Abstract
A sample of 279 European wild rabbits, Oryctolagus cuniculus (141 males, 138 females), captured alive in Andalusia (Spain) and belonging to the two haplotype classes previously described for this species (230 and 49 corresponding with haplotypes A and B, respectively), were tested for the presence of Bartonella alsatica DNA. Two species-specific nested polymerase chain reaction assays targeting for 16S–23S rRNA intergenic spacer region and RNA polymerase β subunit genes have been developed. Forty-eight (17.20%) rabbits were infected with B. alsatica. Two-way contingency table analyses and the calculation of Cramer's V statistic showed no differences in infection rate, considering haplotype lineage or sex. The risk of infection of human population, especially for hunters in close contact with this demonstrated human pathogen, should be considered.
Introduction
Bartonella alsatica is a human pathogen that was formerly isolated from the blood of European rabbit in Alsace Department, France (Heller et al. 1999), and identified as a causative agent of endocarditis and lymphadenitis in humans (Raoult et al. 2006, Angelakis et al. 2008, Jeanclaude et al. 2009). European rabbit (Oryctolagus cuniculus) has been indicated as the natural host of this Bartonella species, but the vector has not yet been determined (Vayssier-Taussat et al. 2009).
European wild rabbit is a keystone species in the Iberian Peninsula biodiversity hotspot, serving as prey or host for a number of predators or parasites (Delibes-Mateos et al. 2008) and a relevant species for hunters in Iberian Peninsula (Virgós et al. 2007). Southwestern Europe has been the origin of the European rabbit. Until the Middle Ages the natural range of this lagomorph was restricted to the Iberian Peninsula and southern France. Posterior spreading and colonization of rabbits across central and northern Europe and the British Isles was mediated by human activity, and today it can be found throughout the world (Ferrand and Branco 2007). Analysis of rabbit blood proteins, immunoglobulin, and mitochondrial DNA variation shows the existence of two parapatric lineages in the Iberian Peninsula (Branco et al. 2002). Lineage A is typical of the southwest and lineage B of the northeast, overlapping in a central region along a northwest–southeast axis. Both rabbit lineages are extensively present in Andalusia.
Material and Methods
DNA from a portion of the pinnae of ear, preserved in 70% ethanol, of 279 adult wild European rabbits captured in Andalusia (Spain), during a study concerning the management and conservation status of this species, was extracted using the Macherey-Nagel DNA Tissue Kit (Macherey-Nagel, Düren, Germany), according to the manufacturer's instructions. Negative controls consisting of sterile water were included, using a ratio of one control for every 23 samples. DNA extracts were stored at −20°C until further processing.
Bartonella detection
Two nested polymerase chain reaction (PCR) assays targeting the 16S–23S rRNA intergenic spacer region (ITS) and the gene coding for RNA polymerase β subunit (rpoB) of B. alsatica were developed.
For ITS gene, DNA was tested by nested PCR assay using the primers URBarto1 (5′-CTT CGT TTC TCT TTC TTC A) (Rolain et al. 2003) and Balsatrev1 (5′-CTT CTC TTC ACA ATT TCA TT) for the first amplification round, and Balsatfor2 (5′-CGT TTC TCT TTC TTC AGA TG) and Balsatrev2 (5′-TCA CAA TTT CAT TAG AAC AAG) for the second amplification round.
For rpoB gene amplification, the oligonucleotides used were rpob1400F and rpob2300R (Renesto et al. 2001) for the first amplification round, and rpobf2 (5′-ACA GTC ATG CCA CAG GAT TTG) and rpobr2 (5′-TGT CAA CAC CAG ATT TTG AAG G) for the second PCR.
New specific primers (Balsatrev1, Balsatrev2, rpobf2, and rpobr2) have been developed for this study using the program Oligo (version 4.0) (Rychlik and Rhods 1989) based on GenBank sequences with accession codes AF312506 (Houpikian and Raoult 2001) and AF165987 (Renesto et al. 2001) for ITS and rpoB genes, respectively.
Negative controls were further processed by PCR as tissue specimens. PCR amplifications were carried out in a MJ Mini Personal Thermal Cycler (Biorad, Hercules, CA). Each first PCR mixture round consisted of the following: 8 μL of DNA, 20 pmol of each primer, 200 μM of each desoxynucleotide triphosphate (dNTP) 2.0 mM MgCl2, 0.033U of DNA polymerase in 1× Taq buffer advanced (5 Prime GmbH, Hamburg, Germany), and sterile distilled water to a final volume of 50 μL. PCR cycles included an initial 90 s denaturation step at 96°C, followed by 25 and 30 cycles, respectively, for first and second amplification rounds, of denaturation at 94°C for 30 s, annealing at 50°C (ITS) and 53°C (rpoB) (52°C for second ITS PCR; 54.3°C in the case of rpoB gene) for 30 s, and extension at 68°C for 60 s. Amplification was completed by holding the reaction mixture at 68°C for 7 min to allow complete extension. PCR products were resolved by electrophoresis in 1.0% SeaKem agarose gel (Cambrex Bio Science, Rockland, ME) with 1× Bionic buffer (Sigma, St. Louis, MO), using a 100-bp ladder as molecular weight marker (Eurogentec, Seraing, Belgium).
To reduce the risk of PCR carryover contamination, working areas were physically divided and positive controls were not included in any amplification assay.
For both DNA markers, products containing positive results were purified using the Montage PCR kit (Millipore, Bedford, MA) prior to sequencing. Positive PCR products were sequenced using PCR primers and the GenomeLab DTCS-Quick Start kit (Beckman Coulter, Brea, CA) and a CEQ 2000XL capillary DNA sequencer (Beckman Coulter) according to the manufacturer's instructions. The resulting ITS sequences were manually aligned and analyzed with Bioedit (version 7.0.1) sequence analysis software (Hall 1999) to obtain consensus sequences and to align and compare with B. alsatica sequences found in GenBank with accession codes AF312506 (Houpikian and Raoult 2001) and AF165987 (Renesto et al. 2001) for B. alsatica ITS and rpoB genes, respectively. Sequences were identified using the BLAST feature of GenBank (
Wild rabbit haplotype characterization and sexing
Haplotype assignation to European wild rabbits was performed by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP), and the amplification of a fragment of cyt b mitochondrial gene of 1129 bp in length was carried out using the primers ORYCITBFOR (5′-GAA CCT AAT GAC CAA CAT TCG) and ORYCITBREV (5′-GTT TTC GAT TAG GCT TGC GAG) corresponding with positions 14168 and 15296 in the GenBank sequence with accession code AJ001588 (Gissi et al. 1998). Amplification products were digested with SalI restriction endonuclease (Sigma-Aldrich, St. Louis, MO), according to instructions provided by the supplier to perform haplotype differentiation. The fragments obtained were separated by electrophoresis on 3% Nusieve GTG (Cambrex Bio Science) agarose gel and RFLP-specific patterns were visualized under UV light.
For molecular sexing of wild rabbits, a fragment of the SRY gene region of Y chromosome was amplified through PCR: the primers MAEYF1086 (5′-GCA GCT AAT CTG CTC ACA GCC) and MAEYR1376 (5′-AAC AAT CAT ACC CAT TGG TCG AG) (Geraldes et al. 2008) were used to amplify a 291-bp fragment with 35 cycles of 25 s at 94°C, 25 s at 58°C, and 25 s at 68°C, preceded by an initial denaturation step at 94°C for 1 min and followed by a final extension of 7 min at 68°C.
Statistical analysis
Two-way contingency table analyses and the calculation of Cramer's V statistic were performed with SPSS Statistic (version 17.0; SPSS, Chicago, IL). Cramer's V statistic provides a quantitative measure of the strength of the association between the two variables in a contingency table (information that cannot be obtained from the p-value).
Results
Overall, PCR amplicons of both DNA markers were obtained from 48 wild rabbits (17.20%) (Table 1). No amplification was observed in negative controls. After sequencing, only B. alsatica was found. The sequence obtained for the complete amplicon in the second amplification round was 663 bp in length for ITS and 790 bp for rpoB. Comparison of the gene sequence showed a similarity of 99–100% with the ITS of B. alsatica (GenBank accession code AF312506), and 100% for rpoB (GenBank accession code AF165987). In this study sample, 25/141 males (18.79%) and 23/138 females (18.4%) were positive. B. alsatica appeared to infect both wild rabbit haplotypes, 38/230 (16.52%) and 10/49 (20.4%) for A and B haplotypes, respectively, over the geographic range of each subspecies in Andalusia (Table 1).
Values within parentheses represent the number of Bartonella alsatica-positive rabbits.
The values of Cramer's V statistic range from 0 to 1, with 0 indicating no relation and 1 indicating a perfect association. Traditionally, values of 0.36–0.49 indicate a substantial relation, and values of 0.50 or more indicate a strong relation. The V statistic is a generalization of the more familiar phi statistic for non–two-by-two contingency tables, and for two-by-two tables, the V statistic is equal to the phi statistic (van Belle et al. 2004). No significant differences in infection were detected in relation with haplotype or sex inter- or intrahaplotype (Cramer's V statistic value is <0.04 in each case).
Discussion
Wild rabbit's ear pinnae have an important role in thermoregulation effected by vasomotion as ambient temperature changes (Williams and Moore 1989). The vascular bed of rabbit ear shows a central auricular branch and a compact network of periarterial capillaries in the tunica media, which regulate body temperature by countercurrent heat-exchange mechanism (Ninomiya 2000). The capillaries of rabbit's ear pinnae are filled with large quantities of blood, which permit the detection of hemotropic pathogens such as Bartonella spp. The low-invasive sampling procedure used requires minimal animal handling, enables access to tissue samples, and provides easy preservation during transport to laboratory.
This study is the first investigation about the prevalence and distribution of B. alsatica in Andalusia (Spain). In summary, this study has provided evidence that B. alsatica is largely widespread in natural populations of wild rabbit in Andalusia. These results could be extended to the entire natural range of wild rabbits, especially if we consider the infection in both classes of haplotypes defined for this mammal species.
Close contact with rabbit has been signaled as a main epidemiological factor of transmission of B. alsatica to humans (Jeanclaude et al. 2009). As Oryctolagus cuniculus is a very important game species in Spain and Portugal (Virgós et al. 2007), hunters' contact with infected animals can be extremely common, and so the possibility of infection with this Bartonella should be extremely high.
The epidemiology of many Bartonella species is not clear so far, especially their wild cycle, which allows the circulation and maintenance of the pathogens in nature (Chomel et al. 2009, Vayssier-Taussat et al. 2009). In the Iberian Peninsula, so far B. alsatica was only found in the flea Spilopsyllus cuniculi, collected on wild cats (Felis sylvestris) from Grenade province (Márquez et al. 2009).
Further research is necessary to determinate the potential vector(s) of this bacteria between wild rabbits in nature, which may help in the evaluation of the risk of human infection in actual epidemiological conditions, especially if we consider the role of Oryctolagus cuniculus as the main hunted mammal species in Iberian Peninsula ecosystems and one of the preferred indoor pet animal for children (Chomel et al. 2006).
The detection of B. alsatica in South Spain raises the possibility that many cases of bartonellosis considered by clinicians and health authorities as caused by B. henselae from cat scratch could, in fact, be due to other Bartonella, including those found in this study. Analyzing heart valve tissue biopsies in patients with endocarditis (Raoult et al. 2006, Jeanclaude et al. 2009) would be, together with continuing entomological surveys, the best way to increase our knowledge of emerging flea-borne bartonellosis in Spain and more generally in the West Mediterranean basin.
Footnotes
Acknowledgments
This study was partially supported by the Universidad de Jaén Research Program Grant 2009/10 and a grant from the Group BIO-294 (Immunogenetics) of Junta de Andalucía Research Program. The author thanks Dr. Ricardo Oya and Ana Jiménez, Research Services of Jaén University, for their technical assistance in sequencing. The author is grateful to two anonymous referees for their comments and suggestions.
Disclosure Statement
No competing financial interests exist.