Rickettsia felis is an obligate intracellular bacterium belonging to the spotted fever group, suspected to cause a murine typhus-like illness in humans 123. Following its first detection 4 in midgut cells of the cat flea Ctenocephalides felis, this organism has been later described as a new species 5.

Biological and genomic investigations have shown that R. felis displays unique features compared with other rickettsieae, in particular in its genomic 67 and laboratory cultivation 8. Molecular epidemiological investigations allowed detection of R. felis in fleas from different geographical areas of the world, further supporting the cosmopolitan distribution of this pathogen 1.

In Italy, R. felis has been detected in cat and dog fleas limitedly to an area of north-eastern Italy 9. The aim of the present study was to obtain further information on the occurrence and distribution of R. felis in fleas from different Italian regions, where the presence of this agent is still unknown.

From March 2008 to March 2009, fleas were collected from owned animals, colony cats, stray and kennelled dogs of four provinces of Veneto region (north-eastern Italy), two provinces of Campania region (south-western Italy) and of Bari province (Apulia region, south-eastern Italy). All fleas, preserved in isopropanol, were morphologically identified and sexed by using light microscopy and following an identification key 10. A range of 1–3 fleas per animal host were randomly chosen for molecular analyses.

DNA extraction was performed using a commercial kit (DNeasy^® Blood&Tissue Qiagen kit). A 401-bp fragment of the rickettsial gltA (citrate synthase) gene was PCR-amplified using CS-78 (forward) and CS-323 (reverse) primers 11.

Briefly, PCRs (50 μl) were performed in an Applied Biosystems Thermocycler (Gene Amp PCR System 9700), adding 5 μl of the DNA template to 31.7 μl of the molecular-grade water, 5 μl of buffer 10×, 3 μl of MgCl₂ 25 mM, 1 μl of dNTPs 10 mM, 2 μl of each primer (CS-78 and CS-323) 10 μM and 3 μl of AmpliTaq Gold 5 U. PCR cycling conditions were as follows: 1 initial cycle at 94°C for 10', 35 cycles of 15" at 95°C, 30" at 55°C, and 30" at 72°C, and 1 final step at 72°C for 7'.

The complete coding sequences were generated using the Big Dye Terminator v3.1 cycle sequencing kit (Applied Biosystem, Foster City, CA, USA). The products of the sequencing reactions were purified using PERFORMA DTR Ultra 96-Well kit (Edge BioSystems, Gaithersburg, MD, USA) and sequenced in a 16-capillary ABI PRISM 3130×l Genetic Analyzer (Applied Biosystem, Foster City, CA, USA). Sequences obtained were compared with those of characterised rickettsieae in Genbank by using BLAST analysis http://www.ncbi.nlm.nih.gov.

Prevalence differences in relation to host species (cats and dogs), provenance and flea sex were tested by χ² test or Fisher's Exact test using the SPSS statistical package (SPSS Inc., USA) for Windows, version 15.0.

Two species of fleas were identified, C. felis (80.3%) and Ctenocephalides canis (19.7%). Cats were infested only by C. felis and dogs by both species, with C. felis again as the predominant flea (75%).

The results of the molecular study are summarised in Table 1. Overall, 320 fleas (257 C. felis and 63 C. canis) collected from 117 animals (73 dogs and 44 cats) were tested. Thirty-eight (11.9%) C. felis fleas, 13 from cats (17.6%) and 25 from dogs (10.2%) were positive for R. felis yielding PCR bands of the expected size (Figure 1). No C. canis was positive. Sequenced amplicons displayed an overall 99% similarity to R. felis sequence available in the Genbank database (Rickettsia felis URRWXCal2; accession number CP000053.1).

Specifically, positive fleas were found from both dogs and cats in all four provinces of north-eastern Italy, from dogs in two provinces of south-western Italy (Naples and Caserta), whereas all fleas from Bari province (southeastern) were negative (Table 1).

Fleas from cats showed a tendency (Fisher's exact test; p = 0.068) to be more positive (17.6%) than fleas from dogs (10.2%). Male and female fleas showed a similar rate of infection (11.6% and 8.9%, respectively). Prevalence of R. felis among areas and within provinces of the same area was extremely variable, ranging from 0 to 35.3% (Table 1). In general and excluding fleas from Bari province (a single site of sampling) prevalence in north-eastern Italy (23.2%) was significantly higher than in south-western Italy (7.1%) (χ² test = 14.956; p < 0.01).

Our results indicate that R. felis is present in C. felis fleas from several geographical locations of Italy. The occurrence of R. felis-positive fleas from dogs and cats in north-eastern and south-western Italy is similar to data available in other European countries, i.e. Germany 12, France 13 and Spain 14.

In this study, R. felis was detected for the first time in south-western Italy, whilst the negative results from south-eastern Italy require further investigations, being fleas collected only in dogs housed in a sole municipal kennel. The higher prevalence of R. felis-infected fleas in north-eastern Italy compared with south-western areas is currently difficult to explain, a sampling bias can not be excluded and only further and more balanced sampling will confirm this trend.

In our study, C. felis was the only flea species infected as reported by several other authors, nevertheless C. canis was previously found as a R. felis carrier in other investigations, along with others flea species, ticks and mites (reviewed in 2). A possible explanation for the negativity of C. canis in our study could be that this flea was found only in dogs not showing mixed infection with C. felis and presumably not infected with R. felis. This hypothesis would explain also why C. felis fleas were found more infected when collected on cats rather than on dogs. However, the mechanism of R. felis transmission to vertebrates and to uninfected fleas in nature is still unknown, even though there is experimental evidence indicating that R. felis is maintained in cat fleas primarily by transovarian and transstadial transmission 15.

Besides vertical transmission, the first identification of the bacterium in salivary glands of C. felis 16, along with the evidence of seroconvertion and R. felis DNA detection in blood from cats exposed to infected fleas 17, strongly supports the potential for horizontal transmission among vectors and to vertebrate hosts. Among the other possible mammal hosts of C. felis, several studies 3 have shown that opossums seroconvert, and are usually heavily infested with the infected cat fleas. Thus, opossums could indirectly serve as a bridge for the transmission of R. felis to vertebrates.

Furthermore, a study conducted in Germany 12 in fleas collected from cats and dogs, provided evidence that Archeopsylla erinacei (the hedgehog flea) carried R. felis in all the specimens tested, compared to a low positivity found in C. felis (9%). The data above suggest that A. erinacei could have played a role in transmitting R. felis to humans in Germany and indicate the hedgehog as another candidate to be a potential reservoir of the infection.

Opossums and other vertebrate hosts may play a role in rickettsial horizontal transmission to other ectoparasites, and this may account for the occasional reports of R. felis infection in other flea species as well as ticks.

Clinical symptoms have not been reported in any animal carrying positive fleas. The current knowledge suggests that the only role of mammals infected by R. felis is likely to amplify the cycle by fleas feeding on their R. felis-infected blood. This hypothesis is supported by the fact that even if vertical transmission of R. felis persists in C. felis for at least 12 generations not feeding in a R. felis-infected host, over successive generations prevalence shows a natural decrease 18.

Actually, R. felis has been associated with diseased hosts only in humans. Infection by R. felis in humans has been reported in USA, Mexico, Brazil, France and Spain 12 both by serological and molecular evidence, but unfortunately the pathogen has never been successfully isolated from humans. The lack of a human isolate of R. felis does not permit the definition of this organism as a confirmed human pathogen.

To our knowledge no human cases of rickettsiosis due to R. felis have been diagnosed or suspected in Italy. However, clinical symptoms for R. felis infections are similar to those of other rickettsial diseases, which make difficult an aetiological diagnosis based only on the clinical presentations. Conversely, spotted fever rickettsiosis due to R. conorii in Italy is endemic in southern regions while some sporadic case of murine typhus due to R. typhi has been reported and a few cases were serologically attributed to R. helvetica 19. Recently, R. slovaca has been identified in ticks removed from humans showing a tick-borne lymphadenopathy in the Tuscany region 20.

This study confirmed the presence and diffusion of R. felis in cat and dog fleas (C. felis) from Italy, similar to other European countries 21. The results also suggest that R. felis should be considered in the human differential diagnosis of any spotted-like fevers in Italy, especially if the patient is known to have been exposed to flea bites. Nonetheless, the role of mammals, mainly dogs and cats, in the epidemiology of this flea-borne infection needs further confirmatory evidence. To unravel the close relationship between R. felis and C. felis more studies are needed in rickettsial infection dynamics in the flea vector, which likely will further clarify the ecology and epidemiology of R. felis transmission in nature.

The authors declare that they have no competing interests.

GC, FM and GM conceived and designed the experiments; EP and GM carried out the molecular genetic analyses; GM, CF, LR, GO and DO have made substantial contributions to acquisition of fleas and related data; GC and FM analysed and interpreted the data and wrote the paper; GO, LR and especially DO revised the article critically for important intellectual content.