Bartonella henselae in Ixodes ricinus Ticks (Acari: Ixodida) Removed from Humans, Belluno Province, Italy

The potential role of ticks as vectors of Bartonella species has recently been suggested. In this study, we investigated the presence of Bartonella species in 271 ticks removed from humans in Belluno Province, Italy. By using primers derived from the 60-kDa heat shock protein gene sequences, Bartonella DNA was amplified and sequenced from four Ixodes ricinus ticks (1.48%). To confirm this finding, we performed amplification and partial sequencing of the pap31 protein and the cell division protein FtsZ encoding genes. This process allowed us to definitively identify B. henselae (genotype Houston-1) DNA in the four ticks. Detection of B. henselae in these ticks might represent a highly sensitive form of xenodiagnosis. B. henselae is the first human-infecting Bartonella identified from Ixodes ricinus, a common European tick and the vector of various tickborne pathogens. The role of ticks in the transmission of bartonellosis should be further investigated.

The potential role of ticks as vectors of Bartonella species has recently been suggested. In this study, we investigated the presence of Bartonella species in 271 ticks removed from humans in Belluno Province, Italy. By using primers derived from the 60-kDa heat shock protein gene sequences, Bartonella DNA was amplified and sequenced from four Ixodes ricinus ticks (1.48%). To confirm this finding, we performed amplification and partial sequencing of the pap31 protein and the cell division protein ftsZ encoding genes. This process allowed us to definitively identify B. henselae (genotype Houston-1) DNA in the four ticks. Detection of B. henselae in these ticks might represent a highly sensitive form of xenodiagnosis. B. henselae is the first human-infecting Bartonella identified from Ixodes ricinus, a common European tick and the vector of various tickborne pathogens. The role of ticks in the transmission of bartonellosis should be further investigated.
B artonella species are facultative intracellular bacteria associated with a number of emerging anthropozoonoses. They have been detected in or isolated from diverse vertebrate hosts, including humans (1-3), various intradomicillary mammals (4)(5)(6)(7), and a wide range of wild animals (8,9), which serve as natural vertebrate hosts. Various hematophagous arthropods have been implicated in the ecoepidemiology of Bartonella species. B. bacilliformis, the etiologic agent of Carrión disease, is transmitted by the sand fly (Lutzomyia verrucarum) in the Andes Mountains in Peru, Columbia, and Ecuador (10). B. quintana, the agent of trench fever and bacillary angiomatosis, is found worldwide and is transmitted by the human body louse (Pediculus humanus) (11).
B. henselae is another cosmopolitan emerging human pathogen. This agent was first reported in 1990 in association with bacillary angiomatosis (12). The organism was later isolated from the blood of a febrile HIV-positive patient and subsequently described as a new species in 1992 (1). B. henselae is now recognized as the causative agent of cat-scratch disease (1), bacillary angiomatosis, peliosis hepatitis, oculoglandular syndrome, and endocarditis (13,14). B. henselae is associated with cats, which serve as its reservoir (13,15); the cat flea (Ctenocephalides felis) was demonstrated to be a vector (16). Other Bartonella-flea associations are apparent: for example, 61% of rat fleas (Xenopsylla cheopis) were found infected with bartonellae, including a known human pathogen, B. elizabethae (7).
Ticks are vectors of more diverse microorganisms than any other arthropod vector (23). The sheep tick (Ixodes ricinus) is the most common hard tick species in western Europe and has been established as the vector of tick-borne encephalitis virus, Babesia sp., Borrelia burgdorferi, Rickettsia helvetica, and the agent of granulocytic ehrlichiosis, Anaplasma phagocytophila, (24). I. ricinus feeds on a large number of vertebrate hosts. The immature stages of I. ricinus are found mainly on small-size vertebrates and can readily feed on humans. Ticks have been suspected to transmit Bartonella (25). However, evidence of Bartonella infection in ticks has only recently been reported (26,27). Although these observations suggest the possibility of Bartonella transmission by ticks, more precise identification of these tick-infecting agents is required to establish their zoonotic potential.

Tick Collection and Identification
During 2000-2001, a total of 271 ticks were removed from asymptomatic persons who visited first aid departments in Belluno Province, Italy, for assistance with tick bites. Ticks were removed with tweezers by grasping their mouthpart and pulling straight out from the skin. The tick-bite site was disinfected, and individual ticks were placed in sterile tubes and kept frozen at -70°C for further study. The ticks were stored on ice during the identification procedure, which was done on the basis of their morphologic features by using standard taxonomic keys.

Tick DNA Extraction
All ticks were disinfected by immersion into a 70% ethanol solution for 5 min, rinsed with sterile water, and dried in a sterile filter paper. Ticks were then subjected to DNA extraction by using the QiaAmp tissue kit procedure (QIAGEN GmbH, Hilden, Germany). DNA was extracted from ticks according to the manufacturer's protocol. To serve as a negative control, DNA of lice from a laboratory colony that had been fed on an uninfected rabbit was extracted, along with tick DNA, to serve as control. DNA was eluted in a final volume of 200 µL and stored at 4°C until studied further.

PCR Screening of Ticks for the Presence of Bartonella
Tick DNA was screened by PCR amplification of the heatshock protein-encoding gene (groEL) sequences for the presence of Bartonella. Primers HSPF1d and BbHS1630.n were used as described (22) and are listed in the Table. Subtyping of Detected Bartonella with pap31 and ftsZ Partial Sequences Amplification Primers used for amplification sequencing of each gene are listed in the Table. PCR reactions were performed in a Peltier model PTC-200 thermal cycler (MJ Research, Inc., Watertown, MA). PCR was carried out in a total volume of 50 µL, consisting of 10 pmol of each primer, 0.5 U of ELONGase mix enzyme (GibcoBRL, Cergy Pontoise, France), 20 mM concentration of each deoxynucleoside phosphate, and 1.8 mM of MgCl2. Two negative controls were included in the reaction: DNA from uninfected lice, and the master mix with sterile water instead of the DNA template. DNA from a culture of B. elizabethae was used as the positive control. The following amplification program was used: a first denaturation step at 94°C for 4 min was followed by 44 cycles of denaturation at 94°C for 30 s, annealing at temperatures corresponding to each gene (53°C for groEL and pap31 genes and 55°C for ftsZ) for 30 s, and a hybridization step at 68°C for 1 min. The amplifi-cation reaction was terminated with a further extension step at 68°C for 10 min. PCR products were visualized under UV illumination after electrophoresis migration on a 1% gel agarose stained with ethidium bromide.
Sequencing PCR products were purified by the QIAquick PCR purification kits (QIAGEN GmbH) as recommended by the manufacturer. Primers used for the sequencing of each gene are listed in the Table. PCR products were sequenced in both directions by using the d-Rhodamine Terminator Cycle Sequencing Ready Reaction kit (PerkinElmer, Inc., Coignières, France) according to the manufacturer's recommendations. Sequencing products were resolved in an Applied Biosystem automatic sequencer model 3100 (PerkinElmer).

Sequence Analysis
Nucleotide sequences were edited with the Autoassembler (version 1.4; Perkin Elmer) package. Multiple alignment with other Bartonella sp. sequences available from GenBank was carried out by using the Clustal W program (28).

PCR Screening of Ticks for Bartonella
By using primers HSPF1d and BbHS1630.n, a single band of PCR product of approximately 1,490 bp was amplified and sequenced in four I. ricinus ticks (two females and two nymphs) (1.48%). No amplification product was yielded from

Subtyping of Bartonella henselae
Amplification of the pap31 and ftsZ partial sequences yielded 257-bp and 885-bp, fragments, respectively. Sequences of these products had 100% identity with those of B. henselae Houston-1 (GenBank accession nos. AF001274 and AF061746, respectively).

Discussion
Recently, vector biologists and epidemiologists have suggested that ticks may have a role in Bartonella transmission (29). In 1996 Kruszewska et al. reported the preliminary finding of a Bartonella strain in I. ricinus ticks from a park in Walz, Poland (26). Unfortunately, the strain has not been further characterized. In a study conducted in the Netherlands, the 16S rRNA gene sequences of an unspecified Bartonella were amplified in >70% of I. ricinus ticks removed from roe deer (27). Such a high prevalence of Bartonella in ticks is surprising and may be because ticks were collected while they were feeding on bacteriemic hosts (30,31). According to Schouls et al., none of the Bartonella organisms detected was a known human pathogen (27). More recently, different Bartonella sp., including B. quintana, B. henselae, Bartonella strain cattle-1, B. washoensis, and B. vinsonii subsp. berkhoffii, have been detected in 19.2% of I. pacificus ticks collected in California by amplification and sequencing of a fragment of the gltA gene (32).
In this study, we report the detection of B. henselae in four I. ricinus ticks (1.4%) removed from persons in Italy. Because the primers used in the screening PCR generate rather large PCR fragments (1,490 bp), this prevalence could be expected to be greater (usually, the longer the PCR product, the lower the sensitivity). DNA from positive samples was further characterized by using the groEL, the pap31, and the ftsZ genes to establish their relationship with known Bartonella sp. and subsp. On the basis of the 16S rRNA genes and immunogenic characteristics, Drancourt et al. (33) suggested the presence of two variants of B. henselae. Ribosomal genes such as the 16S rRNA genes are, however, highly conserved within bacteria and can pose the risk of unspecific amplification. Protein-coding genes exhibit a higher degree of sequence variation and thus can be targeted as tools for differentiating strains of the same species. Although the two genogroups of B. henselae, Marseille and Houston-1, are closely related, and the respective pathogenicity spectrum of the two serotypes has not been established, the serotypes could be differentiated on the basis of sequences of the groEL, C-terminal region of the ftsz gene and the pap31 gene (21,22).
In northwestern Italy, about 89% of the ticks found to parasitize people were I. ricinus (34). In our study, four different species of ticks were recorded from humans, and I. ricinus was recorded most frequently (98.9%). All the active life stages of I. ricinus were represented.
Experimental studies and epidemiologic observations have suggested that ticks may play a role in the transmission of Bartonella sp. Dermacentor andersoni was proven to be a competent vector of B. bacilliformis in the experimental infection of nonhuman primates many years ago (35). B. vinsonii subsp. berkhoffii infection was correlated with heavy tick infestation of dogs (36). The cat flea (Ctenocephalides felis) is the main arthropod vector of B. henselae with cats serving as the main vertebrate reservoirs. Although finding B. henselae in ticks might suggest another possible reservoir, I. ricinus-like ticks have a very broad host range and are known to infest cats. In our study, none of the persons from whom the positive ticks were collected exhibited symptoms associated with B. henselae infection. However, serum specimens from these patients have not been tested. Because bacteremia levels in cat-scratch disease patients have never been consistently demonstrated (37), the tick was unlikely to have acquired B. henselae from feeding on patients with asymptomatic cat-scratch disease. Nevertheless, a number of wild animals, including the preferred hosts of both adults and immature stages of I. ricinus (38) have been found to be infected with Bartonella sp. (8,39). Questing ticks have been found infected with Bartonella, including B. henselae (32). Furthermore, ticks were suspected of being vectors of B. henselae in an epidemiologic study conducted by Lucey et al. in 1992 (25). These authors reported B. henselae bacteremia levels in patients who recalled a tick bite but had no history of contact with cats. Ticks were also reported as possible source of infection in some human cases of concurrent infection of the central nervous system by Borrelia burgdorferi and Bartonella henselae (40). The evidence that ticks may serve as Bartonella vectors appears to be rapidly accumulating.
In conclusion, we have confirmed that ticks feeding on humans were infected with the agent of cat-scratch disease, B. henselae (Houston-1). The source of infection of the ticks was not determined. No case of transmission to humans was observed. However, our findings suggest that the ticks were naturally infected. These results support the argument that ticks are involved in the transmission of Bartonella organisms and represent a potential source of infection for persons exposed to tick bites. Therefore, we encourage further investigation of ticks as vectors of human pathogenic Bartonella strains.
RESEARCH ical entomology at the Faculté de Médecine, Marseille, France. His work focuses on the relationship between arthropods and bacteria.