Infection and re-infection of domestic cats with various Bartonella species or types: B. henselae type I is protective against heterologous challenge with B. henselae type II

Four Bartonella species have been isolated from domestic cats, of which two serotypes/genotypes of Bartonella henselae and possibly B. clarridgeiae are human pathogens, causing cat scratch disease (CSD).Our objectives were to evaluate infection and potential cross-protection during re-infection in domestic cats with various Bartonella species or types.Thirty-six cats were primarily inoculated with B. henselae type I (n=16), B. henselae type II (n=10), B. clarridgeiae (n=6) or B. koehlerae (n=4). They were challenged with B. henselae type I (n=15), B. henselae type II (n=13) or B. clarridgeiae (n=8).All 36 cats became bacteremic (1.25x10(2)-1.44x10(6)CFU/ml) and bacteremia lasted from 37 to 582 days. Duration of bacteremia for cats inoculated with B. henselae type I was shorter than for cats inoculated with either B. henselae type II (P=0.025) or B. clarridgeiae (P=0.011).After challenge, 26 cats became bacteremic. Among the nine cats primarily inoculated with B. henselae type I and challenged with B. henselae type II, six cats stayed abacteremic. The three bacteremic cats had a transient low-level bacteremia. No bacteremia was observed in three cats primarily inoculated with B. henselae type I and challenged with another strain of B. henselae type I. Bacteremia levels in the 26 cats were significantly lower than for primary inoculation (P=0.022) and its duration was shorter (P=0.012). Among the eight cats challenged with B. clarridgeiae, duration of bacteremia in the four cats primarily inoculated with B. henselae type I was shorter than in the four cats primarily inoculated with B. henselae type II (P=0.01). Bartonella clarridgeiae inoculated cats were more likely to have relapses for both primary and secondary infections.This is the first demonstration of cross-protection, evidenced by absence of bacteremia, in cats primarily infected with B. henselae type I and challenged with B. henselae type II, whereas no cross-protection was previously shown for cats primarily infected with B. henselae type II and challenged with B. henselae type I. Such results are of major importance for future feline Bartonella vaccine development.     

Evaluation and use of a nested polymerase chain reaction assay in cats experimentally infected with Bartonella henselae genotype I and Bartonella henselae genotype II.

Cats have been shown to be infected with Bartonella henselae genotype I, B. henselae genotype II, and B. clarridgeiae. Feline bartonellosis infections and the strains involved in these infections are important in both veterinary and human medicine. Nucleic acid amplification methods such as polymerase chain reaction (PCR) are being used in both research and diagnostics as tools for understanding many infectious diseases. Bartonella bacteremia in cats is detected by blood culture; however, because of the limitations of culture (delayed turnaround time and sensitivity limits), PCR may be a more efficient method for identifying infected cats. Three distinct PCR assays that could differentiate among B. henselae genotype I, B. henselae genotype II. and B. clarridgeiae were developed and used to detect as few as 3.2 organisms. Fourteen cats experimentally infected with B. henselae genotype I and B. henselae genotype II were followed by bacterial culture and PCR through the course of infection, including periods of primary and relapsing bacteremia. The PCR assay was positive in 11 of the 14 cats for periods of 1-9 weeks after culture became negative. Of the 223 blood specimens that were culture negative, the PCR assay was positive in 38 (17%) of the specimens. Two of the 14 cats developed relapsing bacteremia. The 2 B. henselae genotypes were amplified in the cats and the bacteremic phase of these infections as determined by PCR lasted for a longer period than previously determined by culture. Using laboratory assays such as PCR to understand the strains involved in feline bartonellosis and the course of the infection is important in the understanding of these zoonotic agents.     

Prevalence of Bartonella infection in domestic cats in Denmark

Whole blood and serum from 93 cats (44 pets and 49 shelter/stray cats) from Denmark were tested for the presence of feline Bartonella species by culture and for the presence of Bartonella antibodies by serology. Bartonella henselae was isolated from 21 (22.6%) cats. Bacteremia prevalence was not statistically different between shelter/stray cats (13/49, 26.5%) and pet cats (8/44, 18.2%), but varied widely by geographical origin of the cats, even after stratification for cat origin or age (p < 0.001). All isolates but one were B. henselae type II. The only cat bacteremic with B. henselae type I was not co-infected with B. henselae type II. None of the cats was harboring either B. clarridgeiae or B. koehlerae. Almost half (42/92, 45.6%) of the cats were seropositive for B. henselae and antibody prevalence was similar in shelter/stray cats (23/49, 46.9%) and pet cats (19/43, 44.2%). This is the first report of isolation of B. henselae from domestic cats in Denmark. This study also indicates that domestic cats, including pet cats, constitute a large Bartonella reservoir in Denmark.     

Immune response of neonatal specific pathogen-free cats to experimental infection with Bartonella henselae

The purpose of this study was to determine whether neonatal cats develop and maintain a persistent bacteremia for longer than do adult cats with a normal mature immune system, and whether neonatal cats are susceptible to infection with Bartonella henselae by oral inoculation. Neonatal specific pathogen-free (SPF) cats were inoculated with B. henselae intradermally (n = 4) or orally (n = 5) or with 0.9% NaCl (n = 2). Blood was collected periodically through 16 weeks post-inoculation (PI) for serology, bacteriology and complete blood count. Cats inoculated orally or intradermally at 3-5 days of age were bacteremic through 12-16 weeks PI, similar to what is documented for adult cats inoculated intradermally or intravenously. One cat inoculated at age 2 weeks was bacteremic through 10 weeks PI; the other was not bacteremic. Intradermally inoculated neonatal cats produced serum IgG antibodies to B. henselae but orally inoculated neonatal cats did not. Infected cats with and without serum IgG antibodies to B. henselae became blood-culture negative simultaneously, suggesting that IgG is not required to clear bacteremia.     

Prevalence of Bartonella henselae, Bartonella clarridgeiae and the 16S rRNA gene types of Bartonella henselae among pet cats in Japan

The authors investigated bacteriologically the prevalence of Bartonella infection among 690 pet cats derived from 10 private animal hospitals in six cities (Sapporo, Hokkaido Prefecture; Sendai, Miyagi Prefecture; Joetsu, Niigata Prefecture; Fujisawa, Kanagawa Prefecutre; Kyoto, Kyoto Prefecture; Sanda, Hyogo Prefecutre) and 4 counties (Mishima, Osaka Prefecture; Hikawa, Shimane Prefecture; Aira, Kagoshima Prefecture; Shimajiri, Okinawa Prefecture) located from the north to the south of Japan. Bartonella species were isolated from 7.2% (50/690) of all the cats examined. No Bartonella species were isolated from the cats in Sapporo or Sendai. The isolation rate varied from 2% in Joetsu and Sanda to 20% in Shimajiri. Bartonella clarridgeiae was isolated from two of 50 cats in Kyoto, three of 50 in Mishima and one of 50 in Shimajiri, but not in cats from the other cities or counties. Though the cats of Joetsu, Fujisawa, Kyoto, Sanda, Aira and Shimajiri were infected with either B. henselae or B. clarridgeiae, one of eight infected cats in Mishima was harboring both Bartonella species. Type I of 16S rRNA gene was the predominant type among the isolates of B. henselae, but only one isolate derived from Shimajiri was found to be of type II. Prevalence of B. clarridgeiae and the 16S rRNA gene type of B. henselae among cats in Japan was demonstrated for the first time in this investigation.     

Experimental infection of specific pathogen free (SPF) cats with two different strains of bartonella henselae type I: a comparative study

Domestic cats are the reservoir of Bartonella henselae, the main causative agent of cat scratch disease. We compared B. henselae type I infection characteristics in 6 SPF cats infected with a feline strain (4.8 x 10(7) colony-forming units (CFU)/mL) and in 6 SPF cats infected with the reference Houston I strain (6.6 x 10(6) CFU/mL to 9.6 x 10(7) /mL). All the cats inoculated with the feline strain, but none of the cats inoculated with B. henselae Houston I, developed a fever within 2-12 days (mean: 5.8 days) post inoculation (PI), which lasted for 1-2 weeks. However, all 12 cats became bacteremic. The duration of bacteremia was significantly longer in the cats inoculated with the feline strain (mean: 237 days) than in the cats inoculated with Houston I strain (mean: 60 days) (p < 0.01). Five (83%) cats inoculated with the feline strain and none of the six cats inoculated with B. henselae Houston I had relapsing bacteremia (p = 0.02). IgG antibodies were detected by IFA within 1-2 weeks for both strains, but peaked later (week 10 versus week 3 PI) for the feline strain. By ELISA, using antigens of each B. henselae strain, all 12 cats developed Bartonella specific IgM and IgG antibodies, but the cats infected with B. henselae Houston I antigen yielded significantly lower optical density values (p < 0.05). By SDS-PAGE, PFGE and Western blotting, protein profile differences (84 to 89% homology) were observed between the two strains. If a feline vaccine is to be developed in order to prevent human infection, the choice of the vaccine strain will be critical, since major differences were identified even between strains belonging to the same sero/genotype.     

Investigation of Bartonella henselae in cats in Ankara, Turkey

The purpose of this study was to determine Bartonella henselae prevalance in cats in Ankara. Whole bloods and sera collected from 256 cats were investigated for the presence feline Bartonella species by culture and sera were tested for the presence of antibodies against B. henselae IgG using immunofluorescence assay. Bartonella species were isolated by blood culture from 24 (9.4%) cats. Bartonella isolates were subjected to restriction fragment length polymorphism (RFLP) by using TaqI and HhaI endonucleases to identify species. Twenty-one isolates were determined as B. henselae and three of 24 isolates were determined as Bartonella clarridgeiae with RFLP. The bacteraemia prevalence and seroprevalence of B. henselae IgG antibodies in cats was detected as 8.2% and 18.6% respectively. This is the first report on B. henselea and B. clarridgeiae in cats in Turkey.     

Genomic diversity of Bartonella henselae isolates from domestic cats from Japan, the USA and France by pulsed-field gel electrophoresis

The genomic DNA diversity of 27 Bartonella henselae and three B. clarridgeiae isolates from 18 domestic cats from Japan, the USA and France was investigated by pulsed-field gel electrophoresis (PFGE) with NotI, AscI and SmaI restriction enzymes. A great diversity of genomic patterns was found for all B. henselae, but none for B. clarridgeiae isolates. The DNA size of B. henselae and B. clarridgeiae isolates were 1.7-2.9 and 1.7Mbp, respectively. All 13 Japanese cat isolates were identified as B. henselae type I. Furthermore, three of the four Japanese cats harbored genetically different B. henselae type I isolates, suggesting for the first time co-infection with various type I isolates.One French cat and one American cat were co-infected with B. henselae and B. clarridgeiae. B. henselae type I and type II were mainly grouped in two different clusters by PFGE using SmaI endonuclease in the dendrogram.     

Bartonella henselae infection in splenectomized domestic cats previously infected with hemotropic Mycoplasma species

Cat scratch disease is caused by Bartonella henselae and the domestic cat represents its main reservoir. In immunocompromised patients, infection with B. henselae is characterized by more severe clinical forms than in non-immunocompromised individuals. The objective of the present study was to investigate the characteristics of B. henselae (Houston-I strain) infection in four splenectomized and three non-splenectomized cats, five of which were chronically infected with 'Candidatus Mycoplasma haemominutum'. No major clinical signs were observed in either group of cats. Cats in both splenectomized and non-splenectomized groups became bacteremic within a week post-inoculation. Although bacteremia was on average 10 days longer in the splenectomized cats, that difference was not statistically significant (P=0.72). In both groups, the level of bacteremia peaked within the same time frame; however, the level of bacteremia was about 10-fold higher in the splenectomized cats (P=0.007). Such a difference could be associated with a reduced immune response to the infection, especially a reduced ability to phagocytize Bartonella organisms in the splenectomized cats. Concurrent infection with 'Candidatus M. haemominutum' did not appear to alter the course of infection.     

Cat-scratch disease in veterinary-associated populations and in its cat reservoir in Taiwan

In Taiwan, the first human case of cat-scratch disease (CSD) was diagnosed by a serologic test in 1998. Since then, no studies have been conducted to understand the epidemiology of the infection in Taiwan. Therefore, this study is the first epidemiologic survey of CSD in cats and humans in this country. Using veterinary-associated individuals as the study population, it was identified that 1.7% of them were seropositive for B. henselae, and residence was the only factor associated with seropositivity. Bartonella species were successfully isolated from 25 (19.1%) of the 131 cats tested. Only B. henselae and B. clarridgeiae were obtained from bacteremic cats. Furthermore, 9.2% of 131 cats were dually-infected with genotypes I and II of B. henselae. It is the highest prevalence of co-infection that has ever been reported worldwide. In cats, the seroprevalence was 23.7% by indirect immunofluorescence antibody assay with B. henselae Houston-1 (type I) as the antigen. When 12 bacteremic but seronegative cats were re-tested by IFA slides coated with B. henselae U-4 antigen (type II), 9 cats were identified to be seropositive. Our study further suggested that using only direct PCR of 16S-23S rRNA intergenic region or the combination of the PCR method and indirect immuno-fluorescence test will be useful to diagnose Bartonella-free cats.     

Detection and characterization of feline Bartonella henselae in the Czech Republic

The aims of the study were to characterize isolates of Bartonella henselae and to determine the prevalence of bacteremic domestic cats in urban and suburban parts of Prague, Czech Republic. Five (18%) gram-negative fastidious bacterial single-cat isolates were recovered from 27 hemocultures incubated without previous freezing. Four of these isolates originated from flea infested stray cats (n=6) and one from a shelter cat without any ectoparasites (n=21). None of the 34 previously frozen specimens from flea free pet cats yielded any bacteria. All five isolates were catalase and oxidase negative. Their enzymatic activity, RFLP profile of citrate synthetase gene (gltA) and DNA-DNA hybridization results were typical of B. henselae. According to their PvuII and BglI ribotypes the isolates could be allocated to two homogeneous groups. Ribotype HindIII and RFLP of 16S-23S rRNA spacer region analysis gave unique profiles different from those of Bartonella quintana, Bartonella elizabethae and Bartonella clarridgeiae. The 16S rRNA type-specific amplification revealed an identical profile typical of B. henselae genotype II for all the cat isolates studied. Pulsed-field gel electrophoresis (PFGE) assigned a different profile to each of the isolates studied. Determination of the enzymatic activity, RFLP of gltA gene, RFLP of 16S-23S rRNA spacer region, and HindIII ribotype could be efficient tools for identification of B. henselae isolates. Ribotyping (PvuII, BglI), 16S rRNA typing and PFGE may be useful methods to prospect ecology and epidemiology of the agent.     

Genomic variations among Bartonella henselae isolates derived from naturally infected cats

The purpose of this study was to understand the mechanisms of persistent infection with Bartonella henselae in cats. Blood samples were collected from three naturally infected cats for 24 months. These cats were confirmed to be persistently infected with B. henselae by serological and bacteriological examination. Relapsing bacteremia was found in all three cats with intervals of 3-19 months. Following the peaks of bacteremia, increases of specific antibody titer were observed in these cats. To examine the genetic differences among the isolates derived from the first and following bacteremia, the genome DNA patterns of the restriction enzyme fragment length polymorphism (RFLP) of the isolates were examined by pulsed field gel electrophoresis. The isolates derived from the first bacteremia showed an identical RFLP pattern in each of the three cats. The isolates derived from the following peaks, however, showed 1-3 of different RFLP patterns in these cats. Furthermore, the isolates showing different RFLP patterns from those of the first bacteremia were also detected at the following bacteremic peaks in all three cats examined. The 16S ribosomal RNA (rRNA) gene type of all isolates was found to be 16S rRNA type I. The emergence of genetically distinct organisms at various peaks of bacteremia may contribute to the establishment of persistent infection in the naturally infected cats.     

Occurrence of Bartonella henselae types I and II in Central Italian domestic cats

Serological and molecular surveys were conducted to determine the occurrence of Bartonella henselae in domestic cats in Central Italy. Samples from 234 pet cats were tested for B. henselae antibodies by indirect immunofluorescence with 78 (33.3%) positive. A PCR assay specific for the Bartonella 16S rRNA gene was carried out on DNA samples extracted from blood of the 234 cats; 26 (11.1%) of the seropositive cats were positive. Two PCR protocols, which discriminate genotypes I and II of B. henselae, were performed on all DNA samples. Sixteen (6.8%) cats were infected by genotype I, 6 (2.5%) by genotype II, and two males (0.8%) by both genotypes. Two female (0.8%) cats which were Bartonella sp. PCR positive, gave negative results with the types I and II PCR. This protocol facilitates the direct and rapid detection of Bartonella DNA in feline blood samples, and differentiates B. henselae genotypes.     

Molecular evidence for Bartonella spp. in cat and dog fleas from Germany and France

Nine hundred and fifty-two fleas were collected from 148 cats and 133 dogs at 18 widely distributed geographic locations in Germany and France and examined for the presence of six different Bartonella spp. (Bartonella bacilliformis, Bartonella clarridgeiae, Bartonella elizabethae, Bartonella henselae, Bartonella quintana, Bartonella vinsonii subsp. berkhoffii) by PCR. Thirty-five specimens (3.7%) tested positive for either B. henselae (14 positive fleas) or B. clarridgeiae (21 positive fleas). DNA of other Bartonella spp. were not detected. Bartonella clarridgeiae was the dominating species in samples from France (19 out of 22 positive fleas), whereas B. henselae was more frequent in Germany (11 out of 13 positive fleas). With 3.5% (22 out of 632 fleas) in France and 4.1% (13 out of 320 fleas) in Germany, the overall prevalences of pathogen did not vary significantly between the flea populations of both countries. 5.4% of cats in France versus 16.1% of cats from Germany were infested by fleas carrying Bartonella, whereas 9.5% of dogs in France but none of the examined dogs from Germany were infested by Bartonella positive fleas. The molecular evidence of Bartonella infections reveals that agents of zoonotic potential are established in flea populations in Germany and France and that the spectrum of species can vary significantly from country to country.     

Clinical disease in kittens inoculated with a pathogenic strain of Bartonella henselae

OBJECTIVE: To evaluate disease in kittens inoculated with Bartonella henselae strain LSU16. ANIMALS: Eighteen 12-week-old specific-pathogen-free kittens. PROCEDURE: Kittens were inoculated with B henselae strain LSU16 or saline (0.9% NaCl) solution. Blood samples were collected from kittens on alternate weeks, and bacteremia, clinical signs, and antibody concentrations were monitored for 6 months after inoculation. RESULTS: Kittens developed raised, erythematous areas at the site of inoculation within 72 hours. Swelling peaked at 14 days and resolved by 28 days after inoculation. Fever had a biphasic pattern, with an episode of 1- to 3-days' duration beginning 6 to 7 days after inoculation followed by an episode of 3- to 8-days' duration beginning 11 to 13 days after inoculation. Kittens were bacteremic by day 14 with peak bacteremia at days 14 to 28. Strong antibody responses to B henselae were detected. Clinical disease resolved before bacteremia became undetectable, but signs of disease correlated with the highest degree of bacteremia. Regional lymphadenopathy also was evident. CONCLUSION AND CLINICAL RELEVANCE: Clinical disease in kittens was similar to that in adult cats infected with B henselae strain LSU16, except that lethargy and anorexia were less severe in kittens, and a biphasic pattern of fever was detected in kittens. Clinical disease after inoculation with B henselae may be strain-dependent. To limit transmission of Bartonella organisms, appropriate flea prevention should be instituted. IMPACT FOR HUMAN MEDICINE: Kittens that are febrile, anorectic, lethargic, and that have lymphadenopathy should be tested for Bartonella organisms, and contact with immunocompromised owners should be discouraged.     

Prevalence of Bartonella infection in wild African lions (Panthera leo) and cheetahs (Acinonyx jubatus)

Bartonella species are emerging pathogens that have been isolated worldwide from humans and other mammals. Our objective was to estimate the prevalence of Bartonella infection in free-ranging African lions (Panthera leo) and cheetahs (Acinonyx jubatus). Blood and/or serum samples were collected from a convenience sample of 113 lions and 74 cheetahs captured in Africa between 1982 and 2002. Whole blood samples available from 58 of the lions and 17 of the cheetahs were cultured for evidence of Bartonella spp., and whole blood from 54 of the 58 lions and 73 of the 74 cheetahs tested for the presence of Bartonella DNA by TaqMan PCR. Serum samples from the 113 lions and 74 cheetahs were tested for the presence of antibodies against Bartonella henselae using an immunofluorescence assay. Three (5.2%) of the 58 lions and one (5.9%) of the 17 cheetahs were bacteremic. Two lions were infected with B. henselae, based on PCR/RFLP of the citrate synthase gene. The third lion and the cheetah were infected with previously unidentified Bartonella strains. Twenty-three percent of the 73 cheetahs and 3.7% of the 54 lions tested by TaqMan PCR were positive for Bartonella spp. B. henselae antibody prevalence was 17% (19/113) for the lions and 31% (23/74) for the cheetahs. The prevalence of seropositivity, bacteremia, and positive TaqMan PCR was not significantly different between sexes and age categories (juvenile versus adult) for both lions and cheetahs. Domestic cats are thus no longer the only known carriers of Bartonella spp. in Africa. Translocation of B. henselae seronegative and TaqMan PCR negative wild felids might be effective in limiting the spread of Bartonella infection.     

Characterization of the first Bartonella henselae strain isolated from a cat in Italy

Bartonella henselae has been identified and characterized for the first time in Italy. A strain, designed Pavia-1, was isolated from the blood of a cat whose owner developed cat scratch disease (CSD). Pavia-1 and two American B. henselae strains (Houston-1, ATCC 49882, type I and strain 269608, UC Davis, type II) were compared by whole-cell fatty analysis (CFA), sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) for protein profiles, Western immunoblotting (WB) for reactivity with polyclonal antibodies, polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP), type-specific 16S rRNA PCRs, and pulsed-field gel electrophoresis (PFGE). Bartonella clarridgeiae (ATCC 51734) was also included for comparison. Pavia-1 was identified as a B. henselae type I. PFGE allowed differentiation between B. clarridgeiae and B. henselae and furthermore, between all the B. henselae strains. The fingerprints of PFGE observed for Pavia-1 were distinct from those of B. henselae type II and also of Houston-1, suggesting that the two type I strains derived from two different clones. These results show the capability of B. henselae to develop genotypic variability between genetically related strains.     

Experimental infection of domestic cats with passaged genotype I Bartonella henselae

The influence of in vitro passage on Bartonella henselae pathogenesis in cats has not been thoroughly evaluated. Our objective was to examine the bacterial kinetics and humoral immune responses in cats experimentally infected with three different in vitro passages of B. henselae F1, a genotype I strain of feline origin. The F1 strain was in vitro passaged 20 and 40 times, and each was inoculated into a group of 5 cats. The kinetics of bacteremia and the feline humoral immune response to bacterial antigens were compared to a previous study involving a group of six cats inoculated with the original F1 strain. Among the three groups of cats, the kinetics of bacteremia profiles and the humoral immune responses to B. henselae lysates were similar. The influence of passage on bacterial membrane proteins was examined. In vitro passage altered the expression of 4/17 (23.5%) bacterial membrane proteins and 6/15 (40%) bacterial membrane antigens. An association between poor seroreactivity to three lysate antigens (15-, 18- and 45kDa), prolonged bacteremia and decreased serum bactericidal activity was noted. Our data show that in vitro passage of B. henselae did not alter the kinetics of bacteremia, including the occurrence of relapsing bacteremia, in experimentally infected cats. This suggests that highly passaged strains may not be suitable for future vaccination studies. Furthermore, in vitro passage results in phenotypic and antigenic changes in the bacterial membrane protein profile, which warrants caution in the interpretation of studies involving passaged B. henselae strains.     

Seroprevalence of Bartonella infection in American free-ranging and captive pumas (Felis concolor) and bobcats (Lynx rufus)

Bartonella henselae is the main agent of cat scratch disease in humans and domestic cats are the main reservoir of this bacterium. We conducted a serosurvey to investigate the role of American wild felids as a potential reservoir of Bartonella species. A total of 479 samples (439 serum samples and 40 Nobuto strips) collected between 1984 and 1999 from pumas (Felis concolor) and 91 samples (58 serum samples and 33 Nobuto strips) collected from bobcats (Lynx rufus) in North America, Central America and South America were screened for B. henselae antibodies. The overall prevalence of B. henselae antibodies was respectively 19.4% in pumas and 23.1% in bobcats, with regional variations. In the USA, pumas from the southwestern states were more likely to be seropositive for B. henselae (prevalence ratio (PR) = 2.82, 95% confidence interval (CI) = 1.55, 5.11) than pumas from the Northwest and Mountain states. Similarly, adults were more likely to be B. henselae seropositive than juveniles and kittens (PR = 1.77, 95% CI = 1.07, 2.93). Adult pumas were more likely to have higher B. henselae antibody titers than juveniles and kittens (p = 0.026). B. henselae antibody prevalence was 22.4% (19/85) in bobcats from the USA and 33.3% (2/6) in the Mexican bobcats. In the USA, antibody prevalence varied depending on the geographical origin of the bobcats. In California, the highest prevalence was in bobcats from the coastal range (37.5%). These results suggest a potential role of wild felids in the epidemiological cycle of Bartonella henselae or closely related Bartonella species.     

Prevalence of Bartonella species causing bacteraemia in domesticated and companion animals in the United Kingdom

Between October 1999 and February 2000, 691 blood samples examined routinely for either haematological or virological assessment were screened by culture for the presence of Bartonella species. They came from 615 animals: 360 cats, 211 dogs, 27 horses, 16 cattle and a gorilla. The samples were incubated for long periods on 10 per cent horse blood agar at 37 degrees C in an atmosphere containing 5 per cent carbon dioxide. Isolates were obtained from 35 samples from 34 (9.4 per cent) of the cats, but not from any of the other animals. Comparison of citrate synthase gene sequences from the isolates indicated that they were all Bartonella henselae. Analysis of 16S rRNA gene fragments indicated that 30 of the cats were infected solely with B henselae genotype II, two were infected solely with B henselae genotype I and two were infected with both genotypes.