LETTERS

Human teria. A chest radiograph, computed ware (www.ebi.ac.uk/Tools/clustalw2/ tomography of the abdomen, and an index.html) and the GenBank/Europe- Anaplasmosis and echocardiograph of the heart showed an Molecular Biology database library Anaplasma ovis unremarkable results. Blood samples (http://blast.ncbi.nlm.nih.gov/Blast. Variant were negative for antibodies against cgi). Phylogenetic trees were con- cytomegalovirus, Epstein-Barr virus, structed by using MEGA 4 software To the Editor: Anaplasmosis is a hepatitis, HIV, mycoplasma, coxackie (www.megasoftware.net). disease caused by of the genus virus, adenovirus, parvovirus, Cox- The fi rst blood sample was posi- Anaplasma. A. marginale, A. centrale, iella burnetii, R. conorii, and R. typhi, tive for A. ovis by PCR; the other 2 A. phagocytophilum, A. ovis, A. bovis, and for rheumatoid factors. A lymph were negative. A 16S rRNA gene se- and A. platys are obligate intracellular node biopsy specimen was negative quence (EU448141) from the posi- bacteria that infect and in- for infi ltration and malignancy. After tive sample showed 100% similarity vertebrate host cells. A. ovis, which is treatment with doxycycline (200 mg/ with other Anaplasma spp. sequences transmitted primarily by Rhipicepha- day for 11 days), ceftriaxone (2 g/day (A. marginale, A. centrale, A. ovis) lus bursa ticks, is an intraerythrocytic for 5 days), and imipenem/cilastatin in GenBank. Anaplasma sp. groEL rickettsial of sheep, goats, (1,500 mg/day for 1 day), the patient and msp4 genes showed a 1,650-bp and wild ruminants (1). recovered and was discharged 17 days sequence (FJ477840, corresponding Anaplasma spp. infections in hu- after hospitalization. to 748 of 1,650 bp) and an 852-bp mans have been reported in Cyprus Three serum samples from the sequence (FJ460443) for these genes, (2,3). We report infection of a human patient were tested in Crete, Greece, respectively. Phylogenetic trees (Fig- with a strain of Anaplasma sp. other for immunoglobulin (Ig) G and IgM ure) were constructed by using A. ovis than A. phagocytophilum, which was against A. phagocytophilum antigen strains detected in sheep and goats in detected by PCR amplifi cation of ana- by using an immunofl uorescent anti- Cyprus (5). plasmatic 16S rRNA, major surface body assay (Focus Diagnostics, Cy- Fever is common in cases of 4 (msp4), and heat shock pro- press, CA, USA). Serologic analysis human infection with A. phagocyto- tein 60 (groEL) genes. showed IgG titers of 0, 0, and 128 and philum (6). We also detected throm- A 27-year-old woman was admit- IgM titers of 20, 20, and 20 against A. bocytopenia and elevated levels of ted to the pathology clinic of a hospi- phagocytophilum in the 3 serum sam- transaminases. However, hepatosple- tal in Famagusta, Cyprus on May 14, ples, respectively. nomegaly, , and 2007, with an 11-day history of fever Because the blood samples were anemia are not common in persons in- (<39.5°C) after a tick bite. Before ad- transported frozen, detection of mo- fected with A. phagocytophilum. mission, the patient was treated with rulae was not possible. DNA was ex- Immunofl uorescent antibody cefi xime (400 mg/d for 3 days) and tracted by using the QIAamp DNA analysis showed weak antibody titers cefradine (2 g/d for 2 days) without Blood Mini Kit (QIAGEN, Hilden, against A. phagocytophilum. Serolog- abatement of the fever. Physical ex- Germany). PCR amplifi cations (My- ic cross-reactivity of Anaplasma spp. amination showed hepatosplenomega- Cycler DNA thermal cycler; Bio-Rad, is caused by conservation of major ly and an enlarged lymph node. Hercules, CA, USA) were conducted surface protein sequences (7). Initial laboratory examinations for the anaplasmatic 16S rRNA gene; A. phagocytophilum infection showed moderate anemia (hemoglobin A. marginale, A. centrale, and A. ovis usually resolves after treatment with 11.5 g/dL), thrombocytopenia (95,000 heat shock protein 60 (groEL) genes; doxycyline for 4 days (8). The patient thrombocytes/mm3), increased lev- and A. marginale, A. centrale, and A. reported here was treated with doxy- els of transaminases (aspartate ami- ovis major surface protein 4 (msp4) cycline for 11 days. However, 1 case is notransferase 178 U/L, alanine ami- genes (4,5). DNA from previous stud- not suffi cient to form conclusions on notransferase 313 U/L, γ-glutamyl ies in Cyprus (4,5) was used as a posi- severity and duration of illness. transferase 79 U/L, lactate dehydroge- tive control. Double-distilled water In a study conducted in Cyprus, nase 698 U/L), an increased level of was used as a negative control. Anaplasma sp. was identifi ed in C-reactive protein (10.4 mg/L), and PCR amplicons were purifi ed by birds (9). Because birds may be car- an increased erythrocyte sedimenta- using the QIAquick Spin PCR Prod- riers of zoonotic , infec- tion rate (80 mm/h). Blood samples uct Purifi cation Kit (QIAGEN) and tion of humans with these pathogens were obtained from the patient at the sequenced on a 4200 double-beam may occur. However, transmission time of admission and 7 days and 3 automated sequencer (LI-Cor, Inc., of A. ovis to humans is unclear. The months later. Results of blood and Lincoln, NE, USA). Sequences were role of R. bursa, a common tick spe- urine cultures were negative for bac- processed by using ClustalW2 soft- cies in sheep and goats in Cyprus (D.

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 16, No. 6, June 2010 1031 LETTERS

5. Psaroulaki A, Chochlakis D, Sandala- kis V, Vranakis I, Ioannou I, Tselentis Y. Phylogentic analysis of Anaplasma ovis strains isolated from sheep and goats us- ing groEL and mps4 genes. Vet Micro- biol. 2009;138:394–400. DOI: 10.1016/j. vetmic.2009.04.018 6. Dumler JS, Choi KS, Garcia-Garcia JC, Barat NS, Scorpio DG, Garyu JW, et al. Human granulocytic anaplasmosis and Anaplasma phagocytophilum. Emerg In- fect Dis. 2005;11:1828–34. 7. Palmer GH, Abbott JR, French DM, McElwain TF. Persistence of Anaplasma ovis infection and conservation of the msp-2 and msp-3 multigene families with- in the genus Anaplasma. Infect Immun. 1998;66:6035–9. 8. Bakken JS, Dumler JS. Clinical diagno- sis and treatment of human granulocy- totropic anaplasmosis. Ann N Y Acad Sci. 2006;1078:236–47. DOI: 10.1196/ annals.1374.042 9. Ioannou I, Chochlakis D, Kasinis N, Anayiotos P, Lyssandrou A, Papadopou- los B, et al. Carriage of Rickettsia spp., Coxiella burnetii and Anaplasma spp. by endemic and migratory wild birds and their ectoparasites in Cyprus. Clin Micro- Figure. Evolutionary trees based on major surface protein 4 (A) and heat shock protein 60 (B) biol Infect. 2009 Mar 11. [Epub ahead of genes sequences of Anaplasma phagocytophilum, A. marginale, and A. ovis. Evolutionary print]. history was inferred by using the neighbor-joining method. H. sulcata; Haemaphysalis 10. Merino FJ, Nebreda T, Serrano JL, Fernán- sulcata. A) Optimal tree (branch length = 0.87919908) is shown. Percentages of replicate dez-Soto P, Encinas A, Pérez-Sánchez trees in which the associated taxa clustered together in the bootstrap test (500 replicates) R. Tick species and tick-borne infections are shown. B) Optimal tree (branch length = 0.34047351) is shown. Percentages of replicate identifi ed in population from a rural area of trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates) Spain. Epidemiol Infect. 2005;133:943–9. are shown. Trees are drawn to scale, with branch lengths in the same units as those of DOI: 10.1017/S0950268805004061 the evolutionary distances used to infer the phylogenetic tree. Evolutionary distances were computed by using the Kimura 2-parameter method. Strains detected in Cyprus are Address for correspondence: Anna Psaroulaki, indicated in boldface. Scale bars indicate number of base substitutions per site. Laboratory of Clinical Bacteriology, Para- sitology, Zoonoses and Geographical Medicine Chochlakis, unpub. data), as a vector References (World Health Organization Collaborating of other pathogens for humans has Center), Medical School, University of Crete, 1. de la Fuente J, Atkinson MW, Naranjo Heraklion, Greece; email: annapsa@med. been proposed (10). Whether these V, Fernandez de Mera IG, Mangold AJ, uoc.gr pathogens include A. ovis is unknown. Keating KA, et al. Sequence analysis of Thus, laboratory testing of human the msp4 gene of Anaplasma ovis strains. blood samples should include univer- Vet Microbiol. 2007;119:375–81. DOI: 10.1016/j.vetmic.2006.09.011 sal primers against all Anaplasma spp. 2. Chochlakis D, Koliou M, Ioannou I, to avoid missing cases such as the one Tselentis Y, Psaroulaki A. Kawasaki disease we report. and Anaplasma sp. infection of an infant in Cyprus. Int J Infect Dis. 2009;13:e71–3. Dimosthenis Chochlakis, DOI: 10.1016/j.ijid.2008.08.001 3. Psaroulaki A, Koliou M, Chochlakis D, Ioannis Ioannou, Ioannou I, Mazeri S, Tselentis Y. Anaplas- Yannis Tselentis, ma phagocytophilum infection in a child. and Anna Psaroulaki Pediatr Infect Dis J. 2008;27:664–6. DOI: Author affi liations: University of Crete, Her- 10.1097/INF.0b013e31816a0606 4. Chochlakis D, Ioannou I, Sharif L, Kok- aklion, Greece (D. Chochlakis, Y. Tselentis, kini S, Hristophi N, Dimitriou T, et al. A. Psaroulaki); and Veterinary Services, Prevalence of Anaplasma sp. in goats and Nicosia, Cyprus (I. Ioannou) sheep in Cyprus. Vector Borne Zoonot- ic Dis. 2009;9:457–63. DOI: 10.1089/ DOI: 10.3201/eid1606.090175 vbz.2008.0019

1032 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 16, No. 6, June 2010