Veterinary Parasitology 137 (2006) 144149 www.elsevier.

com/locate/vetpar

Development and evaluation of a PCR assay for the detection of Cytauxzoon felis DNA in feline blood samples
Adam J. Birkenheuer a,*, Henry Marr a, A. Rick Alleman b, Michael G. Levy c, Edward B. Breitschwerdt a
a

Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, 4700 Hillsborough Street, Raleigh, NC 27606, United States b Department of Physiological Science, College of Veterinary Medicine, University of Florida, Gainesville, FL 32610, United States c Department of Population Health and Pathobiology, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606, United States

Received 4 November 2005; received in revised form 7 December 2005; accepted 7 December 2005

Abstract Cytauxzoonosis is an emerging tick borne infectious disease of domestic cats in the United States, caused by the organism Cytauxzoon felis (C. felis). In naturally infected domestic cats the disease is almost always fatal. Currently there are no commercially available molecular or serologic tests to facilitate the antemortem diagnosis of C. felis infection. Clinical and pathological diagnosis of cytauxzoonosis is based on microscopic identication of parasites in tissues or on blood smears. We have developed and evaluated the sensitivity and specicity of a polymerase chain reaction (PCR) based assay for the diagnosis of C. felis infections in feline blood samples. The assay is sensitive enough to detect one copy of a cloned fragment of the C. felis 18S rRNA gene. This PCR assay can be used for the rapid clinical diagnosis of cytauxzoonosis and for epidemiological studies that will better dene the geographic distribution of C. felis infection in cats. # 2005 Elsevier B.V. All rights reserved.
Keywords: Cytauxzoonosis; Piroplasmosis; Pancytopenia

1. Introduction Cytauxzoon felis is an emerging tick-transmitted apicomplexan protozoal parasite that infects domestic
* Corresponding author. Tel.: +1 919 513 8288; fax: +1 919 513 6336. E-mail address: ajbirken@ncsu.edu (A.J. Birkenheuer).

cats (Felis catus). It is presumed to be transmitted to domestic cats via ticks from wild felids such as bobcats (Lynx rufus) and cougars (Felis concolor). In domestic cats cytauxzoonosis is characterized by fever, pancytopenia, multi-organ failure and often results in the death of affected cats (Criado Fornelio et al., 2004; Greene et al., 1999; Hoover et al., 1994; Meier and Moore, 2000; Meinkoth et al., 2000). Due

0304-4017/$ see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.vetpar.2005.12.007

A.J. Birkenheuer et al. / Veterinary Parasitology 137 (2006) 144149

145

to progressive multi-organ system failure, most domestic cats either die or are euthanized within 1 week of the onset of clinical signs. In contrast, chronic C. felis infection is better tolerated by wild felids in which fatal cytauxzoonosis appears to be a rare occurrence. In recent years, this devastating infection has been recognized with increased frequency in many areas of the United States including areas where C. felis has not been previously documented (Birkenheuer et al., in press). There are also several more recent reports of cats surviving C. felis infections (Greene et al., 1999; Meinkoth et al., 2000). It is unknown whether non-fatal C. felis infection in cats represents a change in the biological behavior of the organism (less pathogenic strains) or its tick vectors or the establishment of infection in new reservoir species. Despite the original disease description in cats from Missouri almost 30 years ago, little to no information exists about the epidemiology of C. felis (Wagner, 1976). The current cumulative C. felis knowledge base is derived from individual case reports, small case series and experimentally induced infections. The absence of user friendly, sensitive and specic tests for C. felis presents a substantial barrier for future epidemiological studies. The clinical diagnosis of cytauxzoonosis has relied on the light microscopic detection of parasites in erythrocytes or macrophages, and due to the low sensitivity of blood smear examinations, infection is frequently only documented at necropsy (Kier et al., 1977). Serologic testing for anti-Cytauxzoon antibodies has been described experimentally, but a serological test is not available commercially. In addition, perpetuation of an antigen supply requires infection of live cats since C. felis has never been cultured in vivo and recombinant diagnostic antigens have not been produced. Due to the acute onset and rapid clinical course of illness, anti-Cytauxzoon antibodies may not be detectable during the early stages of infection or at the time of death. Recently the polymerase chain reaction (PCR) has been used to amplify C. felis 18S rRNA gene sequences (Allsopp et al., 1994; Meinkoth et al., 2000). These broad-range PCR assays were designed to amplify 18S rRNA genes from nearly all piroplasms. Genus and species conrmation required DNA sequencing to conrm the identity of the parasite. A more recent study used PCR to amplify C. felis DNA in pooled tick samples, but the

investigators did not report the limit of detection for their assay (Bondy et al., 2005). As ticks would be expected to harbor a much higher concentration of C. felis than would be found in a feline blood sample, the limit of DNA detection becomes critical when a PCR assay is used diagnostically. Currently, no PCR assay is readily available to facilitate a clinical diagnosis of feline cytauxzoonosis by practitioners. The clinical diagnosis of C. felis infection based on the identication of merozoites in red blood cells has several potential limitations. Despite substantial differences in DNA sequences, the merozoites of hemoprotozoan parasites are often morphologically indistinguishable via light microscopy (Caccio et al., 2002; Kjemtrup et al., 2000; Zahler et al., 2000). For example, morphologically, C. felis merozoites are indistinguishable from Babesia felis merozoites. The use of PCR tests for the diagnostic detection of hemoprotozoan parasites is well documented and is more sensitive than light microscopic identication of parasites on stained blood smears (Almeria et al., 2001; Birkenheuer et al., 2003; Calder et al., 1996). In addition, the specicity of PCR facilitates the detection of genetic differences between morphologically indistinguishable organisms (Birkenheuer et al., 2003; Calder et al., 1996; Gubbels et al., 1999). Polymerase chain reaction based tests can detect parasitemias that are up to 1000-fold lower than the limit of detection that can be achieved by light microscopic observation of thin stained blood smears (Almeria et al., 2001; Birkenheuer et al., 2003; Krause et al., 1996). Due to the limitations of currently available diagnostic tests for C. felis we sought to develop a PCR assay for cytauxzoonosis. The development of a sensitive and specic PCR assay for the detection of C. felis in feline blood samples will be benecial for the clinical diagnosis of feline cytauxzoonosis and can be also used to study the epidemiology of C. felis in wild and domestic felids.

2. Materials and methods 2.1. Patient samples C. felis: Whole blood samples were collected from six cats from North Carolina, four cats from Florida, and two cats from Tennessee. By microscopy

146

A.J. Birkenheuer et al. / Veterinary Parasitology 137 (2006) 144149

merozoites were observed on the blood smears of all 12 cats, supporting infection with C. felis. Detailed clinical information other than outcome (survival versus non-survival) was not available. Cytauxzoonosis was fatal in 11 cats and 1 cat survived. DNA was extracted from 200 ml of anti-coagulated whole blood using QIAmp DNA blood mini kit (Qiagen, Valencia, CA) following the manufacturers protocol. 2.2. DNA samples The following organisms were used as template for the C. felis PCR assay to assess specicity: F. catus (n = 9; one specic pathogen free cat (SPF) and eight cats that had samples submitted for other vector-borne disease testing), Babesia canis vogeli, B. c. canis, B. c. rossi, B. gibsoni (Asian genotype), B. gibsoni (CA genotype), Theileria annae, E. ewingii, E. chafeensis, Neorickettsia risticii, Anaplasma phagocytophilum, A. platys, Mycoplasma haemofelis, M. haemominutm, and Leishmania infantum DNA was extracted from whole blood samples using a QIAmp DNA blood mini kit (Qiagen) following the manufacturers protocol. Bartonella henselae and Bartonella vinsonii subsp. berkhofi DNA was isolated from organisms cultured on blood agar plates using a QIAmp DNA blood mini kit (Qiagen). Briey, bacterial colonies were collected from the plates and suspended in 200 ml of sterile saline. This

bacterial suspension was used as the starting material for DNA extraction. Rickettsia rickettsii (Breitschwerdt et al., 1999) and Ehrlichia canis (Breitschwerdt et al., 1998) DNA was extracted from infected cell cultures using a DNeasy tissue kit according to the manufacturers protocol (Qiagen). These samples had all been characterized previously by PCR and or DNA sequencing. 2.3. PCR Using multiple sequence alignment, apicomplexan, mammalian, and C. felis 18S rRNA gene sequences were compared (Clustal W, Bioedit, Raleigh, NC). Primer sequences were selected to specically amplify a 284 bp fragment of the C. felis 18S rRNA gene (Fig. 1). The selected primer sequences were: 50 GCGAATCGCATTGCTTTATGCT-30 and 50 -CCAA TTGATACTCCGGAAAGAG-30 . Each reaction contained a 1 concentration of PCR Buffer II (Applied Biosystems, Foster City, CA), 1.25 U of Taq polymerase, 5 ml of DNA template, 1.5 mM MgCl2, 25 pmol of each primer, and 200 mM of each dNTP. The thermal cycling conditions consisted of an initial denaturation at 95 8C for 5 min, followed by 40 amplication cycles (95 8C for 45 s, 59 8C for 45 s, and 72 8C for 60 s), and a nal extension step at 72 8C for 5 min (PCR Express; Thermo Hybaid, Middlesex, UK). Partial glyceraldehyde-3-phosphode-

Fig. 1. Cytauxzoon felis DNA was amplied from cats with naturally occurring cytauxzoonosis, but not from other samples.

A.J. Birkenheuer et al. / Veterinary Parasitology 137 (2006) 144149

147

hydrogenase genes were amplied as previously described to rule out the presence of PCR inhibitors in each sample (Birkenheuer et al., 2003). The C. felisspecic amplicons were sequenced directly. Prior to direct sequencing of PCR products, amplicons were puried using the Qiaquick PCR purication kit (Qiagen) following the manufacturers protocol. Amplicons were sequenced bi-directionally using the C. felis-specic primers designed in this study. 2.4. Determination of sensitivity C. felis-specic amplicons were cloned into a plasmid vector (PCR 2.11, Invitrogen, Carlsbad, CA, USA) following the manufacturers protocol. Escherichia coli (TOP100 1, Invitrogen) were transformed and colonies containing inserts were selected using blue/white screening. Plasmid DNA was isolated using Qiagen plasmid mini prep kit (Qiagen) following the manufacturers protocol. Recombinant plasmid DNA was sequenced bi-directionally with using the primers, M13R-700 50 -CAGGAAACAGCTATGACCATG-30 and T7-800 50 -TAATACGACTCACTATAGGGCGA30 bi-directionally using an automated DNA sequencer (DNA Sequencing Facility, University of California, Davis, CA). Plasmid dilutions: Plasmid DNA containing the 284 bp fragment of the C. felis 18s rRNA gene was quantied using spectrophotometry and was serially diluted in water to concentrations of 1000, 100, 10, 1, 0.1, 0.01, and 0.001 copies/ml. To more closely mimic clinical conditions similar serial dilutions of the plasmid in non-infected SPF feline whole blood to concentrations of 1000, 100, 10, 1, 0.1, 0.01, and 0.001 copies/ml were also made prior to DNA isolation. These serially diluted mock-infected samples were used as template for the PCR assay for a minimum of 10 reactions for each dilution.

PCR reaction (Fig. 1). Amplicons were not produced when DNA samples from either blood agar or cell cultures or blood samples infected with the following organisms were used as template for the C. felis PCR assay: F. catus, B. c. vogeli, B. c. canis, B. c. rossi, B. gibsoni (Asian genotype), B. gibsoni (CA genotype), T. annae, B. henselae, B. vinsonii subsp. berkhofi, R. rickettsii, E. canis, E. ewingii, E. chafeensis, N. risticii, A. phagocytophilum, A. platys, M. haemofelis, M. haemominutm, and L. infantum (Fig. 1; data not shown for all organisms). The 284 bp product was amplied from all 12 C. felis naturally infected feline samples (Fig. 1; data not shown for all samples). The DNA sequence of the amplicons from the C. felis infected samples generated in this study was 100% identical to the C. felis sequences in Genbank (Accession numbers L19080, AY67905, AY531524, and AY39930). The presence of PCR inhibitors in samples that tested negative using the C. felis-specic PCR was excluded by either the amplication of mammalian GAPDH genes or specic amplication of the pathogen DNA that was used as template. 3.2. Sensitivity The PCR was able to detect C. felis DNA in samples spiked with as little as 0.01 gene copies/ml in both water and DNA isolated from feline whole blood spiked with plasmids containing the target sequence. The frequency of amplicon detection in the replicate reactions using the mock-infected feline blood samples as the DNA template is presented in Table 1. Consistent detection (100%) required 50 copies of the target molecule per reaction.

Table 1 Sensitivity of C. felis-specic PCR assay for the detection of 18S rRNA gene copies diluted in non-infected feline blood Gene copies per microliter 1000 100 10 1 0.1 0.01 0.001 Success rate (%) (positive PCR reactions/attempts) 100 100 100 39 20 20 0 (10/10) (22/22) (13/13) (7/18) (2/10) (2/10) (0/10)

3. Results 3.1. Specicity The PCR assay specically amplied a 284 base pair product from C. felis infected feline blood samples, and did not produce amplicons when noninfected feline DNA was used as a template for the

148

A.J. Birkenheuer et al. / Veterinary Parasitology 137 (2006) 144149

4. Discussion This report describes the rst C. felis-specic diagnostic PCR assay validated for use with feline blood samples. The C. felis-specic PCR assay described in this study was able to detect C. felis DNA in all 12 samples derived from naturally infected domestic cats from a wide geographic distribution, including 11 fatal and 1 non-fatal case. The assay did not produce C. felis amplicons from blood samples derived from non-infected cats or from samples containing other blood-borne pathogens. B. felis genetic material was not available for testing. However, based on comparisons with the DNA sequences submitted to Genbank, this assay should not amplify B. felis 18S rRNA genes. B. felis infections should be suspected in domestic cats with microscopically identiable merozoites that test negative with this C. felis-specic PCR assay. Merozoite positive, C. felis PCR negative patient samples should be subjected to specic PCR testing for babesiosis. The C. felis-specic PCR described in this study was determined to be very sensitive and was able to detect C. felis DNA in samples containing less than 1 gene copy/ml. The number of circulating organisms or target molecules, in conjunction with the detection limit of the specic PCR assay, are critically important determinants of the clinical diagnostic sensitivity of tests used to detect the presence of infectious organisms. This is especially applicable to PCR assays in which a negative test may not be useful in ruling out infection with a particular pathogen that can induce persistent low-level infection. The lower percentage of PCR positive test results obtained from samples with 1 or less gene copy/ml highlights the effect that sampling has on detection in patient samples containing low copy numbers. Since it is impossible to detect less than one copy per PCR reaction, the number of target molecules that are actually placed in the reaction dictate whether or not this particular sample will result in a positive result. The probability of a target molecule being placed into the reaction is determined by Poissonian statistics since the molecules in solution should have a Poisson distribution (Pfaf, 2004). For this assay we have determined that 50 copies per reaction are necessary for detection 100% of the time.

In addition to the excellent sensitivity and specicity provided by DNA based assays, the potential for a rapid turn around time will provide additional benets to the clinician and to the client who owns a cat infected with C. felis. As cytauxzoonosis is often fatal in domestic cats within 57 days of the onset of clinical signs, early aggressive treatment is believed to be important for survival (Greene et al., 1999). Therefore, rapid diagnosis is important to provide owners with an appropriate prognosis and to quickly initiate treatment of cats with cytauxzoonosis. In conclusion this PCR assay should enhance the clinical diagnosis of cytauxzoonosis and facilitate future epidemiological studies regarding the geographic distribution, reservoir potential and carriership of C. felis in wild and domestic felids. References
Allsopp, M., Cavalier Smith, T., Waal, D.T.d., Allsopp, B.A., De Waal, D.T., Smith, T.C., 1994. Phylogeny and evolution of the piroplasms. Parasitology 108, 147152. Almeria, S., Castella, J., Ferrer, D., Ortuno, A., Estrada Pena, A., Gutierrez, J.F., 2001. Bovine piroplasms in Minorca (Balearic Islands, Spain): a comparison of PCR-based and light microscopy detection. Vet. Parasitol. 99, 249259. Birkenheuer, A.J., Le, J.A., Valenzisi, A.M., Tucker, M.D., Levy, M.G., Breitschwerdt E.B. Cytauxzoon felis infections in cats in the mid-Atlantic States: 34 cases (19982004). J. Am. Vet. Med. Assoc., in press. Birkenheuer, A.J., Levy, M.G., Breitschwerdt, E.B., 2003. Development and evaluation of a seminested PCR for detection and differentiation of Babesia gibsoni (Asian genotype) and B. canis DNA in canine blood samples. J. Clin. Microbiol. 41, 41724177. Bondy Jr., P.J., Cohn, L.A., Tyler, J.W., Marsh, A.E., 2005. Polymerase chain reaction detection of Cytauxzoon felis from eldcollected ticks and sequence analysis of the small subunit and internal transcribed spacer 1 region of the ribosomal RNA gene. J. Parasitol. 91, 458461. Breitschwerdt, E.B., Hegarty, B.C., Hancock, S.I., 1998. Doxycycline hyclate treatment of experimental canine ehrlichiosis followed by challenge inoculation with two Ehrlichia canis strains. Antimicrob. Agents Chemother. 42, 362368. Breitschwerdt, E.B., Papich, M.G., Hegarty, B.C., Gilger, B., Hancock, S.I., Davidson, M.G., 1999. Efcacy of doxycycline, azithromycin, or trovaoxacin for treatment of experimental Rocky Mountain spotted fever in dogs. Antimicrob. Agents Chemother. 43, 813821. Caccio, S.M., Antunovic, B., Moretti, A., Mangili, V., Marinculic, A., Baric, R.R., Slemenda, S.B., Pieniazek, N.J., 2002. Molecular characterisation of Babesia canis canis and Babesia canis vogeli from naturally infected European dogs. Vet. Parasitol. 106, 285292.

A.J. Birkenheuer et al. / Veterinary Parasitology 137 (2006) 144149 Calder, J.A., Reddy, G.R., Chieves, L., Courtney, C.H., Littell, R., Livengood, J.R., Norval, R.A., Smith, C., Dame, J.B., 1996. Monitoring Babesia bovis infections in cattle by using PCRbased tests. J. Clin. Microbiol. 34, 27482755. Criado Fornelio, A., Gonzalez del Rio, M.A., Buling Sarana, A., Barba Carretero, J.C., 2004. The expanding universe of piroplasms. Vet. Parasitol. 119, 337345. Greene, C.E., Latimer, K., Hopper, E., Shoefer, G., Lower, K., Cullens, F., 1999. Administration of diminazene aceturate or imidocarb dipropionate for treatment of cytauxzoonosis in cats. J. Am. Vet. Med. Assoc. 215, 497500 482. Gubbels, J.M., de Vos, A.P., van der Weide, M., Viseras, J., Schouls, L.M., de Vries, E., Jongejan, F., 1999. Simultaneous detection of bovine Theileria and Babesia species by reverse line blot hybridization. J. Clin. Microbiol. 37, 17821789. Hoover, J.P., Walker, D.B., Hedges, J.D., 1994. Cytauxzoonosis in cats: eight cases (19851992). J. Am. Vet. Med. Assoc. 205, 455460. Kier, A.B., Wightman, S.R., Wagner, J.E., Morehouse, L.G., 1977. Diagnostic features of a newly described blood protozoan disease in domestic catsa Cytauxzoon species. In: Proceedings of the 20th Annual Meeting of the American Association of Veterinary Laboratory Diagnosticians. Minneapolis, Minnesota, 1618 October 123130; 8 ref. (Madison, WI, USA).

149

Kjemtrup, A.M., Kocan, A.A., Whitworth, L., Meinkoth, J., Birkenheuer, A.J., Cummings, J., Boudreaux, M.K., Stockham, S.L., Irizarry-Rovira, A., Conrad, P.A., 2000. There are at least three genetically distinct small piroplasms from dogs. Int. J. Parasitol. 30, 15011505. Krause, P.J., Telford 3rd, S., Spielman, A., Ryan, R., Magera, J., Rajan, T.V., Christianson, D., Alberghini, T.V., Bow, L., Persing, D., 1996. Comparison of PCR with blood smear and inoculation of small animals for diagnosis of Babesia microti parasitemia. J. Clin. Microbiol. 34, 27912794. Meier, H.T., Moore, L.E., 2000. Feline cytauxzoonosis: a case report and literature review. J. Am. Anim. Hosp. Assoc. 36, 493496. Meinkoth, J., Kocan, A.A., Whitworth, L., Murphy, G., Fox, J.C., Woods, J.P., 2000. Cats surviving natural infection with Cytauxzoon felis: 18 cases (19971998). J. Vet. Intern. Med. 14, 521 525. Pfaf, M., 2004. AZ of Quantitative PCR, fth ed. International University Line Publishers, La Jolla, CA, 912 pp. Wagner, J.E., 1976. A fatal cytauxzoonosis-like disease in cats. J. Am. Vet. Med. Assoc. 168, 585588. Zahler, M., Rnder, H., Zweygarth, E., Fukata, T., Maede, Y., Schein, E., Gothe, R., 2000. Babesia gibsoni of dogs from North America and Asia belong to different species. Parasitology 120, 365369.