BiP inhibition ELISA

Veterinary Parasitology 173 (2010) 39–47

Contents lists available at ScienceDirect

Veterinary Parasitology journal homepage: www.elsevier.com/locate/vetpar

Serodiagnosis of bovine trypanosomosis based on HSP70/BiP inhibition ELISA Geraldine Bossard a,∗ , Alain Boulange a,b , Philippe Holzmuller a , Sophie Thévenon a,c , Delphine Patrel a , Edith Authie d a b c d

UMR17 Trypanosomes, IRD-CIRAD, Campus International de Baillarguet, F-34398 Montpellier, France Biochemistry, University Eduardo Mondlane, Maputo, Mozambique CIRDES, BP454 Bobo Dioulasso, Burkina Faso Direction de l’évaluation des Risques Nutritionnels et Sanitaires (DERNS), 27-31 Avenue du Général Leclerc, 94701 Maisons-Alfort Cedex, France

a r t i c l e

i n f o

Article history: Received 26 March 2010 Received in revised form 8 June 2010 Accepted 14 June 2010 Keywords: Serodiagnosis Inhibition ELISA Bovine trypanosomosis HSP70/BiP

a b s t r a c t Animal trypanosomosis is a serious constraint to livestock productivity in tropical and subtropical countries. The pathogenic trypanosomes in bovidae are Trypanosoma congolense, T. vivax, T. brucei and T. evansi. Current serological tests to detect trypanosome infections are based on the use of whole trypanosome lysates; their potential is limited by antigen instability, lack of reproducibility and lack of test specificity due to the antibody’s long persistence after treatment. The development of new tests based on recombinant technology that could be standardized and applied on a large scale at low cost would be very helpful. The major invariant antigen recognized by T. congolense infected cattle belongs to the heat shock protein (HSP) 70 family and is closely related to mammalian Immunoglobulin Binding Protein (BiP). To improve the initial ELISA based on a recombinant fragment of HSP70/BiP, we developed an inhibition ELISA using an anti-BiP monoclonal antibody and a full-length fusion protein expressed in E. coli. Here we report on the development of the test and provide an initial assessment of its performance using sets of sera from experimental infections and from naturally infected cattle maintained in tsetse infested areas of Africa. The HSP70/BIP-based inhibition ELISA shows a good sensitivity in cattle experimentally infected with T. congolense, with an improved sensitivity in secondary infections. One major advantage, particularly for its further application in national laboratories, is that one single set of reagents and one single procedure are sufficient to apply on different mammalian host species infected with different trypanosome species. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Animal trypanosomosis is one of the most severe constraints to agricultural development in Sub-Saharan Africa and is also an important disease of livestock in Latin America and Asia. The causative agents are various species of protozoan parasites belonging to the genus Trypanosoma,

∗ Corresponding author at: UMR17 Trypanosomes, IRD-CIRAD, Campus International de Baillarguet, F-34398 Montpellier, France. E-mail address: [email protected] (G. Bossard). 0304-4017/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.vetpar.2010.06.016

among which Trypanosoma congolense, T. brucei, T. vivax and T. evansi are the pathogenic species in ruminants. In sub-Saharan Africa, trypanosomes are most often transmitted by tsetse flies (Glossina spp.), while the trypanosome species that affect other continents are generally transmitted mechanically (Sumba et al., 1998; Lai et al., 2008). African herdsmen rely on anaemia, weakness and weight loss, the most prominent symptoms, to detect trypanosomosis and undertake trypanocidal treatment. However, these clinical features are not specific for trypanosomosis and other anaemia-causing pathogens often co-exist in endemic areas. Biological tools are still

40

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

required for an accurate diagnosis of trypanosome infections. Current diagnostic methods are based on detection of parasites in blood or detection of specific antibodies, antigens or specific DNA sequences (Luckins, 1992). While direct parasitological techniques are not very sensitive, DNA-based approaches are costly and technologically demanding. Serological tests based on detection of trypanosome-specific antibodies are useful for large scale epidemiological studies, particularly the follow-up of control campaigns (Greiner et al., 1997a). A rapid and specific serological test was developed for T. evansi infections in animals (Bajyana Songa and Hamers, 1988), according to the principle of the card agglutination test based on a detection of variant of a T. brucei gambiense clone in human sleeping sickness (CATT test, Bajyana Songa et al., 1987), but this test is not applicable to infections with T. congolense and T. vivax. Thus, most antibody detection tests for bovine trypanosomosis are based on the use of whole trypanosome lysates (Greiner et al., 1997a; Desquesnes et al., 2001, 2009). The potential of these methods is limited by antigen instability, lack of reproducibility, lack of test specificity due to persistence of antibodies after treatment, as well as ethical considerations (when large numbers of rodents are required to produce sufficient amounts of bloodstream forms for antigen preparation). Thus, serological tests based on recombinant technology that could be standardized and applied on a large scale at low cost would be very useful. Diagnostic candidates may be identified among non-variant trypanosome molecules (as opposed to the Variant Surface Glycoprotein) that are highly antigenic in the mammalian host. The availability of kinetoplastid genome sequences will further the search for reagents used to develop new serodiagnostics for African trypanosomosis (Hutchinson et al., 2004). However, few candidate molecules have been validated so far, leading to operational diagnostic tests (Camargo et al., 2004; Claes et al., 2005; Mendoza-Palomares et al., 2008; Tran et al., 2009). Despite their high degree of conservation in eukaryotic and prokaryotic cells, heat shock proteins (HSP) 70 are highly antigenic and at least two members of this family have been proposed as diagnostic candidates for trypanosomosis. A mitochondrial HSP70 of T. congolense may have diagnostic potential since it elicits antibody in infected mice (Bannai et al., 2003). Prior to the latter report, we discovered that the major invariant antigen recognized by T. congolense infected cattle is an HSP70, which is homologous to mammalian Immunoglobulin Binding Protein (BiP) (Authié et al., 1993; Boulangé and Authié, 1994). The T. congolense BiP is closely related to other members of the BiP sub-family though it has little homology with the mitochondrial HSP70 described by Bannai et al. (2003). The antigenicity of the trypanosomal HSP70/BiP in the mammalian host appears to result from a C-terminal stretch of 40 amino acids which is trypanosome-specific (Boulangé et al., 2002). BiP homologues are also present in T. brucei (Bangs et al., 1993) and T. vivax (Boulangé and Authié, 1994). We developed monoclonal antibodies against the native T. congolense HSP70/BiP and expressed HSP70/BiP as recombinant antigens in a bacterial expression system. A 25-kDa fragment (C-25), comprising the C-terminal third of the molecule, was then selected to develop an antibody

detection ELISA (Boulangé et al., 2002). The test detected both T. congolense and T. vivax infections with a sensitivity that was low in primary infections, but improved drastically during secondary infections. The major advantage of the C-25 based ELISA, as compared to the test using whole trypanosome extract, was the return to baseline values after trypanocidal treatment (Boulangé et al., 2002). As antigen preparation involved tedious procedures that limited the test applicability, we have now developed an inhibition ELISA, using an anti-HSP70/BiP monoclonal antibody and a full-length fusion protein expressed in E. coli. Here we present the development of the test and an initial assessment of its performance using large sets of sera from experimental infections (longitudinal studies) and a more limited sampling from naturally infected cattle maintained in tsetse infested areas of Burkina Faso (cross-sectional study). 2. Materials and methods 2.1. Sera from cattle not infected with Trypanosoma sp. A total of 127 negative sera from cattle, raised in a tsetsefree area, were obtained from ILRI-Nairobi (International Livestock Research Institute, Nairobi, Kenya). They consisted of 113 samples from Boran cattle (Bos indicus, zebu type) and 14 sera from N’Dama cattle (Bos taurus). In addition, to test for serological cross-reactivity with major tick-born pathogens frequently found in trypanosome-infected areas, sera were obtained from 6 cattle experimentally infected with either Theileria parva (n = 2), Babesia bigemina (n = 2), or Anaplasma marginale (n = 2). Sera were collected from each individual at 3-day intervals until day 52 post-inoculation and at 7-day intervals until the end of the experiment at day 296. 2.2. Sera from cattle undergoing trypanosome experimental infections 2.2.1. Primary infections with different trypanosome species and clones Adult Boran cattle (Bos indicus) reared in a tsetse-free area of Kenya were infected at ILRI-Nairobi through 5–10 tsetse fly bites, with cloned trypanosomes of different species and origins. Each individual animal was infected with either one of the following clones: T. congolense IL3575 (Nantulya et al., 1984), T. congolense IL3898 (isolated from a Baoule cow in Burkina Faso in 1960, referenced at ILRI, unpublished), T. brucei IL3394 (isolated from a pig in Nigeria in 1962, referenced at ILRI, unpublished), T. brucei IL3303 (Nantulya et al., 1984), T. vivax IL2172 (LeeFlang et al., 1976) and T. vivax IL3769 (isolated in Uganda in 1960, referenced at ILRI, unpublished). Parasitaemia was estimated according to Murray et al. (1977) and packed red cell volume (PCV) or hematocrit was monitored in cattle every 3 days for 60 days post-infection. In a separate experiment, 40 Boran cattle were infected with a T. congolense clone IL1180 (Nantulya et al., 1984), through 5–10 tsetse fly bites. All cattle were monitored twice a week during 8 months for parasitaemia and PCV. Sera were collected weekly. Trypanocidal treatment with

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

diminazene aceturate was administered after 8 months of infection, after which sera were collected every week for a further 4 months. 2.2.2. Rechallenge infections Six Boran cattle that had undergone primary infection with T. congolense were subsequently rechallenged on five occasions with four different serodemes of T. congolense. They were subjected to a sixth rechallenge infection with T. congolense IL13-E3 as previously described (Authié et al., 1993; Williams et al., 1991). Sera were collected at 3-day intervals for 1 month. 2.3. Sera from naturally infected cattle A total of 19 sera were collected from 2- to 3-year old crossbred cattle (Baoulé × West African Zebu), exposed to natural infections in the endemic area of Folonzo, South Burkina Faso. Blood was examined after microcentrifugation for the presence of trypanosomes in the buffy coat (Murray et al., 1977; Paris et al., 1982) and by PCR using the specific primers for T. brucei, T. congolense savannah type and T. vivax (Masiga et al., 1992). Sera were categorized according to the parasitological status (positive after microscopic examination) of each individual and were subjected to both indirect ELISA using trypanosome extracts as antigens according to Desquesnes et al. (2001) and HSP70/BiP inhibition ELISA. 2.4. Production of recombinant maltose binding protein (MBP)-HSP70/BiP fusion protein The construct encoding the complete mature T. congolense BiP (devoid of the signal peptide) was amplified by PCR from a T. congolense (IL3000 strain) full-length BiP cDNA clone (Boulangé and Authié, 1994) using the following primers (where the restriction sites are shown in bold, the restriction enzymes in brackets, the stop codon in italics): GCC ACC GAA TTC CCC GAG AGC GGC GGG (EcoRI) and GCT AAC GGA TCC TTA AAG GTC ATC CAT GGG (BamHI). The PCR product obtained was purified using Promega WizardTM PCR preps (Promega Corp., Madison, WI) and was cloned in a pGEM® -T-vector (Promega). The integrity of the sequence was verified by DNA sequencing and the fragment was subsequently subcloned into the pMAL-cRI plasmid (New England Biolabs, Inc. Ipswich, MA) by directional cloning using EcoI–BamHI sites. The DNA construct was transformed into the BL21 strain of E. coli by the heat shock method according to the manufacturer’s instructions. For protein production, an overnight culture of pMAL-BiP in Luria Broth (LB) was diluted 1:100 in an LB medium containing 50 ␮g/ml ampicillin, allowed to grow to a density corresponding to A600 = 0.6 and expression induced by adding isopropyl-thio-␤-d-galactoside (IPTG) to a final concentration of 0.3 mM. The medium was supplemented every hour with 50 ␮g/ml ampicillin during induction. The bacteria were harvested 4 h after induction by centrifugation and the pellet frozen at −70 ◦ C until use. The pellet was resuspended in a column buffer (100 mM NaCl, 20 mM Tris pH 7.5, 1 mM EDTA, 2 mM DTT) containing 1 mM phenylmethanesulfonyl fluoride and 1 mM

41

E-64 and subjected to sonication at 4 ◦ C. The lysate supernatant was collected by centrifugation at 6000 × g at 4 ◦ C for 30 min, and loaded onto a pre-equilibrated amylose column (New England Biolabs). The column was washed with 12 volumes of column buffer and the MBP-BiP fusion protein eluted with 10 mM maltose. All protein concentrations were measured using Coomassie (Bradford) Protein Assay (Pierce Biotechnology, Inc. Rockford, IL) and verified by SDS-PAGE. 2.5. Inhibition ELISA using mAb 1F3 mAb 1F3, an IgG1 that binds an epitope located in the C-terminal region of the HSP70/BiP molecule (Boulangé et al., 2002), was selected. Preliminary titration experiments were conducted with known positive and negative bovine sera to determine optimal concentrations of mAb 1F3 hybridoma supernatant, serum sample and anti-mouse conjugate. This was carried out using standard checkerboard titration procedures (Crowther, 1995). The optimal concentration of mAb (limiting concentration) was determined by omitting the serum incubation step in the procedure described below. The reaction was conducted in 96-well microtiter plates (Immulon 4HBX Dynex, VI) with a volume of 100 ␮l per well. Each serum sample was tested in triplicate. The plates were coated with 0.1 ␮g per well of purified fusion protein MBP-HSP70/BiP diluted in 20 mM phosphate buffer saline (PBS), pH 7.4, for 1 h at 37 ◦ C. The coating solution was discarded and the plates were blocked with 100 ␮l PBS containing 0.2% Bovine Serum Albumin and 0.05% Tween 20 (blocking buffer) for 1 h at 37 ◦ C. After the blocking step, the plates were incubated for 1 h at 37 ◦ C with bovine sera diluted 1:40 in blocking buffer. The sera were then discarded and the plates were washed one time with PBS. mAb 1F3 hybridoma supernatant diluted 1:2000 in a blocking buffer was added to the plates and incubated for 1 h at 37 ◦ C. The plates were washed twice with PBS containing 0.1% Tween 20 and once with PBS alone. Horseradish peroxidase-conjugated sheep antimouse whole IgG (Amersham GE healthcare, Chalfon St Gilles, UK) diluted 1:2500 in a blocking buffer was added to the plates and incubated for 1 h at 37 ◦ C. At the end of incubation, the conjugate was discarded, the plates were washed and K-Blue substrate (Neogen Corp., Lansing, MI) was added to the plates as the chromogen. The plates were kept in the dark for 30 min at room temperature, after which the absorbance at a wavelength of 620 nm (A620 ) was measured using a spectrophotometer (Wallac Victor2 1420 multilabel counter, PerkinElmer Inc., Wellesley, MA). The mean and standard deviation of the triplicate measurements were calculated for each sample (A620 sample). The 0% inhibition value was given, for each ELISA plate, by the mean absorbance of 4 wells receiving blocking buffer instead of serum (A620 mAb), all other steps being identical. Negative controls were obtained from 2 wells receiving only conjugate and substrate (serum and mAb were omitted). High and low positive control sera were obtained from rechallenged N’Dama cattle, using results from a western blot analysis of responses (Authié et al., 1993). A pool of

42

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

Fig. 1. Percentages of inhibition obtained with 127 negative sera collected from trypanosome-uninfected cattle submitted to HSP70/BiP inhibition ELISA. Histogram bars represent the number of sample with a percent inhibition value included in the corresponding interval.

negative control sera was obtained from Boran cattle born and raised in a trypanosome-free area. Results were expressed as a percentage of inhibition of mAb 1F3 binding, as per the calculation below:

2 (mean + 2SD) covered 96% of the population, while giving 4% of false positive. The latter two cut-off values were used in parallel when analyzing results subsequently reported in this paper.

100 × (A620 mAb − A620 sample) A620 mAb

3.2. Repeatability of the test

Repeatability (agreement between replicates within and between runs of the assay) was determined repeating the test five times on 20 sera, half of which was from naive nonexposed cattle and the other half was taken at the peak of anti-HSP70/BiP antibody response in cattle experimentally infected with T. congolense. In order to obtain an initial assessment of reproducibility (degree of concordance using the identical assay (protocol, reagents, and controls) between experimenters and laboratories data), a set of 20 bovine sera were tested independently by two different experimenters and their concordance was assessed using Cohen’s kappa test (Cohen, 1960). The specificity of the test, namely its ability to give negative results with sera from non-infected cattle, was determined using a set of negative sera from cattle not exposed to a trypanosome challenge. The sensitivity of the test, namely its ability to detect specific antibodies in cattle showing an active infection by T. congolense, was determined using two sets of positive sera: 325 samples from 40 cattle undergoing primary infections and 101 samples from 6 cattle undergoing rechallenge infections. 3. Results 3.1. Determination of the cut-off value of the test Fig. 1 shows the distribution of percent inhibition values given by 127 sera from naive healthy cattle from Kenya. The cut-off value was calculated by the sequential addition of one standard deviation (SD) to the mean percentage of inhibition until reaching a value that allowed inclusion of 100% of the sera, as illustrated in Table 1. Cut-off 3 (mean + 3SD) covered the entire population of 127 negative sera. Cut-off

Depending on the chosen cut-off value, few false positive results were obtained with sera from non-infected bovine (Supplementary data Table 1). When considering cut-off 3, 1 out of 10 animals gave one false positive result out of 5 independent tests. This represents 98% of tests that gave the same result. When considering cut-off 2, 2 out of 10 animals gave two false positive results out of 5 independent tests (repeatability of 92%). Sera from trypanosome-infected cattle were detected positive, independently of the chosen cut-off value (Supplementary data Table 1). 3.3. Reproducibility As shown in Supplementary data Table 2, mean inhibition values obtained with sera from naive animals obtained by experimenter 1 and experimenter 2 were 9.3 (SD = 3.2) and 8.7 (SD = 3.0), respectively. Mean inhibition values obtained with sera from infected animals obtained by experimenter 1 and experimenter 2 were 47.1 (SD = 8.8) and 44.0 (SD = 8.2), respectively (Supplementary data Table 2). All sera from naive animals were assessed as negative Table 1 Determination of cut-off values of T. congolense HSP70/ BiP inhibition ELISA with sera from naive cattle. %Inhibition

Cut-offa

Mean + 1SD Mean + 2SD Mean + 3SD

16.4 21.5 26.6

Number of sera

Negative

Positive

84 122 127

43 5 0

Total

Specificity (%)

127 127 127

66 96 100

a Cut-off value was calculated with 127 sera by the sequential addition of one standard deviation to the mean percent of inhibition until including all naive cattle sera (100% specificity).

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

43

and all sera from infected cattle were assessed as positive for both experimenters and using either cut-off 2 or 3. Kappa value was 1 showing a perfect concordance between the two experimenters. 3.4. Specificity of the test on sera from animals infected with other parasites Sera from cattle experimentally infected with T. parva, A. marginale or B. bigemina consistently gave inhibition percentages below the cut-off values. Individuals were negative for anti-HSP70/BiP antibodies throughout infections (Supplementary data Fig. 1). 3.5. Serological responses in experimentally infected cattle—longitudinal studies 3.5.1. Primary infections Antibody levels were monitored in Boran cattle (Bos indicus) following primary infection with three different clones of T. congolense, namely IL1180, IL3575 and IL3898 as illustrated in Fig. 2. Irrespective of the T. congolense clone used for infection, antibodies directed against HSP70/BiP appeared concomitantly with the first detection of parasitaemia, 10–15 days after the infective tsetse bite. There was a consistent peak of antibodies around day 20 postinfection (Fig. 2). In the 40 animals infected with clone IL1180, the percentage of cattle detected by inhibition ELISA during the course of infection is shown in Fig. 3. For 25 dpi, the sensitivity is 100% whatever the cut-off. Except for 56 dpi, the sensitivity of the test is between 90% and 100% for cut-off 3 and between 95% and 100% for cut-off 2, until trypanocidal treatment at 240 dpi (Fig. 3). For other clones of T. congolense represented in Fig. 2B (one individual infected with clone IL3575 and one individual infected with clone IL3898), there was a transient detection of HSP70/BiP antibodies (mainly from 15 to 25 dpi), concomitantly with the first wave of parasitaemia. Sera from primary infections with other trypanosome species, namely T. vivax (clones IL2172, IL3769) and T. brucei (clones IL3394 and IL3303) cross-reacted with T. congolense HSP70/BiP (Fig. 4). Sera from T. vivax infections exhibit the same overall profiles in detection of antiHSP70/BiP antibodies, despite only the individual infected with clone IL2172 was detected positive with both cut-off 2 and cut-off 3 (Fig. 4A). Anti-HSP70/BiP antibodies were weakly detected in T. brucei infected animals; individual infected with clone IL3303 was positive in serology late after the first wave of parasiteamia (Fig. 4B). 3.5.2. Response after trypanocidal treatment The persistence of anti-HSP70/BiP antibodies response after trypanocidal treatment with diminazene aceturate was examined in the 40 Boran cattle monitored during primary infection with T. congolense IL1180. Anti-HSP70/BiP antibodies decreased in a time-dependent manner after trypanocidal treatment (240 dpi) and the seroconversion (return to negative results in the inhibition ELISA) occurred 3 months after treatment regardless of the cut-off considered (Figs. 2A and 3). Nevertheless, 23% of the animals were

Fig. 2. Serological and parasitological parameters monitored during primary experimental infection of Boran cattle with T. congolense. (A) Clone IL1180, the arrow indicates the day of treatment with diminazene aceturate and (B) clones IL3898 and IL3575. Serology was determined using HSP70/BiP inhibition ELISA and expressed as average percent inhibition plus standard deviation. Parasitaemia was determined on fresh blood and expressed as a score according to Murray et al. (1977).

still positive 3 months post-treatment using cut-off 2 and only 5% with cut-off 3 (Fig. 3). 3.5.3. Secondary infections Following a T. congolense rechallenge infection, cattle developed high levels of anti-HSP70/BiP antibodies as shown by the high inhibition values measured in ELISA (Fig. 5). In contrast to sera from primary infected cattle, in which antibodies were detected from day 13 irrespective of the cut-off considered, sera from rechallenge infections were detected from day 7 (Fig. 5). It is noteworthy that the value of percent inhibition at the antibodies peak in primary infection is 47.3 ± 16.8 and almost doubled in the case rechallenge infection in the same period (74.2 ± 21.8) (Fig. 5). 3.6. Serological responses of cattle under natural exposure—cross-sectional studies Sera were categorized according to the parasitological status of each individual (Table 2).

44

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

Fig. 3. Detection potency of HSP70/BiP inhibition ELISA during experimental infection of Boran cattle with T. congolense IL1180. Individuals detected positive in serology with either cut-off 2 or cut-off 3 are expressed as percent of the total population The arrow indicates the day of treatment with diminazene aceturate.

Three out of five sera from parasitologically negative animals were negative both in indirect ELISA and inhibition ELISA. Two sera from parasitologically negative animals were positive both in indirect ELISA and inhibition ELISA. Sera from 14 trypanosome-infected cattle (parasitologically positive) were all positive in the inhibition ELISA with cut-off 2 (100% sensitivity), irrespective of the try-

panosome species involved (Table 2). Nevertheless, one serum was negative considering cut-off 3 in inhibition ELISA (92.9% sensitivity) and also negative in indirect ELISA (Table 2) while a high parasitaemia due to T. vivax was detected microscopically (Table 2). The sensitivity was thus 100% with cut-off 2 and 93% with cut-off 3. 4. Discussion In the present paper, we evaluated the ability of recombinant HSP70/BiP from T. congolense and an HSP70/BiP-specific monoclonal antibody to discriminate trypanosome-infected from non-infected bovine sera. The sensitivity was higher for experimental secondary infections than with primary infections. Sera from cattle undergoing natural infections were all detected applying cut-off 2 irrespective of the trypanosome species involved.

Fig. 4. Serological and parasitological parameters monitored during primary experimental infection of Boran cattle with T. vivax clones IL2172 and IL3769 (A) and T. brucei clones IL3394 and IL3303 (B). Each curve corresponds to one individual infected with one trypanosome clone. Serology was determined using HSP70/BiP inhibition ELISA and expressed as percent inhibition. Parasitaemia was determined on fresh blood and expressed as a score according to Murray et al. (1977).

Fig. 5. Average percent inhibition (+standard deviation) given by HSP70/BiP inhibition ELISA of sera from primary and rechallenge experimental infections with T. congolense. Boran cattle undergone primary infection and were subsequently rechallenged on five occasions with four different serodemes of T. congolense. They were subjected to a sixth rechallenge infection with T. congolense IL13-E3 as previously described (Authié et al., 1993; Williams et al., 1991).

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

45

Table 2 Comparison of results from HSP70/BiP inhibition ELISA and standard diagnostic tests for bovine trypanosomosis. Animal identificationa

Parasitologyb

Indirect ELISAc

HSP70/BiP-inhibitiond

Microscope

PCRb

TV

TB

ELISA

Cut-off 2

Cut-off 3

F2 F12 F45 F27 F36

− − − − −

− − − − −

17 0 6 10 20

11 0 7 3 17

9 0 5 71 95

13.0 6.9 10.6 79.5 65.2

− − − + +

− − − + +

F5 F11 F14 F22 F23 F28 F34 F37 F38 F43 F46 F47 F55 F56

TC TV TC/TB TC TC TV TV TC TC TC TV TC/TV, TB? TV (150/field) TC, TB, TV

+ + +/+ + + − + + + + − +/+/− nd +/+/+

38 24 70 61 9 36 60 30 32 48 75 10 16 67

51 12 60 22 21 45 45 24 37 45 62 12 18 42

113 77 115 58 59 68 143 80 85 104 134 94 8 116

67.5 49.0 42.9 40.7 31.8 40.7 68.6 65.4 47.3 67.9 47.6 38.0 25.1 49.2

+ + + + + + + + + + + + + +

+ + + + + + + + + + + + − +

TC

a

Identification code attributed to each animal. Parasitological status of animals was determined by both microscopic examination of fresh blood (Murray et al., 1977) and PCR (Masiga et al., 1992). Trypanosome species was first evaluated phenotypically: TC: T. congolense ; TV: T. vivax ; TB: T. brucei and confirmed by PCR “−”: negative; “+”: positive; “nd”: not determined c Indirect ELISA expressed as RPP (relative percentage of positivity) with a cut-off value of 20% (Desquesnes et al., 2001) d Inhibition ELISA expressed as percent inhibition, and associated positivity (+) or negativity (−) depending on the cut-off. b

Only one animal was just under cut-off 3 while negative in indirect ELISA and despite high T. vivax parasitaemia. T. congolense HSP70/BiP used in inhibition ELISA can now be considered as a promising diagnostic antigen candidate according to the definition of Hutchinson et al. (2004) and confirmed its previously established properties (Boulangé et al., 2002; Bossard et al., 2003). The establishment of a reliable cut-off value is essential in a serological test to be useful for differentiating infected from non-infected animals. Despite the fact that test values in populations targeted by the assay did not follow a normal distribution and that the use of two or three standard deviations above the mean to determine a cut-off would not be recommended for interpreting the serodiagnostic assays (Paweska et al., 2005), we showed that it allowed us to discriminate sera from infected animals from that of non-infected animals. In fact, the reference negative population used for calculating the cut-off is more critical. It must be large enough and representative of the target population (Greiner et al., 1994). This is always difficult to achieve with ELISAs designed for tropical diseases, as previously reported for cowdriosis (Mboloi et al., 1999). Control sera from pathogen-free areas are guaranteed to be antibody-free, but they may not be representative of the target population. Conversely, negative sera from endemic areas are not guaranteed to originate from pathogen-free animals. Another complicating factor is that the cut-off value is influenced by age and host factors, such as previous infections and immune status. Despite suggestions to establish age-specific cut-off values (Greiner et al., 1997b; Mahama et al., 2004), this would be very labour intensive and an additional difficulty would be to obtain groups of animals homogeneous in age in the field. Testing a large

set of non-infected cattle, representative of the total population, from endemic areas should be the object of further evaluation. 4.1. Specificity 4.1.1. Absence of cross-reactivity with other pathogens Despite the high conservation of HSP70/BiP molecules, there was no cross-reactivity of the antibody to the trypanosomal HSP70/BiP with either Theileria, Anaplasma or Babesia. Thus, the epitope recognized by mAb 1F3 and the immune system of infected cattle appears to be specific for trypanosomatids. The developed test does not detect antibodies for major anaemia-causing pathogens to which cattle are exposed in endemic areas. This is important in the frame of mass screening during veterinary surveys. 4.1.2. Specificity within the genus Trypanosoma The kinetics of serological responses in cattle undergoing primary infection, with either T. vivax or T. brucei, indicated that antibodies capable of inhibiting mAb binding to HSP70/BiP were elicited during infection. However, the low level of inhibition in these sera suggests only partial cross-reactivity with heterologous HSP70/BiP. The reasons that T. brucei elicits such low responses (little inhibition) in an experimental infection are unclear. Although cross-reactivity is expected with such highly conserved molecules, it is possible that the epitope recognized by mAb 1F3 is slightly different in T. brucei, thus altering both the affinity and the avidity of the mAb for the antigen. Another hypothesis that could explain the weak anti-HSP70/BiP antibodies response in T. brucei infections is the ability of

46

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

the parasite to induce a strong immunosuppression via the induction of B cells apoptosis (Radwanska et al., 2008). However, this is not confirmed by field data, which indicates that at least T. vivax infections may be as well diagnosed by this test as the T. congolense infections, except for one animal that was serologically negative while heavily parasitaemic. This may be related to absorption of antibodies by circulating antigens, a phenomenon that has been reported previously (Authié et al., 1993). 4.1.3. Persistence of antibodies after treatment The persistence of anti-trypanosome antibodies after an infection was cured, whether spontaneous (“self-cure”) or induced by a trypanocidal drug, is one of the major drawbacks of most antibody detection tests. Antibodies to a complex trypanosome extract may persist at high levels for several months in cattle after successful therapy (Boulangé et al., 2002; Desquesnes et al., 2003). In endemic areas, reinfections often occur before an animal has eliminated antibodies from a previous infection and act as repetitive boosters of immunity to parasite antigens. There is a history of anti-trypanosome antibodies in indigenous cattle (Boulangé et al., 2002). This makes it difficult to distinguish animals with an active infection from those with serological traces of past infections. Nevertheless, the HSP70/BiP-based inhibition ELISA was able to show a seroconversion within 3 months after drug treatment in experimentally infected animals, suggesting that it could represent a useful tool for monitoring the efficacy of treatment. In order to estimate the latter precisely, a comparative study of HSP70/BiP-based inhibition ELISA with indirect ELISA on total antigens will be necessary to confirm the seroconversion in naturally rechallenged animals after treatment. 4.2. Repeatability and reproducibility of the test The ability of a diagnostic test to produce consistent results within the limits of analytical error when specimens are retested is one of the prerequisites for any diagnostic device before it can be accepted for routine applications. The repeatability and reproducibility of the test were good when evaluated in our laboratory, giving congruent results between manipulations and experimenters. Nevertheless, reproducibility was assessed only between experimenters within the same laboratory, and it requires further validation in the framework of technology transfer to other reference laboratories. 4.3. Sensitivity Initial assessment of the HSP70/BiP-based inhibition ELISA confirms our previous report that a serological test based on detection of anti-HSP70/BiP antibodies has low or medium sensitivity in primary infections (Boulangé et al., 2002). Presumably, this would be the case with most serological tests due to inter-individual variability and relatively low levels of trypanosome-specific immunoglobulins in primary infections. This is well illustrated by the good (detection of 100% of the infected Boran with clone T. congolense IL1180) though transitory recognition of

HSP70/BiP by antibodies of animals submitted to one single experimental challenge. In comparison, animals subjected to a rechallenge exhibited an early and more sustained antibody response. The high sensitivity of the test during rechallenge infections is essential to its usefulness in the field. Indeed, in endemic areas, cattle are generally exposed to primary infection during the first months of life and are subsequently reinfected several times a year. Actually, the test might miss few primary infections which thus would not be a major drawback for epidemiological purposes. This is well illustrated by the test’s capacity to easily detect infected animals under natural conditions. It is noteworthy that the inhibition ELISA has an overall higher sensitivity than the indirect ELISA directed against a portion of HSP70/BiP that we initially developed (Boulangé et al., 2002). This is probably due to the capacity of the test to detect both IgG and IgM, whereas the indirect ELISA was targeting only IgG. 4.4. Further evaluation under natural conditions There are two major limitations to validation of new diagnostic tests for trypanosomosis. The first limitation is that there is no gold standard/reference test. Secondly, the determination of the test performance should be based on testing samples from individuals with a known infection status, which are extremely difficult to obtain in endemic areas. Procedures for validation of serological assays are subject to many limitations including availability of standards and representative reference sera (Jacobson 1998 in Paweska et al., 2005). The determination of the performance characteristics of a diagnostic assay should be based on testing samples from individuals of known infection status. The accuracy of the initial classification of the subjects based on a diagnostic discriminator has a significant impact on the selection of the optimal threshold and subsequently on estimates of diagnostic performance. Moreover, as was the case in our study, some animals can be positive in serology but negative in parasitology. Although one cannot rule out that these individuals had recently eliminated their parasites, either spontaneously or following treatment, it is also possible that they were infected but omitted by both parasitological and DNA-based techniques. In conclusion, the HSP70/BiP-based inhibition ELISA demonstrated its ability to identify primary infected cattle with T. congolense IL1180, rechallenged cattle with other T. congolense clones and field infected cattle with T. congolense and T. vivax. For T. brucei, an additional experiment with a larger number of cattle deserves further investigation to validate the pan trypanosome potential of the HSP70/BiP-based inhibition ELISA. This test is more sensitive in endemic areas where animals undergo repeated infectious challenges. After optimization of the cut-off values with field data sets from trypanosome-free cattle taken in endemic areas, further assessment of the test will include a large scale validation for T. congolense, T. vivax and T. brucei infections in cattle under various field conditions. The reagents involved in the test are cheap and easy to produce on a large scale. Thus, the proposed ELISA is well suited to the needs of kit production and screening of large numbers

G. Bossard et al. / Veterinary Parasitology 173 (2010) 39–47

of cattle. Besides those stated above, one of the main advantages of the test, particularly in view of its application by national laboratories, is that one single set of reagents and one single procedure are required to test different mammalian hosts for different trypanosome species. Acknowledgements Part of this work was conducted by G. Bossard, A. Boulangé and E. Authié at ILRI-Nairobi (Kenya) under a collaborative agreement with UMR17. We thank Tony Musoke and Joseph Katende for providing bovine serum from primary infections with T. vivax, T. brucei, T. parva, A. marginale and B. bigemina carried out at ILRI. We also thank CIRDES staff for conducting field surveys and providing sera samples. This work was partially funded by the INCO-DEV grant PL003716 of the European Union. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016 /j.vetpar.2010.06.016. References Authié, E., Muteti, D.K., Williams, D.J.L., 1993. Antibody responses to non-variant antigens of Trypanosoma congolense in cattle of differing susceptibility to trypanosomiasis. Parasite Immunol. 15, 101–111 (Erratum in: Parasite Immunol. 15, 185). Bangs, J.D., Uyetake, L., Brickman, M.J., Balber, A.E., Boothroyd, J.C., 1993. Molecular cloning and cellular localization of a BiP homologue in Trypanosoma brucei. Divergent ER retention signals in a lower eukaryote. J. Cell Sci. 105, 1101–1113. Bannai, H., Sakurai, T., Inoue, N., Sugimoto, C., Igarashi, I., 2003. Cloning and expression of mitochondrial Heat Shock Protein 70 of Trypanosoma congolense and potential use as a diagnostic antigen. Clin. Diagn. Lab. Immunol. 10, 926–933. Bajyana Songa, E., Kageruka, P., Hamers, R., 1987. The use of card agglutination test (Testryp CATT) for use in detection of T. evansi infection. Ann. Soc. Belg. Med. Trop. 67, 51–57. Bajyana Songa, E., Hamers, R., 1988. A card agglutination test (CATT) for veterinary use based on an early VAT RoTat 1/2 of Trypanosoma evansi. Ann. Soc. Belg. Med. Trop. 68, 233–240. Bossard, G., Boulangé, A., Authié, E., 2003. Progress towards development of serodiagnostic tests based on the use of recombinant hsp70/BiP antigen. ICPTV News Lett. 8, 30–33. Boulangé, A., Authié, E., 1994. A 69-kDa immunodominant antigen of Trypanosoma congolense is homologous to BiP (immunoglobulin binding protein). Parasitology 109, 163–173. Boulangé, A., Katende, J., Authié, E., 2002. Trypanosoma congolense: expression of a heat shock protein 70 and initial evaluation as a diagnostic antigen for bovine trypanosomosis. Exp. Parasitol. 100, 6–11. Camargo, R.E., Uzcanga, G.L., Bubis, J., 2004. Isolation of two antigens from Trypanosoma evansi that are partially responsible for its crossreactivity with Trypanosoma vivax. Vet. Parasitol. 123, 67–81. Claes, F., Ilgekbayeva, G.D., Verloo, D., Saidouldin, T.S., Geerts, S., Buscher, P., Goddeeris, B.M., 2005. Comparison of serological tests for equine trypanosomosis in naturally infected horses from Kazakhstan. Vet. Parasitol. 131, 221–225. Cohen, J., 1960. A coefficient of agreement for nominal scales. Educ. Psychol. Meas. 20, 37–46. Crowther, J.R., 1995. ELISA theory and practice. In: Methods in Molecular Biology. Humana Process, Totowa, NJ, USA. Desquesnes, M., Bengaly, Z., Millogo, L., Meme, Y., Sakande, H., 2001. The analysis of the cross-reactions occurring in antibody-ELISA for the detection of trypanosomes can improve identification of the parasite species involved. Ann. Trop. Med. Parasitol. 95, 141–155. Desquesnes, M., Bengaly, Z., Dia, M.L., 2003. Evaluation de la persistance des anticorps détectés par Elisa-indirect Trypanosoma vivax

47

après traitement trypanocide chez des bovins naturellement infectés pathologie parasitaire Revue Élev. Méd. Vét. Pays Trop. 56, 141– 144. Desquesnes, M., Kamyingkird, K., Pruvot, M., Kengradomkij, C., Bossard, G., Sarataphan, N., Jittapalapong, S., 2009. Antibody-ELISA for Trypanosoma evansi: application in a serological survey of dairy cattle, Thailand, and validation of a locally produced antigen. Prev. Vet. Med. 90, 233–241. Greiner, M., Franke, C.R., Bohning, D., Schlattmann, P., 1994. Construction of an intrinsic cut-off value for the sero-epidemiological study of Trypanosoma evansi infections in a canine population in Brazil: a new approach towards an unbiased estimation of prevalence. Acta Trop. 56, 97–109. Greiner, M., Kumar, S., Kyeswa, C., 1997a. Evaluation and comparison of antibody ELISAs for serodiagnosis of bovine trypanosomosis. Vet. Parasitol. 73, 197–205. Greiner, M., Bhat, T.S., Patzelt, R.J., Kakaire, D., Schares, G., Dietz, E., Bohning, D., Zessin, K.-H., Mehlitz, D., 1997b. Impact of biological factors on the interpretation of bovine trypanosomosis serology. Prev. Vet. Med. 30, 61–73. Hutchinson, O.C., Webb, H., Picozzi, K., Welburn, S., Carrington, M., 2004. Candidate protein selection for diagnostic markers of African trypanosomosiasis. Trends Parasitol. 20, 519–523. Paris, J., Murray, M., McOdimba, F., 1982. A comparative evaluation of the parasitological techniques currently available for the diagnosis of African trypanosomiasis in cattle. Acta Trop. 39, 307– 316. LeeFlang, P., Buys, J., Blotkamp, C., 1976. Studies on Trypanosoma vivax: infectivity and serial maintenance of natural bovine isolates in mice. Int. J. Parasitol. 6, 413–417. Lai, D.H., Hashimi, H., Lun, Z.R., Ayala, F.J., Lukes, J., 2008. Adaptations of Trypanosoma brucei to gradual loss of kinetoplast DNA: trypanosoma equiperdum and Trypanosoma evansi are petite mutants of T. brucei. Proc. Natl. Acad. Sci. U.S.A. 105, 1999–2004. Luckins, A.G., 1992. Methods for diagnosis of trypanosomiasis in livestock. World Anim. Rev. 70/71, 15–20. Mahama, C.I., Desquesnes, M., Dia, M.L., Losson, B., De Deken, R., Geerts, S., 2004. A cross-sectional epidemiological survey of bovine trypanosomosis and its vectors in the Savelugu and West Mamprusi districts of northern Ghana. Vet. Parasitol. 122, 1–13. Masiga, D.K., Smith, A.J., Hayes, P., Bromidge, T.J., Gibson, W.C., 1992. Sensitive detection of trypanosomoses in tsetse flies by DNA amplification. Int. J. Parasitol. 22, 909–918. Mboloi, M.M., Bekker, C.P., Kruitwagen, C., Greiner, M., Jongejan, F., 1999. Validation of the indirect MAP1-B enzyme-linked immunosorbent assay for diagnosis of experimental Cowdria ruminantium infection in small ruminants. Clin. Diagn. Lab. Immunol. 6, 66–72. Mendoza-Palomares, C., Biteau, N., Giroud, C., Coustou, V., Coetzer, T., Authié, E., Boulangé, A., Baltz, T., 2008. Molecular and biochemical characterization of a cathepsin B-like protease family unique to Trypanosoma congolense. Eukaryot. Cell. 7, 684–697. Murray, M., Murray, P.K., McIntyre, W.I.M., 1977. An improved parasitological technique for the diagnosis of African trypanosomiasis. Trans. R. Soc. Trop. Med. Hyg. 71, 325–326. Nantulya, V.M., Musoke, A.J., Rurangirwa, F.R., Moloo, S.K., 1984. Resistance of cattle to tsetse-transmitted challenge with Trypanosoma brucei or Trypanosoma congolense after spontaneous recovery from syringe-passaged infections. Infect Immun. 43, 735–738. Paweska, J.T., Mortimer, E., Leman, P.A., Swanepoel, R., 2005. An inhibition enzyme-linked immunosorbent assay for the detection of antibody to Rift Valley fever virus in humans, domestic and wild ruminants. J. Virol. Methods 127, 10–18. Radwanska, M., Guirnalda, P., De Trez, C., Ryffel, B., Black, S., Magez, S., 2008. Trypanosomiasis-induced B cell apoptosis results in loss of protective anti-parasite antibody responses and abolishment of vaccine-induced memory responses. PLoS Pathol. 4, e1000078. Sumba, A.L., Mihok, S., Oyieke, F.A., 1998. Mechanical transmission of Trypanosoma evansi and T. congolense by Stomoxys niger and S. taeniatus in a laboratory mouse model. Med. Vet. Entomol. 12, 417–422. Tran, T., Claes, F., Verloo, D., De Greve, H., Büscher, P., 2009. Towards a new reference test for surra in camels. Clin. Vaccine Immunol. 16, 999–1002. Williams, D.J.L., Naessens, J., Scott, J.R., Mc Odimba, F.A., 1991. Analysis of peripheral leucocyte populations in N’Dama and Boran cattle following a rechallenge infection with Trypanosoma congolense. Parasitol. Immunol. 13, 171–218.