Bartonella clarridgeiae in Fleas, Tahiti, French Polynesia

To the Editor: Bartonella species are small, gram-negative, fastidious, and hemotropic emerging pathogens that cause various human diseases and circulate between a large variety of mammalian and arthropod vectors. More than 30 Bartonella species have been isolated from humans as well as from wild and domestic animals worldwide (1). B. clarridgeiae was suggested to be a minor causative agent of cat-scratch disease (CSD) in humans, however, this suggestion remains controversial. Usually, the agent of CSD is B. henselae and its principal reservoir is domestic cats (Felis catus) (1,2). The principal vector of these 2 species is the cat flea (Ctenocephalides felis) (3,4). We report Bartonella species in fleas collected from cats and dogs in Tahiti, French Polynesia. 
 
In October 2009, fleas were collected from 1 cat and 9 dogs in Papeete, capital of Tahiti Island, French Polynesia. Fleas collected were kept in 70% ethanol and sent to the military veterinary service in Marseille, France; these fleas were later sent to Unite de Recherche sur les Maladies Infectieuses et Tropicales Emergentes in Marseilles. The fleas were identified phenotypically by using current taxonomic criteria. DNA from fleas and negative control DNA from noninfected laboratory lice were extracted by using a QIAamp Tissue Kit (QIAGEN, Hilden, Germany), as described (3). 
 
Flea samples were tested for Bartonella spp. DNA by using the 7900 HT Fast Quantitative Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and primers and Taqman probes specific for the 16S–23S rRNA gene intergenic spacer region as described (5). Fleas were considered positive when cycle threshold was <30. All positive fleas at screening were confirmed by using standard Bartonella PCR and sequencing of partial internal transcribed spacer gene fragments by using primers URBarto1 and URBarto2, as described (3). B. elizabethae DNA was used as positive control. DNA sequencing reagents were obtained with BigDye Terminator Cycle Sequencing Ready Reaction Kit (ABI PRISM; Applied Biosystems). The sequences were assembled in Sequencher 4.2 (GeneCodes 2003; www.genecodes.com) and were compared with Bartonella sequences available in GenBank. 
 
Overall, 81 fleas were collected from 1 cat (13 fleas) and 9 dogs (68 fleas). All 81 fleas collected were morphologically identified as C. felis. 
 
Sample fleas were collected from animals visiting a veterinary clinic for neutering or vaccinations. The overall rate of Bartonella-positive fleas by molecular screening with real-time PCR was 7.4% (6/81): 6 fleas from the cat (6/13) and none from a dog (0/68). These positive samples were confirmed after intergenic spacer PCR amplification and sequencing with sequences at 100% identity with B. clarridgeiae (GenBank accession no. {"type":"entrez-nucleotide","attrs":{"text":"EU589237","term_id":"187251975","term_text":"EU589237"}}EU589237). 
 
B. clarridgeiae was first isolated from the pet cat of an HIV-positive patient in the United States (6). However, B. clarridgeiae has never been isolated or detected by molecular methods in humans, and thus its implication as a human pathogen remains controversial. The presence of B. clarridgeiae antibodies has been reported in a suspected case of CSD and in a patient with a chest-wall abscess (4). However, B. clarridgeiae has been detected on fleas from various continents, including Europe, Asia, North America (1), Africa; New Zealand, and recently from New Caledonia (7). 
 
In France, several studies have reported the molecular detection of B. clarridgeiae in the blood of a cat or in cat fleas (C. felis), indicating the potential role of fleas as vectors of this organism (8,9). Prevalence of this bacterium in cat fleas may vary and be as high as 67.9% in cat fleas from France (3). Moreover, DNA of B. henselae and B. clarridgeiae has been reported from cat fleas from New Zealand (10). Similarly, co-infection with B. clarridgeiae and B. henselae has been reported in domestic cats from Europe and Asia (1). In our study, all Bartonella spp.–positive fleas harbored B. clarridgeiae only; all were obtained from cats and none from dogs, similar to findings reported from New Zealand (10), although B. clarridgeiae has been reported from a flea on a dog in Taiwan (2). 
 
Papeete, the capital of Tahiti, is located in the South Pacific Ocean, and remains one of the most visited areas by tourists from all over the world. There are many stray cats and dogs in Tahiti that may be infected with Bartonella species and thus serve as a reservoir for these pathogens. Our result confirms the presence of B. clarridgeiae in Tahiti and is a warning of the presence of flea-borne bartonellosis and the potential risk of B. clarridgeiae or other flea-borne diseases for humans exposed to cat fleas.

To the Editor: Bartonella species are small, gram-negative, fastidious, and hemotropic emerging pathogens that cause various human diseases and circulate between a large variety of mammalian and arthropod vectors. More than 30 Bartonella species have been isolated from humans as well as from wild and domestic animals worldwide (1). B. clarridgeiae was suggested to be a minor causative agent of cat-scratch disease (CSD) in humans, however, this suggestion remains controversial. Usually, the agent of CSD is B. henselae and its principal reservoir is domestic cats (Felis catus) (1,2). The principal vector of these 2 species is the cat fl ea (Ctenocephalides felis) (3,4). LETTERS We report Bartonella species in fl eas collected from cats and dogs in Tahiti, French Polynesia.
In October 2009, fl eas were collected from 1 cat and 9 dogs in Papeete, capital of Tahiti Island, French Polynesia. Fleas collected were kept in 70% ethanol and sent to the military veterinary service in Marseille, France; these fl eas were later sent to Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes in Marseilles. The fl eas were identifi ed phenotypically by using current taxonomic criteria. DNA from fl eas and negative control DNA from noninfected laboratory lice were extracted by using a QIAamp Tissue Kit (QIAGEN, Hilden, Germany), as described (3).
Flea samples were tested for Bartonella spp. DNA by using the 7900 HT Fast Quantitative Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and primers and Taqman probes specifi c for the 16S-23S rRNA gene intergenic spacer region as described (5). Fleas were considered positive when cycle threshold was <30. All positive fl eas at screening were confi rmed by using standard Bartonella PCR and sequencing of partial internal transcribed spacer gene fragments by using primers URBarto1 and URBarto2, as described (3) Overall, 81 fl eas were collected from 1 cat (13 fl eas) and 9 dogs (68 fl eas). All 81 fl eas collected were morphologically identifi ed as C. felis.
Sample fl eas were collected from animals visiting a veterinary clinic for neutering or vaccinations. The overall rate of Bartonella-positive fl eas by molecular screening with real-time PCR was 7.4% (6/81): 6 fl eas from the cat (6/13) and none from a dog (0/68). These positive samples were confi rmed after intergenic spacer PCR amplifi cation and sequencing with sequences at 100% identity with B. clarridgeiae (GenBank accession no. EU589237).
B. clarridgeiae was fi rst isolated from the pet cat of an HIV-positive patient in the United States (6). However, B. clarridgeiae has never been isolated or detected by molecular methods in humans, and thus its implication as a human pathogen remains controversial. The presence of B. clarridgeiae antibodies has been reported in a suspected case of CSD and in a patient with a chest-wall abscess (4). However, B. clarridgeiae has been detected on fl eas from various continents, including Europe, Asia, North America (1), Africa; New Zealand, and recently from New Caledonia (7).
In France, several studies have reported the molecular detection of B. clarridgeiae in the blood of a cat or in cat fl eas (C. felis), indicating the potential role of fl eas as vectors of this organism (8,9). Prevalence of this bacterium in cat fl eas may vary and be as high as 67.9% in cat fl eas from France (3). Moreover, DNA of B. henselae and B. clarridgeiae has been reported from cat fl eas from New Zealand (10). Similarly, co-infection with B. clarridgeiae and B. henselae has been reported in domestic cats from Europe and Asia (1). In our study, all Bartonella spp.-positive fl eas harbored B. clarridgeiae only; all were obtained from cats and none from dogs, similar to fi ndings reported from New Zealand (10), although B. clarridgeiae has been reported from a fl ea on a dog in Taiwan (2).
Papeete, the capital of Tahiti, is located in the South Pacifi c Ocean, and remains one of the most visited areas by tourists from all over the world. There are many stray cats and dogs in Tahiti that may be infected with Bartonella species and thus serve as a reservoir for these pathogens. Our result confi rms the presence of B. clarridgeiae in Tahiti and is a warning of the presence of fl eaborne bartonellosis and the potential risk of B. clarridgeiae or other fl eaborne diseases for humans exposed to cat fl eas.