Evaluation and selection of tandem repeat loci for a Brucella MLVA typing assay.

BACKGROUND: The classification of Brucella into species and biovars relies on phenotypic characteristics and sometimes raises difficulties in the interpretation of the results due to an absence of standardization of the typing reagents. In addition, the resolution of this biotyping is moderate and requires the manipulation of the living agent. More efficient DNA-based methods are needed, and this work explores the suitability of multiple locus variable number tandem repeats analysis (MLVA) for both typing and species identification.
RESULTS: Eighty tandem repeat loci predicted to be polymorphic by genome sequence analysis of three available Brucella genome sequences were tested for polymorphism by genotyping 21 Brucella strains (18 reference strains representing the six 'classical' species and all biovars as well as 3 marine mammal strains currently recognized as members of two new species). The MLVA data efficiently cluster the strains as expected according to their species and biovar. For practical use, a subset of 15 loci preserving this clustering was selected and applied to the typing of 236 isolates. Using this MLVA-15 assay, the clusters generated correspond to the classical biotyping scheme of Brucella spp. The 15 markers have been divided into two groups, one comprising 8 user-friendly minisatellite markers with a good species identification capability (panel 1) and another complementary group of 7 microsatellite markers with higher discriminatory power (panel 2).
CONCLUSION: The MLVA-15 assay can be applied to large collections of Brucella strains with automated or manual procedures, and can be proposed as a complement, or even a substitute, of classical biotyping methods. This is facilitated by the fact that MLVA is based on non-infectious material (DNA) whereas the biotyping procedure itself requires the manipulation of the living agent. The data produced can be queried on a dedicated MLVA web service site.

Background

Brucellosis is a zoonosis affecting animals and humans worldwide. Brucella infections may result in significant economic losses due to abortion and slaughtering of infected animals. Humans are mainly infected through the consumption of contaminated dairy products or by direct contact with infected animals. In addition, certain Brucella spp have to be considered as potential biowarfare agents. Six species are currently recognized, B. abortus (8 biovars), B. melitensis (3 biovars), B. suis (5 biovars), B. ovis, B. canis and B. neotomae [1]. More recently, Brucella strains have been isolated from marine mammals [2], suggesting the existence of additional species [3,4].

The genus Brucella is highly homogeneous (more than 90% DNA/DNA homology [5]). Brucella classification is mainly based on differences in pathogenicity, host preferences, and conventional microbiological tests used for phenotyping (biotyping) [6]. Routine identification of Brucella species and biovars still relies on biotyping (reviewed in [7]). Only a few tools exist for further molecular subtyping, of which none has proven to be fully satisfactory for epidemiologic investigations or tracing back strains to their origin. Tandem repeat (TR) sequences may be an interesting class of markers, since multiple alleles can be present at a single locus, and size differences are easily resolved by electrophoresis (reviewed by [8,9]). Tandem repeats are often classified as microsatellites (repeat units up to 8 bp) and minisatellites [10,11]. Tandem repeat typing has proven to be highly appropriate for the typing of pathogenic bacterial species with a high genetic homogeneity, including the Mycobacterium tuberculosis complex, Bacillus anthracis, and Yersinia pestis [12-15]. Recently, a family of tandem repeats located within a repeated sequence and present in multiple loci in the Brucella genome was used for strain typing [16,17]. The proposed set of eight microsatellite loci is extremely discriminant and highly efficient to distinguish strains within a local outbreak, but is unable to correctly predict the biovar or even the species of an isolate. A possible reason for that is the high mutation rate of these loci. Consequently, this MLVA assay cannot replace classical biotyping methods.

The availability of the whole genome sequences of B. melitensis 16 M, B. suis 1330 and B. abortus strain 9–941 [18-20] greatly facilitates the search for polymorphic DNA sequences [21]. In this report, we evaluated most tandem repeats showing at least two alleles among the three sequenced strains [22]. Eighteen reference strains and 3 strains isolated from marine mammals [23] were typed using these TR candidates to evaluate their associated polymorphism. For routine typing, a subset of 15 markers which enabled to cluster the isolates according to their biotype was selected. This set of markers was further evaluated on a collection of 236 isolates representing the major biovars affecting terrestrial mammals (Table 1) to produce a first reference data set [see Additional file 1] which can be queried via the internet [21,24].

Table 1

Brucella strains studied (reference and field strains)

Reference and marine strains
Species	Biovar	Strain	Host
B. abortus	1	544 (ATCC 23448; BCCN R4)a	Cattle
B. abortus	2	86/8/59 (ATCC 23449; BCCN R5)a	Cattle
B. abortus	3	Tulya (ATCC 23450; BCCN R6)a	Human
B. abortus	4	292 (ATCC 23451; BCCN R7)a	Cattle
B. abortus	5	B3196 (ATCC 23452; BCCN R8)a	Cattle
B. abortus	6	870 (ATCC 23453; BCCN R9)a	Cattle
B. abortus	9	C68 (ATCC 23455; BCCN R11)a	Cattle
B. melitensis	1	16 M (ATCC 23456; BCCN R1)a	Goat
B. melitensis	2	63/9 (ATCC 23457; BCCN R2)a	Goat
B. melitensis	3	Ether (ATCC 23458; BCCN R3)a	Goat
B. suis	1	1330 (ATCC 23444; BCCN R12)a	Swine
B. suis	2	Thomsen (ATCC 23445; BCCN R13)a	Swine
B. suis	3	686 (ATCC 23446; BCCN R14)a	Swine
B. suis	4	40 (ATCC 23447; BCCN R15)a	Reindeer
B. suis	5	513 (BCCN R21)a	Wild rodent
B. ovis		63/290 (ATCC 25840; BCCN R17)a	Sheep
B. canis		RM6/66 (ATCC 23365; BCCN R18)a	Dog
B. neotomae		5K33 (ATCC 23459; BCCN R16)a	Desert rat
B. pinnipediae		B2/94 (BCCN 94-73)	Common seal
B. cetaceae		B1/94 (BCCN 94-74)	Porpoise
B. cetaceae		B14/94 (BCCN 94-75)	Common Dolphin
Overview of the 236 additional isolates
Species	Biovar	Number of isolates investigated
B. abortus	1	14
B. abortus	3	20
B. abortus	4	1
B. abortus	6	5
B. abortus	7	2
B. abortus	9	2
B. abortus	(rough)	1
B. melitensis	1	13
B. melitensis	2	13
B. melitensis	3	11
B. melitensis	(atypical)	1
B. melitensis	(rough)	2
B. suis	1	13
B. suis	2	87
B. suis	3	5
B. suis	4	5
B. suis	5	1
B. suis	(rough)	1
B. ovis		23
B. canis		16
total		236

a: Reference strain

ATCC, American type culture collection

BCCN, Brucella culture collection, Nouzilly, France

Results

Evaluation of tandem repeats polymorphism

Comparison of the three genome sequences [21,22] identifies 107 TRs with a repeat unit larger than 5 bp and predicted to display size polymorphism. Eighty of them were evaluated for polymorphism among 21 reference and marine mammal strains (Table 1). Twenty-two TRs (numbered Bruce01 to Bruce22 in Table 2) have three predicted alleles. Twelve of the 22 are octamers, five of which have been previously characterized [16].

Table 2

List of tandem repeat loci investigated

22 tandem repeats with a predicted different length in the 3 genomes:
vntra	alias namea	Chr	% b	upper primer	lower primer	b.suisc	b.melc	b.aborc	nb of different allelesd	min-max bpe	HGDI f
BRU1938_8bp_371bp_9u	Bruce01 or TR7	1	100	GGTCTGGGAAAACATGAAAAGC	AGCCGATCCTGCAAAACATAAT	395	371	419	12	331–435	0.95
BRU1923_339bp_787bp_3u	Bruce02	1	94	AACGCAGCATCACCAATGT	CCCAGATGTTCGGCTATAGTATG	448	787	2143	6	448–1974	0.8
BRU1627_9bp_199bp_3u	Bruce03	1	82	GGCTATTATTTCACCGGCAAGA	TCTTGATTTCCTTGCGATACCA	208	199	217	4	154–217	0.48
BRU1543_8bp_152bp_2u	Bruce04 or TR6**	1	100	CTGACGAAGGGAAGGCAATAAG	CGATCTGGAGATTATCGGGAAG	184	152	160	8	152–208	0.87
BRU1365_8bp_185bp_3u	Bruce05	1	84	AAGTATCAGGAAGGGCAGGTTC	GGGAGTAGGGGAATAAGGGAAT	193	185	201	4	185–217	0.59
BRU1322_134bp_408bp_3u	Bruce06*	1	94	ATGGGATGTGGTAGGGTAATCG	GCGTGACAATCGACTTTTTGTC	274	408	542	4	140–542	0.73
BRU1250_8bp_158bp_5u	Bruce07**	1	100	GCTGACGGGGAAGAACATCTAT	ACCCTTTTTCAGTCAAGGCAAA	166	158	150	5	150–190	0.78
BRU1134_18bp_348bp_4u	Bruce08*	1	75	ATTATTCGCAGGCTCGTGATTC	ACAGAAGGTTTTCCAGCTCGTC	330	348	366	4	312–366	0.53
BRU588_8bp_156bp_7u	Bruce09 or TR8**	1	94	GCGGATTCGTTCTTCAGTTATC	GGGAGTATGTTTTGGTTGTACATAG	140	156	124	8	124–244	0.72
BRU221_19bp_127bp_2u	Bruce10	1	90	ATCAATTCGCGGATATTTCACT	AGTGCGTTTCATATGTTTGCTG	146	127	165	2	127–146	0.26
BRU211_63bp_257bp_2u	Bruce11*	1	80	CTGTTGATCTGACCTTGCAACC	CCAGACAACAACCTACGTCCTG	509	257	383	6	257–698	0.84
BRU73_15bp_392bp_13u	Bruce12*	2	58	CGGTAAATCAATTGTCCCATGA	GCCCAAGTTCAACAGGAGTTTC	345	392	375	7	302–452	0.82
BRU19_8bp_196bp_2u	Bruce13	2	100	CGAACGATAGACCAGAACATGC	TTGAAAGAATCAGATAAGATAAAGCA	204	196	220	10	196–300	0.85
BRU18_8bp_102bp_7u	Bruce14	2	100	TTGCTTTATCTTATCTGATTCTTTCAA	GGTGTCGTTGGAGATAGAGGTC	142	102	94	11	70–214	0.93
BRU1112_264bp_346bp_1u	Bruce15	2	81	GCGGTGTTGTGTCTGTGGATA	GCCGTCAGTATCCACGTCATAG	875	346	4112	7	346–2458	0.83
BRU548_8bp_152bp_3u	Bruce16**	2	100	ACGGGAGTTTTTGTTGCTCAAT	GGCCATGTTTCCGTTGATTTAT	168	152	176	8	144–240	0.85
BRU344_5bp_110bp_3u	Bruce17	2	100	TTTTCACAGGGCATGTTCTCAG	CGCGTTTCGATTGTGGAAAATA	125	110	115	6	110–135	0.79
BRU339_8bp_146bp_5u	Bruce18**	2	87	TATGTTAGGGCAATAGGGCAGT	GATGGTTGAGAGCATTGTGAAG	138	146	154	6	130–170	0.82
BRU324_6bp_163bp_18u	Bruce19	2	59	GACGACCCGGACCATGTCT	ACTTCACCGTAACGTCGTGGAT	169	163	184	7	76–190	0.79
BRU329_8bp_148bp_7u	Bruce20 or TR4	2	100	AATACTGGGTCCAGTCCGATG	AGCGCAGCGACCATATTCT	100	148	124	ND	ND	ND
BRU329_8bp_148bp_6u	Bruce21**	2	68	CTCATGCGCAACCAAAACA	GATCTCGTGGTCGATAATCTCATT	175	148	164	3	148–175	0.57
BRU322_8bp_158bp_6u	Bruce22 or TR1	2	98	GATGAAGACGGCTATCGACTG	TAGGGGAGTATGTTTTGGTTGC	150	158	134	10	125–214	0.9
56 tandem repeats with a predicted different length in 2/3 genomes:
vntra	alias namea	Chr	% b	upper primer	lower primer	b.suisc	b.melc	b.aborc	nb of different allelese	min-max bpf	HGDI g
BRU1990_9bp_152bp_1u	Bruce23	1	88	ATCAGCGAGTCGAAGGTCAGTT	TTCGACTATGCCAATCCAGATG	161	152	161	2	152–161	0.26
BRU1940_8bp_146bp_8u	Bruce24 or TR5	1	100	AGGGGAGTATGTTTTTGGTTGC	GCTACAAGATCGAAGTGCTCCA	146	146	106	10	106–217	0.88
BRU1915_8bp_215bp_2u	Bruce25 or TR3	1	100	GGGAGTATGTTTTGGTTGCACA	CTATTTCGTCCTGCCATTCGAC	239	215	239	8	215–736	0.9
BRU1704_12bp_189bp_5u	Bruce26	1	68	TCTTCATCCTGCGAGATCATGT	ATTCGTGATCGGGGTGATGAT	162	189	189	2	162–189	0.5
BRU1609_8bp_170bp_6u	Bruce27	1	76	TCGACGTCGTCTGACATTTTCT	GGGAGTAAGGCAGTAGGGGAAT	170	170	162	3	162–178	0.61
BRU1599_11bp_147bp_4u	Bruce28	1	95	TATCTTCCACGGCCATGAATC	GGCAGGATCGGCGTATAGATAA	136	147	136	2	136–147	0.1
BRU1528_15bp_132bp_3u	Bruce29	1	81	TTGCGTTATTGATTGTCAGCAC	GCTGTGGCTCGTCTATGTGG	132	132	117	2	117–132	0.47
BRU1505_8bp_151bp_6u	Bruce30 or TR2**	1	96	TGACCGCAAAACCATATCCTTC	TATGTGCAGAGCTTCATGTTCG	127	151	151	5	119–151	0.7
BRU1475_18bp_120bp_1u	Bruce31	1	100	GCTGAATCTTTTCCGCATCCT	GGAATCTCTGCACTGACAAAGC	138	120	120	2	120–138	0.51
BRU1424_8bp_106bp_5u	Bruce32	1	88	AGGTTTCCGGCGATAATGG	TCGGGATGCGCTCTAGAATATC	106	106	98	3	98–114	0.6
BRU1413_15bp_158bp_4u	Bruce33	1	81	GATGGAGCTTGGTTCCTGCTT	GCATGATCCGTTTTCTTCTCAA	143	158	143	3	128–158	0.5
BRU1409_18bp_83bp_2u	Bruce34	1	84	GCGATCGAAGGAAATATCGAG	CGCTGCCGGGATGTGAAC	101	83	101	2	83–101	0.26
BRU1282_10bp_136bp_4u	Bruce35	1	91	TGCGATAACAGGTGTACCCAAG	GACGGCAGCCATGCTGAT	116	136	116	3	116–136	0.27
BRU1234_15bp_157bp_4u	Bruce36	1	76	TAAGGCTCTTGCGTTTGTATCG	TGCGTATCTTCAGACTGGCAAT	142	157	142	2	142–157	0.38
BRU1176_21bp_124bp_2u	Bruce37	1	95	CCAAGCGTATCATCGATCTGTC	TCGGACGCAGATTGTTTCTATC	103	124	123	2	103–124	0.51
BRU1116_18bp_108bp_2u	Bruce38	1	100	CTGAATTGGGAGGAGGAACCAG	AGCTGAACGACCTTGGCATCT	126	108	108	2	108–126	ND
BRU1112_15bp_164bp_7u	Bruce39	1	66	GAAGGTCTCGAAGGAAGAGCTG	CCATCCATATTGATCGTCAGGA	164	164	146	2	146–164	0.51
BRU1048_15bp_94bp_2u	Bruce40	1	85	AAAAGAAGGGTTTCCCCATACC	GGAAAGGACAGCTTCGAGTACC	109	94	109	3	94–109	0.34
BRU1030_13bp_94bp_1u	Bruce41	1	100	TTATGTCACCGCTGACGAATTT	CTCATTATGGACCCGGTCTTTC	107	94	107	2	94–107	0.1
BRU424_125bp_539bp_4u	Bruce42*	1	96	CATCGCCTCAACTATACCGTCA	ACCGCAAAATTTACGCATCG	538	539	289	5	164–789	0.75
BRU379_12bp_182bp_2u	Bruce43*	1	69	TCTCAAGCCCGATATGGAGAAT	TATTTTCCGCCTGCCCATAAAC	170	182	182	3	170–194	0.55
BRU256_12bp_110bp_3u	Bruce44	1	96	GGCGCAAGATCGGAATGC	AGGCAGGTGCTGATTCTCCT	110	110	122	1	110	0
BRU233_18bp_151bp_3u	Bruce45*	1	70	ATCCTTGCCTCTCCCTACCAG	CGGGTAAATATCAATGGCTTGG	187	151	151	4	133–187	0.65
BRU217_15bp_256bp_4u	Bruce46	1	81	AAAAGCTTCCGAACCAAGTGTC	GGAGCTGGTTGAGCGTTATTTC	256	256	241	2	241–256	0.5
BRU149_15bp_116bp_2u	Bruce47	1	90	CTGCCAAGGGCGAGATAAAC	CATCGTTCTGATCTTCGTGACC	131	116	131	2	116–131	0.26
BRU131_29bp_131bp_2u	Bruce48	1	95	TATAAGTCCAGCCCATGACAGG	GCGGAATATCTGGATGGGATAC	131	131	160	2	131–160	0.51
BRU112_13bp_266bp_2u	Bruce49	1	97	AACCTCGGTCTATGATGCAACC	ACGCAGGGTTAGGTTTCTCAAA	279	266	265	2	266–279	0.52
BRU80_12bp_162bp_3u	Bruce50	1	75	GCAGAACCTGATGAACAACCTG	ATTTTCTGGTCGAGATCGAAGG	174	162	174	2	162–174	0.26
BRU80_15bp_74bp_2u	Bruce51	1	90	TGACATGATGCAGAAAATGCAG	GTCCCTTGCCGCCTTTCAT	74	74	89	2	74–89	0.47
BRU50_15bp_185bp_3u	Bruce52	1	95	CAATGAACCAGATCAGCTTTCG	CGCCATGGTTTCAATATCACC	170	185	170	2	170–185	0.26
BRU50_30bp_110bp_1u	Bruce53	1	70	CGGTTATGGTGTGGAGCAACT	CTTCCAGCGGGCTTTCAG	140	110	110	2	110–140	0.52
BRU28_13bp_150bp_1u	Bruce54	1	100	CCGATCACAGACACAACAACTTC	GCGAAAAGGGAGCAGACATTAT	163	150	163	2	150–163	0.26
BRU2066_40bp_273bp_3u	Bruce55*	1	80	TCAGGCTGTTTCGTCATGTCTT	AATCTGGCGTTCGAGTTGTTCT	234	273	273	5	194–354	0.69
BRU2028_12bp_135bp_4u	Bruce56	1	79	TTGGTCGTTAGAACAAGAGTGG	CTGAACCTGTTCCGTCAAATCA	135	135	123	2	123–135	0.38
BRU69_9bp_301bp_4u	Bruce57	2	90	ATGGGAGCCTATTTCGCTTACA	GGCGGTAGAATGGATAGCTCAC	292	301	292	3	283–301	0.19
BRU33_24bp_98bp_1u	Bruce58	2	82	CATCCTGCTTGGTGTTCTTTTG	GATGGTCGTCACCAAGTCCAG	122	98	122	2	98–122	0.32
BRU33_9bp_256bp_7u	Bruce59	2	78	CGTATCATCCGGCAATGGT	CTTTCTCTTTGTCGTGGGCTTC	265	256	256	2	256–265	0.38
BRU24_12bp_279bp_3u	Bruce60	2	90	AGCAAATGAATATGTCGCGTTG	TTCACCCCGATATCGATGAAT	267	279	279	2	267–279	0.38
BRU22_12bp_162bp_5u	Bruce61	2	75	CCTAATTTCGCCATTCGGTAAC	TTGCGGATTTTCCGAATAGAAC	162	162	150	2	150–162	0.5
BRU979_18bp_140bp_3u	Bruce62	2	85	CTGACGCAGGGAAGCTTTGT	GAAAGAATGGTGAGCAGCAAGA	155	140	155	2	140–155	0.26
BRU833_15bp_145bp_3u	Bruce63	2	77	AGGGTGACATTTGTTGGAGTCA	GTGGACAGACCCATGGTAAACG	160	145	145	2	145–160	0.51
BRU832_14bp_104bp_1u	Bruce64	2	93	GAGACGACGCTTGAGGTTTTTC	CTTCCGGCGCTTCTTTCTTAT	118	104	118	2	104–118	0.26
BRU824_41bp_182bp_2u	Bruce65	2	94	CGCTCTAGGCATAGCTGTTGTG	GTCCAGCCGATAACCTTGCTAT	143	182	143	1	142	0
BRU652_17bp_91bp_1u	Bruce66	2	100	ATATCCATGCGGGAAGGAAGAT	TATGCAGTGGGCATTCTCTACG	108	91	91	2	91–108	0.32
BRU609_31bp_155bp_2u	Bruce67	2	68	GCACCAGTGCGAGAAAATAGTG	AATTTTCCTTCGCGACTCCTTT	124	155	155	2	124–155	0.52
BRU564_18bp_111bp_3u	Bruce68	2	79	GACAACGATCCGCAGAAAGG	GTCAATGCCCTGATCGGTATC	111	111	93	2	93–111	0.47
BRU488_57bp_181bp_1u	Bruce69	2	100	CGATGACAGAGCAAGACCGTTA	TTGACCGATAATTCTGCAATGG	181	181	238	1	181	0
BRU339_21bp_146bp_2u	Bruce70	2	74	GAGTAAGGCGAATAGGGGGAAC	AACTCTTCTTCCGCGACAACAC	146	146	162	2	146–162	0.51
BRU337_12bp_394bp_3u	Bruce71	2	79	GAAGACGGCTATCGACTGGTCT	ACAAGCTCTATTCGCCTTACGC	370	394	370	ND	ND	ND
BRU322_8bp_230bp_8u	Bruce72	2	66	GAAGACGGCTATCGACTGGTCT	GTTTCAATGAAGGCGAGGTGAG	206	230	206	10	198–294	0.87
BRU285_28bp_178bp_3u	Bruce73	2	86	GTGGAAGGCGTTGTCATTCTG	ATCGGTCATGGTCTATCCTTCC	178	178	150	2	150–178	0.51
BRU275_8bp_147bp_6u	Bruce74	2	85	GGATGAGGATTGAGGGCTTTT	ACGCAGACGATTTAACAAAGCA	139	147	139	4	131–155	0.63
BRU250_19bp_82bp_2u	Bruce75	2	100	AGGACTATCAGGTGCGTGACAA	AAGGAAGACGTCGCTGAAAGAC	82	82	101	2	82–10	0.52
BRU181_14bp_122bp_2u	Bruce78	2	87	CTAACAAATGACGGCAGAGTGC	TTGAACGCAAGCTTATCCAAAA	136	122	136	2	122–136	0.32
BRU163_12bp_141bp_4u	Bruce79	2	65	TCCTGTTGAACGCAAGCTAATG	ATACTTCAGGCGGGGAGGAC	153	141	153	ND	ND	ND
BRU542_12bp_178bp_4u	Bruce80	2	87	CGAGGAATGTCAGGAAGATCAC	ACACAGACGCCAAAAGACAAA	166	178	166	2	166–178	0.18

* markers constituting MLVA-15 ; the 8 non-octamers (*) are panel 1, the 7 octamers (**) are panel 2

a naming nomenclature includes repeat unit size, PCR product size in strain 16M, corresponding repeat copy number, and common name;

b internal repeat homogeneity ; c expected PCR product size in each of the three sequenced genomes, Brucella suis 1330, B. melitensis 16M, B. abortus 9–941;

d number of alleles observed in the 21 strains ; e observed size range ; f HGDI : Hunter-Gaston diversity index ; ND, not determined

Typing was done by PCR using the set of primers listed in Table 2, as described [13]. Six markers failed to amplify DNA satisfactorily, and were not included in the further study: they generated multiple band profiles (bruce20-BRU329_8bp_148bp_7u; bruce38-BRU1116_18bp_108bp_2u; bruce71-BRU337_12bp_394bp_3u), or lacked amplification using the selected primers (bruce79-BRU163_12bp_141bp_4u), or no appropriate primers could be designed targeting the flanking regions because of the presence of repeated elements (bruce76-BRU243_21bp_2u; bruce77-BRU195_21bp_2u, not listed in Table 2).

Three markers (bruce44-BRU256_12bp_110bp_3u; bruce65-BRU824_41bp_182bp_2u; bruce69-BRU488_57bp_181bp_1u) turned out to be monomorphic for the 21 reference strains. The results of the clustering analysis using the 71 remaining markers fits very well with the current knowledge of the degree of relationship between Brucella species [25] (Figure 1). We then looked for a subset of markers providing a similar discriminative power as the whole set for the collection of reference strains evaluated. Although extremely informative, the family of octamers, which includes the eight tandem repeats previously investigated [16,17], are not appropriate for species/biovar discrimination because of their hypervariability and more stable markers must be used. Among the other markers, a set of the ten most polymorphic loci clusters the different species as expected. Two of these ten markers display allele size ranges not appropriate for analysis on currently available automated DNA fragments sizing machines such as capillary electrophoresis sequencing machines (Bruce02 and Bruce15 have alleles up to 2 kb and 5 kb respectively). The amplification patterns of the 21 reference strains using the other eight TRs are shown in Figure 2. These 8 markers (Bruce06, 08, 11, 12, 42, 43, 45, 55) will subsequently be called MLVA typing panel 1. These are minisatellites loci with repeat units length above 9 bp [10]. In addition, 7 robust and highly polymorphic octamers (microsatellites) were selected to constitute MLVA typing panel 2. Panel 2 comprises Bruce04 (designated as TR6 in [16]), Bruce07, Bruce09 (TR8), Bruce16, Bruce18, Bruce21 and Bruce30 (TR2).

Figure 1

Maximum parsimony analysis, on 21 reference strains using data from all 71 markers. The different species are represented by different colours, as indicated. Biovars (b) are mentioned wherever relevant.

Figure 2

Amplification patterns of MLVA panel 1 on the 21 reference strains. Lanes 2- 8 = 2: B. melitensis-bv1 (16M reference strain) ; 3: B. melitensis-bv2; 4: B. melitensis-bv3; 5: B. pinnipediae B2/94; 6: B. cetaceae B1/94; 7: B. cetaceae B14/94; 8: B. ovis. Lanes 10-17 = 10: B. melitensis-bv1 (16M) ; 11: B. abortus-bv1; 12: B. abortus-bv2; 13: B. abortus-bv3; 14: B. abortus-bv4; 15: B. abortus-bv5; 16: B. abortus-bv6; 17: B. abortus-bv9. Lanes 19-26 = 19: B. melitensis-bv1 (16M); 20: B. suis-bv1; 21: B. suis-bv2; 22: B. suis-bv3; 23: B. suis-bv4; 24: B. suis-bv5; 25: B. canis; 26: B. neotomae. Lanes 1;9;18;27 = 100bp DNA ladder. The values for strain 16M are deduced from Table 2. The values for the other strains can be deduced from the 16M value, taking into account the indicated tandem repeat unit size. Examples are indicated, the full data can be deduced from the additional file [see Additional file 1].

Evaluation of a MLVA assay comprising 15 markers

The set of 15 TR markers (panel 1 and 2, listed with one or two asterisk in Table 2) was used for typing a larger collection of biotyped isolates including various species and biovars [see Additional file 1]. Among the 257 strains, panel 1 alone resolves 51 genotypes. This panel does not distinguish B. suis biovar 4 and B. canis. All B. canis strains investigated share panel 1 genotype 2 with some of the B. suis biovar 4 strains (Figure 3). Similarly, most B. suis biovar 3 strains share panel 1 genotype 4 with B. suis biovar 1. Panel 2 alone discriminates 200 genotypes. However, the resulting clustering only approximately fits with the expected species and biovar assignment. When using panel 1 and panel 2 together (MLVA-15 assay), 204 genotypes can be differentiated. The clustering analysis is shown in Figure 3, 4 and 5. A number of major clusters weakly connected to each other can be identified: B. suis biovar 1 (Figure 3), B. suis biovar 2( Figure 3 and figure 4), B. abortus (2 clusters, Figure 4 and Figure 5), B. melitensis (3 clusters, figure 5), B. ovis (Figure 3). Brucella suis biovar 5, B. neotomae and the marine mammal strains are quite distinct from the closest strains (Figure 4). Brucella canis and B. suis biovar 4 are closely related and loosely connected to the B. suis biovar 1 cluster (Figure 3). The three B. melitensis clusters fit moderately with the biotyping results. Similarly, B. suis biovar 3 strains do not constitute a consistent group.

Figure 3

Clustering analysis in 257 strains and isolates with the two panels of markers (MLVA-15), genotypes 1 to 68. In the columns the following data are given from left to right : the DNA batch, the genotype, the strain ID including the name of the institution of origin (“BCCN” = Brucella Culture Collection from Nouzilly, INRA, Nouzilly, France ; “BfR” = Federal Institute for Risk Assessment, BfR, Berlin, Germany ; “REF” = the 21 reference and marine mammal strains, prepared by BCCN ; “vacc.” = vaccine strain ), year of isolation, host and geographic origin when known, species and biovar (when relevant). The first genotype number (going from 1 to 204) is the MLVA-15 genotype number. The second (for instance 1.1) indicates the panel 1 genotype number (from 1 to 51) followed by the panel 2 genotype number (from 1 to 200). The corresponding genotyping data can be found in the additional file [see Additional file 1]. Wherever possible, the more precise geographic origin within a country is indicated (for instance France (03) is a strain originating from the French department number 03 (Allier) in the centre of France). The first part of the clustering of the 257 isolates in 204 genotypes is presented. It comprises 68 genotypes, corresponding to B. ovis, B. canis, B. suis biovar 1, 3, 4, and part of the B. suis biovar 2 isolates. The colour code used is as shown in Figure 1.

Figure 4

The columns content is as indicated in Figure 3 legend. The corresponding genotyping data can be found in the additional file [see Additional file 1]. The second part of the clustering is displayed, genotypes 66 to 141, comprising the rest of B. suis biovar 2 isolates, the B. neotomae strain, the three marine strains, the 2 B. suis biovar 5 isolates, and part of the B. abortus isolates. The colour code used is as shown in Figure 1

Figure 5

The columns content is as indicated in Figure 3 legend. The corresponding genotyping data can be found in the additional file [see Additional file 1]. The third part of the clustering (genotypes 140 to 204) is displayed, comprising the rest of B. abortus isolates, and the B. melitensis isolates. The RB51 vaccine strain tested here is genotype 159, S19 is genotype 161, and a number of Rev1 isolates share genotype 201. The colour code used is as shown in Figure 1.

Discussion

The genus Brucella has been divided into species and biovars for a long time, but this classification has been discussed controversially since DNA-DNA hybridization has been applied. The genus proved to be highly monomorphic with a level of relatedness among all species higher than 90% [5]. This homogeneity complicated the development of molecular assays able to efficiently recognise the species-specific entities. This finding led to the proposal of a monospecies genus, i.e. B. melitensis. The classical species would be considered as biovars only. However, most bacteriologists did not accept this concept which has recently been rejected by the subcommittee of taxonomy [26]. The purpose of the present study was firstly to investigate the polymorphism of tandem repeat loci predicted to be polymorphic by comparing the data of the three different Brucella strains already sequenced and secondly to evaluate to which extend tandem repeat typing and classical biotyping clustering fit together. We evaluated most of these loci with a repeat unit of 5 bp or more.

Polymorphism has been confirmed at 71 loci. DNA was amplified at every locus from all 21 reference strains, including the 3 marine mammal strains (except for Bruce04 in the B. melitensis bv 3 reference strain Ether and Bruce01 in the B.ovis reference strain BOW63/290) confirming the very high genetic homogeneity of the genus Brucella.

A MLVA typing assay depends on the selection of markers which individually would not provide a relevant clustering. Taken separately, the TR markers are either not informative enough, or too variable or show a high level of homoplasy. However, the combination of well selected independent loci may be highly discriminatory and to some extend phylogenetically relevant, as shown previously for other species [9], and demonstrated here for Brucella. We propose a selection of 15 markers to be used in a Brucella MLVA assay consisting of two complementary panels, panel 1 (8 markers) and panel 2 (7 markers). The fifteen markers are a combination of moderately variable (minisatellites, panel 1) and highly discriminant (microsatellites, panel 2) loci (Table 2).

The strain clustering achieved is consistent with well-established phenotypic and molecular characteristics (Figure 3, 4 and 5). The biovars 1, 2 and 4 of B. abortus are gathered in agreement with (i) the sensitivity to thionin and (ii) the PCR-RFLP pattern of the omp2a genes specific for these biovars [27]. B. abortus biovar 3 strains are found in a separate group except for 2 strains originated from Africa (BCCN 93-26 and the reference strain Tulya). Strains isolated in Africa often show distinct phenotypes [28] and thus, it is not surprising to find these two strains separated. The two strains do not require CO2 for growing. Their MLVA closest neighbours are two B. abortus biovar 6 strains also isolated in Africa. Assignment to biovar 3 or 6 reflects the H2S production which is the unique phenotypical criteria to differentiate these two biovars. The MLVA assay confirms that some African strains significantly differ from isolates of other origin and that B. abortus biovar 3 is a heterogeneous group.

The B. melitensis group is very heterogeneous using either panel 1 or both panels (MLVA-15), and comprises four main subgroups. Biovar 2 and 3 strains are mixed in two groups, together with a few biovar 1 strains. The other biovar 1 isolates form 2 groups, one including the 16M reference strain, and the other (genotypes 173 and 174, Figure 5) comprising 3 isolates from the United Arab Emirates. B. melitensis BCCN 84-3 strain (MLVA-15 genotype 20) is an isolate from a dog in Costa Rica, which was biotyped as B. melitensis biovar 2, but appears to be distantly related to other B. melitensis strains. This strain is smooth as observed by the agglutination with anti-A serum, and the profile obtained in oxidative metabolism is typical of B. melitensis. Panel 1 analysis (not shown) does associate this strain with B. melitensis, but the full MLVA-15 analysis suggests a position closer to the B. canis group (Figure 3).

B. suis strains are clearly differentiated in three groups (Figures 3 and 4). A first group includes all biovar 1, 3, and 4 strains, and a second group all biovar 2 strains. The two rare biovar 5 strains are very distantly related. The correlation with biovars is good with some interesting exceptions. The five B. suis biovar 3 isolates from Croatia have the same genotype (MLVA-15 genotype 36, Figure 3 [see Additional file 1]), and cluster with B. suis biovar 1 strains but not with the reference B. suis biovar 3 strain. More B. suis strains phenotypically identified as biovar 3 from other geographic origins are required. This may suggest that the biovar 3 phenotype may have appeared independently more than once. Biovar 1 and biovar 3 strains are distinguished by sensitivity to fuchsine and ability to produce H2S. Atypical fuchsine-resistant biovar 1 strains have already been described [6], as well as atypical fuchsine-sensitive B. melitensis strains [29,30]. So both the fuchsine sensitivity, and the H2S production (as suggested above for B. abortus) may appear to be phylogenetically weak markers with some degree of homoplasy. Among biovar 2, strains isolated from Spain and Portugal are related and can be distinguished from other European strains investigated. Biovar 4 strains can be found right beside B. canis. Meyer [31] has previously proposed a model for evolutionary derivation of Brucella organisms on the basis of phenotypic characteristics and proposed a close relationship between B. suis biovar 3/4, and B. canis. PCR-RFLP analyses of the porin genes are in agreement with this finding [27].

Three classical vaccine strains were included, Rev.1 (genotype 201), S19 (genotype 161) and RB51 (genotype 159). Six other isolates, from Israel, share genotype 201. These streptomycin resistant isolates were confirmed as Rev.1 vaccine strains using the previously described assay [32] (data not shown). This is not unexpected since vaccination is used in this country, and simply illustrates the stability of the MLVA assay in the present case.

Strains clustering together frequently have a close or identical geographic origin, e.g. MLVA-15 genotype 16 comprises 2 B. ovis isolates, coming from the same region of France "Provence-Côte d'Azur" (departments 06 and 13). In almost all such instances where the MLVA genotype of two isolates is identical, the available epidemiological data is indeed compatible with a common source of infection. The rare exceptions would then suggest that some strains travel efficiently. MLVA-15 genotype 132 was observed in Germany in 1972 and in the centre of France (department 87) in 1994. MLVA-15 genotype 1 (B. canis) was observed in Greece and Germany. More epidemiological data will be needed in order to draw precise conclusions on the circulation of the strains.

The MLVA-15 results support the current classification of the genus Brucella. In addition, differences found by phenotypic identification and/or by molecular studies are also detected by MLVA. One major advantage of MLVA is the ease of data exchanges. The data itself can be summarized by a very simple flat text file containing the repeat copy numbers for each locus and each strain. This data can also be made accessible and queried across the internet as shown [21,24].

Another advantage is that MLVA typing only depends on the measurement of DNA amplicon sizes, so that a number of electrophoretic techniques can be used, ranging from manual, low-cost, agarose gels, to high-throughput capillary electrophoresis sequencing machines.

In the near future, it is tempting to speculate that international databases containing MLVA data of thousands of strains will be produced, and MLVA will become a routine assay for any new isolate. We believe that the MLVA-15 assay will be one step in this direction. A first use of the assay for a clinical application was recently described [33].

Methods

Bacterial strains

The 257 strains and isolates used for MLVA typing are listed or described globally in Table 1. One hundred and seventeen B. suis, 43 B. melitensis, 52 B. abortus, 24 B. ovis, one B. neotomae, 17 B. canis and 3 strains isolated from marine mammals [2] were investigated. This collection includes the 18 classical reference strains representing the different species and biovars of Brucella. All strains were mainly isolated from animals and in a few cases from humans or unknown species (Figure 3, 4 and 5), and were identified by phenotypical tests based on agglutination with monospecific antisera (serotyping), phage typing, dye sensitivity, CO2 requirement and H2S production [6].

Identification of variable number tandem repeats by genomic sequence comparison

The methods previously described [10,12,21,22] and the genome sequence data for B. suis strain 1330, B. melitensis strain 16 M and B. abortus strain 9–941 [18-20] were used to identify TRs that may help to differentiate closely related genomes.

The different TRs are designated by using the nomenclature previously described [13]. For instance BRU211_63bp_257bp_2u (bruce11) is a TR at position 211 kb in the B. melitensis 16 M genome. Its common laboratory name (alias name) is Bruce11. It has a 63 bp motif, and a total PCR product length of 257 bp in the B. melitensis 16 M strain when using the primer set indicated in Table 2. This allele size corresponds to 2 units.

PCR amplification and genotyping

Brucella DNA was prepared as previously described [27]. PCR amplification was performed in a total volume of 15 μl containing 1ng of DNA, 1× PCR Reaction Buffer, 1 U of Taq DNA polymerase (Qbiogen, Illkirch, France), 200 μM of each deoxynucleotide triphosphate, and 0.3 μM of each flanking primer as described previously [15].

Amplifications were performed in a MJ Research PTC200 thermocycler. An initial denaturation step at 96°C for 5 minutes was followed by 30 cycles of denaturation at 96°C for 30 s, primer annealing at 60°C for 30 s, and elongation at 70°C for 1 min. The final extension step was performed at 70°C for 5 min.

Two to five microliters of the amplification product were loaded on a 3% standard agarose gel for analyzing tandem repeats with a unit length shorter than 10 bp and on a 2% standard agarose gel for all others, and run under a voltage of 8 V/cm until the bromophenol blue dye had reached the 20 cm position. Gels were stained with ethidium bromide, visualized under UV light, and photographed (Vilber Lourmat, Marnes-la-Vallée, France). A 100-bp and a 20-bp ladder (EZ Load 100 pb or 20 bp PCR Molecular Ruler, Biorad, Marnes-la-Coquette, France) were used as molecular size markers depending on the tandem repeat unit length. Gel images were managed using the Bionumerics software package (version 4.0, Applied-Maths, Belgium).

Data analysis

Band size estimates were converted to a number of units within a character dataset using Bionumerics version 4.0 (Applied-Maths, Belgium) [see Additional file 1]. Clustering analyses used the categorical coefficient and UPGMA (unweighted pair group method using arithmetic averages). The use of the categorical parameter implies that the character states are considered unordered. The same weight is given to a large or a small number of differences in the number of repeats at each locus. Maximum parsimony was done using Bionumerics, running 200 bootstrap simulations and treating the data as categorical.

Polymorphism index

The Hunter Gaston diversity index [34] (HGDI) was used.

Authors' contributions

MG, IJ, SAD, KN, HN were in charge of strain selection, collection and checking of related data, preparation and provision of DNAs. PLF did the MLVA genotyping work. GV was in charge of the Bionumerics database, error checking, clustering analyses. FD and PB did the genome sequence analyses for polymorphic tandem repeat searches and the genotyping page. GV wrote the report. IJ and MG helped to draft the manuscript. All authors read, commented and approved the final manuscript.

Additional File 1

MLVA-15 data for each of the 204 genotypes. The first three columns from the left are genotype numbers obtained with the different panels. The subsequent columns are the typing data itself. The first 8 markers (headings, bruce06 to bruce55) constitute panel 1 (minisatellites, tandem repeat unit length above 9 bp). The last 7 columns (starting from bruce04) constitute panel 2 (microsatellites, tandem repeat unit length up to 8 base-pairs).

Click here for file