Molecular Strain Typing of Brucella abortus Isolates from Italy by Two VNTR Allele Sizing...

RESEARCH

Molecular Strain Typing of Brucella abortus Isolates from Italyby Two VNTR Allele Sizing Technologies

Riccardo De Santis • Massimo Ancora • Fabrizio De Massis •

Andrea Ciammaruconi • Katiuscia Zilli • Elisabetta Di Giannatale •

Valentina Pittiglio • Silvia Fillo • Florigio Lista

� Springer Science+Business Media New York 2013

Abstract Brucellosis, one of the most important re-

emerging zoonoses in many countries, is caused by bacteria

belonging to the genus Brucella. Furthermore these bac-

teria represent potential biological warfare agents and the

identification of species and biovars of field strains may be

crucial for tracing back source of infection, allowing to

discriminate naturally occurring outbreaks instead of biot-

errorist events. In the last years, multiple-locus variable-

number tandem repeat analysis (MLVA) has been proposed

as complement of the classical biotyping methods and it

has been applied for genotyping large collections of Bru-

cella spp. At present, the MLVA band profiles may be

resolved by automated or manual procedures. The Lab on a

chip technology represents a valid alternative to standard

genotyping techniques (as agarose gel electrophoresis) and

it has been previously used for Brucella genotyping.

Recently, a new high-throughput genotyping analysis sys-

tem based on capillary gel electrophoresis, the QIAxcel,

has been described. The aim of the study was to evaluate

the ability of two DNA sizing equipments, the QIAxcel

System and the Lab chip GX, to correctly call alleles at the

sixteen loci including one frequently used MLVA assay for

Brucella genotyping. The results confirmed that these

technologies represent a meaningful advancement in high-

throughput Brucella genotyping. Considering the accuracy

required to confidently resolve loci discrimination, QIAx-

cel shows a better ability to measure VNTR allele sizes

compared to LabChip GX.

Keywords Brucella � QIAxcel � Lab chip GX � MLVA �UPGMA

Background

The genus Brucella comprises Gram-negative bacteria

responsible of brucellosis, a worldwide zoonosis disease

resulting in significant economic losses due to abortion and

infertility in livestock (cattle, goats, and sheep) [1, 2]. The

disease can be transmitted to humans directly by contact

with infected animals or indirectly by contaminated dairy

products. Brucellosis occurs as a sub-acute or chronic ill-

ness, that is generally not lethal in previously healthy

patients, and can result in a wide variety of manifestations

and significant morbidity if the diagnosis is unobserved and

treatment is not rapidly started [2]. Ten species are rec-

ognized within the genus Brucella [3]: B. abortus (7 bio-

vars) which preferentially infects cattle; B. melitensis (3

biovars) infecting sheep and goats; B. suis (5 biovars)

preferentially infecting pigs; B. canis the dog; B. ovis,

sheep and goats; B. neotomae the desert wood rat, B.

microti the common vole [4], B. ceti, cetaceans, B. pinni-

pedialis, seals [5] and B. inopinata sp., isolated from a

human breast implant infection [6]. Each species has dis-

tinctive host preferences, pathogenicity, and epidemiology.

Brucellosis is recognized as public health problem in

many countries around the world although the surveillance

plans, and the animal vaccination programs have been

conducted. Currently, the disease is considered to be

endemic in the Eastern Mediterranean basin, the Middle

R. De Santis � A. Ciammaruconi � V. Pittiglio � S. Fillo �F. Lista (&)

Histology and Molecular Biology Section, Army Medical and

Veterinary Research Center, Via Santo Stefano Rotondo 4,

00184 Rome, Italy

e-mail: [email protected]

M. Ancora � F. De Massis � K. Zilli � E. Di Giannatale

Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise

‘‘G. Caporale’’, (Istituto G. Caporale), Teramo, Italy

123

Mol Biotechnol

DOI 10.1007/s12033-013-9659-3

East, the Arabian Peninsula, Mexico, Central and South

America, Central Asia, and the Indian subcontinent [2, 7].

As the prevalence of brucellosis in animal reservoirs is the

key to its control in humans, the knowledge of the pre-

vailing genotypes of Brucella spp. in a country represents

an important epidemiological tool for the formulation of

policies and strategies for the control of brucellosis in

animal populations. Current characterization of Brucella at

species and biovar levels can be performed by differential

tests based on phenotypic characterization of lipopolysac-

charide (LPS) antigens, phage typing, dye sensitivity,

requirement for CO2, H2S production, and metabolic

properties [8] or low-resolution molecular typing methods

[9, 10]. However, these methods have limited value for

epidemiological trace-back investigations due to the

genetically monomorphic nature of Brucella species [11].

The analysis of the published genome sequences of B.

melitensis 16M, B. suis 1330, and B. abortus 9-941 [12, 13]

resulted in the identification of polymorphic loci. This

allowed the development of schemes based on multilocus

sequence typing (MLST) [14] and multiple-locus VNTR

analysis (MLVA) [12, 15, 16]. Current methodologies for

sizing of VNTR PCR products ranging from traditional

agarose gel electrophoresis to latest microfluidics technol-

ogy. The Lab on a chip technology represents a valid

alternative to standard genotyping technique, and was

previously used for Brucella [17, 18]. The aim of this study

is to evaluate the ability of the QIAxcel Advanced System

(Qiagen, Crawley, UK), using capillary gel electrophoresis,

to be used as an automated improvement to the micro-

fluidics technology for sizing of Brucella VNTR products

in MLVA-16 [16]. The discriminatory capability of the

QIAxcel system was compared to the Lab chip GX (Cali-

per Life Sciences, Hopkinton, USA) in their own ability to

distinguish geographic origins in order to resolve rela-

tionships within a group of 66 B. abortus isolated in Ab-

ruzzo and Molise regions, central Italy. Furthermore a

reference collection, whose MLVA fingerprinting profiles

were previously resolved [18, 19], were de novo genotyped

with QIAxcel. This study also compares to the usability of

the two VNTR alleles sizing technologies with regard to

resources, technological needs, scientific expertises, and

time consumption.

Materials and Methods

Bacterial Strains

An appropriate reference collection, previously genotyped

and reasonably covering the existing allele size ranges, was

reanalyzed to both validate the QIAxcel system and com-

pare the two VNTR allele sizing technologies. Therefore,

fifteen samples were collected by the Istituti Zooprofilattici

Sperimentali (IZS) and submitted to the National Refer-

ence Laboratory (NRL) for brucellosis research at the IZS

dell’Abruzzo e del Molise-G. Caporale (Istituto G. Capo-

rale) for typing during the 2001–2008 years [18], nine

DNA samples, collected in UK, provided at the IZS

dell’Abruzzo e del Molise-G. Caporale (Istituto G. Capo-

rale) for B. suis ring-trial 2006 (COST 845-Brucellosis in

man and animals) [18] and 17 Brucella strains isolated

from Sicilian hospitalized patients with acute brucellosis

[18, 19], were analyzed. Furthermore, ten reference strains,

previously analyzed by Lab on a chip [20] were included as

controls: the type strains of the three biovars of B. meli-

tensis (biovars 1, 2 and 3), biovars 1–3 of B. abortus ref-

erence strains and one for each species of B. canis, B. suis,

B. ovis, and B. neotomae. The sixty-six field strains used in

the study were isolated from cattles in the framework of the

National Brucellosis eradication program throughout Ab-

ruzzo and Molise regions (central Italy) from 2007 to 2009.

The strains were isolated from 13 brucellosis outbreaks in

three provinces (L’Aquila, Isernia and Chieti, about

500 sqkm area). Most outbreaks were epidemiologically

connected one another and the main critical point identified

for brucellosis spreading among herds was the use of

common pastures. Isolation and typing was performed by

the National Reference Laboratory for Brucellosis at the

IZS dell’Abruzzo e Del Molise-G. Caporale (Istituto ‘‘G.

Caporale’’ Teramo), according to the bacteriological stan-

dard techniques described in the OIE Manual [21]. All

isolates were identified as B. abortus biovar 3. The field

isolates and the reference strains used in this study are

summarized in Table 1.

DNA Extraction and VNTR Amplification

Total genomic DNA was extracted by Maxwell 16 Cell

DNA purification kit (Promega, Wisconsin, USA),

according to the manufacturer’s instructions and stored at

-20 �C until needed. The MLVA-16 scheme uses 16 loci

divided in panel 1 (Bruce06, Bruce08, Bruce11, Bruce12,

Bruce42, Bruce43, Bruce45, and Bruce55) and panel 2

(Bruce18, Bruce 19, Bruce21, Bruce04, Bruce07, Bruce09,

Bruce16, and Bruce30). Amplification reaction mixtures

were prepared, according to the methods previously

described [15, 16], in 15 ll volumes using 1U FastStart

polymerase Taq and containing 1 ng of DNA, 19 PCR

Roche reaction buffer (10 mM Tris–HCl, 2,5 mM MgCl2,

50 mM KCl pH 8.3), 0.2 mM dNTPs (Roche, Basel,

Switzerland) and 0,3 lM of each flanking primer (Eurofin

MWJ, Edersberg, Germany). Thermal cycling, conducted

on a Peltier thermal cycler DNA engine DYAD (MJ

Research, Massachusetts, US), was performed as follows:

an initial heating at 95 �C for 5 min, 35 cycles denaturation

Mol Biotechnol

123

at 95 �C for 30 s, annealing at 60 �C for 30 s, and exten-

sion at 70 �C for 60 s. A final extension was performed at

70 �C for 5 min.

MLVA Analysis

The Caliper LabChip GX and the QIAxcel Advanced

System were used to examine polymorphisms at 16 dif-

ferent tandem repeat loci of the sixty-six isolates. Lab on a

chip genotyping was used as described previously [18].

Amplification products were loaded into 96-well or

384-well PCR plates. For each run, the chip (Caliper HT

DNA 5 K Kit) and two tubes were prepared, respectively,

for the ladder (containing the MW size standards of 100,

300, 500, 700, 1100, 1900, 2900, 4900 bp) and the buffer,

according to manufacturer recommendations. The number

of samples per chip preparation is 400, equivalent or four

96-well plates or one 384-well plate. After gel preparation,

the sample plate was loaded into the microplate carrier

moved by a stepper motor driven robot of the Caliper

LabChip GX. During the separation of the fragments, the

samples were analyzed sequentially and amplification

product size estimates were obtained by LabChip Gx

Software. Successfully each size was assigned to the cor-

responding allele by a correspondence table, previously

described [18]. The QIAxcel Advanced System was used to

analyze concurrently six samples for MLVA-16 using a 12

capillary QIAxcel DNA high resolution cartridge, prepared

according to manufacturer recommendations. Amplifica-

tion products were loaded into 96-well PCR plate. The

QIAxcel used a standardized volume (\0.1 ll) for each

analysis of the PCR product that was placed on the

QIAxcel. Accurate measurement of the DNA product sizes

was achieved using on QIAxcel runs a QX DNA Size

Marker 25–500 bp (MW size standards of 25, 50, 75, 100,

150, 200, 250, 300, and 400 bp) for all Panel 2 loci and

Bruce 08, 12, 42, and 43 belonging to the Panel 1. For the

remaining Panel 1 loci a DNA Size Marker 50–1,350 bp

(MW size standards of 100, 250, 300, 350, 400, 450, 500,

550, 600, 650, 700, 766, 916, and 1,350 bp) was used.

QIAxcel electrophoretic separation was performed apply-

ing the OM500 and OM800 methods coupled respectively

with the DNA Size Markers 50–1,350 bp and 25–500 bp.

The separation time (the time taken for the sample to run

through the capillary gel matrix) was approximately 8 min

for OM500 method and 13 min for OM800 method. For

each run, the Biocalculator QIAxcel software sizes the

fragment length producing also a virtual gel image. The

allelic sequences of each sample were then imported into

the BioNumerics software package (version 5.10, Applied-

Maths, Belgium) and clustering analysis were based on the

categorical coefficient and the unweighted-pair group

method using arithmetic averages (UPGMA) analysis and

minimum spanning tree (MST) analysis.

Sequencing Analysis

The PCR amplicons were purified and sequenced by CEQ

8000 automatic DNA Analysis System (Beckman-Coulter,

CA, USA) using a commercial Kit (GenomeLabTM DTCS-

Quick Start Kit, Beckman-Coulter) according to the man-

ufacturer instructions.

Results

The aim of our study was to compare the resolving powers

of two different DNA fragment sizing technologies pro-

viding an assessment of the limitations, advantages, and

field applications of the two MLVA assays for genotyping

Brucella spp. The performance of the QIAxcel against

Caliper LabChip GX was assessed analyzing 66 Italian

field isolates of B. abortus with a MLVA-16 panel (Fig. 1).

Furthermore, in order to set up the system, DNA samples,

Table 1 Brucella fields and reference strains used in this work

Field and reference strains Species-biovar

BruIT200a B. melitensis biovar 3



BruIT207a B. abortus biovar 1







BruIT227a B. suis biovar 2





544b B. abortus biovar 1

86/8/59b B. abortus biovar 2

Tulyab B. abortus biovar 3

RM 6/66b B. canis

B. melitensis 16 Mb B. melitensis biovar 1

B. melitensis 63/9b B. melitensis biovar 2

B. melitensis etherb B. melitensis biovar 3

5K33b Brucella neotomae

63/290b B. ovis

686b B. suis biovar 3

a Field strainsb Reference strains

Mol Biotechnol

123

Mol Biotechnol

123

previously genotyped by micro-fluidics technology and

sequencing system [18, 19], were reanalyzed. The elec-

trophoretic separation by LabChip GX was performed

according to the previously described method [18].

Table 2 Comparison between Brucella product sizes estimated respectively by Qiaxcel and LabChip GX software and actual sizes obtained by

direct sequencing of the PCR product (Expected size)

PCR Locus (UL bps)a Allele Expected size Qiaxcel LabChip GX

Observed size x ± rb Bias Observed size x ± rb Bias

Singleplex 1 Bruce08 (18)c 3 330 333–342 336 ± 5 6 346–359 352.6 ± 5.3 22.6

4 348 351 351 ± 0 3 369–383 376 ± 4.6 28

5 366 365–371 367.4 ± 1.5 1.4 385–410 405 ± 4 39

6 384 387 387 ± 0 3 411–413 412 ± 1 28

Singleplex 2 Bruce43 (12)c 1 170 173–177 174.3 ± 1.1 4.3 179–188 183,17 ± 2 13.2

2 182 184–190 186.1 ± 1.4 4.1 191–200 198 ± 2 16

Singleplex 3 Bruce12 (63)d 10 347 344–346 345 ± 1.4 2 359–369 362,8 ± 3.7 15.8

11 362 362–364 363 ± 1 1 379–388 384.1 ± 3.6 22.1

12 377 375–382 377.2 ± 1.7 0.2 391–402 396.8 ± 2.3 19.8

13 392 393–397 394.6 ± 1 2.6 407–416 412.6 ± 2.3 20.6

14 407 410–411 410.6 ± 0.5 3.6 424–440 430.5 ± 5 23.5

Singleplex 4 Bruce18 (8)c 3 130 135 135 ± 0 5 145 145 ± 0 15

4 138 144 144 ± 0 6 153–155 154 ± 1 16

5 146 150–153 151.5 ± 1.7 5.5 161–162 163 ± 1.1 17

6 154 157–162 158.4 ± 0.6 4.4 164–175 173 ± 2 19

7 162 167–169 168.1 ± 0.6 6.1 177–183 181.3 ± 1.7 19.3

8 170 176–177 176.5 ± 0.6 6.5 189–191 190.3 ± 1 20.3

Singleplex 5 Bruce11 (63)d 2 257 258–263 260.6 ± 2.5 3.6 266–270 268 ± 2.82 11

3 320 291–325 316.3 ± 4 4.3 321–344 335.4 ± 3.6 15.4

4 383 386–394 389.1 ± 2.6 6.1 408–422 412.8 ± 5.3 29.8

6 509 532–535 534 ± 1.7 25 509–536 521.3 ± 13.6 12.3

8 635 669–681 675.6 ± 6.1 40.6 623–648 637.7 ± 11.3 2.7

9 698 706–731 715 ± 13.9 17 680–724 702.3 ± 22 4.3

Singleplex 6 Bruce21 (8)c 6 148 151–155 152.6 ± 2 4.6 163 163 ± 0 15

7 156 166–168 167.5 ± 0.5 11.5 178–179 178.5 ± 0.6 22.5

8 164 169–171 169.6 ± 0.7 5.6 180–188 182.3 ± 1.4 18.3

9 172 180–182 181 ± 0.9 9 193–198 194.4 ± 2 22.4

Singleplex 7 Bruce06 (134)d 1 140 141 141 ± 0 1 153 153 ± 0 13

2 274 262–285 271.5 ± 8.2 2.5 282–294 286.8 ± 3.8 12.8

3 408 393–424 411.1 ± 9.2 3.1 432–454 439.5 ± 5.8 31.5

4 542 555–571 561.6 ± 3.5 19.6 570–606 594 ± 6.6 52

Singleplex 8 Bruce42 (125)d 1 164 157–167 162.8 ± 2.8 1.2 172–177 174.7 ± 1 10.7

2 289 286–302 290.4 ± 3.4 1.4 288–293 290 ± 1 1

3 414 406–430 422.2 ± 9.5 8.2 420–442 433 ± 8.1 19

4 539 546–557 550 ± 6 11 504–550 526.7 ± 19 12.3

Singleplex 9 Bruce45 (18)c 3 151 151–159 155.7 ± 1.4 4.7 159–166 162.5 ± 1.5 11.5

5 187 191–195 192.7 ± 2 5.7 196–206 199.4 ± 3.1 12.4

Singleplex 10 Bruce55 (40)d 1 193 186–190 187.6 ± 1.9 6.5 204–208 206 ± 2 13

2 233 223–236 230.5 ± 4.7 2.5 245–259 249.8 ± 5.2 15.8

3 273 264–282 271 ± 4.7 2 275–289 284 ± 2 11

Fig. 1 Dendrogram based on differences in the copy number of

tandem repeat units of 16 loci. The genetic profiles were arranged in

the following order of loci: Bruce 06, 08, 11, 12, 42, 43, 45, 55, 18,

19, 21, 04, 07, 09, 16, 30. The 66 B. abortus isolates were divided into

ten genotypes

b

Mol Biotechnol

123

Table 2 continued

PCR Locus (UL bps)a Allele Expected size Qiaxcel LabChip GX

Observed size x ± rb Bias Observed size x ± rb Bias

Singleplex 11 Bruce30 (8)c 3 127 127–133 129.9 ± 1.6 2.7 136–142 139 ± 1.4 12

4 135 139–140 139.5 ± 0.7 4.5 145–149 146.4 ± 1.3 11.4

5 143 146–150 148.6 ± 1.7 4 158–159 158.6 ± 0.5 15.6

6 151 154–158 155.7 ± 2 4.7 167–168 167.3 ± 0.6 16.3

Singleplex 12 Bruce04 (8)c 2 152 154–157 155.6 ± 1.5 3.6 169–172 170 ± 1.7 18

3 160 163–166 165 ± 1.4 5 179–180 179.6 ± 0.5 19.6

4 168 169–175 171.5 ± 1.7 3.5 186–192 188.3 ± 1.4 20.3

5 176 179–182 180.8 ± 1 4.8 195–199 197 ± 1.4 21

6 184 187–190 188.4 ± 1.3 4.4 206–210 206.8 ± 1.6 22.8

7 192 196–198 197.2 ± 0.9 5.2 214–217 216.2 ± 1.3 24.2

8 200 203 203 ± 0 3 222–224 222.7 ± 0.9 22.7


5 158 160–166 162.3 ± 0.9 4 169–174 171 ± 1 13

6 166 170–173 171.9 ± 1.1 5.9 179–181 179.7 ± 0.8 13.7

7 174 179–181 180 ± 1 6 186–190 188.5 ± 1.7 14.5

8 182 186–187 187 ± 0.7 4.5 197–198 197.5 ± 0.7 15.5

Singleplex 14 Bruce09 (8)c 3 124 123–129 125 ± 1.6 1 134–143 137.3 ± 1.7 13.3

4 132 134 134 ± 0 2 147 147 ± 0 15

5 140 140–141 140.5 ± 0.7 0.5 154–158 155.3 ± 2.3 15.3

6 148 150–153 152.3 ± 1.1 4.3 163–167 164.7 ± 1.2 16.7

7 156 159–161 160 ± 1 3 173–175 174.4 ± 0.9 16.4

8 164 167 167 ± 0 3 183–184 183.5 ± 0.7 19.5

9 172 175–177 176 ± 1 3 194 194 ± 0 22

10 180 183–185 184.2 ± 0.9 3 202–203 202.5 ± 0.6 22.5

11 188 193 193 ± 0 5 211–212 211.3 ± 0.6 23.3

12 196 201–202 201.5 ± 0.7 5.5 220 220 ± 0 24

13 204 209 209 ± 0 5 230 230 ± 0 26

16 228 233–234 233.5 ± 0.7 5.5 249–253 251 ± 2.8 24


4 160 160–163 161.3 ± 1.5 1.3 167–173 168.1 ± 1.6 8.1

5 168 170–171 170.5 ± 0.7 2.5 174–183 178.7 ± 3.8 10.7

6 176 174–179 177.7 ± 2.5 1.7 190–192 191 ± 0.8 15

7 184 188–189 188.5 ± 0.6 4.5 200–202 200.7 ± 0.9 16.7

8 192 193–196 192.2 ± 1.5 0.2 207–210 208.2 ± 1.5 16.2

9 200 204–206 205.1 ± 0.7 5.1 216–219 217.7 ± 1.2 17.7

10 208 213 213 ± 0 4 224–225 224.5 ± 0.7 16.5

11 216 221–222 221.5 ± 0.7 5.5 234–236 235 ± 1.4 19

Singleplex 16 Bruce 19 (6)c 18 163 164–168 165.7 ± 2 2.7 176–178 177 ± 0.7 17

19 169 173 173 ± 0 4 182 182 ± 0 13

20 175 179–182 180.3 ± 1.5 5.3 188 188 ± 0 13

21 181 183–185 184.5 ± 0.6 3.5 189–193 191.6 ± 1.5 10.6

22 187 186–188 187 ± 0.9 0 194–199 197.7 ± 1 10.7

23 193 203 203 10

a Unit lenght sizeb Arithmetic average (x) ± standard deviation (r) of the observed sizesc DNA size markers 25–500 bp coupled with OM800 methodd DNA size markers 50–1,350 bp coupled with OM500 method

Mol Biotechnol

123

Initially, the QIAxcel analysis was performed with differ-

ent electrophoresis methods on a unique ladder

(50–1,350 bps), able to cover all the ranges of the DNA

fragments generated in the 16 different loci. The results

obtained with all the different electrophoresis methods

showed a low discriminatory power for the eight Panel 2

loci and the four loci Bruce 08, 12, 43, and 45 belonging to

the Panel 1. The reason of this failure is probably due to a

few ladder reference bands in the range between 100 and

300 bps. In order to obtain a best resolution of the

ambiguous alleles, a different DNA Size Marker

(25–500 bps) coupled with OM800 method was used. All

loci analyzed by QIAxcel and LabChip GX showed a range

according to the expected sizes respectively from 0 to

40 bp and from 1 to 39 bp (Table 2). The ascertainment of

the number of repeats within loci showed a 100 % corre-

lation between the two instruments QIAxcel and Caliper

LabChip GX except for Bruce 08, Bruce 21, Bruce 16, and

Bruce 19. The alleles 3, 4, and 5 of locus Bruce 16 and 5

and 6 of locus Bruce 08, as previously described [18],

showed also by the LabChip GX equipment a continuity

between neighboring ranges that leads to an ambiguously

assignment of the allele to the observed value (Table 2),

while the QIAxcel showed an higher discriminatory

capacity. Furthermore both the LabChip GX and the

QIAxcel showed a continuity between the neighboring

ranges of the alleles 7 and 8 of locus Bruce 21 (Table 2)

with resulting reduced discriminatory ability. Finally, both

the platforms showed a low discrimination for Bruce 19

(Table 2). In this study, the data discrepancies found by the

comparison between the two platforms were resolved by

sequencing. The publically available MLVA database for

Brucella, MLVA-NET, (http://mlva.u-psud.fr/brucella/)

[22] was used to identify or confirm the identity of all

isolates used in this study.

Stability of the MLVA Loci and Diversity Index (DI)

DI values, affected by the level of genetic diversity within

the test population, were calculated through 66 isolates

from livestock of three different areas of Abruzzo and

Molise region by the Simpson’s diversity index (SDI) using

the online V-DICE bioinformatics tool available at the

HPA website (http://www.hpa.org.uk/srmd/bioinformatics/

tools/tools.htm). The DI values reflect the number of alleles

detected, ranging from 0 (no diversity) to 1 (extreme

diversity). The MLVA-16 assay, developed for epidemio-

logical typing, contains two panels: panel 1 made of eight

conserved markers with larger repeat units ([8 bp)

exhibiting DIs \0.40 in the isolates examined and panel 2

made of eight more-diverse loci with smaller repeat units

(B8 bp). The Bruce 06, 08, 11, 12, 42, 43, 45 belonging to

the panel 1 and the Bruce 18, 04, and 30 belonging to the

panel 2 were monomorphic in the isolates examined. The

invariance we observed at Bruce 18, 04, and 30 is most

likely due to the localized geographic origins of our iso-

lates and represents a level of strains-to-strains conserva-

tion within a geographic area that can be better evaluated in

a larger population of isolates coming from similar loca-

tions. The Bruce 55, 19, 21, 07, 09, and 16 VNTR markers

exhibit DIs ranging from 0.03 to 0.45 among the 16

markers (Table 3) and were shown to have two allelic

patterns with the exception of Bruce 09 that was revealed

to have three allelic profiles (Fig. 1). The allelic type 1 of

Bruce 09 was the most representative of the overall pop-

ulation, at the same time two isolates from AQ1 farm

(2008) showed the allelic type 6 and two isolates one from

IS4 farm (2009) and the other from IS9 farm (2007)

showed the allelic type 5. The Bruce 16 showed the higher

diversity index in the overall population and two different

allelic profiles (3 and 4).

Genetic Relationships

The assigned number of each tandem repeat was imported

into the BioNumerics software package to determine the

genetic relationships among the 66 B. abortus isolates.

Clustering analysis was performed by UPGMA, a method

for inferring genetic relationship when a constant rate of

evolution occurs, which is performed for Brucella species

because of the low genetic recombination and the rare

Table 3 Numbers of alleles and diversity index values of 66 B.abortus isolates from Abruzzo and Molise

Locus SDIa Confidenceinterval

No. ofalleles

Tandemrepeatcopy no.

max(pi)b

Panel 1

Bruce 06 0.000 0.000–0.103 1 4 1.000

Bruce 08 0.000 0.000–0.103 1 5 1.000

Bruce 11 0.000 0.000–0.103 1 3 1.000

Bruce 12 0.000 0.000–0.103 1 12 1.000

Bruce 42 0.000 0.000–0.103 1 2 1.000

Bruce 43 0.000 0.000–0.103 1 2 1.000

Bruce 45 0.000 0.000–0.103 1 3 1.000

Bruce 55 0.114 0.013–0.215 2 1, 3 0.939

Panel 2

Bruce 18 0.000 0.000–0.103 1 6 1.000

Bruce 19 0.059 0.000–0.136 2 22, 23 0.970

Bruce 21 0.087 0.000–0.178 2 7, 8 0.955

Bruce 04 0.000 0.000–0.103 1 4 1.000

Bruce 07 0.030 0.000–0.087 2 5, 6 0.985

Bruce 09 0.143 0.030–0.256 3 3, 5, 6 0.924

Bruce 16 0.454 0.384–0.524 2 4, 3 0.652

Bruce 30 0.000 0.000–0.103 1 3 1.000

a Simpson’s diversityb Fraction of samples that have the most frequent repeat number in this locus(range 0.0–1.0)

Mol Biotechnol

123

http://mlva.u-psud.fr/brucella/

http://www.hpa.org.uk/srmd/bioinformatics/tools/tools.htm

http://www.hpa.org.uk/srmd/bioinformatics/tools/tools.htm

horizontal gene transfer [13, 23]. An equal weight was

given to all markers, irrespective of the number of repeats

because at one locus multiple tandem repeats can be

incorporated in one recombination event [16]. The sixty-six

B. abortus isolates were divided into ten genotypes. As

showed in Fig. 1, the most part of isolates from the dif-

ferent livestock were independently divided into two dif-

ferent genotypes, with a single monomorphic allelic profile

(identified as bv3). Eight isolates showing a single geno-

type slightly different by the two main profiles were also

identified as bv3.

Among the 13 livestock, farm IS2 was particularly

investigated for the changes of genotype in isolates of

2009. The farm seems to be infected by B. abortus strains

originating from different sources as showed in Fig. 1. The

genetic relationship of field isolates were successfully

analyzed by MST analysis comparing their genetic profiles

with those obtained by European and African strains,

whose data have been published [15, 24–26] or down-

loaded from the 2010 version of a related website

(http://minisatellites.u-psud.fr). In MST analysis, (Fig. 2)

the strains TE5736/1, TE5736/2 (farm AQ1) and IS2470

(farm IS4) were grouped with France (BCCN 95 7, 94 17

and 94 18), Spain (BCCN 93 15), Germany (BfR 101, 107

and 106), Portugal (LNIV-122Ba3-03, LNIV-120Ba3-03),

and Italy (BCCN 95 34) strains; the most part of the other

strains were located near this group.

Discussion

Currently, in genotyping analysis for bacterial identification,

the gold standard method is represented by the full

sequencing. However, this is an excessively expensive pro-

cedure requiring sophisticated equipment and consequently

is not appropriate for routine typing. Viable alternatives are

represented by MLST and MLVA. In this paper, we utilized

the last method for its high discriminatory power that has

proven to be appropriate for pathogenic bacterial species

typing with an high genetic homogeneity, as well as Brucella

spp. MLVA genotyping can be performed either on micro-

fluidics technology or capillary gel electrophoresis that are

less expensive and demanding in the set up process than

allele sequencing. In the last years, both methodological [17,

18] and epidemiological [24] studies about the use of Caliper

and QIAxcel for MLVA typing of Brucella spp. have been

published. In this study, the resolving capacity of these two

different automated DNA sizing systems were compared

using a 16 loci panel (MLVA16) and repeat units (RU) from

6 bps to more than 100 bps [15, 16].

Fig. 2 Minimum spanning tree

analysis on 66 field isolates and

reference strains using data

from eight markers (Panel 1:

bru06, 08, 11, 12, 42, 43, 45,

55). The central OTU includes

60 of 66 Abruzzo strains, but six

strains individually indicated

(IS319 IS2, IS1255/IS1, IS2470

IS4, TE11426 IS9, TE5736/1

AQ1, TE5736/2 AQ1)

Mol Biotechnol

123

http://minisatellites.u-psud.fr

The two platforms have different logistical and technical

requirements but both require a conversion table that

allowed to correlate the DNA fragment sizes to the alleles.

Furthermore, both the LabChip GX and QIAxcel softwares

improve the efficiency of data acquiring by automating

data flows. As a matter of fact, the softwares allow to

export the summary of analysis results to a spreadsheet

application, with the consequent elimination of the paper-

based flows. In despite of its lower precision, Caliper

LabChip GX shows the ability to acquire 96 amplification

product sizes in 40 min rather than 100 min of the QIAx-

cel, with an increased time-reduction. However, the Caliper

equipment shows an accurate discrimination only for 12

loci (Bruce 43, 12, 18, 11, 06, 42, 45, 55, 30, 04, 07, 09).

As previously described [18] and showed in the Table 2,

this platform cannot be used to confidently type the four

loci Bruce 08, 21, 16, and 19. The overlapping of some

alleles range could lead to a failure of the allelic calling.

Conversely, the QIAxcel Advanced System is able to dis-

criminate all alleles of Bruce 08 and 16. This capability lies

a greater flexibility, that enables the use of various ladders

coupled with different methods of electrophoresis, over-

coming therefore the reduced discriminatory ability of the

LabChip GX. Anyway QIAxcel fails, as LabChip GX, the

assignment of the alleles 7 and 8 of locus Bruce 21. The

Bruce 19 data, obtained with both platforms, have to be

confirmed by different systems as fluorescent fragments

analysis or sequencing. However, the QIAxcel Advanced

System represents a significant improvement in the high-

throughput sizing equipments.

The molecular genotyping methods based on MLVA-16

was used to perform a trace-back analysis helping the

identification of specific pathogenic bacteria. As reported

in Kang et al. [27], unstable genetic markers can generate

atypical MLVA profiles among common mutants. Actu-

ally, the changes of repeat numbers at Bruce 09 may be

recognized after in vitro passage experiment, probably

caused by simple insertion and deletion of DNA [27].

Therefore, the stability of the genetic marker for each locus

was investigated. Among the 16 loci, Bruce 16 had a

maximum of two allelic types, and the DI value was 0.45

(Table 3), which is similar to the case with Brucella spp.

previously reported [16]. Bruce09 showed various allelic

patterns in isolates, both coming from the same farm (IS2)

and other different farms. Furthermore, the allelic type 6

was originated in 2008 from farm AQ1. Isolates showing

two or three allelic profiles among the genetic markers also

with a high diversity index have been described [28].

Moreover, the TRs copy numbers of some loci may change

in the course of the adaptation in the host. Also strains re-

isolated from infected host showed minor changes [12].

These minor mutants can be caused by genetic events such

as DNA polymerase slippage and recombination.

Therefore, the isolates should be considered originating

from the same source or closely related strains. The geo-

graphical distribution of outbreaks, concentrated in a

500 km2 area, and the most representative risk factor

identified for the spreading of brucellosis in the area (use of

common pastures) are supporting these findings. In con-

clusion, the Brucella MLVA-16 is useful to analyze the

epidemiological correlation of strains, even in case of

analysis including samples showing low genetic diversity

in a restricted area, and can also be applied to trace the

geographic origin by comparing their genetic patterns

respect to isolates previously reported.

References

1. Corbel, M. J. (1997). Brucellosis: an overview. Emerging Infec-tious Diseases, 3, 213–221.

2. Pappas, G., Papadimitriou, P., Akritidis, N., Christou, L., &

Tsianos, E. V. (2006). The new global map of human Brucellosis.

The Lancet Infectious Diseases, 6, 91–99.

3. Euzeby, J.P. (2010). List of prokaryotic names with standing in

nomenclature—genus Brucella. Available: http://www.bacterio.

cict.fr/b/Brucella.html.

4. Scholz, H. C., Hubalek, Z., Sedlaek, I., Vergnaud, G., Tomaso,

H., Al Dahouk, S., et al. (2008). Brucella microti sp. nov., iso-

lated from the common vole Microtus arvalis. InternationalJournal of Systematic and Evolutionary Microbiology, 58,

375–382.

5. Foster, G., Osterman, B. S., Godfroid, J., Jacques, I., & Cloe-

ckaert, A. (2007). Brucella ceti sp. nov. and Brucella pinniped-ialis sp. nov. for Brucella s with cetaceans and seals as their

preferred hosts. International Journal of Systematic and Evolu-tionary Microbiology, 57, 2688–2693.

6. Scholz, H. C., Nockler, K., Gollner, C., Bahn, P., Vergnaud, G.,

Tomaso, H., et al. (2010). Brucella inopinata sp. nov., isolated

from a breast implant infection. International Journal of Sys-tematic and Evolutionary Microbiology, 60(Pt 4), 801–808.

7. Gwida, M., Al Dahouk, S., Melzer, F., Rosler, U., Neubauer, H.,

& Tomaso, H. (2010). Brucellosis—regionally emerging zoonotic

disease? Croatian Medical Journal, 51(4), 289–295.

8. Alton, G. G., Jones, L. M., Angus, R. D., & Verger, J. M. (1988).

Techniques for the Brucellosis laboratory. Paris: Institut National

de la Recherche Agronomique.

9. Bricker, B. J., & Halling, S. M. (1994). Differentiation of Bru-cella abortus bv.1, 2, and 4, Brucella melitensis, Brucella ovis,

and Brucella suis bv. 1 by PCR. Journal of Clinical Microbiol-ogy, 32, 2660–2666.

10. Hinic, V., Brodard, I., Thomann, A., Cvetnic, Z., Makaya, P. V.,

Frey, J., et al. (2008). Novel identification and differentiation of

Brucella melitensis, B. abortus, B. suis, B. ovis, B. canis, and B.neotomae suitable for both conven-tional and real-time PCR

systems. Journal of Microbiological Methods, 75, 375–378.

11. Gandara, B., Merino, A. L., Rogel, M. A., & Martinez-Romero,

E. (2001). Limited genetic diversity of Brucella spp. Journal ofClinical Microbiology, 39, 235–240.

12. Whatmore, A. M., Shankster, S. J., Perrett, L. L., Murphy, T. J.,

Brew, S. D., Thirlwall, R. E., et al. (2006). Identification and

characterization of variable-number tandem-repeat markers for

typing of Brucella spp. Journal of Clinical Microbiology, 44,

1982–1993.

Mol Biotechnol

123

http://www.bacterio.cict.fr/b/Brucella.html

http://www.bacterio.cict.fr/b/Brucella.html

13. Whatmore, A. M., Perrett, L. L., & MacMillan, A. P. (2007).

Characterisation of the genetic diversity of Brucella by multilo-

cus sequencing. BMC Microbiology, 7, 34.

14. Bricker, B. J., Ewalt, D. R., & Halling, S. M. (2003). Brucella‘Hoof-Prints’: typing by multi-locus analysis of variable number

tandem repeats (VNTRs). BMC Microbiology, 3, 15.

15. Le Fleche, P., Jacques, I., Grayon, M., Al Dahouk, S., Bouchon,

P., Denoeud, F., et al. (2006). Evaluation and selection of tandem

repeat loci for a Brucella MLVA typing assay. BMC Microbi-ology, 6, 9.

16. Al Dahouk, S., Le Fleche, P., Nockler, K., Jacques, I., Grayon,

M., Scholz, H. C., et al. (2007). Evaluation of Brucella MLVA

typing for human Brucellosis. Journal of Microbiological Meth-ods, 69, 137–145.

17. De Santis, R., Ciammaruconi, A., Faggioni, G., D’Amelio, R.,

Marianelli, C., & Lista, F. (2009). Lab on a chip genotyping for

Brucella spp. based on 15-loci multi locus VNTR analysis. BMCMicrobiology, 9, 66.

18. De Santis, R., Ciammaruconi, A., Faggioni, G., Fillo, S., Gentile,

B., Di Giannatale, E., et al. (2011). High throughput MLVA-16

typing for Brucella based on the microfluidics technology. BMCMicrobiology, 11, 60.

19. Marianelli, C., Graziani, C., Santangelo, C., Xibilia, M. T., Im-

briani, A., Amato, R., et al. (2007). Molecular epidemiological

and antibiotic susceptibility characterization of Brucella isolates

from humans in Sicily, Italy. Journal of Clinical Microbiology,45, 2923–2928.

20. Lista, F., Reubsaet, F. A., De Santis, R., Parchen, R. R., de Jong,

A. L., Kieboom, J., et al. (2011). Reliable identification at the

species level of Brucella isolates with MALDI-TOF-MS. BMCMicrobiology, 11, 267.

21. OIE World Organization for Animal Health (OIE). (2008).

Manual of Diagnostic Tests and Vaccines for Terrestrial Animals.

Paris: OIE.

22. Denœud, F., & Vergnaud, G. (2004). Identification of polymorphic

tandem repeats by direct comparison of genome sequence from dif-

ferentbacterial strains: a web-based resource. BMCBioinformatics, 5, 4.

23. Foster, J. T., Beckstrom-Sternberg, S. M., Pearson, T., Beck-

strom-Sternberg, J. S., Chain, P. S., Roberto, F. F., et al. (2009).

Whole-genome-based phylogeny and divergence of the genus

Brucella. Journal of Bacteriology, 191, 2864–2870.

24. Kilic, S., Ivanov, I. N., Durmaz, R., Bayraktar, M. R., Ayaslioglu,

E., Uyanik, M. H., et al. (2011). Multiple-locus variable-number

tandem-repeat analysis genotyping of human Brucella isolates

from Turkey. Journal of Clinical Microbiology, 49(9), 3276–3283.

25. Ferreira, A. C., Chambel, L., Tenreiro, T., Cardoso, R., Flor, L.,

Travassos Dias, I., et al. (2012). MLVA16 typing of Portuguese

human and animal Brucella melitensis and Brucella abortus iso-

lates. PLoS ONE, 7(8), e42514.

26. Muendo, E. N., Mbatha, P. M., Macharia, J., Abdoel, T. H.,

Janszen, P. V., Pastoor, R., et al. (2012). Infection of cattle in

Kenya with Brucella abortus biovar 3 and Brucella melitensis

biovar 1 genotypes. Tropical Animal Health and Production,44(1), 17–20. doi:10.1007/s11250-011-9899-9.

27. Kang, S. I., Heo, E. J., Cho, D., Kim, J. W., Kim, J. Y., Jung, S.

C., et al. (2011). Genetic comparison of Brucella canis isolates by

the MLVA assay in South Korea. Journal of Veterinary MedicalScience, 73(6), 779–786.

28. Her, M., Kang, S. I., Cho, D. H., Cho, Y. S., Hwang, I. Y., Heo,

Y. R., et al. (2009). Application and evaluation of the MLVA

typing assay for the Brucella abortus s isolated in Korea. BMCMicrobiology, 9, 230.

Mol Biotechnol

123

http://dx.doi.org/10.1007/s11250-011-9899-9

Molecular Strain Typing of Brucella abortus Isolates from Italy by Two VNTR Allele Sizing...

Documents

Transcript of Molecular Strain Typing of Brucella abortus Isolates from Italy by Two VNTR Allele Sizing...