Detection of the Agent of Heartwater, Cowdria ruminantium, inAmblyomma Ticks by PCR: Validation and Application of the
Assay to Field TicksTREVOR F. PETER,1 ANTHONY F. BARBET,2 ARTHUR R. ALLEMAN,2 BIGBOY H. SIMBI,1
MICHAEL J. BURRIDGE,2 AND SUMAN M. MAHAN1*
UF/USAID/SADC Heartwater Research Project, Harare, Zimbabwe,1 and Department of Pathobiology, College ofVeterinary Medicine, University of Florida, Gainesville, Florida 32611-08802
Received 10 August 1999/Accepted 17 December 1999
We have previously reported that the pCS20 PCR detection assay for Cowdria ruminantium, the causativeagent of heartwater disease of ruminants, is more sensitive than xenodiagnosis and the pCS20 DNA probe forthe detection of infection in the vector Amblyomma ticks. Here, we further assessed the reliability of the PCRassay and applied it to field ticks. The assay detected DNA of 37 isolates of C. ruminantium originating fromsites throughout the distribution of heartwater and had a specificity of 98% when infected ticks were processedconcurrently with uninfected ticks. The assay did not detect DNA of Ehrlichia chaffeensis, which is closelyrelated to C. ruminantium. PCR sensitivity varied with tick infection intensity and was high (97 to 88%) withticks bearing 107 to 104 organisms but dropped to 61 and 28%, respectively, with ticks bearing 103 and 102
organisms. The assay also detected C. ruminantium in collections of Amblyomma hebraeum and Amblyommavariegatum field ticks from 17 heartwater-endemic sites in four southern African countries. Attempts at ticktransmission of infection to small ruminants failed with four of these collections. The pCS20 PCR assay ispresently the most characterized and reliable test for C. ruminantium in ticks and thus is highly useful for fieldand laboratory epidemiological investigations of heartwater.
Xenodiagnosis has long been the standard technique for thedetection in ticks of the tick-borne rickettsia Cowdria ruminan-tium, the causative agent of heartwater (cowdriosis), an eco-nomically important disease of domestic and wild ruminants inAfrica and in the Caribbean (11, 12, 13, 44, 45, 50). By inoc-ulating susceptible small ruminants or mice with homogenatesof the Amblyomma species tick vectors and monitoring clinicaldisease or seroconversion, the existence of C. ruminantiuminfection in the Caribbean has been confirmed (4, 9) and es-timates of tick infection prevalence have been obtained (9, 10,14, 17, 19, 20, 21, 47). However, xenodiagnosis is expensive andcumbersome, particularly in ruminants, and is slow (taking upto 6 weeks to complete). Furthermore, xenodiagnosis in micehas recently been shown to have low sensitivity and is unreli-able for the detection of infection (38). More-expedient DNAprobe and PCR assays based on the pCS20 and MAP 1 DNAsequences of C. ruminantium have been developed recentlyand have been used to detect infections in ticks and animals (1,31, 37, 38, 42, 43, 52, 60, 62). PCR assays, which have greatersensitivity than DNA probe assays (52), may be useful tests forlaboratory and field epidemiological investigations which areneeded for improved understanding of heartwater epidemiol-ogy and the impact of new control measures (49). However,prior to extensive use, PCR assays have to be fully validatedand their ability to detect infection in field ticks needs to beassessed. Here we determine the reliability of the pCS20 PCRassay and apply it to the detection of C. ruminantium in Am-blyomma hebraeum and Amblyomma variegatum ticks, the ma-jor vectors of heartwater, collected from various locations insouthern Africa.
MATERIALS AND METHODS
PCR assay. The PCR assay was performed as previously described (52), withsome modifications. Briefly, DNA was extracted from the individual tick tissuesamples using the QIA-amp PCR DNA extraction tissue kits (Qiagen, Hilden,Germany). The tissues were digested at 55°C for 16 h, followed by a 1-h incu-bation at 70°C. DNA was extracted from the digests as recommended exceptthat, for nymphs, the volumes for digestion and extraction were reduced by half,and elution of DNA from the columns (for all ticks) was performed twice with 50ml of elution buffer at 70°C. The purified DNA eluate was stored at 4°C untilanalysis.
For PCRs, 5 ml (5%) of the purified DNA of each adult tick and 20 ml (20%)of the DNA from each nymph were used as the template in reactions with theprimers AB128 (59-ACTAGTAGAAATTGCACAATCTAT-39) and AB129 (59-TGATAACTTGGTGCGGGAAATCCTT-39). These primers flank a 279-bpfragment within open reading frame 2 of the 1,306-bp pCS20 sequence of C.ruminantium (43, 60). PCRs were performed for 45 cycles, except that the MgCl2and primer concentrations were reoptimized for DNA purified from each tickinstar by the Qiagen method. For PCRs with adult tick DNA the primer andMgCl2 concentrations were 0.3 mM and 2.0 mM, respectively, and for nymphs theconcentrations were 0.5 mM and 2.0 mM, respectively. Additionally, denatur-ation of sample DNA prior to amplification was achieved by incubation of thereaction mixtures for 1 min at 94°C in the PCR heating block (model 480;Perkin-Elmer Cetus Corp., Norwalk, Conn.) before the first cycle. Each set ofPCRs included negative and positive reagent controls (reactions with no DNAand with 0.1 ng of C. ruminantium DNA, Plumtree isolate, respectively) andsample controls containing DNA from laboratory-reared, uninfected male, fe-male, or nymph A. hebraeum or A. variegatum ticks, as appropriate. Amplificationof C. ruminantium DNA was detected by dot blotting 40 ml (80%) of NaOH-denatured PCR products onto nylon membranes followed by hybridization withthe pCS20 DNA probe, as previously described (42, 52, 60, 62). Hybridizationwas detected by exposure of the blots to X-ray film (Kodak Biomax) for 1 to 7days.
Validation of PCR assay (i) Detection of diverse C. ruminantium isolates. Todetermine the ability of the PCR assay to detect DNA from geographicallydiverse C. ruminantium isolates, DNA prepared from cultured C. ruminantiumisolated from ticks collected at 13 study sites in southern Africa was tested in thePCR assay (Tables 1 and 2). The test was also performed on DNA from 24 othercultured C. ruminantium isolates made previously from Amblyomma species ticksor blood collected in heartwater-endemic regions of 10 countries in sub-SaharanAfrica and two islands in the Caribbean (Table 2). Supernatant (0.5 ml) fromterminal bovine endothelial cell cultures of each isolate was centrifuged at20,000 3 g for 20 min, and DNA was extracted from the pellets using QIA-ampkits and tested by PCR, as described above.
* Corresponding author. Mailing address: University of Florida/USAID/SADC Heartwater Research Project, P.O. Box CY 551,Causeway, Harare, Zimbabwe. Phone and Fax: 263-4-794980. E-mail:[email protected].
(ii) Specificity. The specificity of the PCR assay was determined by testing 100laboratory-reared, uninfected adult A. hebraeum ticks (50 males and 50 females)and 100 uninfected A. hebraeum nymphs. Tissue samples from 10 extra unin-fected adult ticks (5 male and 5 female), each spiked with 106 cell culture-derivedC. ruminantium organisms (Plumtree isolate) which were purified on Percollgradients and enumerated by staining with acridine orange (33), were processedconcurrently with the adult tick samples to simulate the processing of realsamples and determine the potential for cross-contamination under routineconditions. Similarly, 10 extra nymph samples spiked with 106 C. ruminantiumorganisms were processed together with the uninfected nymphs. The 95% con-fidence intervals of the specificity estimates for adults and nymphs were calcu-lated by the Bonferroni method.
(iii) Specificity against Ehrlichia chaffeensis. A specificity test against E.chaffeensis, which is phylogenetically and antigenically closely related to C. ru-minantium, was performed (27, 28, 34, 58). PCRs were performed, as describedabove, with C. ruminantium AB128 and AB129 primers on 1 and 10 ng of E.chaffeensis DNA purified from organisms cultured in vitro in canine DH82 cellsand on 1 and 10 ng of C. ruminantium DNA (Zimbabwean Crystal Springsisolate). PCRs were also performed on 1 and 10 ng of E. chaffeensis DNA and onC. ruminantium DNA with the primers 59-GAGTACGGATCCGCAATTTTTCTAGGATATTCC-39 and 59-ATCAGACTGCGGCCGCATGTAATTAGCGATAGAAACACCA-39, which amplify a 727-bp fragment from E. chaffeensis con-taining a gene homologous to the C. ruminantium MAP 2 gene (5, 35). Reactionswith E. chaffeensis primers were performed in 100 ml consisting of 13 buffer (20mM Tris-HCL [pH 8.8], 10 mM KCL, 10 mM (NH4)2SO4, 2 mM MgSO4, 0.1%Triton X-100, 100-mg/ml bovine serum albumin), 1.25 U of pfu DNA polymerase(Stratagene, La Jolla, Calif.), a 200 mM concentration of each deoxynucleosidetriphosphate, and a 1.0 mM concentration of each primer. The reaction mixtureswere overlayed with mineral oil and incubated at 94°C for 3 min before under-going 10 cycles each of 15 s at 94°C, 1 min at 43°C, and 1 min at 72°C followedby 25 cycles each of 15 s at 94°C, 1 min at 49°C, and 1 min at 72°C, and finally7 min at 72°C. As a further positive control, PCRs were also performed, underconditions described above for C. ruminantium primers, on 1 and 10 ng of E.chaffeensis DNA and C. ruminantium DNA with general Ehrlichia primers E2(59-GTGGCAGACGGGTGAGTAATGC-39) and E3 (59-GGTAACGTCAATATCTTCCC-39), which amplify a 350-bp fragment from the conserved region ofthe 16S rRNA gene of members of the tribe Ehrlichia (L. A. Matthewman, N.Lally, K. Sumption, P. J. Kelly, and D. Raoult, unpublished data; 54). Twentymicroliters of the products of the above PCRs were electrophoresed throughagarose gels, Southern blotted, and hybridized with the pCS20 DNA probe.
(iv) Sensitivity. To determine the sensitivity of the PCR assay under routinesample-handling conditions, dissected tissues from laboratory-reared, uninfectedadult male and female A. hebraeum ticks were individually spiked with 10-foldserial dilutions of cell culture-derived, Percoll-purified C. ruminantium (Plumtreeisolate) containing 102 to 107 organisms to simulate these levels of infection. Intotal, 100 ticks (50 male and 50 female) were spiked at each dilution. These weredivided into batches of 8 to 15, which were tested separately. Confidence inter-vals (CI; 95%) for the sensitivity of the PCR assay at each infection level werecalculated by the Bonferroni method, and negative and positive predictive valueswere determined for each level of tick infection.
Detection of C. ruminantium in field ticks. Collections of A. hebraeum or A.variegatum adult male and flat adult female ticks were made from cattle, goats,or vegetation at 17 locations. These locations included large-scale commercial
livestock ranches, livestock quarantine stations, and traditional farming (com-munal) areas in heartwater-endemic regions of Botswana, South Africa, Swazi-land, Zambia, and Zimbabwe (Table 1). The ticks were tested for C. ruminan-tium infection by PCR as described above. To confirm C. ruminantium infectionin each collection, attempts were made to isolate C. ruminantium by transmissionto small ruminants. Sixty to 120 ticks from each site were fed on C. ruminantium-naive sheep or goats, which were then monitored daily for clinical signs ofheartwater. Confirmation of infection was obtained by examination of brainsmears prepared from biopsies performed during the febrile reaction or post-mortem (53, 56). During the febrile stage of infection, plasma was collected fromsome of the animals and inoculated into bovine endothelial cell cultures toattempt isolation of C. ruminantium (8). Infection of the cultures was confirmedby examination of cell smears prepared after the development of cytopathiceffects in the cell monolayers. Thereafter, the isolates were grown in continuousculture, as previously described (7).
RESULTS
Validation of the PCR assay. (i) Detection of diverse C.ruminantium strains. The PCR assay amplified a 279-bp frag-ment from the DNA of each of the 37 isolates of C. ruminan-tium tested (Table 2). In each case, the amplified fragmenthybridized with the pCS20 DNA probe (data not shown), con-firming the conservation of the PCR assay primer target sitesand the ability of the assay to detect C. ruminantium fromthroughout the distribution of heartwater.
(ii) Specificity. Two of 100 uninfected adult ticks and 2 of100 uninfected nymph ticks tested positive by PCR after DNAextraction and PCR testing together with tick samples spikedwith culture-derived C. ruminantium organisms. The specificityof the PCR assay for both adult and nymph ticks was therefore98%.
(iii) Specificity against E. chaffeensis. No amplification wasdetected from 1 and 10 ng of E. chaffeensis DNA with C.ruminantium primers after agarose gel electrophoresis of theproducts and hybridization with the pCS20 DNA probe (Fig. 1and 2). Amplification of a 727-bp product with primers for theMAP 2 gene homologue of E. chaffeensis confirmed the pres-ence of E. chaffeensis DNA in the specificity test. These prim-ers did not produce a product from C. ruminantium DNA thatwas detectable by gel electrophoresis, though the general Ehr-lichia primers amplified a 350-bp product from both C. rumi-nantium and E. chaffeensis (Fig. 1).
(iv) Sensitivity. The sensitivity of the PCR assay varied withthe level of tick infection and was high, ranging from 97 to
TABLE 1. Sites for collection of Amblyomma ticks in southern Africa
Sitea Location Vector sp. Source
Mochudi Communal Land Southeastern Botswana A. hebraeum GoatsSunnyside Quarantine Station Southeastern Botswana A. hebraeum CattleSABS Farm Eastern Cape Province, RSAb A. hebraeum VegetationWarmbaths LSC Northern Gauteng, RSA A. hebraeum VegetationBig Bend Agricultural Station Western Swaziland A. hebraeum CattleMpisi Quarantine Station Eastern Swaziland A. hebraeum CattleGamela Communal Land Southern Zambia A. variegatum CattleLutale Communal Land Central Zambia A. variegatum CattleChinamora Communal Land Northern Zimbabwe A. hebraeum CattleChivhu LSC North central Zimbabwe A. hebraeum VegetationKana E LSC Northeastern Zimbabwe A. variegatum CattleKwekwe Communal Land Central Zimbabwe A. hebraeum CattleMhondoro Communal Land North central Zimbabwe A. hebraeum CattleMutasa Communal Land Eastern Zimbabwe A. hebraeum CattlePlumtree LSC Southeastern Zimbabwe A. hebraeum CattleZvimba Communal Land North central Zimbabwe A. variegatum CattleZvishavane LSC Southern Zimbabwe A. hebraeum Vegetation
a SABS, South African Bureau of Standards; LSC, large-scale commercial farm.b RSA, Republic of South Africa.
88%, with samples containing 107 to 104 organisms (Table 3).However, the sensitivity dropped to 61% and then to 28% withsamples containing 103 and 102 C. ruminantium organisms,respectively. The positive predictive value of the PCR assay
was high, greater than 93%, at all levels of infection (Table 3).The negative predictive value, however, dropped from 93% atthe 105 level to 58% at the 102 level.
Detection of C. ruminantium in field ticks. The PCR assaydetected C. ruminantium DNA in A. hebraeum ticks collectedat 13 sites in four southern African countries, Botswana, SouthAfrica, Swaziland, and Zimbabwe (Table 4). Transmission ofheartwater was attempted with ticks and was successful fromall but two of these sites (Chinamora and Mhondoro Commu-nal Lands in Zimbabwe), confirming the presence of C. rumi-nantium infection in these tick batches. The prevalence ofinfection in the A. hebraeum ticks from different sites rangedfrom 1.6 to 83.3% in males and from 2.1 to 57% in females.The PCR assay also detected C. ruminantium in A. variegatumticks collected at two sites in Zambia (Gamela and Lutale) andat two sites in Zimbabwe (Kana E and Zvimba) (Table 4).Transmission of C. ruminantium with the ticks from Zambiawas successful, while transmission with A. variegatum from theZimbabwean sites failed due to lack of tick attachment. Theprevalence of infection within A. variegatum ticks ranged from6.25 to 39.3% in males and from 2.1 to 14.3% in females.
DISCUSSION
This study expands on a previous preliminary evaluation ofthe pCS20 PCR diagnostic assay for C. ruminantium in ticks(52) by providing data on its operational efficiency that permitthe interpretation of assay results under routine testing condi-tions. The use of the test to detect infections in widespreadcollections of field ticks is also described. Positive features ofthe PCR assay include its high sensitivity, specificity, predictivevalues, and isolate cross-reactivity. The ability of the PCR
FIG. 1. Specificity of the C. ruminantium pCS20 PCR assay against E.chaffeensis. Agarose gel electrophoresis of products from PCRs with C. ruminan-tium primers and C. ruminantium DNA (lanes 2 and 3) or E. chaffeensis DNA(lanes 4 and 5), with E. chaffeensis primers and C. ruminantium DNA (lanes 8 and9) or E. chaffeensis DNA (lanes 10 and 11), and with general Ehrlichia primersand C. ruminantium DNA (lanes 14 and 15) or E. chaffeensis DNA (lanes 16 and17). The 123-bp DNA standard ladder and negative-control reactions for the C.ruminantium, E. chaffeensis, and general Ehrlichia primers are shown in lanes L,1, 7, and 13, respectively. Lanes 6 and 12 are empty.
FIG. 2. Specificity of the C. ruminantium pCS20 PCR assay against E.chaffeensis. Lanes 1 to 5 are lanes 1 to 5 of the gel in Fig. 1, respectively, afterSouthern blotting and hybridization with the pCS20 DNA probe.
TABLE 2. Origin of C. ruminantium isolates tested by PCR
Isolate Origin Reference
Beatrice Zimbabwe 55Crystal Springs Zimbabwe 61Finale Zimbabwe 55Highway Zimbabwe 6Hunyani Zimbabwe 55Kwekwe Zimbabwe 55Lemco Zimbabwe 61Mbizi Zimbabwe 61Mubayira Zimbabwe 55Nyatsanga Zimbabwe 48Palm River Zimbabwe 61Plumtree Zimbabwe 41, 51Rusape Zimbabwe 55Zvimba Zimbabwe 48Ball 3 South Africa 22Blaaukrantz South AfricaKwanyanga South Africa 11Rietgat South Africa 36Skukuza South Africa 51Vosloo South Africa 15Warmbaths South Africa 36Welgevonden South Africa 14Big Bend Swaziland 36Mpisi Swaziland 36Mochudi Botswana 36Sunnyside Botswana 36Gamela Zambia 36Lutale 98 Zambia 36Tanga Tanzania 55Isiolo Kenya 46Kiswani Kenya 29Nigeria D225 Nigeria 23Pokoase 417 Ghana 3Sankat 430 Ghana 3Um Banein Sudan 25Antigua Antigua 4Gardel Gardel 57
TABLE 3. Sensitivity and predictive values of the PCR assay for C.ruminantium at different levels of tick infection
assay to detect DNA of C. ruminantium originating from lo-cales throughout the distribution of heartwater and in fieldticks collected from diverse sites in southern Africa demon-strates conservation of the primer sequences and the wideapplicability of the assay. The test did not detect DNA of E.chaffeensis, which, serologically and on the basis of 16S rDNAanalysis, is closely related to C. ruminantium (27, 28, 58). Theabsence of cross-reactivity with E. chaffeensis is particularlypertinent to the screening of Amblyomma ticks in the UnitedStates, which is under threat of the introduction of heartwaterand where E. chaffeensis infection occurs naturally in Ambly-omma americanum ticks (2, 6) and in a wild host of Ambly-omma ticks, white-tailed deer (Odocoileus virginianus) (32).The PCR assay detected infection in both A. hebraeum and A.variegatum field ticks and has also been used to detect exper-imental infections in Amblyomma gemma, a significant vectorof heartwater in east Africa (unpublished observations), and inAmblyomma maculatum (39), the most important potentialvector in the United States. The pCS20 PCR assay has alsobeen shown, in an independent evaluation, to be more sensi-tive than other C. ruminantium PCR assays based on 16SrDNA and the MAP 1 gene (1). It is thus highly suited tolaboratory and field studies on C. ruminantium transmission, toinvestigations in previously unstudied heartwater regions, andfor surveillance programs in areas under risk of the introduc-tion of infection or attempting eradication of infection. Forexample, the assay would be a valuable tool for defining thedistribution of C. ruminantium infection in ticks in the Carib-bean region, where eradication of the disease is being consid-ered.
The PCR assay demonstrated high sensitivity (88 to 97%)and negative predictive values (89 to 97%) at tick infectionlevels between 104 and 107 organisms. This reliability was,however, not observed at lower infection levels and droppedsubstantially to a 28% sensitivity and 57% negative predictivevalue in ticks carrying 100 organisms. This was not unexpectedas the sample aliquot used for PCR from such ticks contained,on average, five organisms (1/20 of the total). The success ofthe PCR assay with this level of template frequency is likely tobe significantly influenced by inefficiencies in DNA extraction(loss or damage of DNA) or by PCR failure resulting frommispriming or the presence of PCR inhibitors. Amblyomma
tick tissue has been shown previously to contain PCR-inhibi-tory elements which are not always removed during DNApurification, which may partially account for the occurrence ofPCR-negative results on DNA probe-positive samples (52).The reduced sensitivity of the PCR assay at low infection levelsis likely to affect estimates of field tick infection prevalence,though the full implications are difficult to assess. The pCS20DNA probe has a detection limit of approximately 70,000organisms (52) and provides semiquantitative information onthe intensity of infection above this level. The distribution ofinfection intensities below this level is presently not known;therefore it is not possible to determine the proportion ofinfections that are missed by the PCR assay. Further studies,potentially with quantitative PCR techniques, may help eluci-date the nature of low-level infections.
Under sample handling conditions that mimicked the pro-cessing of real infected samples, the specificity of the PCRassay was 98%, suggesting that minor sample-to-sample con-tamination occurred. In situations where no infected samplesare present, the specificity appears to be closer to 100% (38),thus improving the positive predictive value when uninfectedtick populations are screened for the introduction of infection.
The PCR assay was more effective than tick transmissiontrials for the detection of C. ruminantium in field-collectedticks. Four of the 17 tick batches failed to feed or failed totransmit infection. These batches were analyzed by PCR andfound to be infected. PCR can also be applied to dried andfixed ticks (52), obviating the need to transfer ticks to recipientanimals quickly after collection. However, in situations whereproof of infection requires direct demonstration of C. rumi-nantium infection or isolation of the agent, xenodiagnosisshould be used. This should preferably be done in small rumi-nants instead of mice due to easier ante- and postmortemconfirmation of infection and the availability of simple tech-niques for isolation of the organism in cell culture from rumi-nant plasma (8). In addition, certain C. ruminantium isolatesappear to be noninfective for mice (18), though other conclu-sive experiments on this aspect are required.
The PCR analysis of field ticks provided a wide range ofinfection rate estimates for both male and female ticks. In mostcases, the ticks were collected from livestock. These infectionrates, therefore, cannot be considered representative of host-
TABLE 4. Analysis by PCR of field-collected Amblyomma ticks from southern Africa
Country Sitea Vector sp.% PCR positive/no. analyzed Tick transmission of
C. ruminantiumMale Female Total
Botswana Mochudi CL A. hebraeum 19.0/21 33.3/9 23.3/30 1Sunnyside Quarantine Station A. hebraeum 55.6/18 nd 55.6/18 1
South Africa SABS Farm A. hebraeum nd 2.8/36 2.8/36 1Warmbaths LSC A. hebraeum 1.6/60 6.9/43 3.9/103 1
Swaziland Big Bend Agricultural Station A. hebraeum 5.6/18 11.8/17 8.6/35 1Mpisi Quarantine Station A. hebraeum 10.4/67 8.7/46 9.7/113 1
Zambia Gamela CL A. variegatum nd 14.3/49 14.3/49 1Lutale CL A. variegatum 39.3/84 2.1/96 19.4/180 1
Zimbabwe Chinamora CL A. hebraeum 40/25 50/6 41.9/31 2Chivhu LSC A. hebraeum 8/75 17.6/17 9.8/92 1Kana E LSC A. variegatum 6.25/16 13.3/15 9.7/31 2Kwekwe CL A. hebraeum 83.25/17 53.8/13 70/30 1Mhondoro CL A. hebraeum 53.3/15 57.1/7 54.5/22 2Mutasa CL A. hebraeum 34.8/23 57.1/7 40/30 1Plumtree LSC A. hebraeum 13.3/15 13.3/15 13.3/30 1Zvimba CL A. variegatum 9.5/21 nd 9.5/21 2Zvishavane LSC A. hebraeum 12/50 6/50 9/100 1
a SABS, South African Bureau of Standards; LSC, large-scale commercial farm; CL, communal land; nd, not done.
seeking tick populations due to the potential for feeding ticksto acquire infection intrastadially (30). This effect is particu-larly important for male ticks, which can feed for prolongedperiods (weeks to months) (26) and do not lose infection dur-ing feeding. The flat female ticks examined in this study may,however, be more representative due to their short attachmentperiod prior to the start of engorgement (1 to 2 days) and thuslimited exposure to infection. The infection prevalence of theunfed A. hebraeum ticks collected from vegetation ranged from2.8 to 9.8%, providing better estimates of the magnitude of thevector infection reservoir, though the sample sizes were small.
The pCS20 PCR assay is presently the most reliable andbest-characterized test for C. ruminantium infection in ticks,exceeding previous assays in reliability and speed (38). Thelack of cross-reaction with closely related organisms such as E.chaffeensis demonstrated here and with Ehrlichia canis (52)increases the value of this test, particularly as C. ruminantiumserologic assays developed to date are limited by either poorspecificity or low sensitivity (5, 16, 24, 27, 28, 40, 42, 59).Application of the PCR assay in other heartwater-endemicregions of Africa and the Caribbean will provide valuable andaccurate information on infection rates of the Amblyomma tickvectors. Adaptation of the assay to detect C. ruminantium incarrier animals will further facilitate the understanding of anarea of epidemiology of heartwater which is currently not com-pletely understood, and this would significantly improve sur-veillance and control for heartwater.
ACKNOWLEDGMENTS
This study was supported by U.S. Agency for International Devel-opment grant LAG-1328-G-00-3030-00 awarded to the University ofFlorida.
We thank the following for assistance in tick collections: Ian andMargueritte Swannack (Finale Farm, Zimbabwe), Boetie and HendrikO’Neill (Vlakfontein Estates, Zimbabwe), L. Modisa of the BotswanaVeterinary Services, G. Axsel and P. van der Reit of the South AfricanDepartment of Veterinary Services, N. Bryson of the Medical Univer-sity of South Africa, L. van der Merwe of Warmbaths Farm RSA, P.Dlamini of the Swaziland Department of Veterinary Services, and L.Makala and G. Munyama of the Zambian Department of VeterinaryServices. We are grateful for the technical assistance provided byGillian Smith, Godfree Mlambo, Fidelis Mugova, and Lovemore Ka-tivu.
REFERENCES
1. Allsopp, M. E. T. P., C. M. Hattingh, S. W. Vogel, and B. A. Allsopp. 1999.Evaluation of 16S, map 1 and pCS20 probes for the detection of Cowdria andEhrlichia species. Epidemiol. Infect. 122:323–328.
2. Anderson, B. E., K. G. Sims, J. G. Olson, J. E. Childs, J. F. Piesman, C. M.Happ, G. O. Maupin, and B. J. B. Johnson. 1993. Amblyomma americanum:a potential vector of human ehrlichiosis. Am. J. Trop. Med. Hyg. 49:239–244.
3. Bell-Sakyi, L., E. B. M. Koney, O. Dogbey, J. Abbam, and K. G. Aning. 1996.Isolation and in vitro cultivation in Ghana of Cowdria ruminantium, thecausative agent of heartwater, p. 46–51. In E. B. M. Koney and K. G. Aning(ed.), Proceedings of the 22nd Annual Congress of the Ghana VeterinaryMedical Association. The Information Support Unit, Ministry of Food andAgriculture, Accra, Ghana.
4. Birnie, E. F., M. J. Burridge, E. Camus, and N. Barre. 1985. Heartwater inthe Caribbean: isolation of Cowdria ruminantium from Antigua. Vet. Rec.116:121–123.
5. Bowie, M. V., G. R. Reddy, S. M. Semu, S. M. Mahan, and A. F. Barbet. 1999.Potential value of major antigenic protein 2 for serological diagnosis ofheartwater and related ehrlichial infections. Clin. Diagn. Lab. Immunol.6:209–215.
6. Burket, C. T., C. N. Vann, R. R. Pinger, C. L. Chatot, and F. E. Steiner. 1998.Minimum infection rate of Amblyomma americanum (Acari: Ixodidae) byEhrlichia chaffeensis (Rickettsiales: Ehrlichieae) in southern Indiana. J. Med.Entomol. 35:653–659.
7. Byrom, B., and C. E. Yunker. 1990. Improved culture conditions for Cowdriaruminantium (Rickettsiales), the agent of heartwater disease of domesticruminants. Cytotechnology 4:285–290.
8. Byrom, B., C. E. Yunker, P. L. Donovan, and G. E. Smith. 1991. In vitro
isolation of Cowdria ruminantium from plasma of infected ruminants. Vet.Microbiol. 26:263–268.
9. Camus, E., and N. Barre. 1987. Epidemiology of heartwater in Guadeloupeand in the Caribbean. Onderstepoort J. Vet. Res. 54:419–426.
10. Camus, E., and N. Barre. 1992. The role of Amblyomma variegatum in thetransmission of heartwater with special reference to Guadeloupe. Ann. N. Y.Acad. Sci. 653:33–41.
11. Camus, E., N. Barre, D. Martinez, and G. Uilenberg. 1996. Heartwater(cowdriosis). A review, 2nd ed. Office International des Epizooties, Paris,France.
12. Cowdry, E. V. 1925. Studies on the etiology of heartwater. I. Observation ofa rickettsia, Rickettsia ruminantium (n. sp.), in the tissues of infected animals.J. Exp. Med. 42:253–274.
13. Cowdry, E. V. 1925. Studies on the etiology of heartwater II Rickettsiaruminantium (n. sp.) in the tissues of ticks transmitting the disease. J. Exp.Med. 42:231–245.
14. Du Plessis, J. L. 1985. A method for determining the Cowdria ruminantiuminfection rate of Amblyomma hebraeum: effects in mice injected with tickhomogenates. Onderstepoort J. Vet. Res. 52:55–61.
15. Du Plessis, J. L. 1993. The relationship between Cowdria and Ehrlichia:change in the behaviour of ehrlichial agents passaged through Amblyommahebraeum. Rev. Elev. Med. Vet. Pays Trop. 46:131–143.
16. Du Plessis, J. L., J. D. Bezuidenhout, M. S. Brett, E. Camus, F. Jongejan,S. M. Mahan, and D. Martinez. 1993. Serodiagnosis of heartwater: a com-parison of five tests. Rev. Elev. Med. Vet. Pays Trop. 46:123–129.
17. Du Plessis, J. L., P. J. Loock, and C. J. F. Ludemann. 1992. Adult Ambly-omma hebraeum burdens and heartwater endemic stability in cattle. Onder-stepoort J. Vet. Res. 59:75–89.
18. Du Plessis, J. L., L. Van Gas, J. A. Olivier, and J. D. Bezuidenhout. 1989.The heterogeneity of Cowdria ruminantium stocks: cross-immunity and se-rology in sheep and pathogenicity to mice. Onderstepoort J. Vet. Res. 56:195–201.
19. Du Plessis, J. L., and L. Malan. 1987. Problems with the interpretation ofepidemiological data in heartwater: a study on 23 farms. Onderstepoort J.Vet. Res. 54:427–433.
20. Gueye, K. L., M. Mbengue, T. Dieye, A. Diouf, M. Seye, and M. F. Seye. 1993.Cowdriosis in Senegal: some epidemiologic aspects. Rev. Elev. Med. Vet.Pays Trop. 46:217–221.
21. Gueye, K. L., M. Mbengue, and A. Diouf. 1993. Epidemiologie de la cow-driose au Senegal. I. Etude de la transmission et du taux d’infectiond’Amblyomma variegatum (Fabricious, 1947) dans la region des Niayes. Rev.Elev. Med. Vet. Pays Trop. 46:441–447.
22. Haig, D. A. 1952. Note on the use of the white mouse for the transport ofstrains of heartwater. J. S. Afr. Vet. Med. Assoc. 23:167–170.
23. Ilemobade, A. A., and P. Leeflang. 1977. Epidemiology of heartwater inNigeria. Rev. Elev. Med. Vet. Pays Trop. 30:149–155.
24. Jongejan, F., N. de Vries, H. Nieuwenhuys, A. H. M. van Vliet, and L. A.Wassink. 1993. The immunodominant 32 kilodalton protein of Cowdriaruminantium is conserved within the genus Ehrlichia. Rev. Elev. Med. Vet.Pays Trop. 46:145–152.
25. Jongejan, F., A. J. van Winklehoff, and G. Uilenberg. 1984. Isolation andtransmission of Cowdria ruminantium (causal agent of heartwater disease) inthe Blue Nile Province, Sudan. Vet. Res. Commun. 8:141–145.
26. Jordaan, J. O., and J. A. F. Baker. 1981. Survival rate on the host and matingcapacity of Amblyomma hebraeum (Koch) male ticks, p. 115–117. In G. B.Whitehead and J. D. Gibson (ed.), Tick biology and control. Tick ResearchUnit, Rhodes University, Grahamstown, South Africa.
27. Katz, J. B., A. F. Barbet, S. M. Mahan, D. Kumbula, J. M. Lockhart, M. K.Keel, J. E. Dawson, J. G. Olson, and S. A. Ewing. 1996. A recombinantantigen from the heartwater agent Cowdria ruminantium reactive with anti-bodies in some southeastern United States white-tailed deer (Odocoileusvirgininanus), but not cattle sera. J. Wildl. Dis. 32:424–430.
28. Katz, J. B., R. DeWald, J. E. Dawson, E. Camus, D. Martinez, and R.Mondry. 1997. Development and evaluation of a recombinant antigen,monoclonal antibody-based competitive ELISA for heartwater serodiagno-sis. J. Vet. Diagn. Investig. 9:130–135.
29. Kocan, K. M., S. P. Morzaria, W. P. Voigt, W. P. Kiare, and A. D. Irvin. 1986.Demonstration of colonies of Cowdria ruminantium in midgut epithelial cellsof Amblyomma variegatum. Am. J. Vet. Res. 48:356–360.
30. Kocan, K. M., R. A. I. Norval, and P. L. Donovan. 1993. Development andtransmission of Cowdria ruminantium by Amblyomma males transferredfrom infected to susceptible sheep. Rev. Elev. Med. Vet. Pays Trop. 46:183–188.
31. Kock, N. D., A. H. M. van Vliet, K. Charlton, and F. Jongejan. 1995.Detection of Cowdria ruminantium in blood and bone marrow samples fromclinically normal, free-ranging Zimbabwean wild ungulates. J. Clin. Micro-biol. 33:2501–2504.
32. Lockhart, J. M., W. R. Davidson, D. E. Stallknecht, J. E. Dawson, and E. W.Howerth. 1997. Isolation of Ehrlichia chaffeensis from wild white-tailed deer(Odocoileus virginianus) confirms their role as natural reservoir hosts. J. Clin.Microbiol. 35:1681–1686.
33. Mahan, S. M., H. R. Andrew, N. Tebele, M. J. Burridge, and A. F. Barbet.
VOL. 38, 2000 DETECTION OF C. RUMINANTIUM IN AMBLYOMMA TICKS BY PCR 1543
1995. Immunization of sheep with inactivated Cowdria ruminantium providesprotection against heartwater. Res. Vet. Sci. 58:46–49.
34. Mahan, S. M., B. Allsopp, K. M. Kocan, G. H. Palmer, and F. Jongejan.1999. Vaccine strategies for Cowdria ruminantium and their applications toother ehrlichial infections. Parasitol. Today 15:290–294.
35. Mahan, S. M., T. C. McGuire, S. M. Semu, M. V. Bowie, F. Jongejan, F. R.Rurangirwa, and A. F. Barbet. 1994. Molecular cloning of a gene encodingthe immunogenic 21 kD protein of Cowdria ruminantium. Microbiology140:2135–2142.
36. Mahan, S. M., T. F. Peter, B. D. Perry, L. Modisa, G. Axsel, N. Bryson, L. vander Merwe, P. van der Reit, J. Pistorius, P. Dlamini, L. Makala, G.Munyama, A. F. Barbet, and M. J. Burridge. 1998. Characterization ofheartwater challenge at field sites in six SADC countries, p. 30. In Proceed-ings of the 9th International Conference of the Association of Institutions ofTropical Veterinary Medicine (AITVM).
37. Mahan, S. M., T. F. Peter, S. M. Semu, B. H. Simbi, R. A. I. Norval, and A. F.Barbet. 1995. Laboratory reared Amblyomma hebraeum and A. variegatumdiffer in their susceptibility to infection with Cowdria ruminantium. Epide-miol. Infect. 115:345–353.
38. Mahan, S. M., T. F. Peter, B. H. Simbi, and M. J. Burridge. 1998. PCRdetection of Cowdria ruminantium infection in ticks and animals from heart-water-endemic regions of Zimbabwe. Ann. N. Y. Acad. Sci. 849:85–87.
39. Mahan, S. M., T. F. Peter, B. H. Simbi, B. H., K. M. Kocan, E. Camus, A. F.Barbet, and M. J. Burridge. 2000. Comparison of efficacy of American andAfrican Amblyomma ticks as vectors of heartwater (Cowdria ruminantiuminfection) by molecular analyses and transmission trials. J. Parasitol. 86:44–49.
40. Mahan, S. M., S. M. Semu, T. F. Peter, and F. Jongejan. 1998. Evaluation ofthe MAP 1B ELISA for cowdriosis with field sera from livestock in Zimba-bwe. Ann. N. Y. Acad. Sci. 849:259–261.
41. Mahan, S. M., G. E. Smith, and B. Byrom. 1994. Concanavalin A-stimulatedbovine T-cell supernatants inhibit growth of Cowdria ruminantium in vitro.Infect. Immun. 62:747–750.
42. Mahan, S. M., N. Tebele, D. Mukwedeya, S. Semu, C. B. Nyathi, L. A.Wassink, P. J. Kelly, T. Peter, and A. F. Barbet. 1993. An immunoblottingassay for heartwater based on the immunodominant 32-kilodalton protein ofCowdria ruminantium detects false positives in field sera. J. Clin. Microbiol.31:2729–2737.
43. Mahan, S. M., S. D. Waghela, T. C. McGuire, F. R. Rurangirwa, L. A.Wassink, and A. F. Barbet. 1992. A cloned DNA probe for Cowdria rumi-nantium hybridizes with eight heartwater strains and detects infected sheep.J. Clin. Microbiol. 30:981–986.
44. Meltzer, M. I., B. D. Perry, and P. L. Donachie. 1996. Mortality percentagesrelated to heartwater and the economic impact of heartwater disease onlarge scale commercial farms in Zimbabwe. Prev. Vet. Med. 2:187–199.
45. Mukhebi, A. W., T. Chamboko, C. J. O’Callaghan, T. F. Peter, R. L. Kruska,G. F. Medley, S. M. Mahan, and B. D. Perry. 1999. An assessment of theeconomic impact of heartwater (Cowdria ruminantium infection) and itscontrol in Zimbabwe. Prev. Vet. Med. 39:173–189.
46. Ngumi, P. N., R. M. Rumberia, S. M. Williamson, K. J. Sumption, A. C.Lesau, and D. P. Kariuki. 1997. Isolation of the causative agent of heartwa-ter (Cowdria ruminantium) from three Amblyomma species in eight districtsof Kenya. Vet. Rec. 140:13–16.
47. Norval, R. A. I., H. R. Andrew, and C. E. Yunker. 1990. Infection rates with
Cowdria ruminantium of nymphs and adults of the bont tick Amblyommahebraeum collected in the field in Zimbabwe. Vet. Parasitol. 36:277–283.
48. Norval, R. A. I., B. D. Perry, M. I. Meltzer, R. L. Kruska, and T. H. Booth.1994. Factors affecting the distributions of the ticks Amblyomma hebraeumand A. variegatum in Zimbabwe: implications of reduced acaricide usage.Exp. Appl. Acarol. 18:383–407.
49. O’Callaghan, C. J., G. F. Medley, T. F. Peter, and B. D. Perry. 1998. Inves-tigating the epidemiology of heartwater (Cowdria ruminantium infection) bymeans of a transmission dynamic model. Parasitology 117:49–61.
50. Peter, T. F., E. C. Anderson, M. J. Burridge, and S. M. Mahan. 1998.Demonstration of a carrier state for Cowdria ruminantium in wild ruminantsfrom Africa. J. Wildl. Dis. 34:567–575.
51. Peter, T. F., N. R. Bryson, B. D. Perry, C. J. O’Callaghan, G. F. Medley, G. E.Smith, G. Mlambo, I. G. Horak, M. J. Burridge, and S. M. Mahan. 1999.Cowdria ruminantium infection in ticks in the Kruger National Park. Vet.Rec. 145:304–307.
52. Peter, T. F., S. L. Deem, A. F. Barbet, R. A. I. Norval, B. H. Simbi, P. J. Kelly,and S. M. Mahan. 1995. Development and evaluation of PCR assay fordetection of low levels of Cowdria ruminantium infection in Amblyommaticks not detected by DNA probe. J. Clin. Microbiol. 33:166–172.
53. Purchase, H. S. 1945. A simple and rapid method for demonstrating Rick-ettsia ruminantium (Cowdry, 1925) in heartwater brains. Vet. Rec. 57:413–414.
54. Savadye, D. T., P. J. Kelly, and S. M. Mahan. 1998. Evidence to show thatan agent that cross-reacts serologically with Cowdria ruminantium in Zim-babwe is transmitted by ticks. Exp. Appl. Acarol. 22:111–122.
55. Smith, G. E., E. C. Anderson, M. J. Burridge, T. F. Peter, and S. M. Mahan.1998. Growth of Cowdria ruminantium in tissue culture endothelial cell linesfrom wild African mammals. J. Wildl. Dis. 32:297–304.
56. Synge, B. A. 1978. Brain biopsy for the diagnosis of heartwater. Trop. Anim.Health Prod. 10:45–48.
57. Uilenberg, G., E. Camus, and N. Barre. 1985. Quelques observations sur unesouche de Cowdria ruminantium isolee en Guadeloupe (Antilles francaises).Rev. Elev. Med. Vet. Pays Trop. 38:34–42.
58. van Vliet, A. H. M., F. Jongejan, and B. A. M. van der Zeijst. 1992. Phylo-genetic position of Cowdria ruminantium (Rickettsiales) determined by anal-ysis of amplified 16S ribosomal DNA sequences. Int. J. Syst. Bacteriol.42:494–498.
59. van Vliet, A. H. M., B. A. M. van der Zeijst, E. Camus, S. M. Mahan, D.Martinez, and F. Jongejan. 1995. Use of a specific immunogenic region onthe Cowdria ruminantium MAP1 protein in a serological assay. J. Clin.Microbiol. 33:2405–2410.
60. Waghela, S. D., F. R. Rurangirwa, S. M. Mahan, C. E. Yunker, T. B.Crawford, A. F. Barbet, M. J. Burridge, and T. C. McGuire. 1991. A clonedDNA probe identifies Cowdria ruminantium in Amblyomma variegatum ticks.J. Clin. Microbiol. 29:2571–2577.
61. Yunker, C. E., B. Byrom, and S. Semu. 1988. Cultivation of Cowdria rumi-nantium in bovine vascular endothelial cells. Kenya Vet. 12:12–16.
62. Yunker, C. E., S. M. Mahan, S. D. Waghela, T. C. McGuire, F. R. Ruran-girwa, A. F. Barbet, and L. A. Wassink. 1993. Detection of Cowdria rumi-nantium by means of a DNA probe, pCS20, in infected bont ticks, Ambly-omma hebraeum, the major vector of heartwater in southern Africa.Epidemiol. Infect. 110:95–104.
