INTRODUCTION
Toxoplasmosis, caused by the protozoan Toxoplasma gondii, is a worldwide zoonosis [Reference Dubey1]. In general its seroprevalence is very high in South America and low in Asia. Fragmentary reports indicate a high incidence of T. gondii infections in Africa [Reference Dubey1]. Toxoplasmosis is usually asymptomatic in immunocompetent adults, but can cause mortality in the very young and the immunocompromised. Many patients infected with human immunodeficiency virus (HIV) who are untreated die of toxoplasmosis. This is of particular concern in many African countries because of the high prevalence of HIV and lack of resources to manage it.
Humans and animals become infected with T. gondii mostly by ingesting uncooked meat of infected animals or by ingesting soil, food or water contaminated with oocysts. Cats are essential in the life cycle of T. gondii because they are the only hosts that can excrete the environmentally resistant oocysts in nature. Prevalence of T. gondii antibodies varies with age, lifestyle of the cat (stray vs. pet), the serological test utilized, the screening dilution, and other undefined factors. In general, infection in cats increases with age and the prevalence is higher in stray cats. Infected cats can shed millions of oocysts in a matter of a few days, and after sporulation oocysts can survive in the environment for months or even years depending on the moisture and ambient temperature. Cats are thought to become infected by ingesting infected prey, often soon after they begin to hunt. Cats usually shed oocysts only for a short time and once in their life. However, poor nutrition, concurrent infections, and immunosuppression may affect the immune status of the cat and lead to increased oocyst shedding.
Cats have been shown by culture or DNA amplification to be infected by a number of Bartonella spp. [Reference Brunt2–Reference Varanat7]. Bartonella henselae, B. clarridgeiae, and B. koehlerae are transmitted among cats by the cat flea, Ctenocephalides felis. B. henselae is the main aetiological agent associated with cat scratch disease in immunocompetent people, as well as bacillary angiomatosis and peliosis hepatis which are common disorders in people with AIDS. Bartonella spp. infections have also been associated with a number of other chronic disease syndromes in immunocompetent people [Reference Breitschwerdt8]. B. quintana is known to be present in lice in Ethiopia [Reference Cutler9] and in addition, C. felis has been reported in the country [Reference Mulugeta10]. Thus, while prevalence of Bartonella spp. infections of cats has not been reported, there is potential for these infections to exist.
Feline immunodeficiency virus (FIV) is a retrovirus related to HIV, and is known to cause immunosuppression in some cats, depending on the stage of infection. Feline leukaemia virus (FeLV), is related to human T-lymphocytic virus, and can also cause immunosuppression in cats. Here we investigated serological prevalence of T. gondii, Bartonella spp., and FIV and FeLV infections in cats in Ethiopia for the first time.
MATERIALS AND METHODS
Naturally infected cats surveyed
The cats surveyed were from the Addis Ababa area. Addis Ababa is the dense urban capital city of Ethiopia with an estimated population of 3 million in 2010. It experiences generally two main climates: the dry season between October and June and the rainy season between July and September. A total of 48 feral cats were used in this study. The cats were acquired by volunteers during the entire 2011 rainy season. The study was endorsed by the appropriate bodies of the Aklilu Lemma Institute of Pathobiology (ALIPB). Thirty-one of the cats were from the subcity of Lideta (latitude 9°1′21.8496″ N, longitude 38°44′48.4764″ E) of Addis Ababa, which is where ALIPB is located. Twenty-eight cats were female and 20 were male. Age was estimated by dental maturity, status and health; 41 of 48 cats were aged ⩾6 months (adults). Most cats had fleas but appeared to be in good physical condition.
Blood was collected from either a jugular or a femoral vein after mild sedation following the guidelines of the local institutional Animal Care Policies. The serum was separated and kept cold during air transport to the Animal Parasitic Diseases Laboratory (APDL), Beltsville, Maryland; 3 days elapsed between collection of sera and transport to APDL. Sera were subsequently stored at −20 °C until assayed.
Testing for T. gondii antibodies
All cat sera were initially tested for T. gondii IgG antibodies at 1:25, 1:50, 1:100, 1:200, 1:400, 1:800, 1:1600 and 1:3200 dilutions using the modified agglutination test (MAT) as described previously [Reference Dubey and Desmonts11]. A cut-off point of 1:25 dilution was used as indicative of T. gondii infection [Reference Dubey1]. This MAT detects only IgG antibodies because the mercaptoethanol used in the test destroys both specific and non-specific IgM.
Examination for T. gondii IgM and concurrent infections
After MAT testing at APDL, leftover serum from the cats were shipped to the Colorado State University (CSU), Fort Collins, Colorado, USA and stored at −80 °C until assayed for T. gondii IgM antibodies, Bartonella spp. antibodies, FIV antibodies and FeLV antigens at CSU. T. gondii IgM antibodies were assayed as described previously [Reference Lappin12]. Bartonella spp. IgG antibodies in serum were detected using a previously reported technique with a titre of <1:64 considered negative [Reference Lappin13]. While this assay uses B. henselae as the capture antigen, it appears to detect antibodies against B. clarridgeiae, and B. koehlerae; the lowest positive titre is defined as 1:64 [Reference Lappin13]. It is unknown whether antibodies against B. quintana are detected. Forty-six of the 48 cats were tested for T. gondii IgM and Bartonella spp. antibodies (Table 1). Forty-one of the 48 cat sera were assayed for FeLV antigen and FIV antibodies (Table 1) using a commercial kit (SNAP® FeLV/FIV®, IDEXX Laboratories, USA).
Table 1. Serological examination of cats for T. gondii, Bartonella, FIV, and FeLV infections
RESULTS
T. gondii IgG antibodies (MAT, 1:25) were found in 41 (85·4%) of the 48 cats; in 24 (85·7%) of 28 females and 17 (85·0%) of 20 males. The age distribution and titres are shown in Table 2. Subsequently, six of the seven sera negative at 1:25 dilution (sample for cat no. 7 was exhausted) were tested at 1:5, 1:10, and 1:20 dilutions. Four 2- and 4-month-old kittens had no detectable antibodies even at 1:5 dilution, two 2-month-old kittens had a titre of 1:5, and one 4-month-old kitten was seropositive at 1:10 dilution. Thirteen cats had titres of ⩾1:3200 (Table 2).
Table 2. Seroprevalence of T. gondii, Bartonella spp., FIV, and FeLV in cats from Ethiopia
* Of 48 cats tested.
† Two were <1:5, two were 1:5.
‡ Two were <1:5, one was 1:10.
T. gondii IgM antibodies were found in 11 cats; all of them were aged ⩾3 months; all IgM-positive cats also had IgG antibodies (Table 3).
Table 3. Details of Ethiopian cats positive for Toxoplasma gondii IgM antibodies
F, Female; M, male.
Antibodies to Bartonella spp. were found in five cats [aged 8 months (n=1), 1 year (n=2), 3 years (n=1), 8 years (n=1)], with titres of 1:64 in four and 1:128 in one (Table 2). All five Bartonella-infected cats also had T. gondii IgG antibodies.
None of the 41 cats tested were positive for FeLV or FIV, and the negativity was not related to the age of the cat (Tables 1, 2).
DISCUSSION
A high percentage of cats in the present study had T. gondii antibodies. The cats that we sampled represented both young and old cats, and both genders. All four 2-month-old cats were seronegative at 1:25 dilution; the low titre of 1:5 in two of these kittens was probably a result of colostrally acquired T. gondii antibodies. T. gondii transcolostral antibodies disappear in cats usually by age 3 months [Reference Dubey14–Reference Omata16]. Both 3-month-old cats had relatively high titres of 1:200 and 1:3200, and thus probably had active infections. Of the four 4-month-old kittens, two were clearly seronegative at 1:5, one had a high IgG titre of 1:1600; the fourth kitten had a titre of 1:10 that could have been recently infected, but it was negative for IgM antibodies. Cats in the present study were tested for T. gondii IgM antibodies to detect any additional infected cats missed by IgG antibody screening, but all IgM-positive cats also had IgG antibodies. IgG antibodies can be detected by MAT as early 10 days post-inoculation [Reference Dubey and Thulliez17]. A very high IgM titre of 1:16 384 and low IgG titre of 1:50 in cat no. 7 (Table 3) suggests a recently acquired T. gondii infection.
The 85% seroprevalence of T. gondii in cats from Addis Ababa suggests a high level oocyst contamination in the environment, because by the time cats become seropositive they have already shed oocysts [Reference Dubey1]. In addition, a number of cats shedding oocysts are likely to be seronegative. There are no data on the feral cat population in Ethiopia but they are common in public places; most of these stray cats are very wild and difficult to catch. Thus, we could survey only 48 cats.
There is considerable variability with respect to T. gondii prevalence and concurrent infections [Reference Dubey18, Reference Dubey19]. There could be many reasons for this observation, including different serological tests, cut-off titres, and lifestyle of cats surveyed [Reference Dubey18, Reference Dubey19]. Witt et al. [Reference Witt20] and Childs et al. [Reference Childs21] tested stray and pet cats from Baltimore, Maryland, USA and found that 14·7% had Bartonella spp., 2·4% had FIV, and 15·2% were seropositive to T. gondii. Both the magnitude of T. gondii titre and the seroprevalence were higher in FIV-infected cats [Reference Witt20]. In the present study, there was no evidence of FIV and FeLV infection and seroprevalence of Bartonella spp. was low.
To our knowledge, this is the first report of Bartonella spp. infections in cats from Ethiopia. Most of the cats tested in this study were infested by fleas, but the genus and species was not determined. The seroprevalence of Bartonella was lower than most other studies of cats infested with fleas. For example, the Bartonella spp. seroprevalence rates using the assay described here were 37·8%, 59%, and 78·4% in studies performed in Australia, Egypt, and the USA, respectively [Reference Lappin13, Reference Al-Kappany22, Reference Barrs23]. The Bartonella assay performed here used B. henselae antigens. However, it appears that antibodies to B. koehlerae and B. clarridgeiae are also detected in the assay. It is unknown whether the assay detects B. quintana which is known to be present in lice in Ethiopia [Reference Cutler9]. Thus, it cannot be stated with certainty which Bartonella sp. infected the cats in this study. In most studies around the world, B. henselae, B. clarridgeiae, and B. koehlerae are most common in cats because of the association with C. felis. This ectoparasite is present in Ethiopia but prevalence rates in cats are unknown. Thus, the low Bartonella spp. prevalence rate may reflect infestation by a different flea genus and species.
This limited survey in cats indicates potentially high contamination of the environment by oocysts shed by infected cats, and is the first indication of the potential for ingestion of oocysts from the environment as a mode of T. gondii infection and high seroprevalence in humans and food animals in Ethiopia [Reference Dubey24]. Bartonella spp. are likely to infect cats in the region and further work is required to document the infective species.
ACKNOWLEDGEMENTS
Thanks to all the staff at ALIPB for their assistance in handling the cats and with sample collection. The Bartonella spp. ELISA was performed by Arianne Morris in the Center for Companion Animal Studies at Colorado State University, Fort Collins, Colorado. The authors thank IDEXX Laboratories for donating the SNAP® Feline Triple® test kits used in this study and the Center for Companion Animal Studies donors (www.csuvets.colostate.edu/companion) for the Bartonella spp. ELISA reagents and technical support to complete these assays. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Department of Health and Human Services or the Centers for Disease Control and Prevention or the U.S. Department of Agriculture.
DECLARATION OF INTEREST
None.
