C, Nemoto Y, Colman RE, Jay Z, Keim P: Effect of repeat copy number on variable-number tandem repeat mutations in Escherichia coli O157:H7. J Bacteriol. 2006, 188 (12): 4253-4263. 10.1128/JB.00001-06.PubMed CentralView ArticlePubMedGoogle Scholar
  44. Lovett ST: Encoded errors: mutations and rearrangements mediated by misalignment at repetitive DNA sequences. Mol Microbiol. 2004, 52 (5): 1243-1253. 10.1111/j.1365-2958.2004.04076.x.View ArticlePubMedGoogle Scholar
  45. Mrazek J: Analysis of distribution indicates diverse functions of simple sequence repeats in Mycoplasma genomes. Mol Biol Evol. 2006, 23 (7): 1370-1385. 10.1093/molbev/msk023.View ArticlePubMedGoogle Scholar
  46. Zivanovic Y, Lopez P, Philippe H, Forterre P: Pyrococcus genome comparison evidences chromosome shuffling-driven evolution. Nucleic Acids Res. 2002, 30 (9): 1902-1910. 10.1093/nar/30.9.1902.PubMed CentralView ArticlePubMedGoogle Scholar
  47. Watt VM, Ingles CJ, Urdea MS, Rutter WJ: Homology requirements for recombination in Escherichia coli. Proc Natl Acad Sci U S A. 1985, 82 (14): 4768-4772. 10.1073/pnas.82.14.4768.PubMed CentralView ArticlePubMedGoogle Scholar
  48. Vogler AJ, Smith KL: Mutations, mutation rates, and evolution at the hypervariable VNTR loci of Yersinia pestis. Mutation Research. 2007, 616 (1-2 ): 145-158.View ArticlePubMedGoogle Scholar
  49. Pearson T, U'Ren JM, Schupp JM, Allan GJ, Foster PG, Mayo MJ, Gal D, Choy JL, Leadem Daugherty R, Kachur S, Clark Friedman CL, Leadem B, Georgia S, Hornstra H, Vogler AJ, Wagner DM, Keim P, Currie BJ: VNTR analysis of selected outbreaks of Burkholderia pseudomallei in Australia. Infect Genet Evol. 2006,Google Scholar
  50. Vesaratchavest M, Tumapa S, Day NP, Wuthiekanun V, Chierakul W, Holden MT, White NJ, Currie BJ, Spratt BG, Feil EJ, Peacock SJ: Nonrandom Distribution of Burkholderia pseudomallei Clones in Relation to Geographical Location and Virulence. J Clin Microbiol. 2006, 44 (7): 2553-2557. 10.1128/JCM.00629-06.PubMed CentralView ArticlePubMedGoogle Scholar
  51. McCombie RL, Finkelstein RA, Woods DE: Multilocus sequence typing of historical Burkholderia pseudomallei isolates collected in Southeast Asia from 1964 to 1967 provides insight into the epidemiology of melioidosis. J Clin Microbiol. 2006, 44 (8): 2951-2962. 10.1128/JCM.00725-06.PubMed CentralView ArticlePubMedGoogle Scholar
  52. Cheng AC, Godoy D, Mayo M, Gal D, Spratt BG, Currie BJ: Isolates of Burkholderia pseudomallei from Northern Australia are distinct by multilocus sequence typing, but strain types do not correlate with clinical presentation. J Clin Microbiol. 2004, 42 (12): 5477-5483. 10.1128/JCM.42.12.5477-5483.2004.PubMed CentralView ArticlePubMedGoogle Scholar
  53. Benson G: Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 1999, 27 (2): 573-580. 10.1093/nar/27.2.573.PubMed CentralView ArticlePubMedGoogle Scholar
  54. Jay Z: Discovery and analysis of Bacillus anthracis chromosomal VNTRs using universal primers. Department of Biological Sciences. 2005, Flagstaff , Northern Arizona University, M.S.:Google Scholar
  55. Swofford DL: PAUP, Phylogenetic Analysis Using Parsimony (and Other Methods) , Version 4.0 Beta. 1999, Sunderland, MA , SinauerGoogle Scholar

Copyright

© U'Ren et al; licensee BioMed Central Ltd. 2007

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Advertisement