Bartonella henselae isolate
Bartonella henselae strain CSU-1 was initially isolated from a shelter cat in Florida and has been passaged once in cats. Blood in EDTA from a cat infected with CSU-1 by exposure to infected C. felis had been stored at −80°C until used in the current study.
The C. felis used in this study were purchased from an insectary of a local research laboratory.c Two groups of 5 male and 5 female adult C. felis that were not exposed to the cats were purchased (one group before the study began and another group after the completion of this study) and were pooled by group, subjected to total DNA extraction, and shown to be negative for DNA of Bartonella spp. and haemoplasmas by PCR assay[6, 11, 12].
Flea and tick collar
A flea and tick collar that contains imidacloprid 10% and flumethrin 4.5% was placed on designated cats in this study following the manufacturer’s guidelines[9–11].bThe collars were stored in a locked, fireproof cabinet and maintained at 20°C to 25°C until used.
A total of 19 cats (10 male; 9 female) were purchased from a commercial research cattery at approximately 4 months of age and shipped to the study site. The cats had been vaccinated against feline viral rhinotracheitis, feline calicivirus, and feline panleukopenia virus. During the 49 day equilibration period for the study, the male cats were neutered following protocols of the animal facility, all cats were shown to have negative serum test results for FeLV-antigen, FIV-antibody, Dirofilaria immitis antigen,d and Bartonella spp. antibody (ELISA), and all cats were shown to be negative for Bartonella spp and haemoplasma DNA in blood using previously reported PCR assays[12–14]. A temperature sensing microchip was implanted as previously described for use during clinical monitoring over the course of the study.
The study design was approved by the Institutional Animal Care and Use Committee at the research facilitye that housed the cats during the flea infestation portion of the study and the Animal Care and Use Committee at Bayer Animal Health.
After arrival at the facility and entry into the 49 day equilibration period, the cats were randomized into 3 groups and housed together in 1 room that was divided into 3 sections (R1, R2, and R3) beginning on Day −44. The 3 sections were separated from each other by mesh so that C. felis could move amongst the cats, while body contact between cats or fighting with members of other groups was prevented. Sections R1 and R3 were adjacent to the center section R2 but not to each other. Approximately 25% of the floor space in each section was covered in carpet to promote survival and a complete life cycle of C. felis within the room. Cat perches were placed in the corners of each section adjacent to the mesh dividing the sections to encourage cats from the various groups to be close to each other. Based on this design, it seems unlikely the physical structure itself would influence C. felis to preferentially transfer to R1 or R3.
The group in R1 (n = 7 cats) were untreated control cats throughout the study, the group in R2 (n = 4 cats) were infected with B. henselae by intravenous inoculation of 0.2 ml of blood containing B. henselae strain CSU-1 on Day – 39 of the study, and the flea and tick collars were placed on the group in R3 on Day 0 of the study. After placement of the collars on Day 0, a total of 50 males and 50 female C. felis were placed on each of the 4 cats in the B. henselae infected group in R2. Additional fleas (50 males and 50 female) were placed on each of the R2 cats monthly for an additional 6 applications and then every 12 – 14 days for 4 applications. The interval between C. felis application was shortened at the end of the experiment to increase likelihood of B. henselae infection in all cats. Flea counts by flea comb were determined every 12 – 14 days for the duration of the study and the fleas returned to the cat. The flea collars were removed from R3 cats on Day 238 of the study, all cats were administered imidaclopridf once at that time, and then were observed and sampled as described until Day 254. Once shown to be negative for B. henselae by PCR, the cats were returned to the research facility cat colony with biosafety committee approval.
Samples were collected for performance of Bartonella spp. serology and PCR assay for amplification of Bartonella spp. DNA on Days 0, 14, 28, 42, 56, 70, 84, 98, 112, 126, 140, 154, 168, 182, 196, 210, 224, 238, and 252 for all cats as well as on Days −30, -23, -16, -9 and −2 for cats in R2.
Blood samples (2 ml in a clot tube for serum separation; 1 ml in EDTA) from each cat were transported from the research facility to Colorado State University within 2 hours of collection. Samples were prepared for determination of serum titers of IgG against Bartonella spp. and for PCR assay for Bartonella spp. DNA on the day of collection[12, 14]. One aliquot of each blood sample (500 μL) in EDTA was stored at −80°C for bacterial culture, pending results of the Bartonella spp. PCR assay. Samples negative for Bartonella spp. DNA by PCR assay were cultured using the stored blood as previously described. Results for samples that yielded characteristic Bartonella spp. colonies were confirmed by Bartonella spp. PCR assay.
The cats were examined daily for attitude and appetite throughout the study. Mucous membrane color assessment and physical examination (including cardiac auscultation) was made on any cat exhibiting signs of depression or inappetance and for cats with a rectal body temperature of > 102.5°F (39.2°C). Any cat with fever had a complete blood cell count performed. Cats that developed clinical illness consistent with bartonellosis (fever and inappetance of > 2 days duration) during the study were administered enrofloxacing at 5 mg/kg, PO, daily for at least 14 days and supportive care as indicated. Cats requiring antibiotic treatment were administered imidaclopridf and moved to a C. felis free room for continued care.
Cat hair sampling
On Day 224, approximately 0.5 g of hair was collected from each cat. Each sample was placed into an individually labelled plastic bag for determination of imidacloprid concentrations.h The cutoff sensitivity for the imidacloprid assay is 0.1 mg/kg hair.
Cats were diagnosed with B. henselae infection if at least 2 samples over the course of the study were positive for Bartonella spp. IgG, Bartonella spp. DNA by PCR assay, or Bartonella spp. organisms by culture. The proportion of B. henselae-infected cats was compared between groups R1 and R3 by use of a 2-tailed Fisher exact test. Differences among group mean flea counts were compared with the 2-tailed Student t test. A value of P < 0.05 was considered significant for all analyses.