Bartonella are emerging and re-emerging pathogens affecting humans and a wide variety of animals, including rodents. Various ectoparasites have been shown to be competent vectors for Bartonella transmission between mammalian hosts. The body louse (Pediculus humanus corporis) and sandflies of the genus Lutzomyia were shown to be the respective vectors of B. quintana, the agent of trench fever, and B. bacilliformis, the agent of Carrion disease at the beginning of the 20th century (Townsend 1914; Kostrzewski 1949). Fifteen years ago, the cat flea (Ctenocephalides felis) was shown to be the main vector of B. henselae, the agent of cat scratch disease (Chomel et al. 1996). Many rodent fleas have been found to carry Bartonella DNA (Tsai et al. 2011), but very limited data on vector competency by these rodents fleas are available (Bown et al. 2004). Transmission of the bacterial pathogen from cat fleas or from body lice has been shown to be mainly through contaminated faeces (Foil et al. 1998; Raoult & Roux 1999; Finkelstein et al. 2002). In lice, B. quintana multiplies in the gut lumen of naturally or intra-rectally infected lice without interfering with viability and is excreted in their faeces (Fournier et al. 2001). Furthermore, it was shown that B. quintana was present in the gut and faeces but not in the eggs of lice fed artificially on infected rabbits (Fournier et al. 2001). Bartonella schoenbuchensis, which commonly causes bacteremia in roe-deer (Capreolus capreolus) in Europe, was isolated from the deer ked Lipoptena cervi and was shown to localize to the midgut of this blood-sucking arthropod (Dehio et al., 2004). The vertical transmission of Bartonella species has not been fully addressed, at least for fleas.
In their study, Morick et al. placed Bartonella-free laboratory-reared X. ramesis fleas on naturally Bartonella-infected and Bartonella-negative control desert rodents (jirds and gerbils) for 7 days, 12–14 h per day. The fleas that were placed on the Bartonella-positive rodents acquired four different Bartonella genotypes. Eggs laid by fleas and larvae that developed from both rodent groups were collected daily for 7 days and molecularly screened for Bartonella. All eggs and larvae from both groups were found to be negative for Bartonella DNA. Two of five gut voids regurgitated by Bartonella-positive fleas contained Bartonella DNA. The naturally infected rodents remained persistently infected with Bartonella for at least 89 days suggesting their capability to serve as competent reservoirs for Bartonella species. These authors findings indicate that X. ramesis fleas can acquire several Bartonella strains from wild rodents but cannot transmit Bartonella transovarially, as shown for B. quintana and lice. Similarly, in experimental infection of Rhipicephalus sanguineus ticks with B. vinsonii subsp. berkhoffii, no DNA was detected from eggs laid from female ticks fed B. vinsonii subsp. berkhoffii-tainted blood (Billeter et al., unpublished data). However, in another study, among nine pools of eggs laid by females fed on B. henselae-infected blood, three showed amplification of the expected Bartonella spp.–specific 337-bp DNA fragment (Cottéet al. 2008). No amplification was obtained with larvae from eggs positive or negative for Bartonella spp. DNA.
In conclusion, several Bartonella species, including B. quintana, B. henselae and B. schoenbuchensis, appear to have adopted a stercorarian transmission from the vector to its mammalian host and do not appear to rely on vertical transmission within the vector. However, such observations will need to be further documented for other Bartonella species and other natural vectors.