The wild-type strain of E. tenella used in this research was isolated from birds showing clinical signs of cecal coccidia on a broiler farm in Querétaro, Mexico’s central state. The oocysts of E. tenella were purified using the method reported by Stephan et al. (30). Three-week-old Leghorn Specific Pathogen Free (SPF) chickens (SPAFAS Inc., Norwich, CT, United States) were used to propagate oocysts. Following established procedures, cecal oocysts were isolated, sporulated, and cleaned (31).

Sporocysts were released by vortexed sporulated oocysts (2.5 × 10⁷/ml) at 2,000 X g using 1 mm diameter glass beads (Sigma-Aldrich, Inc., Burlington, MA, United States) for 1 min. A 50% Percoll gradient (density 1.13 g/mL, GE Healthcare, Piscataway, NJ, United States) was used to purify sporocysts (32). 1 × 10⁷ purified sporocysts were resuspended in an excystation medium and incubated at 42°C for 150 min. PBS (pH 7.4) with 0.75% (w/v) taurodeoxycholic acid (Sigma-Aldrich, Inc., Burlington, MA, United States) and 0.25% (w/v) trypsin from porcine pancreas Type II-S (Sigma-Aldrich, Inc., Burlington, MA, United States) comprised the excystation medium. Sporozoites (Sz) were washed and purified using a 60% Percoll gradient (density 1.13 g/mL, GE Healthcare, Piscataway, NJ, United States). Sporozoites were resuspended in sterile PBS for the immunization program, gradually frozen at −70°C at a 1°C/min rate, and then stored at −70°C until use.

Three 10-week-old hybrid pullets were gavage-inoculated with 5 × 10⁵ sporulated oocysts of E. tenella and euthanized 112 h post-infection (PI) to collect second-generation merozoites (Mz). These birds’ intestines were treated for merozoite isolation, as described previously by Liu et al. (33). The merozoites were purified using the method outlined by Geysen et al. (34). Each parasite (Sz and Mz) was suspended in sterile PBS with a protease inhibitor (cOmpleteTM Roche Applied Science, Mannheim, Germany). Five freeze/thaw cycles each disrupted both asexual zoite stages. The final suspension was centrifuged at 2,000 x g for 18 min at 4°C. The supernatants were collected, and the protein concentration was measured with the Bradford reagent (BioRad, Hercules, CA, United States) using a bovine serum albumin standard curve (Sigma-Aldrich, Inc., Burlington, MA, United States). Antigen suspensions were kept in 200 μL aliquots at −70°C until needed.

Two white New Zealand rabbits (2.5 kg) and two SPF White Leghorn chickens (1.1 kg) were immunized subcutaneously with 100 μg of purified whole sporozoites (5.3 × 10⁶ sporozoites), diluted 1:1 with the IMS 1313 N VG PR nanoparticle adjuvant (Seppic Montanide™, France) in a total volume of 1 mL as previously was described by Juárez-Estrada et al. (35). Three further immunizations were given at two-week intervals. Both rabbits were euthanized by exsanguination 1 week after the last immunization. After clotting for 1 h at room temperature and overnight at 4°C, the blood was centrifuged at 2000 X g for 5 min, and serum samples were aliquoted and stored at −20°C until use. Both SPF Leghorn chickens were wing-bled, and serum was extracted by clotting it for 1 h at room temperature and overnight at 4°C, then centrifuged at 2000 X g for 5 min, aliquoted, and stored at −20°C until needed. According to the manufacturer’s instructions, the rabbit serum’s IgG fraction was purified by affinity chromatography using Sepharose 4 FF protein G (GE HLS Marlborough, Mass., United States).

The reactogenicity of the bird’s and rabbit’s anti-sporozoite sera, the rabbit anticlone serum, and the purified rabbit IgG fraction to sporozoite and 2nd generation merozoite antigens were evaluated by ELISA, essentially, as Constantinoiu et al. (3) described it. 96-well microtiter plates (MaxiSorb, Nunc, Roskilde, Denmark) were coated overnight at 4°C with 1 μg of sporozoite or merozoite antigen in 100 μL of carbonate buffer (0.1 M sodium bicarbonate and 0.1 M sodium carbonate buffer, pH 9.6). The control wells were incubated with only 100 μL of carbonate buffer. After four washes on a shaker with a saline solution (S) (120 mM NaCl, 25 mM Tris–HCl, pH 7.9) containing 1% Tween 20 (ST), non-specific binding sites were blocked by incubating for 1 h at 37°C in a static oven with 110 μL of 5% skim milk in ST (STM). Sera diluted in STM (1:10 and 1:100) were added to the test and control wells after four ST washes, and both were then incubated for 1 h at 37°C. Each plate included negative control sera from unimmunized SPF Leghorn chickens. Following incubation, the plates were washed four times with ST and incubated with their respective secondary antibody peroxidase conjugate diluted 1:2000 with STM (anti-rabbit, anti-chicken) (Jackson ImmunoResearch Laboratories, Inc. West Grove, PA, United States). After 1 h at 37°C, the plates were washed four times. The enzymatic reaction was developed by adding 100 μL of OPD chromogen (o-phenylendiamine dihydrochloride SIGMA, St. Louis MO, USA) at 5 μg/10 mL in citrate buffer (0.1 M citric acid, 0.1 M sodium citrate p.H. 4.5, and 20 μL of 30% hydrogen peroxide) for 10 min in a shaker under dark conditions. An ELISA microplate spectrophotometer (Epoch, BioTek, Winooski, VT, United States) was used to read the absorbance produced by substrate hydrolysis at 450 nm. All serum samples were examined in duplicate thrice times.

For SDS-PAGE analysis of both asexual zoite phases, 20 μg of each purified fraction was separated by 12% SDS-PAGE under standard reducing conditions. According to Constantinoiu et al. (36), the resolved proteins were stained with Coomassie brilliant blue (CBB) or electrotransferred on polyvinylidene difluoride membranes (PVDF) (Bio-Rad, Hercules, CA, USA). PVDF membranes were probed with anti-sporozoite of E. tenella sera from rabbits and Leghorn fowl. As a secondary antibody, horseradish peroxidase (HRP)-conjugated IgG goat anti-rabbit IgG, and (HRP)-conjugated IgG goat anti-chicken IgY (Jackson ImmunoResearch Laboratories, Philadelphia, PA, United States) were employed at a 1: 1500 (v/v) in 5% (w/v) STM. Tablets of DAB SigmaFast™ (3,3′-diaminobenzidine) (Sigma-Aldrich, St Louis MO, United States) at 10 mg/5 mL were used to visualize the PVDF membranes.

IgG fraction purified from rabbits previously immunized with sporozoites of E. tenella was used to screen commercially available dodecapeptide (Ph.D.-12TM) and cysteine-constrained heptapeptide (Ph.D.-C7CTM) M13 pIII libraries (New England Biolabs Inc., Ipswich, MA, United States). Three rounds of clone selection were performed for each screening using progressively decreasing quantities of antibodies (20 μg, 10 μg, and 5 μg, respectively). Briefly, two wells in a 96-well polystyrene microtiter plate (Maxisorb, Nunc, Roskilde, Denmark) were coated and incubated at 4°C overnight (O/N) with IgG in sterile phosphate chloride buffer (PCB) (0.1 M potassium chloride, 3 mM sodium chloride, 5 mM sodium phosphate dibasic, 1 mM potassium phosphate monobasic). Wells were washed five times with 0.1% (v/v) Tween 20 in 0.01 M phosphate-buffered saline (TPBS), pH 7.2, then blocked with BSA (5 μg/mL in distilled H2O). Every plate was incubated for 1 h at room temperature with 1 × 10¹⁰ plaque-forming units (pfu) of each phage library. After washing with TPBS, the bound phages were eluted with 100 μL of 0.1 M glycine-HCl, pH 2.2 (hereafter referred to as “eluate”) and immediately neutralized with 15 μL of 1 M Tris–HCl, pH 9.1. The eluate phages were amplified and titer through infection of the E. coli strain ER2738 (New England Biolabs Inc., Ipswich, MA, United States). The phages were titled as plaque-forming units by ml (pfu/ml). For the next round of phage-display screening, every phage-display library eluate was amplified and concentrated with polyethylene glycol-8000 PEG/2.5 M NaCl (20%/40%) following the manufacturer’s instructions. The phage input and output titers ratio was used to calculate the specific phage enrichment level.

An aliquot of approximately 100 pfu from the third-round panning was combined with E. coli ER2738 and plated on LB agar with IPTG (isopropyl-β-D thiogalactopyranoside) and X-gal (5-bromo-4-chloro-3-indoyl-β-D-galactopyranoside). After 37°C O/N growth, twenty separate isolated phage plaques per library were randomly collected (n = 40). Each plaque was grown in 2XYT broth containing a dilution of 1/1000 of E. coli strain ER2738 grown O/N and incubated for 5 h at 37°C at 250 rpm. Following standard procedures, single-stranded phage DNA was isolated using an iodide buffer extraction, precipitated with 1 volume of ethanol, and resuspended in nuclease-free distilled water (Thermo Fisher Scientific Inc. Waltham, MA, United States). Purified phage DNA was sequenced with the Big Dye Terminator v1.1 cycle sequencing kit (Thermo Fisher Scientific, Waltham, MA, United States) using the pIII primer (5’-CCAGACGTTAGTAAATG-3′).

Using the Clone Manager Professional Edition program, version 9.0 (Scientific and educational software, Cary, NC, United States), the amino acid sequences of the peptide were determined from the nucleotide sequences. The protein BLAST software¹ was used to screen the selected clones in the published sequences of the Eimeria tenella genome. The settings for the search were: Database: non-redundant protein sequence (nr); organism: Eimeria tenella (taxid: 5802); algorithm: blast (protein–protein BLAST).

ELISA reactogenicity to the selected phage clones was assessed using the pooled anti-sporozoite serum from both rabbits immunized with sporozoites of E. tenella. Briefly, phage–displayed clones were bound to microtiter plates (MaxiSorb, Nunc, Roskilde, Denmark) and developed with the anti-sporozoite sera diluted (1:10 and 1:100). Wild-type M13 phage and a clone of M13 phage (Ph.D. 12mer and c7c libraries) from screening with pseudorabies virus rabbit IgGs were used as negative controls. The total phage elution of the third round panning of both libraries was amplified and used as a positive control. After four washes with SS, the anti-sporozoite antibodies were detected using a secondary anti-rabbit IgG or anti-chicken IgY peroxidase conjugate diluted at 1:2,000 in blocking solution for 1 h at 37°C (Jackson ImmunoResearch Laboratories, Inc. West Grove, PA, United States). The plate was then washed four more times with SS before adding 100 μL of TMB chromogen at 1 mg/10 mL in citrate buffer (3,3′,5,5′-Tetramethylbenzidine, Sigma-Aldrich, Inc., Burlington, MA, United States), the addition of 2 N sulfuric acid stopped this reaction. An ELISA reader (Epoch, BioTek, Winooski, VT, United States) was used to determine the color intensity at 450 nm. Otherwise, the antisera reactivity of phage-immunized rabbits was determined in phage-ELISA format using 1 μg/well of E. tenella sporozoite and merozoite as antigens.

Rabbit antisera for every one of the four most relevant phage clones were obtained. Candidate phage clones were selected based on the following criteria: (i) highest frequency selection during the screening process, and (ii) highest reactivity in the ELISA test (e.g., Ph.D. 12mer 15: HFAYWWNGVRGP; Ph.D. 12 mer 18: AGHTTQFNSKTT; Ph.D. c7c 1: CNTGSPYEC; Ph.D. c7c 7: CSISSLTHC). One rabbit by each clone was subcutaneously immunized with 1 mL containing 2 × 10¹² pfu in sterile PBS, mixed 1:1 with the adjuvant ISA 50 V (Seppic Montanide ™, France). Six immunizations were applied at intervals of 2 weeks each. The immunological reactogenicity and specificity of every antiserum from phage-immunized rabbits were evaluated by ELISA and Western blot, as already described before.

Data were analyzed using the SAS/STAT 9.2 software (SAS Institute Inc., Cary, NC). Overall differences between groups were analyzed using ANOVA test. The mean value with respective standard deviation (SD) was calculated. A post-hoc analysis (Tukey’s test at p < 0.05) was performed after the overall analysis showed significant differences between all groups.

Time-course of the reactogenicity of the anti-sporozoite sera from rabbits and birds immunized with whole-sporozoites of E. tenella was evaluated. Figure 1 shows the kinetics of E. tenella anti-sporozoite sera production in rabbits (Figure 1A) and birds (Figure 1B), assessed by its reactivity to sporozoite and merozoite; we can observe that rabbit and bird antisera cross-react to both asexual zoite stages in a similar pattern. Similar levels of antibody production were observed in rabbits and chickens; however, bird titers reached a plateau after 2 weeks, whereas rabbit titers did not reach a plateau until 4 weeks. Even though only sporozoites were used for immunization schedule, the immune response to the second generation of merozoites exhibited similar reactivity to the sporozoite antigen and followed the same dynamic pattern, indicating the presence of the same or cross-reacted antigens between both asexual zoite stages. Antisera diluted 1:100 showed better reactivity against the sporozoite than the merozoite (Figure 1B). At 7 week post-first immunization titers in rabbits against both asexual zoite stages are more homogenous than titers observed in birds at this same time indicating more affinity from rabbit antibodies toward both antigens.

Reactogenicity of the anti-sporozoite sera from rabbits and birds immunized with whole-sporozoites of E. tenella against polypeptides of two asexual zoite stages of E. tenella was evaluated. Figure 2A shows the sporozoite and merozoite proteins separated on a 12% SDS-PAGE and stained with Coomassie brilliant blue. Although a complex protein profile of sporozoite and merozoite proteins was detected, it is possible to identify some common bands above 35 kDa. In the sporozoite, two specific distinct proteins below 25 kDa are evident, while in the merozoite, a band closer to 10 kDa is notoriously abundant. High molecular weight proteins seem shared among both asexual stages zoites but with some differences in concentration. The Western blot analysis with anti-sporozoite sera from rabbits (Figure 2B) and birds (Figure 2C) showed that both antisera recognized a similar polypeptide pattern in the sporozoite antigen, mainly polypeptides with molecular weight higher than 35 kDa, among which a 40 kDa protein is relevant. In the sporozoite, the most intense bands detected weighted 21 kDa, 23 kDa, and 24 kDa. In contrast, in the 2nd generation merozoite antigen, the antisera from rabbits or chickens indistinctly detected only a few high molecular weight antigens (approximately 70 kDa and 124 kDa). The most relevant here is that each rabbit antiserum had to be proportionally diluted four times more than the bird antiserums in order to obtain the clearest image of the PVDF membrane, which indicates a superlative reactogenicity of these antibodies toward both antigens subject to evaluation.

In order to select recombinant Phage-display clones with specific affinity for rabbit anti-sporozoite antibodies, the IgG fraction was purified from the rabbit’s antisera by affinity chromatography with protein G sepharose. We diluted the purified IgG fraction from 30 μg/well to 0.23 μg/well to test by indirect ELISA their reactivity against the sporozoite or the 2nd generation of the Mz antigens. Figure 3 shows greater reactivity to the Sz antigen compared to the Mz antigen.

Using this same purified IgG fraction three screening rounds were performed for the 12-mer and cyclic 7-mer phage-display libraries. Table 1 shows the number of phage clones selected during each screening round. The selected phages of the 12-mer Ph.D. library increased from 2.0 × 10⁵ pfu in the first round to 6.9 × 10⁶ pfu in the third round. Similarly, phages obtained from the cyclic 7-mer Ph.D. library increased from 1.5 × 10⁵ pfu in the first round to 2.2 × 10⁷ pfu in the third round, indicating a selection of specific phage clones through the three screening rounds.

After the third screening round of each Phage-display random peptide libraries, 20 single phage clones were randomly selected and amplified. DNA sequencing was used to examine all of the selected clones. Tables 2, 3 provide the deduced amino acid sequences of the selected clones. DNA sequencing of the 12-mer Ph. D. library-selected phage clones revealed two immunodominant clones with the AGHTTQFNSKTT (7 times) and HFAYWWNGVRGP (8 times) peptide sequences. One clone with the amino acid sequence PNSAFWAGSER was obtained twice (Table 2). The clones selected from the cyclic 7-mer library were more diverse than those selected from the 12-mer library, four clones were selected twice, and one clone was selected three times (Table 3). All selected clones matched with one or two main proteins of E. tenella previously quoted in the NCBI gen bank. Some of these proteins are crucial in adhesion and invasion events of the sporozoite to the enterocyte, however, most of them have been described as hypothetical proteins.

The E. tenella genome was BLAST screened with the sequence of every selected peptides. The peptides obtained from the 12 mer library found clear identities (Table 2). The AGHTTQFNSKTT peptide matched with a Cullin homolog. It was found that the HFAYWWNGVRGP peptide has an identity with the Dynein beta chain. The phage clone sequence GPNSAFWAGSER showed three matches with the Elongation Factor 2 (EF-2). The peptides obtained from the cyclic 7-mer library also found matches in the E. tenella genome (Table 3). The CNTGSPYEC clone had ten matches with the extracellular domains of EtMIC4’s epidermal growth factor-like repeats. Furthermore, the amino acid sequence of the phage clone CMSTGLSSC matched the thrombospondin type-I repeats of EtMIC4. In contrast, the CSISSLTHC clone showed several matches with a hypothetical protein conserved in E. tenella. Finally, as seen in Table 3, at least three 7-mer peptides share the CLK and KL motifs; interestingly, two of these phage clones showed the highest reactivity in the ELISA test (Figure 4C).

The anti-sporozoite sera from rabbits and Leghorn chicks were used to test the recognizing level of every Phage clone selected from each Phage display random peptide libraries. Figure 4 shows the antigen reactivity of the selected Phage clones to rabbit and bird anti-sporozoite sera evaluated by indirect ELISA. Most of the selected phage clones interacted specifically with anti-sporozoite sera and had greater OD values than negative control sera (Figure 4). Every Phage clone showed different level of reactogenicity toward anti-sporozoite sera. Although with some differences almost all Phage clones showed reactogenicity in the same order to both anti-sporozoite sera, it indicates that the sporozoite has immunodominant antigens recognized by both species.

If the peptide clones selected with the anti-sporozoite antibodies represent mimotopes present in E. tenella antigens, these phage clones will be able to induce antibodies against E. tenella antigens. In that sense, four Phage clones were selected and used to immunize one rabbit each. Every anti-clone rabbit serum was used to assess the recognition of the sporozoite and merozoite native antigens. These four Phage clones were selected for rabbit immunization based on their (i) ELISA reactivity to anti-sporozoite antibodies, (ii) frequency of clone selection, and (iii) bioinformatic similarity with E. tenella proteins. In this way, phage clones 1 and 7 of the Ph.D. c7c library, and phage clones 15 and 18 of the Ph.D. 12 mer library were selected for rabbit immunization (Table 2). Figure 5 shows the time course reactogenicity of the antiserum from every rabbit immunized with the selected phage clones to both asexual zoite stages of E. tenella. Although with different time course reactivity against every life cycle of E. tenella antigen, overall the rabbit anti-clone serums showed high titer recognizing both antigens tested, it was observed a time course light tendency to recognice more to Mz than Sz antigen, however at last of the immunization shedule the reactogenicity of every anti-clone serum agains both antigens was indistinghishable.

The reactivity level and specific polypeptide recognition of every rabbit anti-phage serum to Sz and Mz antigens was investigated by Western blot. Figure 6 shows the reactivity of the anti-phage 1 (Figure 6C), anti-phage 7 (Figure 6D), anti-phage 15 (Figure 6E), and anti-phage 18 (Figure 6F) serums. Figure 6A shows both E. tenella asexual zoite stages protein profile and BSA resolved by 12% SDS-PAGE electrophoresis. It was a control group for every PVDF membrane studied. Figure 6B shows the reactivity of the antiserum pool from two rabbits immunized with wild-type M13 used as a negative control.

All anti-phage serums notably recognized both Sz and Mz antigens (Figure 6). The anti-M13 negative control sera recognized bands of approximately 38 kDa, 20 kDa, and 13 kDa in the Mz proteins and only a ≈ 25–27 kDa range band in the Sz extract. These bands were also detected in all blots tested with all anti-phage serums (Figure 6). The anti-phage 1 serum detected prominent bands of 78 kDa, 68 kDa, 48 kDa, and 40 kDa in the Sz proteins (Figure 6C, lane Sz). In addition, a pair of approximately 23 and ≈25–27 kDa antigens (as was already seen with M13 antiserum) were recognized. Two bands of 68 kDa and 60 kDa were recognized in the Mz proteins; both share the same molecular weight with antigens detected in the sporozoites. A band of 40 kDa was observed, which is also shared in the Sz antigens (Figure 6A).

The phage clone 7 antiserum (Figure 6D) detected at the Mz extract a pattern of bands closely similar to the anti-phage 1 serum. In contrast, very few bands were detected in the Sz proteins. A light band of 40 kDa (also present in the Mz) and a band of ≈25–27 kDa (present in the anti-M13 sera). The 40 kDa antigen was detected in the Sz and Mz proteins by the anti-phage 1 and anti-phage 7 serums. Anti-phage 15 (Figure 6E) and anti-phage 18 (Figure 6F) serums detected a similar antigenic pattern in the Mz proteins and a pair of bands of approximately 68 kDa and 70 kDa. These bands were also detected in the Sz antigen. A specific band of 40 kDa was detected in the Sz and Mz proteins with the anti-phage 18. This antiserum recognized a prominent band of 45 kDa. This same antiserum also recognized bands of 34, 33, 27, 24, and 17 kDa in the Sz extract (Figure 6F, lane Sz). Although with different rank of reactivity and molecular mass, the four anti-phage clone serums recognized more bands in the Sz and Mz antigens than recognized by the sera from rabbits immunized with a wild-type M13 strain that was used as a negative control.