Veterinary Parasitology 196 (2013) 258–264

Contents lists available at ScienceDirect

Veterinary Parasitology

journa l homepage: www.e lsev ier .com/ locate /vetpar

Sero-surveillance for surra in cattle using native surfaceglycoprotein antigen from Trypanosoma evansi

Krishnendu Kundua, Anup Kumar Tewaria,∗, Samarchith P. Kurupa,1,Surajit Baidyaa,2, Jammi Raghavendra Raoa,3, Paritosh Joshib

a Division of Parasitology, Indian Veterinary Research Institute, Izatnagar 243122, UP, Indiab Division of Animal Biochemistry, Indian Veterinary Research Institute, Izatnagar 243122, UP, India

a r t i c l e i n f o

Article history:Received 9 July 2012Received in revised form 12 March 2013Accepted 6 April 2013

Keywords:GlycoproteinELISA

a b s t r a c t

Surra, caused by Trypanosoma evansi affects a wide range of domestic and wild animals inthe tropics, taking a huge toll on the already impoverished economy here. In bovines surranormally develops into a chronic infection that is often associated with severe productionlosses, yet with no distinct clinical signs making its adequate diagnosis vital. Though directmicroscopic observation of T. evansi in circulation may be the diagnostic gold standardfor surra, it is insensitive and impractical for population prevalence studies, making sero-diagnosis the preferred choice for the latter. In this study, we standardize an ELISA withConcanavalin-A (Con-A) affinity purified T. evansi surface glycoprotein antigen and compare

IndiaPrevalenceTrypanosoma evansi

its sensitivity and specificity to direct microscopy of stained thin smears and molecular(PCR) diagnostics. The ELISA was then put on field trial for sero-surveillance of cattle forsurra in three geographically distinct populations in the Indian subcontinent, to yield anoverall sensitivity and specificity of 100% and 89.15% compared to standard stained thinsmear examinations and 95.23% and 90.84% compared to blood PCR examinations.

1. Introduction

Surra, caused by Trypanosoma evansi is a disease syn-drome affecting a wide range of domestic and wild animalswith high prevalence in the tropical world, leading to

severe economic losses. The disease pathology varieswith the host, with an acute, clinically severe disease incanines and equines, and a more chronic, latent disease in

∗ Corresponding author. Tel.: +91 5812315194; fax: +91 5812302368.E-mail addresses: anuptewari1@rediffmail.com,

tewarianup@gmail.com (A.K. Tewari).1 Current address: Center for Tropical and Emerging Global Diseases,

The University of Georgia, Athens, GA 30605, USA.2 Current address: Department of Veterinary Parasitology, West Ben-

gal University of Animal and Fisheries Sciences, 37&68 Kshudiram BoseSarani, Kolkata, WB, India.

3 Current address: Emeritus Scientist, NAARM, Hyderabad, AP, India.

0304-4017/$ – see front matter © 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.vetpar.2013.04.013

© 2013 Elsevier B.V. All rights reserved.

ruminants. Though an acute form of the disease can occur inbovines, the latent form of the infection is most widespreadand responsible for the production losses in dairy herds(Tuntasuvan et al., 1997). There may also be sudden clini-cal outbreaks in the chronically infected individuals, owingto the various physiological, climatic or nutritional stress-ors (Otte et al., 1994; Seed et al., 1984), which may thenserve as foci for future epidemics. The semi-intensivenature of animal husbandry practices in the tropics putsother susceptible animal species also under severe risk insuch outbreaks. Moreover, the general immunosuppress-ive character of surra (Bajyana-Songa et al., 1987; Hollandet al., 2003; Tewari et al., 2009) also contributes to vacci-nation failures (Onah et al., 1997; Holland et al., 2001) andco-infections. These considerations make adequate diag-

nosis of latent surra in cattle a priority in the field.

Since the standard trypanosome detection methods(STDMs) like direct microscopy of stained thin smearsand mice hemo-transfer assays lack adequate diagnostic

Parasito

sdat2Escwusstfistt

2

2

asuwtublwJmsia

2

W2fp6av4e

w(ebsfdaat

K. Kundu et al. / Veterinary

ensitivity and is time consuming and ethically avoidable,evelopment and standardization of serological assaysre considered vital for effective diagnosis and control ofhe infection (OIE – Office International des Epizooties,004). T. evansi whole cell lysate antigen (Te-WCL) basedLISA has been conventionally used for sero-diagnosis ofurra (Luckins, 1977), but its low sensitivity and antigenicross reactivity has been an inconvenience. In this study,e develop, and validate a diagnostic ELISA for surrasing Concanavalin-A (Con-A) affinity purified T. evansiurface glycoprotein antigen (Te-GP). Using bovine seraamples from experimental infections and from suscep-ible cattle populations in three geographically distincteld locations in the Indian subcontinent we demonstrateuperior diagnostic sensitivity for Te-GP ELISA, comparedo Te-WCLELISA, blood PCR or direct microscopy of stainedhin smears.

. Materials and methods

.1. Experimental animals

Ten crossbred (Bos taurus × Bos indicus) male calves,ged 2–3 months, housed in fly/tick-proof large animalhelters, on standard feeding and watering regimen weresed in each trial. T. evansi infection free status of the calvesas confirmed by hemotransfer into susceptible mice and

esting for parasitemia by IFAT and PCR. Calves were inoc-lated sub-cutaneous (s.c.) with 5 × 105 viable T. evansilood stage forms (n = 6), or with PBS (n = 4). A buffalo iso-

ate of T. evansi maintained as cryostock in the laboratoryas used in the present study following revival in mice.

ugular blood was collected aseptically from the experi-entally infected calves at weekly interval for separation of

erum or for extraction of DNA for PCR. All animal exper-mentation protocols were approved by the institutionalnimal ethics committee of IVRI, Izatnagar.

.2. Antigen preparation for ELISA

The soluble whole cell lysate antigen of T. evansi (Te-CL) was prepared following the standard protocol (OIE,

004). In short, 6 × 108 T. evansi trypomastigotes isolatedrom infected Swiss mice were washed thrice in cold phos-hate saline glucose (PSG) buffer (pH 8.0), re-suspended in00 �l of PBS (pH 7.2) and sonicated (Soniprep 150, UK)t 8 Hz for 0.5 min on ice, three times with 5 min inter-al. The soluble fraction was separated by centrifugation at0,000 × g at 4 ◦C for 60 min and its protein concentrationstimated (Lowry et al., 1951), before storing at −80 ◦C.

The T. evansi surface glycoprotein antigen (Te-GP)as extracted by a two-step process as described before

Reinwald et al., 1981; Cook et al., 1985). In short, 1 × 109 T.vansi trypomastigotes were washed three times with PSGuffer, re-suspended in dioxane lysis buffer (1 ml, 0.01 Modium phosphate, 0.145 M NaCl, 0.1 M phenyl methyl sul-onyl fluoride (PMSF) with dioxane content of 1/20, v/v, in

eionized (DI) water, pH 7.2 and incubated at room temper-ture for 60 min, for partial lysis. The lysate was centrifugedt 4000 × g for 10 min at 4 ◦C to separate the soluble frac-ion.

logy 196 (2013) 258–264 259

To purify the Te-GP from this soluble fraction, Con-A sepharose column was used, following the standardprotocol (Pharmacia, 1979). The equilibrated (with theequilibration buffer: 150 mM sodium chloride, 10 mM Trishydrochloride, calcium chloride 1 mM and manganesechloride 1 mM in DI water, pH 7.5) column was loadedwith the soluble fraction of T. evansi partial lysate andwashed with equilibration buffer to remove the Con-Aunbound fractions. The Con-A bound Te-GP fraction waseluted with 0.1 M �-d-methyl mannoside and collectedas 1.5 ml aliquotes. The protein concentrations of all theeluted fractions were estimated using U.V. spectropho-tometer (Perkin Elmer, USA) and stored at −80 ◦C.

The T. evansi whole cell lysate, T. evansi partial (dioxane)lysate, the Con-A unbound or bound proteins in T. evansipartial (dioxane) lysate were analyzed by SDS-PAGE underdenaturizing conditions using 6–15% gradient gels. The gelswere stained with 0.1% Coomassie brilliant blue R-250.

2.3. ELISA

ELISA was performed by modifying the protocoloriginally standardized by Luckins (1977). Flat bottompolystyrene ELISA plates (Greiner, Germany) were coatedovernight with 0.5 �g Te-WCL or Te-GP antigens, blockedand 100 �l of sera samples loaded in duplicate. The plateswere incubated at 37 ◦C for 1 h, followed by the additionof the secondary, rabbit anti-bovine IgG-HRP conjugatedantibody (Sigma, USA) at a dilution of 1:10,000. Theplates were incubated further at 37 ◦C for 60 min, washedand developed with freshly prepared O-phenylenediamine(Sigma). The plates were read at 492 nm in an ELISA reader(Microscan, USA) after stopping the reaction with 50 �l of3 M HCl. The cut-off value for designating a sample sero-positive or to be from a reactor was determined by addingthree standard deviations to the mean O.D. values of thenaive calf sera obtained from the experimental trials.

2.4. Reference sera for ELISA

Positive reference sera were obtained from calvesexperimentally infected with T. evansi. Reference negativesera were from naïve (confirmed by PCR and mice hemo-transfer assays) bovine calves. All the infected calves weredrug cured at the end of the study with diminazine ace-turate (Berenil RTU, Intervet) at 10 mg/Kg body weight.Drug-cure was confirmed by PCR done in blood (DNA) atweekly intervals for seven consecutive weeks.

2.5. Stained thin smears for diagnosis of T. evansi

Blood smears made, fixed with methanol, stained withGiemsa stain and screened for T. evansi under a compoundlight microscope.

2.6. PCR for diagnosis of T. evansi

Approx. 1.5 ml of jugular blood was collected asep-tically from cattle, in sterile eppendorf tubes and thewhole DNA was extracted using a blood DNA extrac-tion kit (Promega, WI, USA). T. evansi infection status

Parasito

260 K. Kundu et al. / Veterinary

was determined by PCR using the Trypanozoon sub-genus specific diagnostic primers (forward primer, TeF1:5′TGCAGACGACCTGACGCTACT3′ and reverse primer, TeR1:5′CTCCTAGAAGCTTCGGTGTCCT3′ (Wuyts et al., 1994),given that T. evansi is the only reported member of thesubgenus prevalent in the region.

2.7. Collection of bovine blood samples from the field

The blood samples were collected from adult cattleof either sex, reared in either intensively housed (orga-nized) or semi-intensive/free ranging (un-organized) cattlefarms in three different states of India, viz. Kerala (n = 350),Odisha (n = 165) and Uttar Pradesh (n = 85) falling underthree different agro-climatic zones of southern, eastern andnorthern parts. Blood was collected from the jugular veinunder sterile conditions to separate the sera, prepare bloodsmears or to do PCR.

2.8. Statistical analysis

The sensitivity, specificity, positive and negative pre-dictive values were determined as previously described(Lalkhen and McClusckey, 2008). Te-GP ELISA was usedto test the samples, with true positive or negative refer-ence/control sera obtained from experimental infectionsin cattle, as confirmed by PCR, mice-hemotransfer or bymicroscopy in stained thin smears.

In short, Sensitivity = True positives/True posi-tives + False negatives (Samples positive by ELISA andreference test)/Samples positive by both tests + Samplespositive by reference test but negative by ELISA). Speci-ficity = True negatives/True negatives + False positives(Samples negative by both tests/Samples negative by bothtests + Samples negative by reference tests but positiveby ELISA). Positive Predictive Value = True positives/Truepositives + False positives (Samples positive by referencetests/Samples positive by both tests + Samples negative byELISA but positive by reference test. Negative PredictiveValue = True negatives/True negatives + False negatives(Samples negative by both tests/Samples negative by bothtests + samples negative by ELISA but positive by referencetest). All degrees of significance were determined byStudent’s t-test or chi-square test.

3. Results

3.1. Purification of native Te-GP

Native Te-GP was fractionated from the soluble portionof the partially lysed T. evansi by Con-A affinity chromatog-raphy. The soluble proteins of T. evansi that failed to bindto the Con-A matrix ran off the column to form the firstpeak in the elution profile, whereas the Con-A bound (Te-

GP) fraction was eluted to make the second peak (Fig. 1a).Te-GP appeared un-contaminated with irrelevant proteins,as determined by the elute’s resolution into a single bandof expected molecular weight (Fig. 1b).

logy 196 (2013) 258–264

3.2. Comparative evaluation of Te-GP and Te-WCL ELISAin the diagnosis of acute experimental surra in cattle

Sera samples were collected at weekly intervals fromcalves experimentally infected with T. evansi, and the Te-GP or Te-WCL antigen specific antibody response generatedwas determined by ELISA. Te-GP and Te-WCL antigensshowed significantly higher sero-reactivity with T. evansiinfected, than with naive cattle sera. Moreover, the abso-lute antibody titers observed with Te-GP antigen weresignificantly higher than with Te-WCL antigen (Fig. 2a). Allthe T. evansi infected calves in the acute phase of infection(up to 49 days post infection) showed detectable T. evansiDNA in circulation (Fig. 2b), though the parasites wereonly inconsistently observed in blood (data not shown).However, cattle in the chronic phase of experimental T.evansi infection (>300 days post infection (dpi)) thoughappeared sero-positive by ELISA with Te-GP or Te-WCLantigens, showed inconsistently detectable T. evansi DNAor parasites in blood (data not shown). Given that a self-cure of T. evansi infection is unlikely (Atarhouch et al.,2003) these results indicated that Te-GP antigen ELISA maybe a more dependable diagnostic tool to detect surra incattle.

3.3. Te-GP ELISA as a tool to assess sero-prevalence at apopulation scale

We employed Te-GP antigen to determine seropreva-lence of surra in three geographically distinct areas:Kerala, Odisha and Uttar Pradesh in the Indian subcon-tinent, primarily as a pilot trial for its application at thefield level. Theileriosis and babesiosis being the otherprotozoan infections with significant prevalence in theregion, we confirmed the absence of any sero-cross reac-tivity of sera from cattle experimentally infected withBabesia bigemina or Theileria annulata, with Te-GP (data notshown)

A total of 600 bovine blood/sera samples collected fromorganized or unorganized farms were tested for surra-by Te-GP ELISA (for anti-Te-GP IgG in sera), direct lightmicroscopy of stained thin smears (for T. evansi in blood)or by PCR (for T. evansi DNA in blood). Overall prevalenceof surra, assessed based on ELISA seemed to be signifi-cantly (Pearson chi-square test value: 2.888 at 2 degreesof freedom; p ≤ 0.05) higher than by direct blood smearexamination (Table 1). The pattern was similar, irrespec-tive of the geographical region or the farming practice. PCRbased diagnosis of surra showed prevalence levels, inter-mediate to that with Te-GP ELISA or direct light microscopyof stained thin smears, consistently in the three regions andwith either farming systems (Table 1). This was expectedgiven the sensitivity of PCR, yet the unlikelihood of T.evansi being in circulation at significantly high levels inchronic bovine surra. Te-GP ELISA yielded high sensitivityand specificity compared to direct microscopy of stainedthin smears or blood PCR, with adequate positive and neg-

ative predictive values (Table 2). Overall, the prevalence ofsurra was recorded at 10.66% with Te-GP ELISA, 1.66% bydirect light microscopy of stained thin smears or 3.5% withblood PCR.

K. Kundu et al. / Veterinary Parasitology 196 (2013) 258–264 261

Fig. 1. (A) Purification of native Te-GP from T. evansi partial lysate. Native Te-GP bound with Con-A fractionated from the soluble portion of T. evansi partiallysate with 0.1 M �-d-mannose elution. Total protein in the eluted fractions assayed by spectrophotometry. (B) Most of the Con-A bound fraction of T.e TeGP (∼( proteinr

4

wiccscclmoco(fim2c

FtdD

vansi partial lysate is native Te-GP. The relative representation of nativedioxane) lysate ((lane 2), Con-A unbound protein (lane 3) or Con-A boundesolution in SDS-PAGE.

. Discussion

The absence of pathognomonic clinical signs, coupledith the evasive nature of T. evansi, especially in chronic

nfections, make detection of surra in cattle extremelyhallenging (Nantulya, 1990; Ngaira et al., 2003). Thoughonventional diagnostics like direct light microscopy oftained thin smears and hemotransfer may be ideal inlinical cases of surra, they are unreliable in detectinghronic carriers, making the role of serological or molecu-ar diagnostics relevant to the diagnosis of surra. Given that

icroscopy or PCR based diagnostics rely on the presencef T. evansi (and hence its DNA) in circulation, their effi-iency may be compromised with lower levels or absencef parasitemia (OIE, 2011), yielding false negative resultsBengaly et al., 2001; Holland et al., 2001, 2004). There-ore, sero-diagnostics for specific anti-T. evansi antibodies

n circulation may be more reliable, considering that ani-

als once infected will remain infected (Atarhouch et al.,003) and hence sero-positive for its entire life, unless drugured. However, naturally diminishing serum antibody

ig. 2. Te-GP antigen yields stronger sero-reactivity than by Te-WCL antigen in ache sera of T. evansi infected or naïve cattle using Te-GP or Te-WCL antigens. Dataeviation. (B) T. evansi DNA specific amplification detected in the blood of T. evansiNA specific amplification observed in naïve (sero-negative) calves (lane 2).

50 kDa) in the total soluble fractions of Te-WCL ((lane 1), T. evansi partial(lane 4) from T. evansi partial lysate, as determined by molecular weight

titers associated with progressing surra in cattle demandssuperior antigens to improve the sensitivity of the sero-logical assays. Multiple antigens have been tested to thisend since the first description of ELISA for the diagnosis ofAfrican trypanosomes (Luckins, 1977). Individual native orrecombinant T. evansi antigens may be more sensitive andspecific, in addition to being economical and applicable tothe large-scale evaluations of the infection status in a herdor population (Reyna-Bello et al., 1998).

The T. evansi whole cell lysate antigenic preparations,routinely used for sero-diagnosis of surra (as recom-mended by OIE) contain many diagnostically irrelevantproteins often contributing to lower sensitivity and higherbackgrounds in serological tests. The variable surface gly-coprotein (VSG) coat of T. evansi is perhaps the mostimmunogenic among those and represents at least 10% ofthe total protein (Shapiro, 1989). Higher test sensitivity

was obtained in the sero-surveillance for surra in camels,where T. evansi VSG antigen was used (Shahardar, 1999). Inthis study, we found that sero-diagnosis using native Te-GPantigen yielded significantly stronger antibody titers with

ute T. evansi infection in cattle. (A) Antibody titers determined by ELISA inrepresentative of 3 separate experiments, error bars represent standardinfected sero-positive (Te-GP antigen) calves, 49 dpi (lane 1). No T. evansi

262K

.Kundu

etal./V

eterinaryParasitology

196(2013)

258–264

Table 1A comparative analysis of the prevalence of bovine surra in cattle reared in organized and unorganized farms determined by Te-GP ELISA, stained thin smears or PCR from three separate regions in the Indiansubcontinent.

Farming system Kerala Odisha Uttar Pradesh Overall prevalence PCR

Samples Prevalence PCR Samples Prevalence PCR Samples Prevalence PCR ELISA Stained thinsmears

ELISA Stained thinsmears

ELISA Stained thinsmears

ELISA Stained thinsmears

Organized farms 250 21 (8.4%) 3 (1.2%) 6 (2.4%) 105 12 (11.4%) 2 (1.90%) 4 (3.80%) 50 7 (14%) 1 (2%) 1 (2.0%) 40/405 (9.87%) 6/405 (1.48%) 11/405 (2.71%)Unorganized farms 100 11 (11%) 2 (2%) 5 (5%) 60 7 (11.66%) 1 (1.66%) 2 (3.3%) 35 6 (14.28%) 1 (2.85%) 3 (8.57%) 24/195 (12.30%) 4/195 (2.05%) 10/195 (5.1%)Total 350 32 (9.14%) 5 (1.42%) 11 (3.14%) 165 19 (11.51%) 3 (1.82%) 6 (3.63%) 85 13 (15.29%) 2 (2.85%) 4 (4.7%) 64/600 (10.66%) 10/600 (1.66%) 21/600 (3.5%)

Table 2Relative sensitivity, specificity, positive and negative predictive values of Te-GP ELISA compared to direct microscopy or PCR inbovine sera or blood samples collected from three separate regions in the Indiansubcontinent.

Stained thin smears Total samples PCR Total samples Sensitivity(estimated)

Specificity(estimated)

Positive predictivevalue

Negative predictivevalue

Negative Positive Negative Positive Stainedthin smears

PCR Stainedthin smears

PCR Stainedthin smears

PCR Stainedthin smears

PCR

ELISAPositive 64 10 74 53 20 73Negative 526 0 526 526 1 527 100% 95.23% 89.15% 90.84% 13.51% 27.39% 100% 99.81%

Total samples 590 10 600 579 21 600

Parasito

mW

tTdbssia(tndpftHanbecpuatprirasTt–rd7s(ci

bserepUfrordf2Ec

K. Kundu et al. / Veterinary

inimal non-specific binding, compared to with native Te-CL in acute surra of cattle.Any sero-diagnostic test developed for protozoal infec-

ions in animals have come with its own shortcomings.hough various tests have been recommended for theiagnosis of trypanosomosis, most of them appear toe inconsistent in reliably detecting the infection. Thetandard latex agglutination and the CATT failed to deliveratisfactory results in chronic experimental infections orn field studies but ELISA and immuno-trypanolysis testsppeared better, making immunological assays favoredHolland et al., 2005) in distinguishing true negatives fromrue positives in these cases. Holland et al. (2005) alsooted that ELISA using the RoTat 1.2 VSG was better atifferentiating the sero-negative and sero-positive sam-les in T. evansi infected pigs. Since cattle trypanosomosisollow a chronic course of infection as in pigs, we hopedo extrapolate the VSG based ELISA in cattle infections.owever, though RoTat 1.2 VSG is the most predominantntigen, all the T. evansi isolates do not express it, withon-reactivity to RoTat 1.2 specific antibodies in westernlot, probably contributing to false negative reactions inpidemiological studies (Ngaira et al., 2003). Given thatross reactivity of T. evansi RoTat 1.2 VSG with other try-anosome species (e.g. T. equiperdum (Claes et al., 2002)),sing T. evansi with non RoTat1.2 VSG appeared more suit-ble. The soluble antigens of T. evansi have been showno identify immunoglobulins from different isolates of thearasite (Laha and Sasmal, 2008) but with evidence of crosseactivity with other protozoal antigens in a multi-speciesnfection among livestock (OIE, 2011). The present studyevealed sero-conversion in experimentally infected calvess early as 1-week post T. evansi infection that progres-ively increased in titers, as determined by Te-GP ELISA ande-WCL ELISA. Further, Te-GP ELISA showed no cross reac-ivity with either the B. bigemina or T. annulata infectionsthe two predominant protozoal infections in cattle of the

egion. Recently Tran et al. (2009) reported that the sero-etection potential of a recombinant invariant glycoprotein5 (ISG75) for detection of heterologous trypanosomepecies was quite comparable to the immune-trypanolysisTL) test, which is believed to be very sensitive, though theross reactivity with other protozoans may hinder its utilityn an epidemiological survey.

Historically, the prevalence data for surra in cattle haseen inadequate from the Indian subcontinent thoughurra in cattle is thought to be widely prevalent in thentire south-east Asia. T. evansi prevalence of 1.42% wasecorded from Andhra Pradesh by stained thin smears (Dast al., 1998). Buffaloes in the Punjab region showed a sero-revalence for surra of 7.92% (Singla et al., 2004). Similarly,lHasan et al. (2006) reported a prevalence of 3.3% or 4%

or surra in camels from Pakistan by STDM or serologyespectively. The use of defined antigens in the diagnosisf cattle surra in the Indian subcontinent has also beenare. Glycoprotein antigen based competitive ELISA andouble antibody sandwich ELISA has been reported earlier

or sero-detection of T. evansi in camels (Shahardar et al.,004; Singh et al., 2004). Recombinant glycoprotein basedLISA was shown to have similar diagnostic performanceompared to native glycoprotein in detection of surra in

logy 196 (2013) 258–264 263

camels (Lejon et al., 2005). The lack of a standardized sero-diagnostic assay for surra in cattle that has been adequatelyvalidated in the sub-continent prompted us to develop andevaluate a T. evansi native antigen based serological assay.We here used Te-GP ELISA to assess sero-prevalence ofsurra in cattle from three geographically distinct regionswith varying farming practices, in the Indian subcontinent.

Diagnosis of chronic surra in cattle is complicated, asis in any other chronic infection. Though T. evansi maybe undetectable by PCR, stained thin smears or micehemo-transfer assays, cattle may still be infected, withthe parasites sequestered in tissue fluids, with only occa-sional re-appearances in circulation. Serological studieshave an edge in such cases, since animals once infected;remain infected for life (Atarhouch et al., 2003). In fieldapplications, where majority of animals may be chronicallyinfected, Te-GP ELISA would offer an overall better sensi-tivity, specificity, positive and negative predictive valuescompared to stained thin smears or PCR based diagno-sis. Nevertheless, mere presence of specific anti-T. evansiantibodies in sera may not be suggestive of an active infec-tion; drug-cured animals may also appear sero-positivefor a considerable time. Trypanosome specific antibodieswere detected up to 83 days post drug treatment in cattle(Luckins, 1977) and up to 69–78 days post drug treat-ment in horses (Monzon et al., 2003). Though false positiveresults with Te-GP ELISA are a possibility, we believe theadvantages of only seldom missing a true positive sampleoutweigh this deficiency, principally from an epidemiolog-ical standpoint. Moreover, Te-GP ELISA may be of greatervalue in the herd testing for quarantine measures and thedetermination of Surra free status of a population. Overall,we propose that Te-GP ELISA may be suitable for verifyingthe disease-free status of a population prior to movementor during quarantine, as proposed by OIE, with the pos-itive animals being further assessed by other diagnosticmethods.

Conflict of interest

We report no conflict of interests of any kind among theauthors.

Acknowledgements

The authors acknowledge the Director, IVRI for the facil-ities provided. This research was supported by the researchproject grant awarded by Indian Veterinary Research Insti-tute.

References

Atarhouch, T., Rami, M., Bendahman, M.N., Dakkak, A., 2003. Camel try-panosomosis in Morocco 1: results of a first epidemiological survey.Vet. Parasitol. 111, 277–286.

Bajyana-Songa, E., Kageruka, P., Hamers, R., 1987. The use of card agglu-tination test (Test tryp CATT) for the detection of Trypanosoma evansi

infection. Ann. Soc. Belge. Med. Trop. 67, 51–57.

Bengaly, Y.Z., Kasbari, M., Desquesnes, M., Sidibe, I., 2001. Validationof a polymerase chain reaction assay for monitoring the therapeu-tic efficacy of diminazeneaceturate in trypanosome-infected sheep.Vet. Parasitol. 96, 101–113.

Parasito

264 K. Kundu et al. / Veterinary

Cook, G.A., Honigberg, B.M., Zimmerman, R.A., 1985. Isolation and cellfree synthesis of variant surface glycoproteins from Trypanosomacon-golense. Mol. Biochem. Parasitol. 15, 281–294.

Das, A.K., Nandi, N.C., Mohankumar, O.R., 1998. Prevalence of bovine surrain Guntur district, Andhra Pradesh. Indian Vet. J. 75, 526–529.

Holland, W.G., My, L.N., Dung, T.V., Thanh, N.G., Tam, P.T., Vercruysse,J., Goddeeris, B.M., 2001. The influence of T. evansi infection on theimmuno-responsiveness of experimentally infected water buffaloes.Vet. Parasitol. 102, 225–234.

Holland, W.G., Do, T.T., Huong, N.T., Dung, N.T., Thanh, N.G., Vercruysse,J., Goddeeris, B.M., 2003. The effect of Trypanosoma evansi infectionon pig performance and vaccination against classical swine fever. Vet.Parasitol. 111, 115–123.

Holland, W.G., Thanh, M.G., My, L.N., Magnu, S.E., Do, T.T., Goddeeris, B.M.,Vercruysse, J., 2004. Prevalence of Trypanosoma evansi in water buf-faloes in remote areas of Northern Vietnam using PCR and serologicalmethods. Trop. Anim. Health. Prod. 36, 45–48.

Holland, W.G., Thanh, N.G., Do, T.T., Sangmaneedet, S., Goddeeres, B.,Vercruysse, J., 2005. Evaluation of diagnostic tests for Trypanosomaevansi in experimentally infected pigs ans subsequent use in field sur-veys in North Vietnam and Thailand. Trop. Anim. Health. Prod. 37,457–467.

Laha, R., Sasmal, N.K., 2008. Characterisation of immunogenic proteins ofTrypanosoma evansi isolated from three different Indian hosts usinghyperimmune serum and immune sera. Res. Vet. Sci. 85, 534–539.

Lalkhen, A.G., McClusckey, A., 2008. Clinical tests: specificity and sensitiv-ity. Continuing education in anaesthesia. Crit Care Pain. 8, 221–223.

Lejon, V., Claes, F., Verloo, D., Maina, M., Urakawa, T., Majiwa, P.A., Buscher,P., 2005. Recombinant Rotat 1.2 variable surface glycoprotein as anti-gen for diagnosis of Trypanosoma evansi on dromedary camels. Int. J.Parasitol. 35, 455–460.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Proteinmeasurement with the Folin-Phenol reagents. J. Biol. Chem. 193,265–275.

Luckins, A.G., 1977. Detection of antibodies in trypanosome-infectedcattle by means of microplate enzyme-linked immunosorbent assay.Trop. Anim. Health Prod. 9, 53–62.

Monzon, C.M., Mancebo, O.A., Russo, A.M., 2003. Antibody levels inindirect ELISA test in Trypanosoma evansi infected horses followingtreatment with Quinapyramine sulphate. Vet. Parasitol. 111, 59–63.

Nantulya, V.M., 1990. Trypanosomiasis in domestic animals: the problemsof diagnosis. Rev. Sci. Technol. 9, 357–367.

Ngaira, J.M., Bett, B., Karanja, S.M., Njagi, E.N., 2003. Evaluation of antigenand antibody rapid detection tests for Trypanosoma evansi infectionin camels in Kenya. Vet. Parasitol. 114, 131–141.

OIE, (Chapter 2.1.17)2004. Trypanosoma evansi infection (Surra). In: OIE

Manual of Standards for Diagnostic Tests and Vaccines for TerrestrialAnimals. World Organisation of Animal Health (Ed.), Paris, France.

OIE,2011. Trypanosoma evansi infection (Surra). In: OIE Manual of Stan-dards for Diagnostic Tests and Vaccines for Terrestrial Animals. WorldOrganisation of Animal Health (Ed.), Paris, France (Chapter 2.1.17).

logy 196 (2013) 258–264

Onah, D.N., Hopkins, J., Luckins, A.G., 1997. Effects of Trypanosoma evansion the output of cells from a lymph node draining the site of Pasteurellahaemolytica vaccine administration. J. Comp. Pathol. 117, 73–82.

Otte, M.J., Abuabara, J.Y., Wells, E.A., 1994. Trypanosoma vivax in Colombia:epidemiology and production losses. Trop. Anim. Health Prod. 26,146–156.

Pharmacia, 1979. Affinity Chromatography: Principles and Methods.Pharmacia Fine Chemicals, Uppsala, Sweden.

Reinwald, E., Rautenberg, P., Risse, H.J., 1981. Purification of the variantantigens of Trypanosoma congolense new approach to the isolation ofglycoproteins. Biochem. Biophys. Acta 668, 119–131.

Reyna-Bello, A., Garcia, F.A., Rivera, M., Sanso, B., Aso, P.M., 1998.Enzyme-linked immunosorbent assay (ELISA) for detection of

anti-Trypanosoma evansi equine antibodies. Vet. Parasitol. 80,149–157.

Seed, J., Edwards, R., Sechelski, J., 1984. The ecology of antigenic variation.J. Protozool. 31, 48–53.

Shahardar, R.A., 1999. Detection of Trypanosoma evansi in Dromedarycamels by enzyme immunoassays and Polymerase Chain Reaction torDNA target. Ph.D. Thesis, Submitted to Deemed University, IVRI.

Shahardar, R.A., Rao, J.R., Mishra, A.K., 2004. Detection of antibodiesagainst Trypanosoma evansi in dromedary camels by competitive inhi-bition enzyme-linked immunosorbent assay. J. Appl. Anim. Res. 24,41–48.

Shapiro, S.Z., 1989. The potential for Trypanosoma vaccine development.In: Wright, I.G. (Ed.), Veterinary Protozoan and Haemoparasite Vac-cines. CRC det Press Inc, Boca Raton, FL.

Singh, N., Pathak, K.M.L., Kumar, R., 2004. A comparative evaluation ofparasitological, serological and DNA amplification methods for diag-nosis of natural Trypanosoma evansi infection in camels. Vet. Parasitol.126, 365–373.

Singla, L.D., Aulakh, G.S., Juyal, P.D., Singh, J., 2004. Bovine trypanosomo-sis in Punjab, India. In: Proceedings of 11th International Conferenceof the Association of Institutions for Tropical Veterinary Medicineand 16th Veterinary Association Malaysia Congress, Petaling Jaya,Malaysia, pp. 283–285.

Tewari, A.K., Rao, J.R., Singh, R., Mishra, A.K., 2009. Histopathologicalobservations on experimental Trypanosoma evansi infection in bovinecalves. Ind. J. Vet. Pathol. 33, 85–87.

Tran, T., Claes, F., Verloo, D., De Greve, H., Buscher, P., 2009. Towards aNew Reference Test for Surra in camels. Clin. Vaccine Immunol. 16,999–1002.

Tuntasuvan, D., Sarataphan, N., Nishikawa, H., 1997. Cerebral trypanoso-mosis in native cattle. Vet. Parasitol. 73, 357–363.

UlHasan, M., Muhammad, G., Gutierrez, C., Iqbal, Z., Shakoor, A., Jabbar,A., 2006. Prevalence of Trypanosoma evansi infection in equines and

camels in the Punjab region, Pakistan. Ann. N. Y. Acad. Sci. 1081,32–324.

Wuyts, N., Choke Sajjawattee, N., Panyim, S., 1994. A simplified andhighly sensitive detection of Trypanosoma evansi by DNA amplifica-tion. Southeast Asian J. Trop. Med. Publ. Health 25, 266–327.