Journal of Medical Microbiology (2005), 54, 833–841
DOI 10.1099/jmm.0.45956-0
Characterization of a tandem repeat polymorphism in Rickettsia strains Liliana Vitorino,1,2 Rita de Sousa,2 Fatima Bacellar2 and Lı´bia Ze´-Ze´1 Correspondence Lı´bia Ze´-Ze´
1
Universidade de Lisboa, Faculdade de Cieˆncias, Centro de Gene´tica e Biologia Molecular and Instituto de Cieˆncia Aplicada e Tecnologia, Edifı´cio ICAT, Campus FCUL, Campo Grande. 1749-016 Lisboa, Portugal
2
CEVDI, Instituto Nacional de Sau´de Dr Ricardo Jorge, 2965 A´guas de Moura, Portugal
[email protected]
Received 10 November 2004 Accepted 2 June 2005
Mediterranean spotted fever (MSF) is a tick-borne rickettsiosis caused by ‘Rickettsia conorii complex’ strains. In Portugal, R. conorii and Israeli tick typhus (ITT) are the aetiological agents of this disease. A novel 65 bp tandem repeat was identified by the analysis of the R. conorii Malish 7 whole genome sequence with an appropriate algorithm for searching for repeated sequences. The variable number tandem repeat (VNTR) was named VNTR Rc-65 and this locus was amplified by PCR and sequenced in order to characterize the repeat diversity within different rickettsial strains including Portuguese strains isolated from clinical and vector samples. The VNTR Rc-65 has seven alleles within the rickettsial strains studied and a diversity index value of 0.71, meaning that this locus has a great discriminatory capacity and therefore can be used for identification of closely related strains. PCR amplification of the Rc-65 locus can be used to differentiate between the Portuguese R. conorii Malish-like and Israeli tick typhus strains, enabling a more accurate and rapid identification of these rickettsial isolates.
INTRODUCTION Bacteria of the genus Rickettsia are obligate intracellular parasites that grow in the cytoplasm or occasionally in the nucleus of eukaryotic host cells. They are transmitted to vertebrates by ectoparasitic arthropod vectors: ticks, mites and fleas (Weiss & Dasch, 1991). Some species are pathogenic, causing characteristic clinical illness with general symptoms, such as sudden onset of high fever, myalgia, maculopapular rash and inoculation eschar (tache noire) at the tick-bite site (Raoult & Roux, 1997). Mediterranean spotted fever (MSF), also known as Boutonneuse fever, is an acute, febrile tick-transmitted rickettsiosis caused by strains of ‘Rickettsia conorii complex’. MSF is caused by two strains: R. conorii Malish, which is endemic in the Mediterranean area, and Israeli tick typhus (ITT). The latter strain was believed to be restricted to Israel, but was isolated from a patient in Portugal for the first time in 1997 (Bacellar et al., 1999). In Portugal, MSF is endemic and is an obligatory reported disease, with an incidence rate of 9.8 per 105 inhabitants (de Sousa et al., 2003). The number of reported cases and the fatality rate of the disease has been Abbreviations: ITT, Israeli tick typhus; assay; MLVA, multilocus VNTR analysis; MSF, Mediterranean spotted fever; SSM, slipped-strand mispairing; VNTR, variable number tandem repeat. The GenBank/EMBL/DDBJ accession numbers for the VNTR Rc-65 locus sequences of the 24 strains tested are AY820021–AY820045.
45956 & 2005 SGM Printed in Great Britain
increasing (Bacellar et al., 2003). Moreover, severe forms of the disease have been described in several patients. To date, the diagnosis of a rickettsial illness has been often carried out by microimmunofluorescence assay and identification is accomplished by ompA PCR-RFLP gene analysis (Regnery et al., 1991), and sequencing of the citrate synthase (gltA) (Roux et al., 1997) and ompA genes (Fournier et al., 1998). Although ompA gene analysis is suited for accurate identification of rickettsiae from spotted fever group, it cannot be used to identify several other rickettsial species. Likewise, the gltA gene sequence has little nucleotide variation to distinguish among all rickettsial species (Roux et al., 1997). Therefore, an accurate, discriminatory, reproducible and faster typing test should lead to an improvement in the epidemiological surveillance and diagnosis of rickettsiosis. Variable number tandem repeats (VNTRs) consist of multimeric repeats that can often vary in copy number, showing inter-individual length polymorphisms that can be detected by a PCR assay since the regions flanking the repeats are generally well-conserved targets for primer design (van Belkum et al., 1998). Polymorphisms at a tandem repeat locus can also occur as a result of nucleotide sequence changes between individual repeat units. VNTR sequences display high variability and diversity, and hence they have great discriminatory capacity. The variability observed in VNTR loci in repeat numbers and sequence degeneracy is thought to be primarily caused by slipped-strand mispairing (SSM) 833
L. Vitorino and others
increased by inadequate DNA mismatch repair or, alternatively, explained by DNA recombination between multiple loci of similar repeat motifs (van Belkum et al., 1998). The availability of whole genome sequences and appropriate algorithms for searching for repetitive sequences has led to the application of the VNTR typing approach, namely the development of multilocus VNTR analysis (MLVA) typing systems for several pathogens, such as Bacillus anthracis (Keim et al., 2000), Yersinia pestis (Klevytska et al., 2001), Francisella tularensis (Farlow et al., 2001), Borrelia burgdorferi (Farlow et al., 2002), Legionella pneumophila (Pourcel et al., 2003) and Mycobacterium tuberculosis (Spurgiesz et al., 2003). The MLVA approach has been proven to present high discriminatory power, reproducibility and portability, making it a strong candidate for the increased development of reference databases that allow online strain identification services (Supply et al., 2001; Onteniente et al., 2003). In this study, we identified a 65 bp tandem repeat by screening for tandem repeat regions in the R. conorii Malish 7 genome sequence data. PCR amplification and sequencing of this locus was performed to characterize the repeat diversity within different rickettsial strains including Portuguese clinical and vector isolates.
METHODS Bacterial strains. A set of Portuguese clinical and tick isolates
belonging to ‘R. conorii complex’ were used. These Portuguese isolates were previously identified by serology and ompA gene sequencing as described elsewhere (Fournier et al., 1998). Reference rickettsial species were also included in this study. Overall, a total of 44 rickettsiae were tested (Table 1). The genome sequences of R. conorii Malish 7 (accession no. AE006914) and Rickettsia prowazekii Madrid E (accession no. AJ235269) were downloaded from the GenBank database. Culture conditions and DNA extraction. Rickettsial isolation from
human total blood and from tick haemolymph was achieved by shellvial technique (Marrero & Raoult, 1989), followed by 7 days of incubation in VERO cells (E-6 clone) in Eagle’s minimal essential medium (MEM; GibcoBRL) supplemented with 1 % glutamine and 10 % fetal bovine serum at 32 8C for 5–6 days. Infected cell cultures were monitored for the presence of rickettsiae. Harvesting was done when the degree of infection was optimal, which was determined by Gimenez staining (Gimenez, 1964). Infected cells were mechanically suspended in the medium. The suspension was frozen and unfrozen three times, then centrifuged at 1000 r.p.m. for 30 min. Supernatants were recovered and rickettsiae were harvested by centrifugation at 8000 r.p.m. for 30 min and resuspended in ultrapure water. DNA was extracted using Qiagen columns (DNAeasy Protocol for Animal Tissue Kit) according to the manufacturer’s instructions. Tandem repeat identification. The complete genome sequence of R.
conorii Malish 7 (GenBank accession no. AE006914) was screened for tandem repeat loci using the software Tandem Repeat Finder (TRF, version 3.21), developed by Benson (1999). The TRF software was downloaded from the web page of the Department of Biomathematical Sciences at Mount Sinai School of Medicine (http://tandem.bu.edu/trf/ trf.html). A tandem repeat locus with a repeat unit of 65 bp in length, 98 % similiarity between adjacent copies overall and five copies of the repeat unit was selected for further analysis. The locus was named Rc-65 and is located in position 1 168 767 to 1 169 087 of the R. conorii Malish 7 834
genome. BLASTN (Altschul et al., 1997) searches within R. prowazekii Madrid E genome (GenBank accession no. AJ235269), Rickettsia sibirica 246 (GenBank accession no. AABW00000000) and Rickettsia rickettsii (GenBank accession no. AADJ00000000) were performed to determine the presence of related repeat sequences in these genomes. PCR amplification and sequencing. PCR amplification of the VNTR
Rc-65 locus was accomplished using specific primers Rc-65-Fw (59TTGAGAAGGTTTATATCCCATAG-39) and Rc-65-Rv (59-TAC TACCGCATATCCAATTAAAAA-39) located 31 nucleotides upstream and 78 nucleotides downstream of the Rc-65 tandem repeat in the R. conorii Malish 7 genome. PCR was performed in a 50 ìl reaction mixture containing 1 pmol of each primer, 200 ìM (each) dATP, dGTP, dCTP and dTTP (Invitrogen), 1.75 U Taq polymerase (Invitrogen), 2 mM MgCl2 , 0.53 BSA, 13 Taq buffer and 50–100 ng genomic DNA. Amplification was carried out in a DNA thermocycler (TGradient; Biometra) under the following conditions: 4 min of initial denaturation at 96 8C, then 35 cycles of 94 8C for 1 min, 45 8C for 1 min and 72 8C for 1 min. The amplification was completed by holding for 5 min at 72 8C to allow complete extension of the PCR products. PCR products were visualized by ethidium bromide staining after electrophoresis in a 1 % agarose gel and their sizes were estimated by comparison with a molecular mass standard (1 kb plus DNA ladder; GibcoBRL). The PCR products were purified using Jet Quick-PCR Purification Kit (Genomed) as described by the manufacturer. The purified PCR products were sequenced in an automated DNA capillary sequencer CEQ 2000-XL (Beckman Coulter) by a dye-labelled dideoxy-termination method (DTCS, Dye Terminator Cycle Sequencer start kit, Beckman Coulter). All sequences were determined by the consensus of the forward and the reverse sequence analysis. Low quality sequences were repeated using different conditions for the PCR sequencing reaction, namely denaturation temperature and DNA concentration. GenBank accession numbers for the nucleotide sequences are listed in Table 2. Data analysis. The discriminatory power of the VNTR Rc-65 locus was estimated by the number of alleles detected as well as by its diversity index (D), which can be calculated as 1Ó(allele frequency)2 (Weir, 1990). The complete VNTR Rc-65 region was used to assess the genetic relationships among the isolates using the UPGMA clustering method of the software package PAUP4a (D. Swofford, Sinauer Associates). UPGMA analysis was performed with the simple matching coefficient to estimate genetic distances.
RESULTS Using the TRF software (Benson, 1999) we were able to identify 16 tandem repeats within the R. conorii Malish 7 genome (Ogata et al., 2000, 2001a, b), ranging in size from 2 to 216 bp. Since a larger degree of sequence heterogeneity is observed among the longer repeat units (van Belkum et al., 1998) and they can be easily analysed by agarose gel electrophoresis, a tandem repeat unit of 65 bp was selected for VNTR analysis. This locus is located in an intergenic region in the R. conorii Malish 7 genome, between the dksA gene, which is a dnaK suppressor protein homologue, and the xerC gene, which encodes an integrase/recombinase protein. However, the VNTR begins in the 39 end of the dksA gene (overlapping the last 12 nucleotides). This locus proved to be polymorphic by PCR amplification of several rickettsial strains (Fig. 1), and hence can be referred to as a VNTR locus. However, five species failed to yield VNTR Rc-65 PCR amplification, Rickettsia typhi, Rickettsia bellii, Rickettsia canadensis, Rickettsia australis and Rickettsia Journal of Medical Microbiology 54
VNTR-based typing of Rickettsia
Table 1. Rickettsia strains Abbreviations: PoHuR, Portuguese human Rickettsia; PoTiR, Portuguese tick Rickettsia. Species R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii R. conorii Astrakhan fever rickettsia Thai tick typhus rickettsia R. slovaca R. massiliae Rickettsia sp. Bar29 R. aeschlimannii R. montanensis R. helvetica R. helvetica R. sibirica R. rhipicephali R. akari R. bellii R. canadensis R. africae R. japonica R. australis R. parkeri R. prowazekii R. typhi R. rickettsii R. felis
Strain Malish 7 PoTiR12‡ PoHuR1258‡ PoHuR1450§ PoHuR4649‡ PoHuR4991§ PoHuR4993§ PoHuR6143‡ PoHuR6571‡ PoHuR6647§ PoHuR6648§ PoHuR6733‡ PoHuR8216‡ PoHuR8122§ PoHuR8557‡ PoHuR8584§ PoHuR8618‡ PoHuR8619§ PoHuR8621‡ PoHuR8643§ PoHuR8732‡ PoHuR8756§ PoHuR8958§ PoHuR23469‡ A – 167 TT-118 PoTiR30 PoTiR66 Bar29 MC16T M/56 C3 PoTiR43 246 3-7-6 MK 369L42-1 #2678 EFS-5 YH Philips Maculatum 20 Madrid E WilmingtonT R Marseille
Geographic origin
Source of strain*
South Africa Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Russia Thailand Portugal Portugal Spain Morocco USA Switzerland Portugal Russia USA USA USA Canada Ethiopia Japan Australia USA Spain USA USA USA
AE006914† This study (1992) This study (1997) This study (1997) This study (2001) This study (2001) This study (2001) This study (2002) This study (2002) This study (2002) This study (2002) This study (2002) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) This study (2003) D. Raoultk D. Raoultk This study (1993) This study (2003) D. Raoultk D. Raoultk D. Raoultk D. Raoultk This study (1995) D. Raoultk D. Raoultk D. Raoultk D. Raoultk D. Raoultk D. Raoultk D. Raoultk D. Raoultk D. Raoultk AJ235269† D. Raoultk D. Raoultk D. Raoultk
*Numbers in parentheses indicate the year of isolation. †GenBank accession number. ‡Identified as R. conorii Malish-like. §Identified as Israeli tick typhus (ITT). kUnite´
des Ricketsies, Faculte´ de Me´decine, Universite´ de la Me´diterranne´e, Marseille, France.
http://jmm.sgmjournals.org
835
L. Vitorino and others
Table 2. Characteristics of VNTR Rc-65 locus Abbreviations: PoHuR, Portuguese human Rickettsia; PoTiR, Portuguese tick Rickettsia. Strains
R. montanensis R. conorii PoTiR12 R. conorii PoHuR 8216 R. conorii PoHuR 8557 R. conorii Malish 7 Thai tick typhus rickettsia ITT PoHuR 6647 ITT PoHuR 8122 ITT PoHuR 8756 ITT PoHuR 1450 ITT PoHuR 8584 ITT PoHuR 8643 ITT PoHuR 8958 R. astrahkan R. slovaca PoTiR30 R. parkeri R. japonica R. africae R. rickettsii R. sibirica R. helvetica Rickettsia sp. Bar29 R. massiliae PoTiR66 R. aeschlimannii R. rhipicephali R. akari
Consensus size (bp)
No. of repeats
Sequence similarity between repeats (%)*
Sequence size (bp)†
GenBank accession no.
65 63 63 63 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65 65
10 8 8 8 5 4 4 4 4 3 3 3 3 3 3 3 2 2 2 1 1 1 1 1 1 1
98 97 97 97 98 99 100 100 100 100 100 100 100 100 99 100 98 100 98 – – – – – – –
750‡ 650‡ 650‡ 650‡ – 400‡ 371 371 371 305 305 306 306 306 306 305 240 250‡ 240 175 180 180 200‡ 180 180 171
AY820021 AY820022 AY820023 AY820024 AE006914 AY820025 AY820026 AY820027 AY820028 AY820029 AY820030 AY820031 AY820032 AY820033 AY820034 AY820035 AY820036 AY820037 AY820038 AY820039 AY820040 AY820041 AY820042 AY820043 AY820044 AY820045
*Percentage of matches between adjacent copies overall (calculated using the TRF software). †The amplicon lengths were obtained by direct sequencing of the PCR products. ‡Values inferred by agarose gel electrophoresis analysis (Fig. 1).
felis. Among all the rickettsiae tested, six different amplicon lengths were identified, corresponding to six distinct alleles. Taking into account the analysis of R. conorii Malish 7, a total of seven alleles are considered in the present study. Based on the number of alleles and the frequency of each allele, a diversity index value of 0.71 was determined, which provides a measure of the VNTR Rc-65 discriminatory power. The Rc-65 loci of 25 rickettsiae, which consist of distinct alleles, as determined by agarose gel electrophoresis analysis (Fig. 1), were sequenced to assess the nucleotide consensus sequences and the number of repeats within each allele (Table 2). To consider the sequence variability among individual units and the number of arrays, the entire VNTR region was compared by sequence alignment, and cluster analysis was performed using the UPGMA method (Fig. 2). In the majority of the rickettsiae species tested, the repeatunit length was 65 bp, as predicted by TRF software (Benson, 1999). Rickettsia rhipicephali, Rickettsia massiliae and Rick836
ettsia sp. Bar29 revealed a tandem repeat unit of 64 bp (Fig. 2), which corresponds to the minimal amplicon length detected, 180 bp. Two distinct alleles were detected within ITT Portuguese isolates; three isolates revealed an additional repeat unit (Fig. 1, Table 2). Eight tandem repeats within the Rc-65 locus were identified in all of the R. conorii Portuguese isolates, giving them a different array size to the R. conorii Malish 7 genome (GenBank accession no. AE006914), which consisted of five tandem repeat units of 65 bp (Table 2). Moreover, the consensus pattern of R. conorii Malish-like Portuguese isolates determined by the TRF software (Benson, 1999), which applies the majority rule, consisted of 63 bp repeat units (Fig. 2). However, within the eight tandem repeats there were four that had 65 bp, which reflects the high level of nucleotide sequence polymorphism between individual repeats (Table 2, Fig. 2) detected in these strains. The nucleotide sequence structure of the Rc-65 locus is Journal of Medical Microbiology 54
VNTR-based typing of Rickettsia 1
2
3
4
5
6
7
8
9
10
11 12
13
14
15
16
17
18
19
Fig. 1. Length polymorphisms at the VNTR Rc-65 locus in rickettsial species and strains. The VNTR Rc-65 locus was amplified by PCR and the products were resolved in 1 % agarose gel electrophoresis. Lanes: 1 and 19, 1 kb plus DNA ladder (GibcoBRL); 2, R. montanensis; 3, R. conorii PoTiR12; 4, R. conorii PoHuR8216; 5, ITT PoHuR8122; 6, ITT PoHuR6647; 7, ITT PoHuR8958; 8, ITT PoHuR8643; 9, R. slovaca PoTiR30; 10, R. japonica; 11, R. africae; 12, R. rickettsii; 13, R. helvetica; 14, Rickettsia sp. Bar29; 15, R. massiliae PoTiR66; 16, R. aeschlimannii; 17, R. rhipicephali; 18, R. akari.
illustrated by dot plot analysis (Fig. 3). The centre diagonal line in each panel represents the sequence identity with itself, while parallel diagonals indicate directly repeating sequences. The difference in the number of repeats between the R. conorii Malish-like and ITT strains is easily observed in these plots. The nucleotide structure of R. conorii PoTiR12, which has eight tandem repeats, is represented by one central diagonal and seven parallel lines (Fig. 3c), while the ITT strains show one central diagonal and two or three parallel lines (Fig. 3a, b). It is also possible to observe the nucleotide sequence degeneracy between the repeats, by the lack of continuity in the parallel lines. We also tested whether the VNTR Rc-65 locus could be used to identify Rickettsia species directly from vector and clinical samples. An expected amplicon size was achieved by PCR amplification from tick lysates previously screened for rickettsiae by ompA gene amplification (Regnery et al., 1991). The Rc-65 locus was also specifically amplified in ticks coinfected with Borrelia species (data not shown). Considering the skin biopsies or blood samples, which were previously known to be positive for rickettsiae, the VNTR Rc65 locus was not successfully amplified by PCR. Even so in some cases a faint band was observed (data not shown). This could probably be due to the low number of bacteria present in these kinds of samples. Indeed, the amplification of other genes, such as the ompA or gltA genes, in these samples was also difficult, and was only achieved by nested PCR. In the future, a nested-PCR approach using an additional primer pair will be attempted in these clinical samples.
DISCUSSION We have identified and characterized a VNTR locus, consisting of a 65 bp repeat motif, that can be used for http://jmm.sgmjournals.org
genotyping rickettsiae. Seven alleles of the VNTR Rc-65 locus were represented within the rickettsiae tested. This VNTR has a diversity index value of 0.71, which means that this locus is highly informative, and hence possesses great discriminatory power for identification of genetically similar strains. Indeed, the main finding of this study is the accurate discrimination of ‘R. conorii complex’ strains, R. conorii Malish and Israeli tick typhus, solely by PCR amplification of this novel polymorphic VNTR locus. This is particularly useful in the MSF diagnosis of Portuguese patients, since recent data from the National Institute of Health Dr Ricardo Jorge have pointed out that half of MSF cases occurring in Portugal are caused by ITT strains, revealing a prevalence of infection similar to R. conorii Malish-like strains (Bacellar et al., 2003), which were previously thought to be the most common agent of MSF disease in Portugal. Indeed, the VNTR genotypes of the Portuguese isolates are in agreement with the serotyping and sequencing data previously obtained, which validates the usefulness of this approach as a diagnostic tool to distinguish between R. conorii Malish-like and ITT isolates. Regarding ITT isolates two distinct alleles were detected (Table 2), which points out the capacity of this methodology to reveal some heterogeneity within different ITT isolates. The usefulness of the Rc-65 VNTR region for typing R. conorii strains was also observed in a recent study developed by Fournier et al. (2004), which was published after we finished our analysis. That study, based on the analysis of R. conorii and R. prowazekii genomes, compares the capacity of three different types of sequences (variable coding genes, genes degraded in R. conorii but intact in R. prowazekii, and conserved and variable intergenic spacers) for R. conorii 837
A
0.005 substitutions/site
GTATTTCTAGCATGTTAATATCTCAATATTAGCTTTAATATATTATGTTTATAAGATTTTTAAAA 65
R. aeschlimannii
GTGTTTCTAGCATATTAATATCTCAATATTAGCTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
Rickettsia sp.Bar29
GTGTTTCTAGCATATTAATATCTCAATATTAGCTTTAATATATTA-GTTAATAAGGTTTTTAAAA 64 GTGTTTCTAGCATATTAATATCTCAATATTAGCTTTAATATATTA-GTTAATAAGGTTTTTAAAA 64 GTGTTTCTAGCATATTAATATCTCAATATTAGCTTTAATATATTA-GTTAATAAGGTTTTTAAAA 64
R. massiliae PoTiR66 R. rhipicephali R.conorii Malish7
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. conorii PoTiR12
GTGTTTCTAGCATATTAATATCTCAA--TTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 63
R. conorii PoHuR8216 R. conorii PoHuR8557 Thai tick typhus rickettsia R. conorii PoHuR6647
GTGTTTCTAGCATATTAATATCTCAA--TTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 63
R. conorii PoHuR8122
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. conorii PoHuR8756
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. conorii PoHuR8958
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. conorii PoHuR1450
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. conorii PoHuR8643
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. conorii PoHuR8584
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. parkeri R. astrakhan
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
GTGTTTCTAGCATATTAATATCTCAA--TTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 63 GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65 GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
Journal of Medical Microbiology 54
R. slovaca PoTiR30
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. africae
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. sibirica
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAAGTTTTTAAAA 65
R. japonica
GTGTTTCTAGCATATTAATATCTCAATATTAACTTTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. rickettsii
GTGTTTCTAGCATATTCATATCTCAATATTAACTRTAATATATTATGTTTATAAGGTTTTTAAAA 65
R. helvetica
GTGTTTCTAGCATATTAATATCTCAATATTAGCTTTAACATATTATATTTATAAGGTTTTTAAAA 65
R. akari
GTGTTTCTATCATAGAAATATCTCAATATTAGCTCTAGTATATATAGTTTATAAGGTTTTTAAAA 65 ** ****** *** ********* *** ** ** **** ** **** *********
Fig. 2. Dendrogram based on UPGMA analysis of sequence data from the complete VNTR Rc-65 region. Genetic distances were inferred by simple matching coefficient with PAUP4a software. The cluster corresponding to the R. rickettsii group is labelled ‘A’ and the cluster corresponding to the R. massiliae group is labelled ‘B’. The alignment of the repeat unit consensus sequence is shown. The open boxes represent nucleotides that differ from the VNTR consensus pattern but are conserved in all the repeats. The shaded boxes indicate nucleotides that differ from the VNTR consensus pattern and also between individual units. Grey shadows indicate nucleotides that vary only in some arrays but are present in the majority of the tandem repeats. Nucleotide indetermination, R, indicates the presence of different nucleotides within both repeat units (according to the IUB code, adenine in one repeat and thymine in the other).
L. Vitorino and others
838
B
R. montanensis
VNTR-based typing of Rickettsia
1
30 60 90 120 150 180 210 240 270 300
1
55
110
165
220
275
330
385
440
495
550
1
35 70 105 140 175 210 245 280 315 350
150 180
245
210
280
240
315
270
350
300
11 0 16 5
175 210
22 0
120
27 5
90
140
33 0
60
105
38 5
30
70
0
35
55
1
44
1
(c)
49 5
1
(b)
55 0
(a)
Fig. 3. Dot plot homology analysis of VNTR Rc-65 nucleotide sequences. Each VNTR was aligned against itself using BioEdit Sequence Alignment Editor, vers. 6.0.6. The window size was 15 and the stringency was 3. (a) ITT PoHuR8122; (b) ITT PoHuR1450; (c) R. conorii PoTiR12.
strain typing. Although some differences can be pointed out in the consensus analysis determined (mainly due to the types of approaches employed) the dksA–xerC intergenic spacer is also considered one of the most polymorphic regions to be used in genotyping rickettsiae. The VNTR typing results showed that this assay can be used to distinguish between different rickettsial species, mainly those belonging to the R. rickettsii group, which is comprised of many different species including the ‘R. conorii complex’ strains (Sekeyova et al., 2001). As predicted by a BLASTN search within the R. prowazekii genome (Andersson et al., 1998), the Rc-65 locus was not amplified in this species nor in R. typhi. Interestingly, the other species that did not yield amplification results were those previously described as being more distant phylogenetically (Sekeyova et al., 2001; Roux et al., 1997; Fournier et al., 1998; Vitorino et al., 2003). Moreover, it has been suggested that R. canadensis and R. bellii, as well as an unidentified bacterium from Adalia bipunctata, were the first representatives of the genus Rickettsia to diverge, and hence they have major genetic differences (Stothard et al., 1994). In view of this fact, the unsuccessful PCR amplification of the Rc-65 locus in these rickettsiae species could mean that they may have different nucleotide sequences in the primer regions. Therefore, a new set of primers designed from an alignment consensus sequence of the dksA and xerC genes might enable the VNTR Rc-65 analysis for these species. In the dendrogram obtained from the UPGMA clustering analysis of the VNTR region some rickettsiae groups were observed, mainly the R. massiliae group (Fig. 2, indicated by ‘B’) and the R. rickettsii group (Fig. 2, indicated by ‘A’). R. rickettsii revealed mismatches between both adjacent copy repeats, and so the presented nucleotide indetermination (according to IUB code) in the VNTR consensus sequence means that different nucleotides are present in the two repeats in the identified position. Thus, this VNTR consensus sequence represents distinctive nucleotide differences when http://jmm.sgmjournals.org
compared with other rickettsiae from the R. rickettsii group (Fig. 2). Rickettsia akari shows the highest variability within the consensus sequence. Concerning the R. conorii strains, there is 3 % sequence variation between tandem repeat arrays. R. conorii Malish 7 displays sequence variation of 2 % (Table 2). This variation is due to mismatches and indels (insertions/deletions) (1 or 2 nucleotides) between adjacent copies of the sequence. Heterogeneity among repeats is thought to be indicative of more frequent recombination (van Belkum et al., 1998), and thus R. conorii Malish-like isolates may be more prone to suffer recombination events than the other rickettsiae species included in this study. Within the Portuguese R. conorii Malish-like isolates there is the same indel, which accounts for a 63 bp consensus array. In fact, Fournier et al. (2004) described a repeated sequence array ranging in size from 63 bp to 102 bp within the dksA– xerC intergenic spacer of R. conorii strains. In that study there are also some strains possessing eight tandem repeat units, including a Portuguese isolate, which is in agreement with our results. Nevertheless, there are two other Portuguese isolates included in the study that revealed different VNTR lengths. Hence, these strains must be tested by this approach, and a wide range of local isolates should be screened to find out if there is a different VNTR Rc-65 locus in the R. conorii Malish-like Portuguese strains. Regarding the ITT strains, no sequence variation among individual units is detected, though there is some variation in the number of repeats within three different isolates, which could reflect some heterogeneity among the ITT Portuguese isolates. However, no correlation could be made between this VNTR profile and the geographic source, the isolation year or even the patient’s clinical manifestations. Nevertheless, further analysis of other potential VNTR loci as well as a multilocus sequencing study based in several genes should be developed to elucidate the heterogeneity of these isolates. 839
L. Vitorino and others
As pointed out by Fournier et al. (2004), the Rc-65 locus has a low G+C content, in contrast to other intergenic regions scattered across the R. conorii Malish 7 genome (Ogata et al., 2001b), and since the rickettsial genome is AT-rich and is thought to be in a reductive evolution process (Andersson et al., 1998; Ogata et al., 2001b), this locus may represent remnants of a rickettsial gene on its way to decay. From this perspective it would be reasonable to consider that VNTR polymorphisms are most likely to occur by deletion of repeat units, either by DNA recombination between repeats or by SSM. Indeed, it seems that a higher percentage of adenosine and thymine may increase the possibility of SSM and therefore enhance deletion events (van Belkum et al., 1998).
scientific supervision and engagement in providing the necessary resources and facilities. L. Vitorino and L. Ze´-Ze´ are the recipients of FCT research grants SFRH/BD/10676/2002 and SFRH/BPD/3653/2000, respectively.
Nevertheless, this tandem repeat region could be somehow important to the rickettsiae, given the fact that sequence repeats have been described as important elements in bacterial adaptation and pathogenesis, particularly when located within surface-protein coding genes or within potential virulence genes, respectively (van Belkum et al., 1998). Even in the case of extragenic VNTRs, they may contribute to a high mutation rate of flanking genes, allowing the bacterium to react quickly to deleterious environmental conditions (Moxon et al., 1994). Since the 59 end of the VNTR Rc-65 contingency gene locus encodes the dnaK suppressor protein homologue, whose function is related to stress conditions, it would be interesting to study whether array repeat number variation plays any role in the regulation of gene expression, as for many pathogenic bacteria that have VNTR-mediated phase variation of virulent factor expression (Peak et al., 1996).
Bacellar, F., Beati, L., Franc¸a, A., Poc¸as, J., Regnery, R. & Filipe, A. (1999). Israeli tick typhus Rikettsiae (R. conorii complex) causing disease
REFERENCES Altschul, S. F., Madden, T. L., Scha¨ffer, A. A., Zhang, J., Zhang, Z., Miller, W. & Lipman, F. J. (1997). Gapped BLAST and PSI-BLAST: a new
generation of protein database search programs. Nucleic Acids Res 25, 3389–3402. Andersson, S. G. E., Zomorodipour, A., Andersson, J. O & 7 other authors (1998). The genome sequence of Rickettsia prowazekii and the
origin of mitochondria. Nature 396, 133–134.
in Portugal. Emerg Infect Dis 5, 835–836. Bacellar, F., Sousa, R., Santos, A., Santos-Silva, M. & Parola, P. (2003).
Boutonneuse fever in Portugal: 1995–2000. Data of state laboratory. Eur J Epidemiol 18, 275–277. Benson, G. (1999). Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res 27, 573–580. De Sousa, R., Nobrega, S. O., Bacellar, F. A. & Torgal, J. (2003).
Epidemiologic features of Mediterranean spotted fever in Portugal. Acta Med Port 16, 429–436. Farlow, J., Smith, K. L., Wong, J., Abrams, M., Lytle, M. & Keim, P. (2001).
Francisella tularensis strain typing using multiple-locus, variable-number tandem repeat analysis. J Clin Microbiol 39, 3186–3192. Farlow, J., Postic, D., Smith, K. L., Jay, Z., Baranton, G. & Keim, P. (2002). Strain typing of Borrelia burgdorferi, Borrelia afzelii, and Borrelia
garinii by using multiple-locus variable-number tandem repeat analysis. J Clin Microbiol 40, 4612–4618.
This VNTR analysis has shown to be a valuable approach for rickettsiae typing, in agreement with the results of Fournier et al. (2004) obtained only for R. conorii strains, with excellent potential to be further optimized and used in a MLVA approach for diagnostics and application to large-scale epidemiological studies. Future work will involve the improvement of rickettsiae detection in clinical samples (biopsy and blood) and the analysis of several other tandem repeat loci, as well as automated gel analysis, through the use of fluorescent-labelled primers, to accomplish a high-throughput system that reduces time and determination errors in results. The combination of several VNTR loci should improve the discriminatory capacity of the VNTR typing system, providing greater resolution and probably resulting in the development of a novel and accurate diagnostic PCRbased approach for rickettsiae detection and identification.
Fournier, P. E., Roux, V. & Raoult, D. (1998). Phylogenetic analysis of the
This new assay constitutes an improvement in the accuracy and swiftness of rickettsiae strain identification, since the result can be analysed directly by gel agarose electrophoresis because there is no need for further time expenses in restriction analysis or sequencing, unlike ompA or gltA gene typing assays.
rapid detection of Mediterranean spotted fever rickettsia in blood culture. Am J Trop Med Hyg 40, 197–199.
ACKNOWLEDGEMENTS The ICAT/FCUL research team involved in this work strongly acknowledges its scientific leader, Professor Roge´rio Tenreiro, for the general 840
spotted fever Rickettsiae by study of the outer surface protein rOmpA. Int J Syst Bacteriol 48, 839–849. Fournier, P. E., Zhu, Y., Ogata, H. & Raoult, D. (2004). Use of highly
variable intergenic spacer sequences for multispacer typing of Rickettsia conorii strains. J Clin Microbiol 42, 5757–5766. Gimenez, D. F. (1964). Staining rickettsiae in yolk-sac cultures. Stain
Technol 39, 135–140. Keim, P., Price, L. B., Klevytska, A. M., Smith, K. L., Schupp, J. M., Okinaka, R., Jackson, P. J. & Hugh-Jones, M. E. (2000). Multiple-locus
variable-number tandem repeat analysis reveals genetic relationships within Bacillus anthracis. J Bacteriol 182, 2928–2936. Klevytska, A. M., Price, L. B., Schupp, J. M., Worsham, P. L., Wong, J. & Keim, P. (2001). Identification and characterization of variable-number
tandem repeats in the Yersinia pestis genome. J Clin Microbiol 39, 3179–3185. Marrero, M. & Raoult, D. (1989). Centrifugation-shell vial technique for
Moxon, E. R., Rainey, P. B., Nowark, M. A. & Lenski, R. E. (1994).
Adaptative evolution of highly mutable loci in pathogenic bacteria. Curr Biol 4, 24–33. Ogata, H., Audic, S., Barbe, V., Artiguenave, F., Fournier, P. E., Raoult, D. & Claverie, J. M. (2000). Selfish DNA in protein-coding genes of
Rickettsia. Science 290, 347–350. Ogata, H., Audic, S., Renesto-Audiffren, P. & 8 other authors (2001a).
Mechanisms of evolution in Rickettsia conorii and R. prowazekii. Science 293, 2093–2098.
Journal of Medical Microbiology 54
VNTR-based typing of Rickettsia
Ogata, H., Audic, S., Barbe, V., Artiguenave, F., Fournier, P. E. & Claverie, J.-M. (2001b). Selfish DNA in protein-coding genes of
Spurgiesz, R. S., Quitugua, T. N., Smith, K. L., Schupp, J., Palmer, E. G., Cox, R. A. & Keim, P. (2003). Molecular typing of Mycobacterium
Rickettsia. Science 291, 252–253.
tuberculosis by using nine novel variable-number tandem repeats across the Beijing family and low-copy-number IS6110 isolates. J Clin Microbiol 41, 4224–4230.
Onteniente, L., Brisse, S., Tassios, P. T. & Vergnaud, G. (2003).
Evaluation of the polymorphisms associated with tandem repeats for Pseudomonas aeruginosa strain typing. J Clin Microbiol 41, 4991–4997.
Stothard, D. R., Clarck, J. B. & Fuerst, P. A. (1994). Ancestral divergence
Peak, I. R., Jennings, M. P., Hood, D. W., Bisercic, M. & Moxon, E. R. (1996). Tetrameric repeat units associated with virulence factor phase
of Rickettsia bellii from spotted fever and typhus groups of Rickettsia and antiquity of the genus Rickettsia. Int J Syst Bacteriol 44, 798–804.
variation in Haemophilus also occur in Neisseria spp. and Moraxella catarrhalis. FEMS Microbiol Lett 137, 109–114.
Supply, P., Lesjean, S., Savine, E., Kremer, K., van Soolingen, D. & Locht, C. (2001). Automated high-throughput genotyping for study of
Pourcel, C., Vidgop, Y., Ramisse, F., Vergnaud, G. & Tram, C. (2003).
global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units. J Clin Microbiol 39, 3563–3571.
Characterization of a tandem repeat polymorphism in Legionella pneumophila and its use for genotyping. J Clin Microbiol 41, 1819–1826.
van Belkum, A., Scherer, S., van Alphen, L. & Verbrugh, H. (1998).
emerging infectious diseases. Clin Microbiol Rev 10, 694–719.
Short-sequence DNA repeats in prokaryotic genomes. Microbiol Mol Biol Rev 62, 275–293.
Regnery, R. L., Spruill, C. L. & Plikaytis, B. D. (1991). Genotypic
Vitorino, L., Ze´-Ze´, L., Sousa, A., Bacellar, F. & Tenreiro, R. (2003).
identification of rickettsiae and estimation of interspecies sequence divergence for portions of two rickettsial genes. J Bacteriol 173, 1576–1589.
rRNA intergenic spacer regions for phylogenetic analysis of Rickettsia species. In Rickettsiology: Present and Future Directions, vol. 990, pp. 726–733. Edited by K. E. Hechemy and others. New York: Annals of the New York Academy of Sciences.
Raoult, D. & Roux, V. (1997). Rickettsioses as paradigms of new or
Roux, V., Rydkina, E., Eremeeva, M. & Raoult, D. (1997). Citrate
synthase gene comparison, a new tool for phylogenetic analysis and its application for the rickettsiae. Int J Syst Bacteriol 47, 252–226.
Weir, B. S. (1990). Genetic Data Analysis. Sunderland, MA: Sinauer
Sekeyova, Z., Roux, V. & Raoult, D. (2001). Phylogeny of Rickettsia spp.
Weiss, E. & Dasch, G. A. (1991). Introduction to the rickettsiales and
inferred by comparing sequences of ‘gene D’, which encodes an intracytoplasmatic protein. Int J Syst Evol Microbiol 51, 1353–1360.
other parasitic or mutualistic prokaryotes. In The Prokaryotes, 2nd edn, pp. 2402–2406. Edited by A. Balows and others. New York: Springer.
http://jmm.sgmjournals.org
Associates.
841
