<i>Ehrlichia muris</i> Infection in a Dog from Minnesota

A 6-year-old-male Labrador Retriever from northern Minnesota was examined by a veterinarian in June 2011 because of decreased activity, reluctance to climb stairs, and a history of having 2 attached ticks, approximately 2 months earlier. The dog was febrile (rectal temperature 103.5°F) and had a stiff gait, particularly noticeable in the front legs. On palpation, there was pain in both elbow joints. Because of the high prevalence of tick-borne arthropathies in the practice area, Lyme borreliosis or anaplasmosis was suspected. Because a commercially available in-house assay1 was strongly positive for Anaplasma species antibodies, canine anaplasmosis was diagnosed serologically and doxycycline was administered at a dosage of 6 mg/kg PO q12h for 21 days. Carprofen (2 mg/kg PO q24h–q12h) was dispensed for control of pain as needed and topical fipronil was dispensed for tick control. When re-examined 5 days later, overall activity was decreased, the dog frequently licked the paws and remained stiff, but the owner had not administered carprofen. Rectal temperature was 102.2°F. Amoxicillin (11 mg/kg PO q12h for 10 days) and Fortiflora were added to the treatment regimen. When re-examined in July, the dog had returned to normal activities and normal mobility. On September 7, 2011, a second veterinarian collaborating with the Intracellular Pathogens Research Laboratory (IPRL) examined the dog because of decreased appetite, lethargy, and recurrent bouts of vomiting. Rectal temperature was 103.3°F. SNAP 4Dx again indicated the presence of Anaplasma species antibodies. The dog was thrombocytopenic (platelet count, 132,000/μL; reference range, 160,000–525,000/μL), according to an in-house platelet count, but a CBC and serum biochemical profile performed by a commercial diagnostic laboratory2 on a blood sample drawn the same day identified no hematologic or serum biochemical abnormalities (platelet count, 243,000/μL; reference range, 140,000–540,000/μL). Initial treatment included 1 dose of oxytetracycline (1 mg/kg, IV). Doxycycline was dispensed with instructions to administer 8 mg/kg PO q12h for 30 days along with a digestive support supplement.3 Blood and serum samples collected on September 7, 2011 were sent to the IPRL for serological and PCR testing for Anaplasma and Ehrlichia spp. Anaplasma species antibodies were confirmed using SNAP 4Dx. PCR amplification was performed using previously described GEP 16S Ehrlichia genus primers and species-specific primers for Ehrlichia canis, Ehrlichia chaffeensis, Ehrlichia ewingii, Anaplasma phagocytophilum, and Anaplasma platys.1 An amplicon was obtained with the Ehrlichia genus primers, but amplification using the above-mentioned species-specific primers, including A. phagocytophilum, did not generate amplicons. Sequencing4 of the Ehrlichia genus amplicon resulted in the highest DNA similarity (100%) to Ehrlichia muris (361 of 361 base pairs when compared to GenBank Accession number NR_025962, a strain from a mouse in Japan). After designing a 16S rDNA, E. muris primer pair Emu58 (sense): 5′ ATAGCTACCCATAGCTTTTTTAGCTATAGGTT 3′ and SEP (anti-sense): 5′ CTTCTRTRGGTACCGTCATTATCTTCCCY 3′ (as forward and reverse primers respectively), a 395 base pair amplicon (expected amplicon size) was obtained from the September 7, 2011 blood sample. The dog was examined on September 23, 2011 because of vomiting and diarrhea that was temporally associated with administration of doxycycline. A fecal flotation was negative for parasite ova. Clostridium enteritis was diagnosed by the attending veterinarian using fecal smear examination. Doxycycline was discontinued and amoxicillin (20 mg/kg POq12h for 14 days), metronidazole (20 mg/kg PO q12h for 14 days), and metoclopramide (0.9 mg/kg PO q24 for 5 days) were dispensed. Within 3 days, the vomiting and diarrhea had resolved and the dog was acting normally. PCR amplification5 using Ehrlichia genus and E. muris-specific primers from a September 27, 2011 blood sample were negative, no additional doxycycline was dispensed. On October 29, 2011, the dog was somewhat less active, had bloody urine, and occasional blood clots were observed dripping from the prepuce. The prostate gland was slightly enlarged, nonpainful, and symmetrical. The packed cell volume was 48%, the platelet count was 450,000/μL (reference range, 166,000–575,000/μL), and hypoglobulinemia (2.1 g/dL; reference range, 2.5–4.5 g/dL) was present. Urine was alkaline (pH8.0) with 3+ proteinuria and specific gravity 1.040, and hematuria was present. Radiographs of the abdomen disclosed no abnormalties. Cephalexin (20 mg/kg PO q8h for 10 days) was dispensed for presumptive urinary tract infection. Despite treatment with cephalexin, hematuria persisted. On November 2, 2011, the platelet count was 118,500/μL and packed cell volume was 50%. Urine specific gravity was 1.040 and sediment examination identified too numerous to count erythrocytes and occasional leukocytes. Because of the thrombocytopenia and because the dog had not completed the 30-day course of doxycycline because of vomiting and diarrhea in September, doxycycline again was administered (5 mg/kg PO q12h for 30 days). In addition, the dog was referred for an abdominal ultrasound examination. No ultrasonographic structural abnormalities were found within the urinary tract. The hematuria resolved and the dog did not experience any additional medical problems. On November 16, 2011, post-treatment blood sample for PCR and convalescent serum for testing against a panel of vector-borne pathogens were submitted to the IPRL. Again, seroreactivity to the Anaplasma peptide was the only SNAP 4Dx finding. By IFA testing, there was no antibody reactivity at a titer of 1 : 16 (testing scale, 1 : 16–1 : 8,192) to Babesia canis, Bartonella henselae, Bartonella vinsonii subsp. berkhoffii, E. canis, and Rickettsia rickettsii antigens. Ehrlichia genus and E. muris-specific PCR were negative. In addition, PCR amplification was retrospectively performed for the 3 sample collection dates using previously described groEL Ehrlichia genus primers.2 A GroEL PCR amplicon was obtained from the blood sample collected on September 7, 2011, but not from blood samples collected on September 27, 2011 and November 16, 2011. Sequence analysis of the GroEL PCR amplicon showed 100% homology (547/547 bp) with E. muris (GenBank accession number AF210459, from the E. muris type strain AS145). The E. muris 16S and GroEL sequences derived from this dog have been deposited into the GenBank data base under the accession numbers JQ10629 and JQ10630, respectively. Sequence phylogeny analyses of the 16S and GroEL gene sequences conducted with Molecular Evolutionary Genetics Analysis software supported a close phylogenetic relationship with E. muris. (Fig 1). In collaboration with 2 veterinary clinicians (Drs Koskinen and Eberts) in Minnesota, who routinely examine dogs with acute, febrile illness, we were able to provide the first molecular diagnostic evidence to support a potential role for E. muris as a pathogen in dogs from the United States. In 2 previously published studies, seroreactivity to E. canis-derived peptides using SNAP 3DX or 4DX assays in dogs from the Minnesota was unexpectedly more prevalent than expected based on tick vectors found in this region.1, 3 Exposure to Rhipicephalus sangineus, the vector for E. canis, and Ambylomma americanum, the vector for E. chaffeensis and E. ewingii, occurs infrequently in colder regions of the United States, therefore infection with another bacterial pathogen, potentially transmitted by Ixodes scapularis, a tick that is plentiful throughout this region, was suspected. In addition, E. canis and E. chaffeensis are the only 2 recognized Ehrlichia species in North America that are known to induce seroreactivity to the E. canis peptides used in the SNAP 3Dx and 4Dx assays.4, 5 In a previous study, neither E. canis nor E. chaffeensis DNA was amplified from Minnesota dogs, whereas the B. burgdorferi and A. phagocytophilum seroprevalence was 55%, indicating frequent exposure to I. scapularis.1 Based upon serology, the dog in this study had been exposed to A. phagocytophilum, supporting exposure to I. scapularis. In 2011, after the publications by Bowman and Beall,1, 3 human infection with an E. muris-like agent was reported for the first time in the medical literature in patients from Wisconsin.6 In addition, Telford et al. retrospectively documented the presence of E. muris DNA in I. scapularis ticks collected in the 1990s from northern Wisconsin.7 Collectively, these observations suggested that pet dogs exposed to ticks in this region also might be infected with this novel pathogen within the genus Ehrlichia. Potentially, cross-reacting antibodies resulting from exposure to another genus of bacteria provided a plausible explanation for the high SNAP E. canis seroreactivity in dogs from this region, but it seemed more likely that an organism within the genus Ehrlichia might be responsible for the seemingly disparate seroepidemiological findings. Interestingly, the E. muris-infected dog in this study never seroconverted to the E. canis peptides using SNAP 4Dx or E. canis IFA assays. Thus, some E. muris-infected dogs may not seroconvert, particularly if antibiotic treatment is started very early in the course of infection, or alternatively, E. muris may not be responsible for the sero-epidemiologic observations in previous studies from Minnesota.1, 3 In the context of human illness, the 4 patients described to date, most of whom have been immunocompromised, presented with fever, headache, thrombocytopenia, lymphopenia, and increased liver enzyme activities.6 Experimentally, E. muris-infected mice developed splenomegaly and lymphadenopathy.8 Although the role of E. muris in the overall pathogenesis of disease manifestations reported in this dog between June and November 2011 is impossible to assess with the available clinical, serologic and microbiological data, fever and thrombocytopenia are features of illness in human patients. Moreover, the expected therapeutic response to doxycycline treatment in dogs infected with A. phagocytophilum is generally dramatic, with rapid resolution of disease manifestations occurring generally within 24–48 hours.9, 10 Given the fact that this dog remained stiff and inactive for at least 5 days after initially being treated with doxycycline in June, it is possible that the dog was co-infected with A. phagocytophilum and E. muris at the time of the initial evaluation and that the E. muris infection persisted until documented in September 2011. Alternatively, A. phagoctyophilum may have been solely responsible for the June presentation, with E. muris transmission occurring in early September, just before onset of the second documented febrile episode. Recently, to further complicate clinical interpretation of disease outcomes, chronic infection with A. phagocytophilum has been documented in experimentally infected dogs in which treatment with doxycycline did not eliminate the infection.11 Despite this research observation, to date, molecular evidence supporting chronic canine anaplasmosis in naturally infected dogs is lacking in the literature from North American and European regions, where A. phagocytophilum transmission is endemic, and efforts to PCR amplify A. phagocytophilum DNA from 3 blood samples from the dog of this study were not successful. Whether E. muris causes an acute, short duration infection in dogs or is capable of inducing chronic long-lasting (months to years) infection in dogs, as occurs with other Ehrlichia spp., such as E. canis and E. ewingii, remains unknown. Historically, because Ehlrichia spp. could only be cultured using cell culture systems in a research setting, the advent of diagnostic DNA testing has greatly facilitated enhanced understanding of the host range and the pathogenic potential of these obligate intracellular bacteria. Ehrlichia muris was first isolated from rodents in Japan in the early 1990s.12 An experimental infection study involving 2 dogs failed to demonstrate evidence of transmission and the dogs did not develop signs of disease.8 Subsequently, monocytic Ehrlichia sp. DNA with sequence homology to E. muris was PCR-amplified from Ixodes persulcatus and Ixodes ricinus ticks found in Eastern Europe.13, 14 Unlike the DNA sequences from human patients, which were reported as E. muris-like,6 the 16S rDNA and rpoB sequences obtained from the dog in this study were 100% homologous to the E. muris type strain from Japanese rodents, and differed from the sequences reported from human patients in Wisconsin. The routine diagnostic use of PCR amplification and DNA sequencing practices are likely to identify other as yet unknown Ehrlichia sp. and an expanded host range in which infection with various Ehrlichia sp. occurs.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.