Rickettsia felis in Fleas, Germany

Among 310 fleas collected from dogs and cats in Germany, Rickettsia felis was detected in all specimens (34) of Archaeopsylla erinacei (hedgehog flea) and in 9% (24/226) of Ctenocephalides felis felis (cat flea). R. helvetica was detected in 1 Ceratophyllus gallinae (hen flea).

R ickettsia felis, the causative agent of the fl ea-borne spotted fever rickettsiosis, is pathogenic for humans (1)(2)(3)(4). Since the fi rst detection of R. felis from midgut epithelial cells of the cat fl ea, Ctenocephalides felis felis, in 1990 (5), interest in the role of this fl ea species as its main vector has increased. R. felis has been found in cat fl eas on all continents (6). Because R. felis is not lethal for cat fl eas and is transmitted transovarially by these fl eas (4), C. felis could be a vector and a reservoir of this pathogen. For these reasons, the cat fl ea was considered the only fl ea species with a major role in the epidemiology of fl ea-borne spotted fever rickettsiosis. However, R. felis has been reported in other fl ea species (4,(6)(7)(8), and fl ea-borne spotted fever rickettsiosis is now considered an emerging human infectious disease. We analyzed the presence of R. felis in different fl ea species collected from naturally infested cats and dogs in different locations in Germany.

The Study
A total of 310 fl eas were collected from 49 dogs and 54 cats in 11 widely distributed locations in Germany (Berlin, Munich, Brandenburg, Leipzig, Chemnitz, Rostock/Laage, Bremen, Osnabrück, Münster, Freising, and Schongau) (Figure) in 2007. Specimens collected were recorded and kept at -20°C. Samples were shipped on dry ice to our laboratory, and species identifi cation was performed by using light microscopy and following the determination key of Hopkins and Rothschild (9). Because of infestation variations (1-150 fl eas per animal), 3 fl eas per animal host were chosen randomly for species differentiation.
Fleas were homogenized individually in 80 μL of phosphate-buffered saline with a RETSCH Tissue Lyser Mixer Mill 300 (QIAGEN, Hilden, Germany) by using 5-mm steel beads. A 100-μL volume of ATL buffer and 20 μL of proteinase K (QIAGEN) were added, and homogenates were incubated at 56°C in an Eppendorf Thermomixer (Eppendorf, Hamburg, Germany) until tissues were completely lysed. DNA was extracted from each fl ea by using a QIAamp DNA Mini Kit (QIAGEN) according to the manufacturer's instructions (tissue protocol) and stored at -20°C until used.
Five species of fl eas were identifi ed in the study. The most prevalent species was C. felis (93% of fl eas in cats and 78% in dogs) ( Table 1). Archaeopsylla erinacei (hedgehog fl ea), was the second most abundant species, with 26 specimens collected from dogs and 8 from cats. A few specimens of Ctenocephalides canis (dog fl ea), Pulex irritans (human fl ea), and Ceratophyllus gallinae (hen fl ea) were also identifi ed (Table 1). Eight dogs had mixed populations of fl eas; 5 had C. felis and A. erinacei, 2 had C. felis and C. gallinae, and 1 had C. felis and C. canis. Mixed populations of fl eas were also detected in 3 cats; 2 were infested with C. felis and A. erinacei, and 1 with P. irritans and C. felis.
Thirty-six (25%) of 146 fl eas collected from dogs and 24 (15%) of 164 fl eas collected from cats were positive for the gltA gene. Positive fl eas were found in 6 of 11 sampled locations. Proportions of infected fl eas collected from dogs ranged from 25% (Berlin) to 56% (Münster), and propor-tions of infected fl eas collected from cats ranged from 10% (Freising) to 100% (Münster) ( Table 2).
Of 60 fl eas positive for the gltA gene (for dogs and cats), only 2 were negative for the ompA and ompB genes. Sequencing analysis of the gltA gene for these 2 samples showed that 1 sequence (from C. gallinae) was 99% homologous with part of the Rickettsia helvetica gltA gene (AM418450.1) from an Ixodes persulcatus tick isolated in Russia; the other sequence (from C. gallinae) was 94% homologous with the Rickettsia sp. citrate synthase gene (U76908.1). Thus, we report R. helvetica in C. gallinae ticks.
Of the other 58 gltA-positive samples, 2 were positive for the ompA gene in the fi rst round; 56 fl eas were positive for the ompB gene. The 2 ompA-positive samples were   (8).

Conclusions
Our study confi rms that C. felis remains the most common fl ea species infesting cats and dogs in Germany. Nevertheless, only 24 of 266 cat fl eas collected were infected with R. felis. Infected cat fl eas were found only in 4 of 11 studied sites, in contrast with a recent study in France, where R. felis-infected C. felis were present in all locations studied (14). In the 4 positive sites in Germany, 3 had positive A. erinacei specimens and 1 had positive C. gallinae (Table 2). In the other sites where no positive fl eas where found, only C. felis was present either alone or in association with P. irritans and C. canis (Table 2).
Although C. felis seems to be the main vector of R. felis, our fi ndings indicate that A. erinacei may be a vector for human fl ea-borne rickettsiosis in Germany. Because hedgehogs may act as a reservoir of pathogens (15), further studies will be conducted to investigate the role of hedgehogs and hedgehog fl eas in maintenance and transmission of R. felis in Germany. Dr Gilles is National Institutes of Health project leader at the University of Kentucky in Lexington. His research interests focus on medical entomology, arthropod-borne diseases, vector biology, ecology, and vector control.