PIROPLASMS IN FREE-RANGING BOBCATS AND COUGARS IN THE UNITED STATES: DISTRIBUTION, PREVALENCE, AND INTRASPECIFIC VARIATION BARBARA C.

Transcription

1 PIROPLASMS IN FREE-RANGING BOBCATS AND COUGARS IN THE UNITED STATES: DISTRIBUTION, PREVALENCE, AND INTRASPECIFIC VARIATION by BARBARA C. SHOCK (Under the Direction of Michael Yabsley) ABSTRACT Cytauxzoon felis, a tick-borne protozoal parasite (family Theileridae), is the causative agent of cytauxzoonosis in domestic cats in the United States. The parasite was first identified in domestic cats from Missouri, Texas, and Arkansas in the 1970 s and is now common in the southeastern and Midwestern US. The bobcat (Lynx rufus) has been identified as the natural reservoir of C. felis. The overall goal of this project was to better understand the natural history of C. felis in bobcats and cougars (Puma concolor). Based on PCR testing of >700 wild felids, infected bobcats were identified in numerous eastern and central states, with higher prevalence rates being found in states where Amblyomma americanum is present in addition to Dermancentor variabilis. Sequence analysis of internal transcribed spacer (ITS)-1 and ITS-2 regions revealed extensive genetic variation. Interestingly, one bobcat was infected with a Babesia sp. which is the first report in a bobcat. INDEX WORDS: Cytauxzoon felis, Babesia, Lynx rufus, Puma concolor, piroplasm

2 PIROPLASMS IN FREE-RANGING BOBCATS AND COUGARS IN THE UNITED STATES: DISTRIBUTION, PREVALENCE, AND INTRASPECIFIC VARIATION by BARBARA C. SHOCK B. S., West Virginia University, 2008 A Thesis Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE ATHENS, GEORGIA 2010

3 2010 Barbara C. Shock All Rights Reserved

4 PIROPLASMS IN FREE-RANGING BOBCATS AND COUGARS IN THE UNITED STATES: DISTRIBUTION, PREVALENCE, AND INTRASPECIFIC VARIATION by BARBARA C. SHOCK Major Professor: Committee: Michael J. Yabsley Fred Quinn David S. Peterson Electronic Version Approved: Maureen Grasso Dean of the Graduate School The University of Georgia August 2010

5 DEDICATION I would like to dedicate this manuscript to my grandmothers, Betty Fern Weaver and Jean Shock. Your support and love have been unending. iv

6 ACKNOWLEDGEMENTS First I would like to thank Dr. Michael John Yabsley for being my major professor. I have learned so much in the past two years from you. You are an engaging and entertaining researcher, and I just hope that I can someday read a paper before you have already. I look forward to working with you in the future. My biggest thanks goes to my parents, Mark and Linda Shock for your support and love. It s always helpful to call home and have someone tell you they re proud of you. I would also like to especially thank my dad for helping me to collect samples for this project and telling me about SCWDS. Although I have missed my family and friends from home, they have always shown me the utmost support and love (this means you Chelsie). I have to give a special thank you to Jenn and Todd Stueckle for first taking me on in their lab as an undergrad and now for being my friends. There are so many people at SCWDS that I am grateful to (and for), but I prefer to thank you in person. In addition, I d like to thank numerous personnel from state agencies who collected felid samples. The studies in this thesis were primarily funded by the Morris Animal Foundation (DO8FE-003). Additional support was provided by the Federal Aid to Wildlife Restoration Act (50 Stat. 917) and through sponsorship from fish and wildlife agencies in Alabama, Arkansas, Florida, Georgia, Kansas, Kentucky, Louisiana, Maryland, Mississippi, Missouri, North Carolina, Oklahoma, Puerto Rico, South Carolina, Tennessee, Virginia, and West Virginia. v

7 TABLE OF CONTENTS Page ACKNOWLEDGEMENTS...v LIST OF TABLES... vii LIST OF FIGURES... viii CHAPTER 1 INTRODUCTION and LITERATURE REVIEW...1 Introduction...1 Literature Review: Tick-borne protozoa of felids...2 Literature Cited DISTRIBUTION AND PREVALENCE OF CYTAUXZOON FELIS IN BOBCATS, LYNX RUFUS, FROM THIRTEEN STATES EXTENSIVE GENETIC VARIABILITY OF CYTAUXZOON FELIS FROM BOBCATS (LYNX RUFUS) AND COUGARS (PUMA CONCOLOR) NOVEL BABESIA IN A BOBCAT, GA CONCLUSIONS...87 vi

8 LIST OF TABLES Page Table 2.1: PREVALENCE OF C. FELIS IN BOBCATS...63 vii

9 LIST OF FIGURES Page Figure 1.1: PHYLOGENETIC RELATIONSHIPS BETWEEN GLOBAL SAMPLES OF CYTAUXZOON FROM FELIDS...4 Figure 1.2: APPROXIMATE DISTRIBUTION OF DERMACENTOR VARIABILIS...21 Figure 1.3: APPROXIMATE DISTRIBUTION OF AMBLYOMMA AMERICANUM...21 Figure 2.1: DISTRIBUTION OF C. FELIS IN BOBCATS...64 viii

10 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW INTRODUCTION Vector-borne pathogens are a significant cause of morbidity and mortality among people and animals worldwide. Ticks are second only to mosquitoes as vectors of pathogens which include a wide range of viruses, bacteria, and protozoa as well as some contribution to fungal infection. Currently, there are three recognized genera of tick-borne protozoans, Theileria, Cytauxzoon, and Babesia. The three are considered piroplasms because of their signet ring form in erythrocytes. Babesia is the most numerous of the three and can be found all over the world in mammals and birds. Theileria and Cytauxzoon are very closely related, and both also have a global distribution. Infection with the piroplasms causes clinical signs which range from subclinical parasitemia to mortality, depending on the type and strain of the parasite as well as the host immune function. Among felids, only Cytauxzoon and Babesia have been reported. Cytauxzoon felis and Babesia felis are emerging significant pathogens in the United States and Africa, respectively. Numerous other infections of felids with Cytauxzoon and Babesia spp. have been reported from a variety of wild and exotic worldwide, but little work has been done to fully characterize these parasites. In the United States, the reservoir species for C. felis is the North American bobcat (Lynx rufus). Bobcats have been documented dying from infection with C. felis, but unlike the high rates of mortality seen in domestic cats, most infected bobcats are subclinical carriers who show little to no signs of infection. Due to the high prevalences of C. felis previously reported in 1

11 bobcats and the general lack of clinical signs, it is assumed that C. felis has been endemic in the bobcat population for a long time. The overall goal of this project is to better understand the natural history of C. felis in its wild felid reservoir. This goal had two focuses: quantifying the distribution and prevalence of C. felis in wild felid populations; and characterizing the strains of C. felis which are circulating in wild felids. LITERATURE REVIEW: TICK-BORNE PROTOZOA OF FELIDS Genera Cytauxzoon and Theileria Cytauxzoon and Theileria species are two closely related tick-transmitted protozoan parasites in the phylum Apicomplexa, class Aconoidasida, order Piroplasmorida, and family Theilieridae (Kocan and Waldrup, 2001). Theileria are found in a wide range of hosts, primarily ruminants, while Cytauxzoon is restricted to felids (Nijhof et al., 2005). Cytauxzoon are distinguished from Theileria by the location of schizogony, with the cytauxzoons in felids replicating in mononuclear phagocytes while theilerias in ungulates replicate in either lymphocytes or mononuclear phagocytes (Ferris, 1997; Cowell et al., 1988; Preston et al., 1999; Nijhof et al., 2005). Transmission of Cytauxzoon and Theileria occurs when sporozoites are transferred to a vertebrate host during tick feeding. Experimental trials indicate transmission occurs 3-5 days after tick-feeding commences (Hazen-Karr et al., 1987). Within the vertebrate host, sporozoites first infect endothelial macrophages and undergo merogeny (Kocan et al., 1992). Merozoites are released within 14 days, which then infect erythrocytes to form intraerythrocytic piroplasms. These piroplasms are pleomorphic and can be round, oval, bipolar or rod-shaped and range from 0.3 to 2.0 μm in diameter. Rarely, Maltese crosses and paired 2

12 piroplasms may be observed (Simpson et al., 1985; Kier et al., 1987; Kocan et al., 1992). Transmission is completed when infected erythrocytes are ingested by a tick during feeding. Although not well described, it is believed that sporogony occurs in the tick vector and that the parasites migrate to the tick salivary glands prior to transmission. Experimental studies have shown that the parasites are maintained transstadially (Blouin et al., 1984; Blouin et al., 1987; Kocan et al., 1988). The genus Cytauxzoon was first described in a gray duiker (Sylvicaprae grimmia) from South Africa (Nietz and Thomas, 1948). Although numerous Theileria spp. had been described before, the new genus was proposed because the parasite replicated in mononuclear phagocytes, which had not been observed at that time in Theileria. The division of these parasites into separate genera was a source of contention between Levine, who had classified the cytauxzoons with the theilierias, and Brocklesby, who supported the separate genera based on the distinct cell lineages used in schizogony (Brocklesby, 1979; Levine et al., 1980). Several other Cytauxzoon spp. have been reported from African ungulates including: kudu, Tragelaphus strepsiceros; eland, Taurotraugs oryx; and giraffe, Giraffa camelopardalis; however, these parasites are now considered to be Theileria spp. (Neitz, 1964; Brocklesby, 1962; McCully et al., 1970; Nijhof et al., 2005). Clinical morbidity and mortality ascribed to Cytauxzoon spp., but more likely Theileria spp., has been reported in: tsessebe, Damaliscus lunatus; roan antelope, Hippotragus equinus; sable antelope, H. niger; and suspected in an impala, Aepyceros melampus (Jardine, 1992; Bigalke, 1989; Wilson et al., 1974; Wilson et al., 1977; Carmichael and Hobday, 1975). Currently Cytauxzoon spp. have been reported from domestic cats and wild felids in seven countries including the United States, Brazil, Germany, France, Mongolia, Spain, and Zimbabwe (Foggin and Roberts, 1982; Ketz-Riley et al., 2003; Mendes-de-Almeida et al., 2004; 3

13 Reichard et al., 2005; Luaces et al., 2005; Millan et al., 2007; Peixoto et al., 2007; Andre et al., 2009; Criado-Fornelio et al., 2009). Molecular characterization suggest that multiple species of Cytauxzoon infect felid species worldwide (Figure 1.1) Spain Iberian Lynx EF Spain Iberian Lynx EF France Domestic cat EU Spain Iberian Lynx EF Spain Domestic cat AY Spain Iberian Lynx AY Mongolia Pallas cat AF Mongolia Pallas cat AY Mongolia Pallas cat AY Stabilate 153 (MO) L19080 Texas Domestic cat AY Oklahoma Domestic cat AF Theileria youngi Woodrat AF Babesia sp. cougar DQ FIGURE 1.1. Phylogenetic relationships between global samples of Cytauxzoon from felids. Sequences obtained from GenBank. Genus Babesia Babesia spp. are small piroplasms that have been detected in numerous mammalian hosts, including several wild felid species. Babesia spp. are in the phylum Apicomplexa, class Aconoidasida, order Piroplasmorida, and family Babesiidae. Similar to the Theileridae, Babesia spp. are obligate intraerythrocytic protozoan parasites and are transmitted by Ixodid ticks. They are distinguished from members of the Theileridae by the absence of tissue schizogenous stage. Additionally Babesia spp. can also be transmitted transovarially in ticks while Theileridae are 4

14 only transstadially transmitted between stages of ticks. Historically, Babesia are divided into two morphologic groups, the large Babesia, which measure more than 2.5 μm, and the small Babesia which measure 1.0 to 2.5 μm in size. Because of morphologic similarity among the large and small piroplasms, genetic characterization is the best method to identify species (Kocan and Waldrup, 2001). A number of Babesia spp. have been recognized in felids from around the world. Large Babesia include B. herpailuri and B. pantherae from wild felids in Africa and small Babesia include B. felis and B. leo from domestic cats and other felids in Africa, B. cati in domestic cats from India, B. canis canis from domestic cats in Spain, B. canis presentii from domestic cats in Israel; a Babesia sp. from domestic cats in Portugal (called Theileria annae), and a Babesia sp. from Florida panthers in the United States (Jacobson et al., 2000; Criado-Fornelio et al., 2003; Penzhorn et al., 2004; Criado-Fornelio et al., 2004; Baneth et al., 2004; Yabsley et al., 2006). Domestic feline babesiosis is generally a mild chronic disease, although the pathogenicity depends on the species or sometimes the strain of the parasite (Penzhorn, 2006). The most common clinical signs include anorexia and lethargy. When disease is present, Babesia spp. generally cause hemolytic anemia, but most felids do not develop clinical disease unless presented with a secondary or immunocompromising infection or stressor (Shoeman et al., 2001). However, exceptions occur as B. felis is a significant pathogen of cats in South Africa (Penzhorn et al., 2004). 5

15 Cytauxzoon spp. in felid species from the United States Domestic cats Cytauxzoonosis was first recognized in domestic cats (Felis domesticus) in Missouri, Arkansas, and Texas during the early 1970 s by the presence of piroplasm-infected erythrocytes and microscopic lesions (schizont-laden cells lining the walls of blood vessels) as well as piroplasm-infected erythrocytes (Bendele et al., 1976; Wagner, 1976; Wightman et al., 1977). A similar syndrome had been previously recognized in African ungulates for many years (Neitz and Thomas, 1948; Neitz, 1957; Martin and Brocklesby, 1960; McCully et al., 1970; Wilson et al., 1974); thus these feline infections represented the first reports of a Cytauxzoon-like illness in the United States as well as the first report in a felid. Based on previous research involving Theileria and Cytauxzoon, the parasite was presumed to be tick-transmitted. The parasite, Cytauxzoon felis, has now been described in domestic cats from Alabama, Arkansas, Florida, Georgia, Kansas, Kentucky, Louisiana, Mississippi, Missouri, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia (Bendele et al., 1976; Wagner, 1976, Wightman et al., 1977; Ferris, 1979; Glenn and Stair, 1982; Hauck, 1982; Kocan and Kocan, 1991; Meier and Moore, 2000; Birkenheuer et al., 2006; Haber et al., 2007). Bobcats Piroplasms morphologically consistent with C. felis was described in bobcats as early as 1930 (Wenyon and Hamerton, 1930) and were again observed in two free-ranging bobcats when C. felis was first recognized in domestic cats (Wagner, 1976). To better characterize this piroplasms, researchers at the University of Missouri, College of Veterinary Medicine tested the ability of C. felis to infect 30 different domestic, laboratory, and wildlife species (Kier et al., 6

16 1982a). Inoculation of blood, spleen and lymph node or tissue homogenate from experimentally infected domestic cats into laboratory mice (Mus musculus, ICR), immunosuppressed nude laboratory mice (M. musculus, BALB/c-nu), rats (Rattus norvegicus), gerbils (Meriones unguiculatus), hamsters (Mesocricetus auratus), guinea pigs (Cavia porcellus), a chinchilla (Chinchilla laniger), rabbits (Oryctolaugs cuniculus), squirrel monkeys (Saimiri sciureus), a dog (Canis familiaris), a cow (Bos Taurus), a goat (Capra hircus), a sheep (Ovis aries), a pig (Sus scrofa), coyotes (Canis latrans), a fox (Vulpes fulva), skunks (Mephitis mephitis), raccoons (Procyon lotor), a woodchuck (Marmota monax), a marmot (Marmota flaviventris), opossums (Didelphis marsupialis), ground squirrels (Citellus tridecemlineatus), a grey squirrel (Sciurus carolinensis), a vole (Microtus ochrogaster), white-footed mice (Peromyscus maniculatus), bats (Myotis lucifugus), cottontail rabbit (Sylvilagus floridanus), a deer (Odocoileus virginianus), an ocelot (Felis pardalis), a mountain lion (P. concolor), and bobcats resulted in infections only in domestic sheep and the two bobcats. It is unknown why the domestic cats, ocelot, and mountain lion failed to develop infections. The inoculated sheep developed a low but persistent parasitemia, but no clinical signs of disease. However, domestic cats inoculated with tissue homogenates from the sheep did not develop cytauxzoonosis. Numerous studies have established that bobcats are experimentally and naturally susceptible to infection with C. felis. In addition, variability in disease progression in experientially infected bobcats has been observed that may be related to infectious dose, route, or stage or subspecies of bobcat, a combination of both, or some other factor (Kier et al., 1982b; Kier et al., 1983). The first experimental inoculation trial gave very different results for two different bobcats (Kier et al., 1982a; Kier et al., 1982b). One bobcat from Florida (here termed Lynx rufus floridanus), developed clinical signs of cytauxzoonosis including anorexia, 7

17 depression, and parasitemia, and died two weeks post-inoculation. Histological examination of tissues revealed large numbers of schizonts occluding vessels. The second bobcat (here termed Lynx rufus rufus) developed a subclinical parasitemia. The eastern bobcat, which survived the initial infection with C. felis, was inoculated four subsequent times, none of which resulted in clinical signs. However, the bobcat did develop transiently elevated infected red blood cell (IRBC) counts and after splenectomy and steroid treatment, the bobcat developed moderate anemia, reticulocytosis, and elevated IRBC counts. This bobcat maintained a subclinical parasitemia for 1,372 days (approximately 4 years) before dying of congestive heart failure. Upon necropsy, no schizogenous states of C. felis were observed and the final parasitemia was 8.5% (Kier et al., 1982b). Domestic cats inoculated with blood from these two bobcats also presented with variable disease: cats inoculated with blood from the surviving eastern bobcat survived while cats inoculated with blood from the Florida bobcat died. This study had three important findings: 1) bobcats can develop disease, 2) bobcats can have long-term parasitemias, and 3) domestic cats can survive experimental infection. Simultaneously, a survey of blood smears from bobcats from Oklahoma showed that 13 of 21 (62%) were positive for Cytauxzoon-like piroplasms (Glenn et al., 1982). The average parasitema observed was 1-3%, but ranged from 0.5% to 5%. Of the 13 positive bobcats, one exhibited anemia, a classic sign of cytauxzoonosis in the domestic cat. A subsequent study collected blood from four wild-caught bobcats which had a naturally occurring infection with an intraerythrocytic piroplasm and showed that the parasites caused fatal cytauxzoonosis in a single domestic cat (Glenn et al., 1983). Three other inoculated cats developed subclinical infections (Glenn et al., 1983). A single cat that was subsequently inoculated with a domestic cat-origin virulent inoculum of C. felis developed fatal cytauxzoonosis within 14 days. Two of the original 8

18 wild-caught bobcats were also inoculated with the same strain of C. felis as the domestic cat, but they did not develop clinical disease. Histologically, schizogenous forms of C. felis were not observed in any tissues from the bobcats or from the two subclinically infected domestic cats. Examination of 10 naturally-infected bobcats revealed that schizonts of C. felis were absent from the liver, lungs, spleen, and lymph nodes. Spleen homogenates from four of the bobcats were inoculated into domestic cats. Schizonts of C. felis were not seen in any tissues of the naturally infected free-ranging bobcats. Additionally, the domestic cats inoculated with the spleen homogenate did not develop clinical signs of cytauxzoonosis, but did develop subclinical parasitemias. Later the same group studied the clinical progression of C. felis in bobcats following exposure to infected D. variabilis (Blouin et al., 1987). In the second part of the study (Blouin et al., 1987), the researchers splenectomized a naturally infected bobcat which had a previous parasitemia of 1%. After the splenectomy, a spleen homogenate was inoculated into a domestic cat. D. variabilis nymphs were allowed to feed on this bobcat. Two bobcats (no. 2, 3) which had been determined to be uninfected with C. felis were parasitized by the infected ticks. Prescapular lymph nodes were removed from the two bobcats at 11 days post-exposure for impression smears and inoculation into domestic cats. A second prescapular lymph node was removed from bobcat 3 after 30 days post-exposure, and smears and inoculation were performed. The first, splenectomized, bobcat displayed no clinical signs of C. felis and schizogenous stages of C. felis were not observed in a histological examination of the spleen. Schizogenous stages were observed in reticuloendothelial macrophages in the 11 day post-tick attachment lymph node impression smears of bobcats 2 and 3 and the domestic cats which were inoculate with these tissue homogenates died of acute cytauxzoonosis 11 and 14 days post-inoculation respectively. Interestingly, bobcat 2 developed 9

19 clinical signs of cytauxzoonosis 19 days post-tick attachment and died. Gross and microscopic lesions revealed classical signs of cytauxzoonosis. The lymph node impression smear from bobcat 3 which was made 30-days post-tick exposure revealed no schizogenous stages of C. felis and the domestic cat inoculated with this tissue homogenate developed a parasitemia of 2-3% but showed no clinical signs of C. felis. Gross and microscopic evaluation of bobcat 3 after euthanasia 60 days post-tick attachment revealed no schizogenous stages of C. felis and only a 8% erythrocytic parasitemia. This paper reveals that development of the schizogenous phase in bobcats can be initiated by sporozoites transmission from the tick vector. It is also evidence that, in general, schizogenous development of C. felis in bobcats is limited and the erythrocytic stage is maintained and dominant. Due to the death of one bobcat (no. 2) in this study from ticktransmitted C. felis, and a previous experimental death of a bobcat (Kier et al., 1982a; Kier et al., 1982b), it can be assumed that while most bobcats in the wild probably have limited schizogenous stages, some may die of naturally acquired C. felis infections. Although a few experimental studies (Kier et al., 1982b; Blouin et al., 1984) indicated that bobcats could develop clinical disease, evidence for natural mortality is limited. Only a single case has been reported (Nietfeld and Pollock, 2002). In 2000, a moribund free-ranging bobcat kitten collected from an open field near Wamego, KS had severe anemia and respiratory difficulty (Nietfeld and Pollock, 2002). Gross examination revealed multifocal petechiae, splenomegaly, and pericardial effusion. Histological examination revealed subacute pulmonary thrombosis, mild vasculitis in the brain, and schizont-filled macrophages occluding the blood vessels of all examined tissues. The first study to examine the prevalence of C. felis in bobcat populations based on PCR testing was Birkenheuer et al. (2008) who tested bobcats from North Carolina and Pennsylvania. 10

20 Currently, C. felis has not been reported from domestic cats in Pennsylvania, but has been reported numerous times from domestic cats in North Carolina (Birkenheuer et al., 2006a). Based on PCR, samples were collected from legally-trapped bobcats in 33% of 32 bobcats from North Carolina and 7% of 70 bobcats from Pennsylvania were positive for C. felis. This represented the first report of C. felis from any felid in Pennsylvania and the authors alerted veterinary practitioners to the possibility of C. felis infection in domestic cats in that state. The high prevalence and low pathogenicity of C. felis for bobcats suggests that C. felis has been endemic in the populations for some time. As no Babesia or Theilieria spp. have yet been described in bobcats, it can be assumed the piroplasm described in a bobcat in 1930 by Wenyon and Hamerton was in fact C. felis and that the infection has been endemic in the population since at least the early 1900s. As the distribution of A. americanum and D. variabilis changes over time, it is probable that naïve populations of bobcats will become exposed to this parasite. As Birkenheuer et al. (2008) stressed, further studies are warranted to better understand the natural history and epidemiology of C. felis in bobcat populations. Florida panthers The first report of C. felis in a Florida panther (Puma concolor coryi or Felis concolor coryi) was in The detection was accidental and occurred when researchers with the Florida Game and Fresh Water Fish Commission and Cornell University unknowingly infected a domestic cat with C. felis during a study to determine if a 3 yr-old female panther had feline immunodeficiency virus (FIV) (Butt et al., 1991). The researchers intraperitoneally inoculated a FIV antigen and antibody-negative adult domestic cat with 1.5x10 6 mononuclear cells from the panther and 11 days later the cat became depressed and febrile. Although supportive care was 11

21 administered, the cat died the next day and upon necropsy, gross lesions such as petechial hemorrhages and splenic enlargement were noted. Microscopic examination of tissues revealed schizont-filled mononuclear cells which completely occluding vessels. Examination of blood smears indicated that the cat had a 10% parasitemia (Butt et al., 1991). Retrospective analysis of blood smears from the infected Florida panther revealed she was infected; however, the panther had received two blood transfusions from other P. concolor donors, so it was unclear if this infection was natural or iatrogenic. A subsequent study was conducted to determine the prevalence of C. felis in the Florida panthers and introduced Texas cougars (Puma concolor stanleyana) (Rotstein et al., 1999). The Texas cougars included in the study were translocated into Florida to introduce additional genetic variability and to increase the population size, which at that point was estimated to be ~50 panthers (Maehr, 1997). A retrospective analysis ( ) of blood smears revealed that 39% of the Texas cougars (11/28) and 35% of the Florida panthers (22/63) were infected with C. felis. Interestingly, a 7 day-old kitten was positive for C. felis. No difference in prevalence was noted between the sexes. Infected Florida panthers had significantly lower mean cell hemoglobin and monocytes counts and significantly higher neutrophil and eosinophil counts compared with infected Texas cougars. However, all the values were within the normal range of expected values so the authors concluded that there was no biological significance associated with the differences (Rostein et al., 1999). These data suggested that the Florida panther may be a reservoir for C. felis. To better understand the role of Florida panthers as a potential reservoir, a PCR-based survey was conducted (Yabsley et al., 2006). Amplification of a conserved segment of the 18S rrna gene revealed that 39 of 41 panthers were infected, which was significantly more than 12

22 indicated by blood smear analysis (only 15 infected). Surprisingly, after sequence analysis, the majority of the panthers (95%, 32 of 39 infected panthers) were infected with a novel Babesia sp. and only seven were infected with C. felis (Yabsley et al., 2006). Two of the cougars were coinfected with C. felis and the Babesia sp. This represented the first report of a Babesia sp. in a felid from North America. Because two morphologically similar piroplasms had been reported in the Florida panther, the conclusions of Rotstein et al. (1991) were re-examined. Although, fatal cytauxzoonosis has not been described in a Florida panther, Harvey et al. (2007) described three cases of cougars that had acute C. felis infections that resulted in mild clinical signs and mild hematologic and biochemical abnormalities. Of the three case reports in Harvey et al. (2007), one cougar demonstrated clinical signs after infection (year 1989) including anorexia, depression, and dehydration. This cougar was also infested with A. americanum, a confirmed vector of the C. felis (Reichard et al., 2009). Sequential testing of blood indicated that the cougar remained infected for at least one year while in the Florida facility and was still infected in 2005 when it was euthanized due to diabetes at her new home in Savannah, Georgia. It is unknown if she maintained the same subclinical C. felis infection for at >16 years or if she acquired a new infection while in Georgia. The other two cases were western cougars that were part of the genetic restoration study for the Florida panthers. These cougars were housed in northern Florida in 1995 before their release into southern Florida and were negative for C. felis (and Babesia sp.) during the initial 3-week quarantine period. After their third week at the facility, the cougars developed anemia and had become infected with C. felis. Neither cougar showed clinical signs and both were released in southern Florida. The cougars died in 1999 and 2001 of illnesses unrelated to cytauxzoonosis, although both were PCR positive for C. felis at the time of death. 13

23 These data indicated that during acute infections with C. felis, cougars can develop mild hemolytic anemia as well as liver injury, but they to survive and develop subclinical infections. Exotic felids Acute fatal cytauxzoonosis has been reported in a white tiger (P. tigris) (Garner et al., 1996) that was housed at the same facility which housed many of the infected cougars from previous studies (Butt et al., 1991; Rotstein et al.,1991; Harvey et al., 2007). The 7-year-old female had a two-day history of anorexia and lethargy and on the third day of illness, two female A. americanum were found and removed. She subsequently developed icterus, a low hematocrit (26%), and was mildly dehydrated. Two days later, she was recumbent and developed petechiae and profuse bleeding at puncture sites. She also had a parasitemia of 5% and a mild nonregenerative anemia, moderate leucopenia, neutropenia, lymphopenia, and severe thrombocytopenia. Histologically, C. felis was present in mononuclear phagocytes which were found in large quantities occluding capillaries as well as small arteries and veins (Garner et al., 1996). Infection dynamics, clinical signs, and pathology of C. felis The classic clinical signs observed from C. felis in domestic cats begin with anemia and depression and are quickly followed by fever, dehydration, icterus, splenomegaly, and hepatomegaly. The pathognomonic signs of C. felis infection are erythrocyte hemolysis and occlusion of the lumen of blood vessels by large schizont-laden mononuclear phagocytes in the lungs, liver, lymph nodes, and spleen (Simpson et al., 1985; Kier et al., 1987; Kocan and Kocan, 1991; Kocan et al., 1992). 14

24 A large group of researchers from the University of Missouri, which had first identified the illness in domestic cats, and the Plum Island Animal Disease Center (PIADC) investigated experimental routes of infection in the domestic cat (Wagner et al., 1980). Experimental infection trials with 131 domestic cats were conducted with parenteral administration of either fresh or liquid nitrogen-frozen blood or tissue homogenates that were collected from domestic cats with acute cytauxzoonosis. Two cats were splenectomized to determine if splenectomization enhanced the infection. Two uninoculated cats were held with the inoculated cats until they died of cytauxzoonosis and then a further 60 days to determine if the parasite could be transmitted directly from cat to cat. Additionally, one cat was administered tissue homogenate via gastric lavage, while the same suspension was intraperitoneally inoculated into another cat to determine if the parasite could be obtained by ingestion. This study determined that domestic cats do not acquire C. felis via contact or ingestion of infectious tissues. This study also determined that the minimum infectious dose of C. felis for a domestic cat is a 0.25mL subcutaneous inoculation of a 1:10 dilution of frozen standardized spleen inoculum. The standard progression to acute cytauxzoonosis was between 17 and 20 days, with an average of 18.4 days. Of the 131 cats included in this study, 54% died of clinical cytauxzoonosis and 46% were euthanized due to acute cytauxzoonosis. The most frequent clinical signs were pyrexia, depression, anorexia and dehydration initially followed by fever, lethargy, severe dehydration, icterus then coma and death (Wagner et al., 1980). Concurrent with the Wagner et al., (1980) study, researchers at the PIADC experimentally infected >500 domestic cats and numerous domestic livestock species with C. felis to elucidate the relationship between this piroplasm and the organisms observed in African ungulates and to address the potential threat of this parasite to livestock in the United States 15

25 (Ferris, 1979). The study was discontinued after C. felis was proven to not be a threat to domestic livestock; however, these data significantly added to the knowledge of C. felis in the United States. Importantly, one of over 500 domestic cats that were infected in this larger study survived which was the first report of a domestic cat surviving infection with C. felis. Additionally, researchers concluded that C. felis can be experimentally transmitted by inoculation of as little as 0.2 ml of blood, splenic, hepatic, pulmonary and lymph node homogenates by any route of injection (eg. intravenous, intraperitoneal, subcutaneous, intradermal, etc). They confirmed that initial clinical signs, depression and anorexia, were typically seen 5-7 days post infection and that fever slowly rose to C and stays at that level an average of 3-5 days before dropping. Before death, pyrexia, anemia, icterus and dehydration as well as dyspnea were observed and death usually occurred one to two weeks after development of clinical signs. After necropsy, gross examination revealed splenomegaly and petechial hemorrhages on lymph nodes and lungs. Microscopic evaluation revealed schizogenous stages of the parasite in reticuloendothelial histocytic cells which occluded blood vessels. The lungs were the organ most heavily affected, followed by the spleen, and then the liver and lymph nodes. The highest parasitemia observed in this study was 4% with an average of 1%, but researchers in Missouri had reported a parasitemia as high as 25% (Ferris, 1979; Wagner et al., 1980). The pathology of experimental infections of domestic cats with C. felis was more completely described by Kier et al. (1987). Experimentally infected cats were sacrificed from day 1 to day 19 post-inoculation (PI) for histological evaluation. By day 16 PI cats were exhibiting classic clinical signs of infection. Parasitemias were first noted 10 days PI. The first schizogenous stage of C. felis was observed at day 12 PI and there was a significant correlation between levels of parasitemia with temperature rise, presence of tissue stages, and decreased 16

26 white blood cell counts. Recently, a retrospective study of the pulmonary histopathology of C. felis infection was conducted on 148 domestic cats from January 1995 to June 2005 in Oklahoma (Snider et al., 2010). Interstitial pneumonia was found to be moderate in most cases, numbers of alveolar macrophages and intra-alveolar hemorrhages were low, and in many cases, infiltrations of neutrophils were noted. Similar to previous studies, extensive vascular occlusion, one of the hallmark signs of acute cytauxzoonosis, was noted (Snider et al., 2010). Both in situ hybridization and immunohistochemical techniques have been used to study the pathogenesis of cytauxzoonosis in domestic cats (Susta et al., 2009). Using a riboprobe targeting the 18S rrna region of Babesia microti, C. felis-infected cells were most often observed in the pulmonary intravascular macrophages and alveolar macrophages and macrophages in the spleen, but ISH-positive cells were also seen in the kidneys, heart, and brain. Using immunohistochemistry with a monoclonal antibody (Mac387), C. felis-infected cells were found to be negative for calprotein which indicates a decrease in diapedesis which would provide more circulating parasites available to the tick. Immunohistochemistry of two proliferation markers, the proliferating cell nuclear antigen and p53, also showed that infected cells replicate more frequently. While C. felis is in the cytoplasm, it blocks the translocation of the p53 proapoptotic protein, to the nucleus which would prevent the cell from undergoing apoptosis (Susta et al., 2009). A single case of abortion and death due to cytauxzoonosis in a domestic cat from Georgia has been reported (Weisman et al., 2007). The 1-year old cat was in her 5 th or 6 th week of gestation when she aborted and died. Fetal tissue, including skeletal muscle, developing bone and bone marrow, and placenta were negative for piroplasms and schizogenous stages of C. felis. Although it is unknown if C. felis can be transmitted transplacentally, there is evidence for 17

27 transplacental transmission of T. equi and T. sergenti in horses and cows, respectively (Baek et al., 2003; Phipps and Otter, 2004; Allsopp et al., 2007). Diagnostic testing for C. felis Previously, the standard method of diagnosis for C. felis infection was blood smear analysis to detect the piroplasm infected red blood cells (Glenn et al., 1982; Kier et al., 1982a; Ferris, 1979). Recently, molecular methods have been used to detect infection with greater sensitivity. Polymerase chain reaction (PCR) can be used to detect low numbers of parasites and to characterize the piroplasms found (Birkenheuer et al., 2006; Brown et al., 2008). Serologic tests for C. felis have been investigated, but domestic cats typically have high numbers of intraerthrocytic forms that are easily observed during acute infections. With the increased recognition of chronically infected cats, however, serologic testing might be useful for population-based studies (Shindel et al., 1978; Cowell et al., 1988). Treatment and survival in C. felis Numerous treatment regimes have been tested including parvaquone, imidocarb, dimiazene aceturate, buparvaquone; however, imidocarb is currently the drug of choice (Green et al., 1999). Although several cats survived infection during these clinical trials (4 of hundreds), no treatment, even imidocarb, has been considered 100% effective and consistent in managing the protozoan infection. In 1987, Uilenberg et al. inoculated a domestic cat with C. felis and treated it with parvaquone. This cat was subsequently immune to the same inoculum. Another study used a combination of parvaquone and buparvaquone in 17 cats, and although one infected cat survived, another cat not treated also survived (Motzel and Wagner, 1990). Importantly, data 18

28 indicate that cats that are treated can remain parasitemic which might allow them to continue the life cycle by infecting ticks (e.g. three cats in Arkansas that were treated with imidocarb dipropionate remained parasitemic for 7 months, 15 months, and 29 months) (Brown et al., 2008). Historically, this disease was considered by be nearly uniformly fatal for domestic cats, but over the years there have been indications (during experiment studies or field-based studies) that some cats could survive and develop subclinical chronic infections (Ferris, 1979; Meinkoth et al, 2000; Brown et al., 2008; Haber et al., 2007). Increased recognition of subclinical infections suggests that either cats are adapting to the parasite, less virulent strains of the parasite are beginning to emerge in domestic cats, or better diagnostic assays are detecting these chronic asymptomatic cases. From 1997 to 1998, Meinkoth et al. (2000) reported 18 cats from Arkansas and Oklahoma which had survived natural infections with C. felis. Four of the cases were asymptomatic and were identified only after one of their housemates showed clinical signs of cytauxzoonosis; these cats had shown no clinical signs. Only one of the cats was treated with imidocarb and all the cats were still parasitemic after 154 days. A single cat tested after 6 yrs was still infected (from Walker and Cowell, 1995). Three other studies have reported a number of cats that either survived infection or were chronic carriers including two of 34 cats in North Carolina, two of 961 cats from Florida, three cats in Arkansas, and one of 75 cats from Tennessee (Birkenheuer et al., 2006; Haber et al., 2007; Brown et al. 2008). Importantly, these data suggest that domestic cats can develop long-term parasitemias which could allow them to serve as reservoirs for the parasite. 19

29 Vectors of Cytauxzoon felis In experimental studies, C. felis has been transmitted by two Ixodid tick species, Dermacentor variabilis, the American dog tick, and Amblyomma americanum, the lone star tick. D. variabilis is a moderate-sized tick (2-6 mm long) that has a three-host lifecycle involving small mammals for the larval and nymph stages and large mammals for the adult stage (Allan, 2001). D. variabilis has a distribution involving the entire Eastern United States, the Pacific coast, and parts of Idaho and Montana (Allan, 2001) and is primarily found in areas with forest undergrowth (Allan, 2001; Figure 1.2). Ticks of the genus Amblyomma are larger ticks (4-8mm), that tend to be catholic in their feeding habits, with three all stages feeding on a variety of different hosts. In general, the genus is restricted to areas of warmer temperature and high humidity for development of each life stage (Semtner et al., 1973; Koch and Dunn, 1980b). The distribution of A. americanum ranges primarily in the Southeastern and south-central United States, although new reports indicate range expansion (Allan, 2001, Figure 1.3) into coastal regions of Maine and other Northeastern states (Keirans and Lacombe, 1998). The preferred habitat of A. americanum is subclimax forests (Sonenshine and Levy, 1971). 20

30 FIGURE 1.2. Approximate distribution of Dermacentor variabilis (CDC, 2009) FIGURE 1.3. Approximate distribution of Amblyomma americanum (CDC, 2009) C. felis was first experimentally transmitted from a bobcat to a domestic cat by D. variabilis (Blouin et al., 1984). Lab-raised nymphs were fed to repletion on a splenectomized 21

31 bobcat with a parasitemia of 40%, allowed to molt to the adult stage, and then were fed on splenectomized domestic cats. Both domestic cats died of acute cytauxzoonosis at 13 and 17 days post-tick engorgement. Tissue impression smears displayed the schizogenous stage of C. felis and the domestic cats displayed clinical and microscopic evidence of infection with C. felis. Inoculation of a domestic cat with parasitemic blood from the same bobcat resulted in a subclinical infection which was maintained for 6 months. This study was the first to show that D. variabilis transstadially maintains C. felis and can transmit the parasite to felids. The study also showed that clinical outcome in domestic cats may be related to the route of infection because blood-inoculation (as was seen previously) caused subclinical infection whereas ticktransmission produced acute cytauxzoonosis (Blouin et al., 1984). D. variabilis also successfully transmitted C. felis infection from a splenectomized parasitemic bobcat to two un-splenectomized bobcats which were previously determined to be uninfected with C. felis by both blood smears and inoculation of whole blood into domestic cats (Blouin et al., 1987). One of the bobcats died 19 days post tick-attachment of clinical cytauxzoonosis which was the first indication that bobcats could develop clinical disease. The first survey of ticks for C. felis infection was conducted by Bondy et al. (2005) who screened 1, 362 ticks (Rhipicephalus sanguineus, D. variabilis, A. americanum) from Missouri. Unfortunately, all the ticks for this study were collected from domestic dogs and cats which complicated interpretation of the results. Of all the samples included in the study, only three (0.93%) A. americanum nymphs were PCR positive for C. felis; however, all three were collected from a domestic cat which was confirmed to be infected with C. felis. Although this study suggested a role of A. americanum in the transmission of C. felis to domestic cats, there were several caveats which prevented the A. americanum from being confirmed as a vector. 22

32 First, since the ticks were removed from a C. felis positive individual, it is impossible to determine if the felid acquired the infection from the A. americanum, if the ticks were positive because they were engorged with C. felis-infected feline blood, or if the vectors were positive because they had acquired the parasite from a previously infected felid. Secondly, clinical signs of infection usually do not develop for at least two weeks post-tick engorgement so the likelihood of one of these nymphs being the vector is very low (Bondy et al., 2005; Blouin et al., 1987; Blouin et al., 1984). Nevertheless, A. americanum was suspected to be a vector because the ubiquitous nature of the tick in regions where C. felis had been identified in domestic cats (Reichard et al., 2008) and epidemiologically, peaks of cytauxzoonosis in domestic cats in May and September correlated with natural peaks in A. americanum activity (Reichard et al., 2008). A. americanum was confirmed as a competent vector when Reichard et al., (2009) conducted a transmission trial using lab-raised A. americanum, D. variabilis, R. sanguineus, and Ixodes scapularis nymphs that had fed on a subclinically C. felis-infected cat. Only the cat infested with adult A. americanum exhibited clinical signs of cytauxzoonosis (11 dpi), while none of the other cats infested with adult D. variabilis, R. sanguineus, and I. scapularis became infected. Similarly, in a subsequent study D. variabilis failed to transmit C. felis (Edwards et al., 2010). Although D. variabilis did not transmit C. felis in these two studies, there are a number of differences between these studies. The parasitemia of the initial felids used in the studies (40% in Blouin et al., (1984) vs % in Reichard et al., (2008)) and the immune status of the subject felids (splenectomized vs. unsplenectomized). After A. americanum was confirmed to be a vector, a field-based study of wild-caught questing ticks from natural habitats surrounding Stillwater, Oklahoma was conducted. C. felis 23

33 was detected in A. americanum (MIRs of 0.5% (1 of 178) for adult males, 1.5% (3 of 197) for adult females, and 0.8% (3 of 393) for nymphs (Edwards et al., 2010). All 160 D. variabilis were negative. Currently, both D. variabilis and A. americanum are considered competent vectors of the parasite (Blouin et al. 1984; Blouin et al., 1987; Reichard et al., 2008); however, very little is understood about the importance of each vector species in the overall ecology of C. felis. Based on A. americanum densities in areas where cytauxzoonosis is common and the epidemiologic association with A. americanum activity, A. americanum likely represents the primary vector of C. felis. Seasonality and risk factors for C. felis Because cytauxzoonosis is a tick-borne disease, a marked seasonality in diagnosed cases has been noted. A retrospective analysis of cases submitted to the Oklahoma Animal Disease Diagnostic Laboratory (n=180) during and the Boren Veterinary Medical Teaching Hospital (n=52) during found a bimodal pattern with a large peak in the spring and early summer months of April, May and June, and a small peak in the autumn months of August and September (Reichard et al., 2009). These peaks correspond with the activity of the tick vectors, especially A. americanum. The authors were able to identify the geographic coordinates and landscape characteristics for 41 of the cases and 68.3% were reported to occur in low density residential areas. More cases (19.5%) occurred in urban edge habitat than expected at random. Significantly more cases of cytauxzoonosis were associated with wooded cover and proximity to natural or unmanaged areas. This study is a confirmation of observations clinicians and researchers had 24

34 been making for years about the risk factors associated with cytauxzoonosis in domestic cats. Basically, cats which reside in habitats that support tick vectors, bobcats or both are more likely to become infected with C. felis. Importantly, risk of C. felis infection is greatly reduced (nearly preventable) by limiting tick exposure by keeping cats indoors. Genetic characterization of Cytauxzoon felis from domestic cats There have been at least three reasons postulated for the increased recognition of chronically infected asymptomatic cats: 1) better treatment strategies, 2) better diagnostics, or 3) variable strains of C. felis that differ in their virulence for domestic cats. Although many treatment strategies have been attempted in reducing the parasitemia in domestic cats (Green et al., 1999; Motzel and Wagner, 1995), none have been consistently effective and recently there have been increasing reports of natural subclinical chronically-infected domestic cats (Birkenheuer et al., 2006; Brown et al., 2008). Although diagnostics, particularly molecular techniques, have improved, historically experimental studies have suggested that different strains of C. felis had variable pathogenicity for cats as some strains induced clinical disease while others induced subclinical chronic infections (Kier et al., 1982b). If genetic markers could be identified to identify clinically different strains, then the clinical outcome and treatment protocols might be more easily predicted (Brown et al., 2009a). One of the most commonly used markers for this type of analysis is the noncoding first and second internal transcribed spacer regions of the ribosomal RNA operon (ITS-1 and ITS-2). These targets are more likely to have genetic variability compared with the conserved regions of this operon (e.g., 18S, 5.8S, and 28S rrna) (Hills and Dixon, 1991). The ITS-1 and ITS-2 regions have been useful in examining variability among a related hemoparasite, Babesia canis, 25