Parasitol Res (2002) 88: 217224 DOI 10.

1007/s00436-001-0524-0

O R I GI N A L P A P E R

Aimdip Noutossi Mouafo Andreas Weck-Heimann Jean-Francois Dubremetz Rolf Entzeroth

Monoclonal antibodies specic for the two types of wall-forming bodies of Eimeria tenella macrogametes (Coccidia, Apicomplexa)
Received: 17 September 2001 / Accepted: 18 September 2001 / Published online: 15 November 2001 Springer-Verlag 2001

Abstract Two monoclonal antibodies (mAbs) raised against the macrogamonts of Eimeria tenella identied antigens located in the wall-forming bodies of type I (WF I) and type II (WF II) by indirect immunouorescence and by immunoelectron microscopy. With these mAbs, the involvement of both types of wall-forming body at the protein level in the formation of the inner and outer oocyst walls of E. tenella was shown by indirect immunouorescence assay. On Western blots of pure macrogamont, mAb E1D8 against WF I reacted with a series of bands between 42 kDa and 105 kDa. In pure, unsporulated extract, this mAb recognized a complex of bands between 26 kDa and 153 kDa. mAb E2E5 against WF II, on Western blots of pure extract of macrogamonts, recognized an antigen of 51 kDa. Later in the development, after the formation of the inner oocyst wall, mAb E2E5 reacted with three polypeptide of 23, 25 and 30 kDa. Proteolytic processing may be forwarded as the mechanism regulating the distinct regulation protein involved in the oocyst wall.

Introduction
Coccidiosis is a disease caused in domestic animal by apicomplexan parasites belonging to the genus Eimeria. Seven Eimeria spp are known to infect chicken (Pellerdy

A.N. Mouafo R. Entzeroth (&) Institute of Zoology, TU-Dresden, Mommsenstrasse 13, 01062 Dresden, Germany E-mail: entz@rcs.urz.tu-dresden.de Fax: +49-351-46337241 A. Weck-Heimann Internet-Multimedia, Kipsdorferstrasse 187, 01279 Dresden, Germany J.-F. Dubremetz Institut de Biologie de Lille 1, FRE 2377CNRS, rue du Professeur Calmette, 59021 Lille cedex, France

1974). Under natural conditions, chickens become infected by ingesting oocysts present in the environment. After being ingested, the sporulated oocyst releases motile sporozoites that actively penetrate the intestinal epithelium and give rise to schizonts, which produce infectious merozoites. After three rounds of merogony, merozoites dierentiate into macrogametocytes and microgametocytes. Sexual reproduction then occurs, leading to a zygote, which becomes an oocyst with a smooth outer layer and a granular inner layer divided by a longitudinal suture (Mouafo et al. 2000). These unsporulated oocysts are gradually excreted with the feces and undergo sporulation in the external milieu, leading to the formation of infectious sporozoites (Hammond 1973). Macrogametes of Eimeria spp can be ultrastructurally identied by two types of wall-forming bodies (WF I, WF II), involved in the building of the oocyst wall (Scholtyseck and Voigt 1964; Scholtyseck 1973). Electron microscopic studies of cell wall formation during macrogametogenesis in E. mivati (Wheat et al. 1976), Eimeria spp (Wang 1982), Sarcocystis spp (Vertterling et al. 1973) and Toxoplasma gondii (Ferguson et al. 1975) have provided evidence for the role of WF I and WF II in oocyst wall genesis. Monne and Honig (1954) postulated a two-layer structure and suggested the inner layer consisted of protein or a lipidprotein matrix. Then, Stotish et al. (1978) showed a composition of 67% peptide, 14% lipid and 19% carbohydrate, after analyzing the puried, unsporulated oocyst wall. The same authors suggested that lipids were in a 10-nm thick outer layer (long-chain alcohols, phospholipids, sterols, triglycerides), covering a 90-nm thick layer of glycoprotein (disulde-linked glycoprotein). Successful propagation of the parasite is due partly to this highly resistant wall, which provides an eective protective barrier for oocyst survival in the environment. Few studies have reported the production of monoclonal antibodies (mAbs) against the oocyst and gametocytes of Eimeria spp: Speer et al. (1983) described an ultrastructural study of antigenic sites on

218

E. tenella oocysts, sporocysts and sporozoites; and protein antigens were extracted from the oocyst wall (Karkhanis et al. 1991; Linda et al. 1991; Eschenbacher et al. 1996; Karim et al. 1996). Wallach et al. (1989) and Larsen et al. (1991) produced mAbs against the macrogamonts of E. maxima and E. tenella, respectively. But none of these mAbs was shown to react with WF and later with the oocyst wall. In the present study, we produced mAb specic for antigens contained in WF I and WF II of E. tenella. Using immunouorescence and immunoelectron microscopy techniques, we were able to follow the involvement of WF in oocyst wall formation.

To analyze the reactivity of antibodies on intracellular macrogamonts and oocysts, single- and double-IFAT were performed on semithin sections of infected chicken ceca embedded in LRWhite (see Immunoelectron microscopy, below). IFAT slides were coated with 2% 3-aminopropyltriethoxysilane (AES; Sigma, Germany) in acetone. After 60 s, the slides were removed from the AES solution and dried for 2 h under a laminar ow. After drying, glass were dipped into 10 ml of distilled water and air-dried for 12 h at room temperature. The slides were then stored in an appropriate container at room temperature. Semithin sections were prepared and air-dried on silane-coated IFAT glass slides. The sections were briey washed with PBS and processed as above, except that they were not treated with acetone.

Double-labeling of semithin sections Semithin sections obtained from the LR-White-embedded material were double-labeled successively with the two anti-WF mAbs in order to demonstrate the anity of each mAb to each type of wallforming body of the macrogametes. The parasites were rst incubated with mAb E2E5 and labeled with the anti-mouse IgG conjugated to FITC). The section was then labeled with mAb E1D8, revealed with rhodamine-conjugated goat anti-mouse IgM secondary antibody (Jackson Immunoresearch Laboratories) and examined with a Zeiss Axioskop 2 microscope under UV illumination. Control wells were probed with culture medium or PBS and also incubated with the secondary antibodies. Immunoelectron microscopy of resin-embedded material Infected ceca were xed with 4% paraformaldehyde and 0.05% glutaraldehyde in PBS for 15 min, dehydrated in ethanol and then embedded in LR-White (London Resin). Ultrathin sections were collected on Formwar-coated nickel grids. The grids were oated for 30 min on 2% bovine serum albumin (BSA) in PBS (PBS/BSA) and were then transferred successively for 1 h onto drops of a hybridoma cell culture supernatant, a dilution of anti-mouse IgG goat antibodies (Sigma, Germany) and protein A/gold (10 nm) in PBS/BSA. Sections were stained with 2% uranyl acetate in water and lead citrate and observed with a Philips 208 transmission electron microscope. Cryo-immunoelectron microscopy Infected chick ceca were xed for 90 min with 4% paraformaldehyde and 0.05% glutaraldehyde in 0.2 M phosphate buer, pH 7.2. The samples were washed with 10% FBS in PBS (PBS/FBS) and inltrated overnight with a solution containing 2.3 M sucrose and 20% polyvinylpyrrolidone (Sigma) in 0.1% sodium phosphate buer, pH 7.2. They were then frozen in liquid nitrogen. Ultrathin cryosections were obtained on a Leica Ultracut equipped with a cryo-attachment operating at 100 C. Sections were oated successively on PBS/FBS, undiluted hybridoma culture supernatant, anti-mouse IgG rabbit serum diluted 1:400 in PBS/FBS, 8 nm Protein A/gold diluted in PBS to an optical density (at 525 nm) of 0.05 units, with 53-min washings in PBS between each step. Sections were then embedded in methylcellulose (2%)-uranyl acetate (0.4%) and observed with a Philips EM 420 electron microscope.

Materials and methods

Preparation of Eimeria tenella (Bayer strain) gametocytes for immunization Gamonts were isolated from the ceca of 3-week-old, experimentally infected chickens, according to Wallach et al. (1989). Gamonts were concentrated in a minimal volume of phosphate-buered saline (PBS). The homogenate was then puried by ultracentrifugation on a percoll gradient (60%, 50%, 30%) centrifuged at 17,000 g for 15 min, followed by another centrifugation at 10,000 g for 5 min in a 65% percoll solution. Pure gamonts were collected and washed four times for 5 min at 800 g. The pellet was resuspended in PBS/phenylmethylsulfonyl uoride and frozen at 20 C or dissolved in sample buer for electrophoresis.

Production of hybridoma cells The equivalent of 5105 pure macrogamonts/ml was homogenized, using 0.5-mm glass beads, and the suspension was mixed with an equal volume of Hunters TiterMax. Then, 50 ll of the emulsied antigen in Hunters TiterMax were injected subcutaneously into 6-week-old BALB/c mice on day 0. The same amount of antigen in Hunters TiterMax was further injected on day 30 and day 60. Three boosters at days 90, 91 and 92 were given by injecting intravenously 20 ll of the homogenized antigen in PBS and the fusion was performed on day 93. The fusion was carried out according to Curtis and Handman (1993), by mixing the splenocytes of the immunized BALB/c mice with mouse myeloma PAI cells (obtained from Dr. Goerlich, University of Bonn) at a 1:2 ratio. Positive hybridoma cells were detected by the indirect uorescent antibody test (IFAT) against macrogamonts and were cloned by limited dilutions, according to Harlow and Lane (1988). Isotyping was performed with a commercial kit (ISO-2) from Sigma. Immunouorescence assay For the screening of hybridoma cell culture supernatants, isolated macrogamonts were air-dried on glass slides and permeabilized at 20 C in acetone. The slides were then dried, washed for 10 min in PBS and blocked in 10% fetal bovine serum (FBS) at room temperature for 10 min. The slides were then probed with undiluted cell culture supernatant at 37 C for 30 min and gently rinsed in 5% FBS/PBS. The bound antibodies were detected using a uorescein-isothiocyanate (FITC) conjugate [anti-mouse Ig(A, G, M, E); Sigma, Germany]. In the control well, the culture medium or PBS was used. After being rinsed with PBS, the slides were mounted (0.1 g 1,4-phenylendiamin, 90 ml glycerol, 10 ml PBS) and examined under an Axiovert 135 uorescence microscope (Zeiss, Germany).

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblotting and periodate oxidation To extract antigen from the cells, scrapings of infected chicken or puried macrogamonts were mixed in PBS supplemented with protease inhibitors (Sigma, Germany) and cells were disrupted by grinding with a Potter homogenizer. The suspension was then centrifuged at 12,000 rpm for 5 min at 4 C and the supernatant was mixed with a volume of sample buer. The sample was then

219 frozen at 20 C until used for electrophoresis. To disrupt the oocysts wall, glass beads were used. The sample was vortexed ve times for 2 min, followed by two cycles of freezing and thawing of the samples at 80 C and then mixed with an equal volume of sample buer. The Samples were always maintained on ice and were always inspected by light microscopy to monitor cell disruption. Samples were heated for 5 min in sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) sample buer and analyzed on 12% and 10% SDS-PAGE gels (Laemmli 1970). After separation, antigens were transferred onto nitrocellulose membrane for 2 h at 100 mA. Molecular weight markers (Sigma and Bio-Rad) were used for calibration. For immunoblotting, the nitrocellulose was saturated for 60 min in 5% non-fat, dried milk in PBS/Tween 20 (0.05%). It was then incubated in mAb (hybridoma culture supernatant) for 60 min at room temperature. The sheet was then washed and incubated in alkaline phosphatase-conjugated anti mouse IgG or IgM (Sigma) diluted 1/1,000 in PBS/Tween 20/milk and then revealed with nitro blue tetrazolium and 5-bromo-4-chloro-3indolyl phosphate (Sigma). To determine the carbohydrate epitope (Woodward et al. 1985), nitrocellulose strips were saturated for 30 min in PBS/Tween 20 (0.05%) at room temperature, briey washed with 50 mM sodium acetate buer (pH 4.5) and then incubated in 50 mM sodium acetate buer (pH 4.5) containing various concentrations (5, 10, 15, 20 mM) of sodium mperiodate for 60 min at room temperature in the dark. After being exposed to freshly dissolved 50 mM sodium borohydride in PBS (pH 7.2) for 30 min at room temperature, nitrocellulose strips were washed for 10 min in 5% non-fat, dried milk in PBS/ Tween 20 and probed with mAb, as described above. Controls consisted of incubating strips in the same buers, without periodate treatment.

Results
Two mAbs reacting against macrogamonts and oocysts of Eimeria tenella were obtained: The mAb E1D8 was of the IgM isotype, while E2E5 was of the IgG2a isotype. When probed on LR-White semithin section by singleIFAT, E1D8 labeled the granules present in the macrogametes (Fig. 1a1, a2). E1D8 also reacted with the oocyst present in the section. By observing the labeled parasite under bright-eld illumination, we realized that the labeled structure within the oocyst was the oocyst wall. As the oocyst wall was recognized by the mAb E1D8, the cytoplasmic granules were no longer labeled. From this result, it is clear that the labeled bodies present within the macrogametes are WF bodies. In young macrogamonts [observed at 136 h post-infection (p.i.)] no reaction was observed with the mAb. Using mAb E2E5 and probing on young macrogamonts (observed within the infected ceca at 136 h p.i.), a discrete uorescent signal was observed within the macrogamont on the semithin section (gure not shown). Fluorescence signals were also observed in mature macrogamonts and oocysts (Fig. 1b1, c). When the corresponding parasites were observed under bright-eld illumination (Fig. 1b2) it was clearly seen that the labeled structure within the oocyst was the inner oocyst wall. This result indicates that the mAb E2E5 is also specic against the WF bodies. In the mature macrogamonts, uorescent bodies were located at the periphery of the parasites. To differentiate between Fig. 1parts a1 and c and also to

Fig. 1ac Immunouorescence labeling of the wall-forming bodies (within the macrogamonts) and oocyst wall on semithin sections using specic monoclonal antibodies (mAb) against Eimeria tenella macrogamonts. a1, a2 mAb E1D8 recognizing bodies within the macrogamont and the oocyst wall. a2 is the bright-eld light micrograph of a1. Note the absence of uorescent bodies within the cytoplasm of the oocyst (see the stars), 630. b1, b2 Involvement of the bodies recognized by mAb E1D8 (macrogamont) in the oocyst wall. b2 is the bright-eld light micrograph of b1, 630. c Reactivity of mAb E2E5 with macrogamont wall-forming bodies and the oocyst wall. Note the absence of uorescent bodies within the macrogamont cytoplasm, 630

conrm the specicity of both mAbs by indirect immunouorescence assay against two dierent set of bodies, both mAbs were used simultaneously in a double-IFAT. They clearly reacted with two independent sets of granules that were easily dierentiable by their colors, depending on the lter and the specic antibody (Fig. 2).

220

Fig. 2ac Simultaneous labeling of bodies (double-labeling) within the macrogamont on semithin sections. a The section was probed with mAb E2E5 and labeled with uorescein-isothiocyanate. Note the peripheral disposition of the green-stained bodies within the macrogamont, 630. b The same parasite on the section was probed with mAb E1D8 and labeled with rhodamine. Note the peripheral disposition of the green-stained bodies within the macrogamont, 630. c Micrograph of the macrogamont, showing the disposition of both types of body recognized by mAbs E1D8 and E2E5 within the macrogamont after taking photos with both lters to show combined staining, 630

Red uorescent signals were due to the presence of antigen recognized by mAb E1D8, as revealed by anti-mouse IgM-rhodamine. Green uorescent signals were observed between the red bodies; and their presence indicated the specicity of mAb E2E5 with the antigen present within them, as revealed by anti-mouse IgG-FITC. In the control wells, no uorescence was observed in the parasites (macrogametes, oocysts). The ndings by IFAT were conrmed by immunoelectron microscopy. When ultrathin sections containing immature macrogamonts (136 h p.i.) were probed with mAb E1D8 and labeled with anti-mouse-IgM-gold, no gold particle was observed by transmission electron microscopy. In similar parasite stages (136 h p.i.) probed with mAb E2E5 and labeled with protein A-gold, many gold particles were present on the immature WF II (Fig. 3a). The absence of gold particles on the younger macrogamonts after probing with mAb E1D8 suggested that WF I were not yet formed. On the mature macrogamonts containing mature WF I and WF II probed with mAb E2E5 and labeled with protein A-gold, many gold particles were also observed on the mature sponge WF II (Fig. 3b). Gold particles were also observed in WF I within the mature macrogamonts after

probing the parasites with the mAb E1D8 and labeling with anti-mouse-IgM-gold (Fig. 3c). Samples collected at dierent times p.i. and sporulated oocysts (either treated with sodium hypochloride or not) were used for Western immunoblots. At 138 h p.i., mAb E2E5 reacted with a protein band of approximately 51 kDa (Fig 4). The mAb used on samples collected at 144 h p.i. recognized a similar 51-kDa band, together with four additional proteins bands (at 23, 25, 30, 46 kDa, respectively; gure not shown). On samples collected from the cecal lumen of infected chicken at 168 h p.i., only three protein bands were found (at 23, 25, 30 kDa; Fig. 4A). No reaction was observed with the scrapings from non-infected ceca (Fig. 4A). In sporulated oocysts, two proteins bands were recognized (at 31 kDa, 81 kDa), whether they had been treated with sodium hypochloride or not (Fig. 4B). When electrophoresis was performed under reducing or non-reducing conditions and Western immunoblotted, no dierence was observed in the mobility of antigens. Also, the immunoblotting reactivity was not abolished by prior treatment of blots with sodium m-periodate. mAb E1D8 did not react with samples of infected gut taken at 138 h p.i. However at 140 h p.i., a complex prole including four major groups of bands (at 105, 76, 57, 42 kDa) was observed (Fig 5A). With samples collected in the cecal lumen of infected chickens at 168 h p.i., seven bands (of about 153, 111, 100, 50, 43, 29, 26 kDa) were identied with our mAb (Fig 5B). The sporulated oocyst extracts probed with the same mAb revealed a major band at 43 kDa; and the reactivity of mAb E1D8 on sporulated oocysts was abolished by treatment with sodium hypochloride (gure not shown). The proles obtained with E1D8 were not modied, whether samples were reduced or not. Treatment of the nitrocellulose membrane with sodium m-periodate did not aect the recognition of bands by mAb E1D8 after immunoblotting.

Discussion
By double-IFAT, it was possible to dierentiate between two types of wall-forming bodies within the mature macrogamonts, using two mAbs. The specicity of these antibodies towards wall-forming bodies was conrmed by immunoelectron microscopy. Characterization of the antigens recognized by mAb E1D8 During macrogametogenesis, WF I is the last structure to be formed (Scholtyseck 1973). This observation was conrmed by the negative reactivity of the specic mAb against WF I at 137 h p.i., despite the fact that macrogamonts were present in the preparation and WF II was detected at the same stage by the other mAb, E2E5.

221 Fig. 3ac Immunoelectron micrographs showing the specicity of mAbs E2E5 and E1D8 against wall-forming bodies of type two (WF II) and type one (WF I), respectively. a Ultrathin section through a macrogamont of E. tenella in the ceca of an experimentally infected chicken at 136 h post infection (p.i.). The sections were probed with mAb E2E5 and labeled with protein A-gold (10 nm). Bar 0.2 lm. b Mature macrogamont (140 h p.i.). The sections were probed with mAb E2E5 and labeled with anti-mouse protein A-gold (10 nm). Note the specic labeling of WF II and the absence of gold particles on WF I. Bar 0.5 lm. c mAb E1D8 probed on mature macrogamont (cryopreserved section) labeled with gold particles (8 nm). Note the specic labeling of WF I and the absence of gold particles from WF II. Bar 0.5 lm

The mAb E1D8 recognized bands at 42, 57, 76 and 105 kDa in the infected cecal samples extracted at 140 h p.i. Bands at 56 kDa and 54 kDa were detected in Eimeria tenella gamonts (Wallach et al. 1995a, b) and those authors suggested that they epitopes were on wall-forming bodies, without specifying the type of wall-forming body. The 57 kDa of WF I appeared as a doublet at 57/56 kDa. Similarly, the 56-, 82- and 230-kDa gametocyte antigens of E. maxima (Fried et al. 1992; Wallach et al. 1995a, b) and the 250 kDa of the Cryptosporidium parvum (Bonnin et al. 1991) oocyst sometimes appeared as doublets and even triplets. This appearance of several bands in WF I antigens is not very well understood. The reactivity of the anti-WF I

antibody in the present study was not diminished by sodium m-periodate (Woodward et al. 1985), leading us to conclude that this antigen was not glycosylated. Thus, the appearance of the complex antigen prole was not due to glycosylation, which has been shown in C. parvum to cause such cross-reactivity between multiple bands (Bonnin et al. 1991). Considering that macrogamonts during and after hyaluronidase treatment were present together with the intestinal extract (and therefore were accessible to intestinal enzymes and then to degradation during purication) we cannot exclude the idea that the patterns observed might be due to degradation. However, protease inhibitors were used during extraction and, furthermore, only one 51-kDa

222

Fig. 4A, B Western blot analysis of mAb E2E5 reacted with E. tenella antigens. A Immunoblot of the total proteins of macrogamonts and oocysts probed with mAb E2E5, using crude extract (scraping) of the ceca from an experimentally infected chicken at 136 h p.i. (lane 1), crude extract of oocysts collected at 168 h p.i. from the luminal contents of the ceca (lane 2) and scraping of the mucosae of non-infected chicken (control, lane 3). The standard molecular weight markers are indicated. B Immunoblot of the total proteins of E. tenella oocysts probed with mAb E2E5, using extract of sporulated oocysts without sodium hypochloride treatment

Fig. 5A, B Western immunoblots showing the reactivity of mAb E1D8 with the macrogamonts and oocyst wall extracts of E tenella. A Immunoblot analysis of puried macrogamonts (percoll gradient). Pure macrogamonts probed with mAb E1D8 (lane 1) and the same sheet of Western blot probed with mAb E2E5 (lane 2). The standard molecular weight markers are indicated. B Immunoblot of a non-sporulated oocyst extract collected at 168 h p.i. from the lumen of ceca from an experimentally infected chicken. Proteins were separated by 10% SDS-PAGE

protein of WF II was observed using the same purication protocol. The outer layer of the oocyst wall in Eimeria spp is formed by electron-dense WF I, which appears later in the development of macrogamont (Scholtyseck and Voigt 1964). With mAb E1D8 as a marker, seven bands (of about 230, 111, 100, 50, 43, 29, 26 kDa) were identied in the samples collected at 168 h p.i. At this time, only oocysts were present in the samples. This led us to conclude that all these seven bands originated from the oocyst extract. Therefore, materials that were previously located within the macrogamont were later transferred into the outer oocyst wall. This result is in agreement with the ndings of Bonnin et al. (1991) and Entrala et al. (2001), but with the dierence that, in the present study, the WF I antigens were directly incorporated into the outer oocyst wall while, in the case of Cryptosporidium, the electronlucent vesicles were released into the parasitophorous vacuole and later incorporated in the outer oocyst wall (Bonnin et al. 1991). Protein proles recognized by mAb E1D8 were different between sporulated and non-sporulated oocysts. Similar results have been observed in E. maxima (Linda

et al. 1991). A common 43-kDa protein was detected in the outer wall of the sporulated and unsporulated oocysts using mAb E1D8. Minor bands present on Western blots of the sporulated oocyst might be due to contamination of the sample with unsporulated oocysts. During this study, protein extract from the sporulated oocysts pre-treated with sodium hypochloride gave a negative result with mAb E1D8. This conrms that mAb E1D8 detected antigens present in the outer oocyst wall, which is removed by the treatment with sodium hypochloride. Characterization of the antigens recognized by mAb E2E5 Western blots with cecal extracts at 136 h p.i. from chickens experimentally infected with E. tenella led to the identication of a single 51-kDa protein. The appearance of this antigen coincided with the presence of macrogamonts in the infected ceca and the development of WF II within the macrogamonts. The same 51kDa antigen was also observed at 138 h p.i., a time when mature WF II were observed by transmission electron

223

microscopy within the macrogamonts (gure not shown). The inner wall of the oocyst is formed by material derived from WF II Scholtyseck and Voigt 1964). In the current study, the involvement of WF II antigens in oocyst formation was studied with the help of mAbs. mAb E2E5 (anti-WF II) reacted with dierent bands, depending on the time of sampling. At 144 h p.i., when oocysts were forming and gamonts were still present, four bands (of about 23, 25, 30, 51 kDa) were observed. Later, at 168 h p.i., when only oocysts were present, three bands (23, 25, 30 kDa) were recognized on Western blots of cecal lumen content. These results were conrmed by light-microscopic observation of the same preparation, in which macrogametes and oocysts were identied at 144 h p.i. while, at 168 h p.i., only oocysts were present in the cecal lumen. This clearly indicates that the 23-, 25- and 30-kDa polypeptides correspond to unsporulated oocyst proteins. Similarly, a 30-kDa antigen has been described in the unsporulated oocysts of E. acervulina, E. maxima and E. tenella (Stotish et al. 1976; Eschenbacher et al. 1996). These data strongly suggest that a proteolytic processing occurs for the 51-kDa protein at the time of release, to form the inner layer of the oocyst wall and give rise to the smaller molecular species recognized by E2E5. This processing is unlikely to be due to artifactual proteolysis, since it does not occur in samples taken at 138 h p.i. Whether it occurs inside the cell before exocytosis or during the deposition of the inner wall remains to be established. No dierence in the mobility of antigen was observed when the samples were treated with the reducing agent, dithiothreitol (DTT). This result indicated the absence of a disulde bridge in the antigens. Stotish et al. (1978) found a single polypeptide chain of 10 kDa only after treating the sample with DTT, a nding which led the authors to suggest that the inner oocyst wall was composed of cross-linked disulde glycoproteins. Because of the absence of disulde bridges and also the insensitivity of the polypeptides recognized by E2E5 to periodate treatment, the antigen described by Stotish et al. (1978) is clearly dierent from the one we describe here. Two bands of about 31 kDa and 81 kDa were observed with mAb E2E5 on Western blots of sporulated oocysts. In the sporulated oocyst extract of E. tenella, 26-, 35- and 110-kDaproteinshavebeencharacterized(Karkhanisetal. 1991; Michalski et al. 1993). The protein pattern on Western blots of the sporulated oocyst wall diers from the 23-, 25- and 30-kDa proteins bands that were recognized by mAb E2E5 (anti-WF II) in the unsporulated oocyst wall. Dierences between the protein prole of the sporulated and unsporulated oocyst of E. maxima have also been described (Linda et al. 1991). The authors found that the sporulated oocyst contains protein bands at approximately 21, 22, 45 and 200 kDa while, in the unsporulated oocyst, one protein of approximately 31 kDa was detected. Since a certain percentage of the unsporulated oocysts is always present in

the sample after sporulation, we could not exclude that the 31 kDa is derived from unsporulated oocysts. Thus, the absence of the 23 kDa and 25 kDa may be due to the lower concentration of unsporulated oocysts in the sample. Therefore, the 81-kDa protein might be specic for sporulated oocyst wall. Whether this could be due to some polymerization of the polypeptide during sporulation remains to be studied. Considering the fact that, by IFAT, no cytoplasmic material was stained after the complete formation of the oocyst wall, which was also conrmed by transmission electron microscopy, immunoelectron microscopy and Western blots, it can be stated that: 1. The WF antigens contribute to oocyst wall formation and the total WF antigens are integrated in the inner and outer oocyst walls. 2. The protein of WF II of E. tenella macrogamonts undergoes a proteolytic processing and gives rise to three polypeptides before or after formation of the oocyst wall. 3. The protein composition of the inner and outer oocyst walls of E. tenella is dierent, as the composition of the WFs is dierent. 4. During sporulation, there are changes in the oocyst wall antigens. From the present results, it could be concluded that there are changes in the conformation of protein during oocyst sporulation within the inner and outer oocyst walls. It has been suggested that a burst of translation was associated with sporulation, leading to the modication of oocyst wall proteins after sporulation (Sutton et al. 1989). Furthermore, working on 14C-leucine incorporation during E. tenella sporulation led to the suggestion that protein synthesis occurred during the 7 h of sporulation (Wang and Stotish 1975), which may lead to the alteration of oocyst wall proteins. To understand the function of WF proteins during macrogametogenesis, further investigations are needed, for example purifying the proteins and screening the cDNA. It will be also be useful to study the crossreactivity of the present monoclonal antibodies with other chicken Eimeria spp.
Acknowledgements We wish to thank Dr. G. Greif (Bayer AG) for providing parasites used in this work and Mrs G. Guhrn for her technical assistance. This study was supported by Bayer AG, Leverkusen, Germany and the German Academic Exchange, which provided a research scholarship.

References
Bonnin A, Dubremetz J-F, Camerlynck P (1991) Characterization and immunolocalization on an oocyst wall antigen of Cryptosporidium parvum (Protozoa: Apicomplexa). Parasitology 103:171177 Curtis J, Handman E (1993) In: Goding JW (ed) Monoclonal antibodies: principle and practice, 3rd edn. Academic Press, London

224 Entrala E, Sbihi Y, Sanchez-Moreno M, Mascaro C (2001) Antigen incorporation on Cryptosporidium parvum oocyst walls. Mem Inst Oswaldo Cruz 96:233235 Eschenbacher K-H, Eggli P, Wallach M, Braun R (1996) Characterization of a 14-kDa oocyst wall protein of Eimeria tenella and E. acervulina. Parasitology 112:169176 Ferguson DJ, Birch-Andersen A, Hutchison WM, Siim JC (1975) The ultrastructural development of the macrogamete and formation of the oocyst wall of Toxoplasma gondii. Acta Pathol Microbiol Scand Sect B 83:491505 Fried M, Mencher D, Sar-Shalom O, Wallach M (1992) Developmental gene expression of a 230-kilodalton macrogametespecic protein of the avian coccidial parasite Eimeria maxima. Mol Biol Parasitol 51:251262 Hammond DH (1973) Life cycles and development of coccidia. In: Hammond DM, Long PL (eds) The coccidia. Eimeria, Isospora, Toxoplasma and related genera. University Park Press, Baltimore, pp 4579 Harlow E, Lane D (1988) Antibodies: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Karim MJ, Shibesh CB, Stree AJ (1996) Characterization and immunoprotective properties of monoclonal antibody against the major oocyst wall protein of Eimeria tenella. Infect Immun 64:12271232 Karkhanis YD, Nollstadt KA, Bhogal BS, Ravino O, Pellegrino R, Crane MS, Murray PK, Tunner MJ (1991) Purication and characterization of protective antigen from Eimeria tenella. Infect Immun 59:983989 Laemmli UK (1970) Cleavage of structural proteins during the assembly of heat of bacteriophage T4. Nature 227:680685 Larsen NC, Rasmussen KR, Healey MC (1991) Production and partial characterization of monoclonal antibodies specic for the gamonts of Eimeria tenella. J Parasitol 77: 10121015 Linda MP, Ainsworth AJ, Brown JE, Haney JA (1991) Characterization of proteins in sporulated and unsporulated Eimeria maxima oocysts. Am J Vet Res 52:7274 Michalski WP, Prowse SJ, Bacic A, Fahey KJ (1993) Molecular characterization of peanut agglutinin-binding glycoproteins from Eimeria tenella. Int J Parasitol 33:985995 Monne L, Honig G (1954) On the properties of the shells of the coccidian oocysts. Ark Zool 7:251256 Mouafo AN, Fleig R, R. Entzeroth (2000) Observation of sutures in the oocyst wall of Eimeria tenella (Apicomplexa). Parasitol Res 86:10151117 Pellerdy LP (1974) Coccidia and coccidiosis, 2nd edn. Akademiai Kiado, Budapest Scholtyseck E (1973) Ultrastructure. In: Hammond DM, Long PL (eds) The coccidia. Eimeria, Isospora, Toxoplasma and related genera. University Park Press, Baltimore, pp 81144 Scholtyseck E, Voigt WH (1964) Die Bildung der Oocystenhulle der bei Eimeria perforans (Sporozoa). Z Zellforsch Mikrosk Anat 62:279292 Speer CA, Wong RB, Schenkel RH (1983) Ultrastructural localization of monoclonal IgG antibody for antigenic sites of Eimeria tenella oocysts, sporocyst, and sporozoites. J Protozool 30:548554 Stotish RL, Wang CC, Hichens M, VandenHeuvel WJA, Gale P (1976) Studies of a glycoprotein in the oocysts of Eimeria tenella. J Biol Chem 25:302307 Stotish RL, Wang CC, Meyenhofer M (1978 ) Structure and composition of the oocyst wall of Eimeria tenella. J Parasitol 64:10741081 Sutton CA, Shirley MW, Wisher MH (1989) Characterization of coccidial proteins by two-dimensional sodium dodecyl sulphatepolyacrylmide gel electrophoresis. Parasitology 99:175187 Vetterling JM, Pacheco ND, Fayer R (1973) Fine structure of gametogony and oocyst formation in Sarcocystis sp. in cell culture. J. Parasitol 20:613621 Wallach M, Mencher D, Yarus S, Pillemer G, Halabi A, Pugatsch T (1989) Eimeria maxima: identication of the gametocyte protein antigens. Exp Parasitol 68:4956 Wallach M, Smith NC, Braun R, Eckert J (1995a) Potential control of chicken Coccidiosis by maternal immunization. Parasitol Today 11:262267 Wallach M, Smith NC, Petracca M, Catherine MDC, Eckert J, Braun R (1995b) Eimeria maxima gametocyte antigens: potential use in a subunit maternal vaccine against coccidiosis in chickens. Vaccine 13: 347354 Wang CC (1982) Biochemistry and physiology of coccidia. In: Long PL (ed) The biology of the coccidia. University Park Press, Baltimore, pp 167228 Wang CC, Stotish RL (1975) Changes in nucleic acids and proteins in the oocysts of Eimeria tenella during sporulation. J Protozool 22:438443 Wheat BE, Jensen JB, Ernst JV, Chobotar B (1976) Ultrastructure of macrogametogenesis of Eimeria mivati. Z Parasitkd 50:125136 Woodward MP, Young WW, Bloodgood RA (1985) Detection of monoclonal antibodies specic for carbohydrate epitopes using periodate oxidation. J Immunol Methods 78:143153