Pattern of recognition of Besnoitia besnoiti tachyzoite and bradyzoite antigens by naturally infected cattle

Veterinary Parasitology 164 (2009) 104–110

Contents lists available at ScienceDirect

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

Pattern of recognition of Besnoitia besnoiti tachyzoite and bradyzoite antigens by naturally infected cattle A. Ferna´ndez-Garcı´a 1, G. A´lvarez-Garcı´a *, V. Risco-Castillo, A. Aguado-Martı´nez, V. Maruga´n-Herna´ndez, L.M. Ortega-Mora SALUVET, Animal Health Department, Faculty of Veterinary Sciences, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain

A R T I C L E I N F O

A B S T R A C T

Article history: Received 9 March 2009 Received in revised form 4 June 2009 Accepted 15 June 2009

Bovine besnoitiosis is caused by the protozoan parasite Besnoitia besnoiti. Many recent cases have been described in different European countries, which may be indicative of expansion of the disease in the next few years. Many infected animals remain asymptomatic; therefore, serological tests are essential tools for diagnosis. The objective of the present work was to identify B. besnoiti tachyzoite and bradyzoite immunodominant antigens (IDAs). IDAs were recognised by SDS-PAGE under reducing conditions and Western blot analysis. Positive sera from symptomatic (n = 18) and asymptomatic (n = 18) cattle came from herds endemically infected by B. besnoiti and were confirmed positive by IFAT, whereas negative sera (n = 4) came from besnoitiosis-free herds and were also confirmed negative by IFAT. Up to 28 tachyzoite antigens in the range of 8.5–190.8 kDa were recognised. Based on the frequency of recognition, six IDAs (14.2, 33, 37.1, 39.6, 46.3 and 190.8 kDa) were identified. The 37.1 kDa antigen was recognised by 100% of sera, usually as an intense band. On the other hand, 30 bradyzoite antigens in the range of 8.5– 187.9 kDa were detected. Seven bradyzoite IDAs (8.5, 15.1, 16.8, 19.0, 34.7, 38.6 and 124.4 kDa) were identified and two of them (15.1 and 16.8 kDa) were considered the most immunogenic ones. Additionally, sera from animals with clinical symptoms recognised a significantly higher number of bradyzoite antigens. Finally, significant cross-reactions with other closely related apicomplexan parasites were not detected. This is the first description of B. besnoiti bradyzoite antigens. In addition, the identification of tachyzoite and bradyzoite IDAs may be useful for the development of vaccines and diagnostic tools for differentiating between acute and chronic infections. Further proteomic studies are needed in order to identify stage-specific proteins. ß 2009 Elsevier B.V. All rights reserved.

Keywords: Besnoitia besnoiti Tachyzoites Bradyzoites Bovines IDAs IFAT Western blot

1. Introduction Bovine besnoitiosis is a disease caused by the cystforming apicomplexan parasite Besnoitia besnoiti (Marotel, 1912), which has been reported in Asia and Africa (Bigalke, 1981). In Europe, the disease has been described in Italy (Agosti et al., 1994), France (Besnoit and Robin, 1912),

* Corresponding author. Tel.: +34 913944095; fax: +34 913944098. E-mail address: [email protected] (G. A´lvarez-Garcı´a). 1 Present address: Viral Pathogenesis Department, CNM, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain. 0304-4017/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.vetpar.2009.06.020

Portugal (Franco and Borges, 1915; Leitao Da Silva, 1963; Cortes et al., 2005), Spain (Juste et al., 1990; Ferna´ndezGarcı´a et al., 2009) and, recently, in Germany (Mehlhorn et al., 2009). Currently, bovine besnoitiosis in Europe behaves as a re-emerging disease. Clinical signs in the early stages of the disease include fever, lymphadenitis, subcutaneous oedema, loss of body condition and testicular inflammation, and are associated with the fast replication of tachyzoites. During the chronic stage, when cysts containing bradyzoites appear predominantly in subcutaneous tissue in the skin, mucosal membranes and scleral conjunctiva, typical symptoms are cutaneous lesions such as thickening and folding of the skin and hypotrichia or

A. Ferna´ndez-Garcı´a et al. / Veterinary Parasitology 164 (2009) 104–110

alopecia (Bigalke, 1981). Severe necrotising orchitis may result in permanent infertility (Bigalke, 1968) and cause high economic losses. Apparent cases of besnoitiosis are easily diagnosed by a combination of clinical signs and the identification of tissue cysts by skin biopsy or cytology. However, many infected animals do not show clinical signs (Bigalke, 1981) so serological diagnosis could be a useful tool for detecting infection in asymptomatic cattle, particularly to avoid the entry of the infection into B. besnoiti-free herds and to control bovine besnoitiosis in infected herds. Detection of B. besnoiti-specific antibodies using the indirect fluorescent antibody test (IFAT) and ELISA have been reported previously (Frank et al., 1977; Goldman and Pipano, 1983; Janitschke et al., 1984; Shkap et al., 1984). However, there is no agreement on the optimal test to be employed. In addition, the use of an ELISA test complemented with a Western blot (WB), in which three antigenic areas have been reported, has been recently suggested in order to reliably diagnose both Besnoitia-infected symptomatic and asymptomatic bovines (Cortes et al., 2006). Further work is needed in order to adopt a common strategy for serological diagnosis among labs. In this sense, stage-specific antigens have proved to be very useful for serological differentiation between acute and chronic infections caused by closely related apicomplexan parasites (Aguado-Martı´nez et al., 2008). Thus, a better characterisation of B. besnoiti tachyzoite antigens together with an accurate identification of B. besnoiti bradyzoite antigens is needed to improve early diagnosis and differentiate acute from chronic infections when clinical signs are not evident. With this objective, the pattern of recognition of B. besnoiti tachyzoite and bradyzoite antigens by naturally infected cattle was identified in the present study, which helped to accurately establish an IFAT cut-off. 2. Materials and methods

105

confirmed by serological tests. B. besnoiti infection was discarded in these serum samples by IFAT (Shkap et al., 2002), which was developed as explained below. Once tachyzoite IDAs were determined, seven additional sera samples with an IFAT titre of 1:100 from asymptomatic animals belonging to herds with a previous history of besnoitiosis were checked by WB in order to determine the IFAT cut-off. 2.2. Parasite production and antigen preparation Tachyzoites of a B. besnoiti Bb-Spain1isolate (Ferna´ndez-Garcı´a et al., 2009) were maintained in vitro by serial passages in MARC-145 cell cultures with Dulbecco’s modified Eagle medium (DMEM) supplemented with 5% fetal bovine serum (FBS), 15 mM HEPES (pH 7.2), 2 mM glutamine, penicillin (100 U/ml), streptomycin (100 mg/ ml) at 37 8C in a 5% CO2 humidified incubator. Tachyzoites were transferred to fresh media every 4 days at a 1:5 host:parasite ratio. To obtain tachyzoite extracts for Western blot tachyzoites were harvested, passed through a 25-gauge needle, separated on a PD10 column (GE Healthcare, Chalfont St. Giles, UK) and pelleted by centrifugation at 1350  g for 10 min at 4 8C. This method has been used previously for harvesting tachyzoites from N. caninum (Ferna´ndez-Garcı´a et al., 2006). Bradyzoites were released by trypsinisation of a skin biopsy from the vulvar region containing numerous macroscopic tissue cysts from the chronically infected cow from which the B. besnoiti isolate had been obtained (BbSpain-1) (Ferna´ndezGarcı´a et al., 2009) using a previously described method with a few modifications (Shkap et al., 1987). Bradyzoites were separated from tissue debris with a 40-mm sieve and collected by centrifugation at 1350  g for 15 min at 4 8C. Pellets with zoites were frozen at 80 8C until use or resuspended in PBS to a final concentration of approximately 107 tachyzoites per ml and formalin-fixed for their use in IFAT.

2.1. Experimental design and serum samples

2.3. IFAT

In total, 36 positive sera previously tested by IFAT (titre  1:200) were collected from cattle naturally infected with B. besnoiti. The animals came from herds with a previous history of endemic besnoitiosis. Samples consisted of sera from animals with a clinical sign consistent with either an acute (fever, swelling of superficial lymph nodes and oedema) or a chronic (scleroderma, hyperkeratosis, alopecia or orchitis in males and tissuecysts in scleral conjunctiva) infection (n = 18), or asymptomatic animals (n = 18). For negative controls, four sera from besnoitiosis-free herds with an IFAT titre of 1:50 were used. A set of samples consisting of six sera from cows infected naturally with either N. caninum or Sarcocystis spp. and one serum sample from a T. gondii experimentally infected calf (Esteban-Redondo et al., 1999) was used to evaluate any cross-reactivity with tachyzoite and bradyzoite B. besnoiti antigens. Sarcocystis spp. infection was confirmed histologically by visualisation of tissue cysts in the heart, and N. caninum and T. gondii infections were

Sera were analysed by IFAT in double serial dilutions starting at 1:50 using the method previously described for N. caninum (Trees et al., 1994), with a few modifications to adapt it to B. besnoiti. Unbroken tachyzoite membrane fluorescence at titres equal to or higher than 1:100 was considered positive. Sera with titres equal to or less than 1:50 were considered negative. 2.4. SDS-PAGE and Western blot Samples containing 2  107 B. besnoiti tachyzoites or bradyzoites were detergent-disrupted at 95 8C for 5 min with Laemmli buffer (Laemmli, 1970), sonicated in an ultrasonic bath at 15 8C for 15 min and then heated for 5 min at 95 8C prior to use. Electrophoresis was performed in 12.5% polyacrylamide-DATD minigels and then transferred to a nitrocellulose membrane (Mini Trans-Blot Cell, Bio-Rad Laboratories, CA, USA). Precision Plus Kaleidoscope weight standards (Bio-Rad Laboratories, CA, USA) were subjected to electrophoresis to estimate the apparent

106

A. Ferna´ndez-Garcı´a et al. / Veterinary Parasitology 164 (2009) 104–110

molecular weights of the different antigens recognised by sera. WB was carried out essentially as described for N. caninum by A´lvarez-Garcı´a et al. (2002) and serum samples were used at a 1:20 dilution. Images from the WB membranes were obtained using a GS-800 Scanner (Bio-Rad Laboratories, CA, USA) and analysed with Quantity One1 quantification software v. 4.0 (Bio-Rad Laboratories, CA, USA).

parametric two-tailed Mann–Whitney U-test. Statistical calculations were computed using STAT-VIEW v.4.0 (Abacus Concepts, Inc. Berkeley, CA, USA).

2.5. Analysis of data

Up to 28 B. besnoiti tachyzoite antigens in the range of 8.5–190.8 kDa were recognised by cattle (Fig. 1A). The frequency of B. besnoiti tachyzoite antigens recognised is summarised in Table 1. Seronegative sera used as negative controls did not bind to any B. besnoiti protein. IDAs were selected based on their frequency of recognition. The most immunogenic antigen was a 37.1 kDa protein that was recognised by 100% of infected animals, usually as an intense band. Additionally, five more IDAs were detected with a frequency of recognition higher than 75% (14.2, 33, 37.1, 39.6, 46.3 and 190.8 kDa), and nine antigens of medium frequency were detected (>50–75%) (8.5, 15.3, 21.5, 30, 64, 72, 86.9, 102.3 and 149.2 kDa). Sera from cattle infected with other apicomplexan parasites such as N. caninum, T. gondii and Sarcocystis spp.

Apparent molecular weight was determined by comparing the bands obtained with the molecular weight markers and the immunodominant antigens (IDAs) and calculated for each sample analysed. Frequency of recognition was expressed as the percentage of animals detecting the same antigen. B. besnoiti tachyzoite or bradyzoite antigens were classified according to their frequency of recognition in WB analysis. An antigen was considered as an IDA when it was recognised by more than 75% of samples analysed. Agreement between WB and IFAT was determined by the kappa statistic (Thrusfield, 1995). Differences in antigen recognition between animals with or without clinical signs were compared by a non-

3. Results 3.1. Pattern of tachyzoite antigen recognition by B. besnoiti-infected cattle

Fig. 1. Pattern of recognition of B. besnoiti tachyzoite and bradyzoite antigens by sera from naturally infected cattle by Western blot. Panel A: tachyzoite extract. Panel B: bradyzoite extract. Lanes 1–17: sera with IFAT titre  1:200. Lane 18: B. besnoiti seronegative serum. IDAs (>75% of frequency of recognition) are indicated on the left of the figure.

A. Ferna´ndez-Garcı´a et al. / Veterinary Parasitology 164 (2009) 104–110 Table 1 Frequency (%) of recognition of B. besnoiti tachyzoite antigens by naturally infected cattle. MW (kDa) 190.8 149.2 119.4 102.3 86.9 72.0 64.0 53.1 46.3 44.7 44.2 39.6 37.1 33.0 30.0 27.9 24.9 22.8 21.5 19.1 17.6 16.5 15.3 14.2 12.6 11.0 9.9 8.5

IDAs (frequency)a

Medium antigens (frequency)a

Minor antigens (frequency)a

77.78 72.22 44.44 58.33 58.33 72.22 72.22 41.67 77.78 5.56 13.89 80.56 100.00 91.67 61.11 33.33 33.33 13.89 63.89 11.11 41.67 41.67 63.89 75.00 27.78 8.33 30.56 61.11

a

Frequency of recognition is expressed as percentage of animals in which the different Besnoitia besnoiti antigens were detected. IDAs appear in bold.

did not show the B. besnoiti tachyzoite characteristic antigen patterns, showing few cross-reactive bands. Of those cross-reactive bands, one was a 64 kDa antigen from B. besnoiti tachyzoite that was recognised by one serum from animals infected with N. caninum. In addition, sera

107

from cattle infected with Sarcocystis spp. recognised three antigens of 77 kDa (not considered in the pattern of B. besnoiti tachyzoite antigen recognition), 64 and 44.7 kDa with a frequency of 66.67%, 66.67% and 100%, respectively. The only sample from an experimentally T. gondii-infected animal did not show any cross-reaction (Fig. 2A). When WB results of sera from infected cattle either with (n = 18) or without (n = 18) clinical signs were compared, we did not find any difference in the pattern of recognition of tachyzoite B. besnoiti antigens in relation to clinical manifestations of the disease. 3.2. Pattern of bradyzoite antigen recognition by B. besnoiti-infected cattle Up to 30 B. besnoiti bradyzoite antigens in the range of 8.5–187.9 kDa were recognised. The frequency of B. besnoiti bradyzoite antigens recognised by cattle is summarised in Table 2. Four bradyzoite antigens (8.5, 15.1, 16.8 and 124.4 kDa) were considered the most immunogenic according to their frequency of recognition by the sera of infected animals (higher than 75%) and the strong intensity of the band. Of these four antigens, two were intensely detected by 100% of the animals. Additionally, three more IDAs were identified (19, 34.7 and 38.6 kDa). Negative controls did not bind to any B. besnoiti protein (Fig. 1B). When B. besnoiti tachyzoite and bradyzoite IDAs were compared, four antigens (14.2, 37.1, 46.3 and 190 kDa) were clearly detected only at the tachyzoite stage, whereas four other antigens (15.1, 16.8, 19 and 124.4 kDa antigens) were only detected at the bradyzoite stage. On the other hand, IDAs with the same (8.5 kDa) or similar apparent molecular weight in the range of 33–39.6 kDa were detected by sera in both tachyzoite and bradyzoite extracts. Regarding cross reactions with N. caninum, T. gondii and Sarcocystis spp. we detected higher number of B. besnoiti

Fig. 2. Western blot analysis with sera from animals infected with other apicomplexan parasites. Panel A: tachyzoite extract. Panel B: bradyzoite extract. Lanes 1 and 2: B. besnoiti seropositive sera. Lane 3–5: sera from Sarcocystis spp.-infected cattle. Lanes 6 and 7: sera from N. caninum-infected cattle. Lane 8: serum from a T. gondii-infected calf.

A. Ferna´ndez-Garcı´a et al. / Veterinary Parasitology 164 (2009) 104–110

108

Table 2 Frequency (%) of recognition of B. besnoiti bradyzoite antigens by naturally infected cattle. MW (kDa) 187.9 159.7 141.7 124.4 93.0 80.2 74.0 68.8 64.7 61.8 57.9 54.0 50.2 48.0 45.7 40.6 38.6 34.7 32.6 28.6 25.7 23.7 19.0 17.8 16.8 15.1 13.1 12.5 11.2 8.5

IDAs (frequency)a

Medium antigens (frequency)a

Minor antigens (frequency)a

61.11 19.44 13.89 94.44 13.89 30.56 27.78 22.22 19.44 11.11 13.89 22.22 5.56 27.78 19.44 5.56 75.00 83.33 13.89 55.56 8.33 16.67 88.89 19.44 100.00 100.00 19.44 5.56 11.11 91.67

a

Frequency of recognition is expressed as percentage of animals in which the different Besnoitia besnoiti antigens were detected. IDAs appear in bold.

bradyzoite cross-reactive bands compared with Western blot results obtained with B. besnoiti tachyzoite extracts. However, most of bands were detected with a low intensity of recognition. Minor cross-reactive antigens were detected by sera from Sarcocystis infected animals with a molecular weight of 28.7 (n = 1), 45.7 (n = 1) and 93 kDa (n = 1). In addition, the 124.4 kDa IDA was also detected by one serum. Of those B. besnoiti bradyzoite cross-reactive bands with sera from animals infected with N. caninum the 16 kDa IDA was slightly detected by one serum, together with other minor antigens of 25.7 and 32.6 kDa. In addition, sera from cattle infected with Sarcocystis spp. recognised two antigens of 17 kDa (n = 1) and 49 kDa (n = 1) that were not considered in the pattern of B. besnoiti bradyzoite antigen recognition. Similar to this finding the only sample from an experimentally T. gondii-infected animal faintly recognised two antigens not considered in the pattern of B. besnoiti bradyzoite antigen recognition (Fig. 2B). Finally, animals with clinical signs (n = 18) detected a higher number of bradyzoite antigens (P < 0.05; U = 73.5) and bradyzoite IDAs (P > 0.05; U = 113.5) than animals without clinical signs (n = 18). 3.3. IFAT cut-off selection All sera with IFAT titres of 1:100 (n = 7) recognised the 37.1 kDa IDA. In addition, all sera with a titre of 1:50 did

Fig. 3. IFAT cut-off determination based on B. besnoiti tachyzoite IDAs recognition. The 37.1 kDa IDA appears in bold.

not bind to any B. besnoiti protein (Fig. 3). According to these results, the kappa estimation gave a perfect agreement between Western blots and IFAT for a 1:100 cut-off point. 4. Discussion As bovine besnoitiosis spreads rapidly once it enters a herd, appropriate control measures should involve serological diagnosis of purchased animals, particularly in terms of replacement policies. However, the main limitation in the diagnosis of bovine besnoitiosis is the detection of infected asymptomatic cattle. Currently used serological tests are useful when clinical signs are not apparent. However, sensitivity can be improved following additional approaches. Thus, in the present work, the pattern of recognition of B. besnoiti bradyzoite antigens by naturally infected cattle has been described for the first time. In addition, tachyzoite IDAs were also determined so that identification of antigens of diagnostic significance may help in the development of more sensitive and specific tests for differentiating between acute and chronic stages of infection. Up to 28 B. besnoiti tachyzoite antigens in the range of 8.5–190.8 kDa were recognised by sera from B. besnoiti naturally infected cattle. Previous studies have shown that B. besnoiti tachyzoite antigens ranging from 14 to 210 kDa were recognised by an anti-B. besnoiti serum

A. Ferna´ndez-Garcı´a et al. / Veterinary Parasitology 164 (2009) 104–110

produced in gerbils (Shkap et al., 2002). However, the large number of antigenic fractions in the 8.5–190 kDa area and the variation in migration positions (due to nonstandardised interlaboratory procedures) cited in the literature make correlations with previous studies difficult (Staubli et al., 2006). Additionally, based on the frequency of recognition (>75%), six IDAs from B. besnoiti tachyzoites were identified. Recently, another group has characterised three main areas of B. besnoiti tachyzoite antigens under non reducing conditions as important for immunodiagnosis; areas I, II and III ranging from 12 to 20 kDa, 23 to 38 kDa and 60 to 90 kDa, respectively (Cortes et al., 2006). The minimal criteria considered for besnoitiosis diagnosis was the detection of four bands in each area. In our study, a higher reactivity was observed against antigens located in an area between 33 and 39.6 kDa. In this range, the humoral IgG responses were mainly directed against 33, 37.1 and 39.6 kDa antigenic fractions. In particular, the 37.1 kDa IDA was recognised by all B. besnoiti naturally infected cattle with an intense and homogeneous recognition. Although it is not possible to make comparisons between antigens detected in both studies due to different technique conditions (non-reducing versus reducing conditions) we suggest a more simplified interpretation of Western blot results. In that sense a WB positive result was recorded when the 37.1 kDa antigen was recognised. Moreover, the intense recognition of the 37.1 kDa IDA by 100% of sera gives evidence for the potential use of this antigen in the development of diagnostic tests. The WB based on tachyzoite antigens has been previously recommended as a complementary technique in ruminants not only for the diagnosis of B. besnoiti infection (Cortes et al., 2006), but also for infections caused by other cyst-forming coccidia such as N. caninum (Schares et al., 1998; A´lvarez-Garcı´a et al., 2002; Von-Blumro¨der et al., 2004) and T. gondii (Cha´vez-Vela´squez et al., 2005). Putative cross-reactivity of B. besnoiti tachyzoite antigens with sera from animals infected with other apicomplexan parasites was also analysed. To our knowledge, N. caninum and Sarcocystis spp. do not appear to critically or significantly cross react with B. besnoiti. A few minor bands in the B. besnoiti antigen were recognised by sera from N. caninum- and Sarcocystis spp.-infected cattle. Moreover, animals infected with these apicomplexan parasites did not recognise the 37.1 kDa IDA. Two proteins of 26 and 73 kDa from soluble B. besnoiti antigens were recognised by an anti-N. caninum serum produced in gerbils (Shkap et al., 2002). In addition, Cortes et al. (2006) have reported a non-specific binding pattern of sera from N. caninum-infected cattle with B. besnoiti tachyzoite antigens between 60 and 90 kDa. Similar to our results, sera from T. gondii infected animals did not show nonspecific binding, probably due the origin of the sample coming from an experimental infection, which can also explain the low cross-reactivity with Besnoitia bradyzoite extracts. Regarding this issue a higher level of crossreactions was observed with bradyzoite extracts and sera from animals infected with Sarcocystis spp. and N. caninum. Although cross-reactions were observed with several IDAs (16 and 124.4 kDa), the intensity of recognition was significantly diminished compared with the positive

109

controls. However, in order to avoid false positive results with perspectives for diagnosis at least the recognition of three IDAs should be considered. Cross-reactions with several Besnoitia bradyzoite antigens among tissue cyst forming coccidian parasites are likely to occur since antibodies directed against Toxoplasma bradyzoite (TgBAG1) and tissue cysts TgCC2 antigens also detected similar antigens in N. caninum (McAllister et al., 1996; Risco-Castillo et al., 2004). With regard to the B. besnoiti bradyzoite stage, up to 30 antigens in the range of 8.5–187.9 kDa were recognised by sera from naturally infected cattle. Seven bradyzoite IDAs (8.5, 15.1, 16.8, 19.0, 34.7, 38.6 and 124.4 kDa antigens) were identified. Based on the frequency and intensity of recognition, the most important tachyzoite IDAs are of medium apparent molecular weight (33, 37.1 and 39.6 kDa), whereas bradyzoite IDAs of major interest are located in a lower migration rate (15.1 and 16.8 kDa). Moreover, a high molecular weight IDA (124.4 kDa) was recognised by almost all sera (94.4%) in bradyzoite extracts. However the bradyzoite antigenic pattern might be altered as a consequence of the treatment of skin biopsies with trypsin during the process to release bradyzoites. In the present study tripsinization was done for a short period of time so that protease activity might have only altered proteins located on the surface of the parasite. Despite this issue a substantial modification of recognition of surface antigens is unlikely to occur since B. besnoiti bradyzoites are viable in cell culture after this treatment (data not shown). These clear and distinct antigenic patterns observed between tachyzoite and bradyzoite extracts lead us to speculate about the differential expression of stagespecific antigens. Moreover, the recognition of an 8.5 kDa IDA in both bradyzoite and tachyzoite extracts gives evidence for the presence of shared antigens in both stages of B. besnoiti. The differential expression of antigens within tachyzoites and bradyzoites of other apicomplexan parasites as T. gondii and N. caninum has been previously reported. In this sense 2D-PAGE followed by mass spectrometry analysis of B. besnoiti bradyzoite and tachyzoite extracts might help to determine different patterns of protein expression and to identify stage specific proteins. Additionally, our results did not show any difference in tachyzoite antigen recognition between asymptomatic and symptomatic B. besnoiti-infected cows in accordance with the observations made by others (Cortes et al., 2006). However, we detected a significantly higher number of bradyzoite antigens in symptomatic cows, which may be explained by a longer exposure to the parasite. Finally, the identification of tachyzoite IDAs helped to define an accurate IFAT cut-off. Previously, several authors have used a 1:256 serum dilution cut-off by IFAT (Goldman and Pipano, 1983; Shkap et al., 2002; Cortes et al., 2006). In the present study, sera with titres equal to or higher than 1:100 were regarded as positive based on the detection of the 37.1 kDa IDA. Moreover, at a 1:64 or lower serum dilution, there was a one-way cross-reactivity between N. caninum antibodies and B. besnoiti antigens by IFAT (Shkap et al., 2002).

110

A. Ferna´ndez-Garcı´a et al. / Veterinary Parasitology 164 (2009) 104–110

The identification of B. besnoiti tachyzoite and bradyzoite IDAs notably improve serological diagnosis since it showed perfect agreement with IFAT, with particular emphasis on early diagnosis, and may help to distinguish acute from chronic infections. Moreover, the present study set the basis for further proteomic studies of stage-specific antigens. The role of these proteins in tachyzoite-tobradyzoite conversion should also be a subject of further research. Acknowledgements We would like to thank Dr. L. Innes (Moredun Research Institute, Scotland, UK) and Dr. J.A. Castillo (Dpto. de Patologı´a Animal, Sanidad Animal, Facultad de Veterinaria, Universidad de Zaragoza, Spain) for kindly providing serum from a T. gondii experimentally infected animal and sera from B. besnoiti naturally infected cattle, respectively. We also thank veterinarians A. Morales and J.M. Pereda (SAT Coagral, Guadalajara, Spain) for supplying the skin biopsies and sera from naturally infected cattle. We wish to acknowledge J.M. Marce´n, Bele´n Rodrı´guez and Vanessa Navarro for their excellent technical assistance. This work has been supported by a research grant from the Santander-Universidad Complutense de Madrid. References Agosti, M., Belloli, A., Morini, M., Vacirca, G., 1994. Segnalazione di un focolaio di Besnoitiosi in bovini da carne importati. Praxis 15, 5–6. Aguado-Martı´nez, A., A´lvarez-Garcı´a, G., Ferna´ndez-Garcı´a, A., Risco-Castillo, V., Arnaı´z-Seco, I., Rebordosa-Trigueros, X., Navarro-Lozano, V., Ortega-Mora, L.M., 2008. Usefulness of rNcGRA7- and rNcSAG4-based ELISA tests for distinguishing primo-infection, recrudescence, and chronic bovine neosporosis. Vet. Parasitol. 157, 182–195. A´lvarez-Garcı´a, G., Pereira-Bueno, J., Go´mez-Bautista, M., Ortega-Mora, L.M., 2002. Pattern of recognition of Neospora caninum tachyzoite antigens by naturally infected pregnant cattle and aborted foetuses. Vet. Parasitol. 107, 15–27. Besnoit, C., Robin, V., 1912. Sarcosporidioses cutane´e chez une vache. Rec. Vet. 37, 649. Bigalke, R., 1968. New conception of the epidemiological features of bovine benoitiosis as determined by laboratory investigations. Onderstepoort J. Vet. Res. 35, 3–137. Bigalke, R.D., 1981. Besnoitiosis and globidiosis. Diseases of cattle in the tropics. Economic and zoonotic relevance. Curr. Top. Vet. Med. Anim. Sci. 6, 429–442. Cha´vez-Vela´squez, A., A´lvarez-Garcı´a, G., Go´mez-Bautista, M., CasasAstos, E., Serrano-Martı´nez, E., Ortega-Mora, L.M., 2005. Toxoplasma gondii infection in adult llamas (Lama glama) and vicunas (Vicugna vicugna) in the Peruvian Andean region. Vet. Parasitol. 130, 93–97. Cortes, H., Leitao, A., Vidal, R., Vila-Vicosa, M.J., Ferreira, M.L., Caeiro, V., Hjerpe, C.A., 2005. Besnoitiosis in bulls in Portugal. Vet. Rec. 157, 262–264. Cortes, H.C., Nunes, S., Reis, Y., Staubli, D., Vidal, R., Sager, H., Leitao, A., Gottstein, B., 2006. Immunodiagnosis of Besnoitia besnoiti infection by ELISA and Western blot. Vet. Parasitol. 141, 216–225. Esteban-Redondo, I., Maley, S.W., Thomson, K., Nicoll, S., Wright, S., Buxton, D., Innes, E.A., 1999. Detection of T. gondii in tissues of

sheep and cattle following oral infection. Vet. Parasitol. 86 (3), 155–171. Ferna´ndez-Garcı´a, A., Risco-Castillo, V., Pedraza-Dı´az, S., Aguado-Martı´nez, A., A´lvarez-Garcı´a, G., Go´mez-Bautista, M., Collantes-Ferna´ndez, E., Ortega-Mora, L.M., 2009. First isolation of Besnoitia besnoiti from a chronically infected cow in Spain. J. Parasitol. 95 (2), 474–476. Ferna´ndez-Garcı´a, A., Risco-Castillo, V., Zaballos, A., A´lvarez-Garcı´a, G., Ortega-Mora, L.M., 2006. Identification and molecular cloning of the Neospora caninum SAG4 gene specifically expressed at bradyzoite stage. Mol. Biochem. Parasitol. 146, 89–97. Franco, E.E., Borges, I., 1915. Nota sobre a sarcosporidiose bovina. Rev. Med. Vet. XIV, 255–299. Frank, M., Pipano, E., Rosenberg, A., 1977. Prevalence of antibodies against Besnoitia besnoiti in beef and dairy cattle in Israel. Refuah Vet. 34, 83–86. Goldman, M., Pipano, E., 1983. Serological studies on bovine besnoitiosis in Israel. Trop. Anim. Health Prod. 15, 32–38. Janitschke, K., De Vos, A.J., Bigalke, R.D., 1984. Serodiagnosis of bovine besnoitiosis by ELISA and immunofluorescence tests 196. Onderstepoort J. Vet. Res. 51, 239–243. Juste, R.A., Cuervo, L.A., Marco, J.C., Oregui, L.M., 1990. La besnoitiosis ˜ a. Med. Vet. 7, 613–618. bovina: desconocida en Espan Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Leitao Da Silva, J.L., 1963. Para´sitas dos Animais Dome´sticos em Portugal Metropolitano. Inst. Alta Cultura. Fundac¸ao Galouste-Gulbekian, Lisboa. Marotel, M., 1912. Discussion of paper by Besnoit e Robin. Bull. et Me´m. de la Soc. des Sciences Ve´t. de Lyon et de la Soc. de Me´d. Ve´t. de Lyon e du Sud-Est 15, 196–217. McAllister, M.M., Parmley, S.F., Weiss, L.M., Welch, V.J., McGuire, A.M., 1996. An immunohistochemical method for detecting bradyzoite antigen (BAG5) in Toxoplasma gondii-infected tissues cross-reacts with a Neospora caninum bradyzoite antigen. J. Parasitol. 82, 354–355. Mehlhorn, H., Klimpel, S., Schein, E., Heydorn, A.O., Al-Quraishy, S., Selmair, J., 2009. Another African disease in Central Europa: besnoitiosis of cattle. I. Light and electron microscopical study. Parasitol. Res. 140 (4), 861–868. Risco-Castillo, V., Ferna´ndez-Garcı´a, A., Ortega-Mora, L.M., 2004. Comparative analysis of stress agents in a simplified in vitro system of Neospora caninum bradyzoite production. J. Parasitol. 90, 466–470. Schares, G., Peters, M., Wurn, W., Ba¨rwald, A., Conraths, F.J., 1998. The efficiency of vertical transmission of Neospora caninum in dairy cattle analysed by serological techniques. Vet. Parasitol. 80, 87–98. Shkap, V., Ungarwaron, H., Pipano, E., Greenblatt, C., 1984. Enzyme linked immunosorbent-assay for detection of antibodies against Besnoitia besnoiti in cattle. Trop. Anim. Health Prod. 16, 233–238. Shkap, V., Pipano, E., Greenblatt, C., 1987. Cultivation of Besnoitia besnoiti and evaluation of susceptibility of laboratory animals to cultured parasites. Vet. Parasitol. 23 (3–4), 169–178. Shkap, V., Reske, A., Pipano, E., Fish, L., Baszler, T., 2002. Immunological relationship between Neospora caninum and Besnoitia besnoiti. Vet. Parasitol. 106, 35–43. Staubli, D., Nunez, S., Sager, H., Schares, G., Gottstein, B., 2006. Neospora caninum immunoblotting improves serodiagnosisof bovine neosporosis. Parasitol. Res. 99, 648–658. Thrusfield, M., 1995. Diagnostic testing. In: Veterinary Epidemiology, The University Press, Blackwell Science Ltd., Cambridge, Great Britain, pp. 266–285. Trees, A.J., Guy, F., Low, J.C., Roberts, L., Buxton, D., Dubey, J.P., 1994. Serological evidence implicating Neospora species as a cause of abortion in British cattle. Vet. Rec. 134, 405–407. Von-Blumro¨der, D., Schares, G., Norton, R., Williams, D.J.L., EstebanRedondo, I., Wright, S., Bjo¨rkman, C., Fro¨ssling, J., Risco-Castillo, V., Ferna´ndez-Garcı´a, A., Ortega-Mora, L.M., Sager, H., Hemplhill, A., Van Maanen, C., Wouda, W., Conraths, F.J., 2004. Comparison and standardisation of serological methods for the diagnosis of Neospora caninum infection in bovines. Vet. Parasitol. 120, 11–22.