All animal experiments were performed in strict accordance with the China Agricultural University Institutional Animal Care and Use Committee guidelines (CAU20160628-2) and followed the International Guiding Principles for Biomedical Research Involving Animals. Experiments were approved by the Beijing Administration Committee of Laboratory Animals.

Eimeria tenella (XJ strain), E. maxima (BJ strain), Eimeria mitis (CAU strain), and E. acervulina (CAU strain) used in this study were maintained by propagating them in coccidian-free, 2- to 5-week-old Arber Acres broilers. The procedures for collection, purification, and sporulation were carried out as previously described (16).

One-week-old SPF chickens were purchased from Merial Animal Health Co., Ltd. (Beijing, China) and were fed a pathogen-free diet and water ad libitum.

Total RNA was isolated from E. maxima sporozoites was isolated using the TRIzol reagent (Invitrogen, USA). cDNA was synthesized through the utilization of random primers and a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, USA). The open reading frame of E. maxima IMP1 was amplified by PCR using IMP1-F/IMP1-R1 primers (Table S1 in Supplementary Material) designed according to the IMP1 sequences (GeneBank Accession number: FN813228.1). Flag-tag was targeted at the 3′ end of IMP1 gene by two rounds overlapping PCR using IMP1-F/IMP1-R2 and IMP1-F/IMP1-R3, respectively (Table S1 in Supplementary Material). After PCR amplification, Nde I and Sac II restriction enzyme sites were introduced into IMP1 fragment. The transfection vector, pSDEP2AIMP1S, was constructed based on pSDEP2ARS (17). Briefly, the ssRFP-His tag fragment of pSDEP2ARS was replaced by IMP1-Flag-tag fragment based on Nde I and Sac II restriction enzymes digestion and DNA ligase ligation. The transfection vector was linearized by SnaB I restriction enzyme for further use.

A restriction enzyme-mediated integration was adapted for the transfection of E. tenella sporozoites as previously described (18). Stably transfected E. tenella line was obtained by pyrimethamine selection combined with fluorescence activated cell sorting (Table S2 in Supplementary Material). Western blot and indirect immunofluorescent assay (IFA) were conducted to confirm foreign antigens expression and distribution in transgenic parasites based on the previous described protocols, respectively (15). Sporozoites were extracted from fresh collected sporulated oocysts. First and secondary schizonts and gametocytes were collected from variable dose EmIMP1 (5 × 10⁷, 1 × 10⁷, and 1 × 10³ oocysts, respectively) infected chickens at 3, 5, and 7 dpi using the previous methods (19). Soluble antigens of sporozoites, first-generation schizonts, second-generation schizonts, gametocytes, and unsporulated oocysts were resolved by SDS-PAGE and immunoblot analysis following standard protocols with mouse anti-flag monoclonal antibody and HRP-conjugated goat anti-mouse IgG (Proteintech, USA) as primary and secondary antibody, respectively. Poly-antibodies against E. tenella GAPDH and enhanced yellow fluorescent protein (EYFP) served as controls.

Indirect immunofluorescent assays were conducted to detect the distribution and relative expression level of foreign antigen. The primary and secondary antibody were mouse anti-flag monoclonal antibody and Cy3-conjugated goat anti-mouse IgG (Proteintech, USA). Poly-antibodies against endogenous IMP1 of E. tenella served as control.

Three groups of four inbred SPF chickens were inoculated with 100 sporulated wild-type E. tenella, EtER (transgene control) and Et-EmIMP1 oocysts at 1 week of age. The output of oocysts of the transgenic lines and wild-type parasite in chickens was measured daily by using McMaster egg counting chamber between 4 and 14 days’ postinoculation, respectively (20).

Six groups of two inbred SPF chickens were inoculated with 5 × 10⁷, 1 × 10⁷, 1 × 10⁶, 1 × 10⁵, 1 × 10⁴, and 1 × 10³ sporulated Et-EmIMP1 oocysts at 1 week of age. Chickens were restricted to feed to reduce the amount of cecum. For each infection dose groups, chickens were euthanized every 24 h for necropsy. The cecum tissue was washed three times using cold PBS. Fresh smears were made from pieces of tissues from the mucous membrane of the cecum and visualized using a confocal laser scanning microscope (SP5, Leica, Germany) for the detection of EYFP-expressing parasites.

Four groups of six inbred SPF chickens were either left naïve or were immunized by infection with 200 sporulated wild-type E. tenella, EtER, and Et-EmIMP1 oocysts at 1 week of age. Secondary immunization was administrated at 2-week intervals with 1,000 oocysts as immunization dosage for each bird. All chickens were housed in the same conditions of temperature and humidity and fed a coccidian-free diet and water ad libitum. Serum was collected when chickens were 1, 3, and 5 weeks old and stored at −20°C until use. Challenge infections with wild-type E. tenella (50,000 oocysts) or E. maxima (50 oocysts) were conducted 2 weeks after the last immunization (at 5 weeks of the chicken age). Oocyst output and body weight gain of the chickens were evaluated after challenge infection.

ELISA was conducted as previously described (15, 21). Briefly, 5 µg/ml E. tenella or E. maxima oocysts antigens or 1 µg/ml recombinant EmIMP1 were coated onto the individual wells of the plates followed by a reaction with serum (1:100) and collected when chicken were 1, 3, and 5 weeks old. The secondary antibody used in this experiment was the HRP-conjugated goat anti-chicken IgY Fc fragment (Bethyl Laboratories, Inc.). Foreign antigen and parasites specific cellular immune responses revealed by IFN-γ secreting cells present in peripheral blood mononuclear cells (PBMCs) after immunization with the transgenic parasite or its wild-type parasite were evaluated by ELISPOT following established protocols in our laboratory (22). Briefly, 10⁶ PBMCs from the naïve, the wild-type E. tenella, EtER, and the Et-EmIMP1 oocysts immunized birds were stimulated with 10 µl PBS, 10 µg E. tenella or E. maxima oocysts antigen or 2 µg recombinant EmIMP1, and 10 µl PMA plus ionomycin (10 ng/ml PMA plus ionomycin 5 µg/ml), respectively. The spots in which IFN-γ secretion lymphocytes were present were detected after 24 h stimulation.

Groups of 12 inbred SPF chickens were either left naïve (Ctrl) or were immunized by infection with 200 sporulated wild-type E. tenella (WT), EtER, and Et-EmIMP1 oocysts at 1 week of age. Chickens were housed in the same conditions of temperature and humidity. New litter of chopped straw was spread over cages’ bottom 5 cm, and chickens were fed a coccidian-free diet and water ad libitum. Six chickens of each group were removed to new cages and challenged with E. maxima (50 oocysts/bird) at 14 and 28 dpi, respectively. Oocyst outputs were evaluated after each challenge infection.

To test the species-specific protective immunity elicited by Et-EmIMP1, groups of 12 inbred SPF chickens were either left naïve (Ctrl) or were immunized by infection with 200 sporulated wild-type E. tenella (WT), Et-EmIMP1 oocysts at 1 week of age. Chickens were housed in the same conditions of temperature and humidity and fed a coccidian-free diet and water ad libitum, while all the chickens were infected with E. acervulina or E. mitis (50 oocysts/bird) at 14 and 28 dpi.

GraphPad Prism 6.01 (GraphPad Software) was used for statistical analysis. Differences between control and treated groups were analyzed using SPSS 12.0 (SPSS Institute Inc.). Differences in the experimental treatments were tested using Duncan’s Multiple Range Test following ANOVA with significance reported at p ≤ 0.05.

Techniques for the selection of stably transfected Eimeria parasites have been successfully developed in the last decade (23–25). To improve the transfection efficiency, here we constructed a relatively short plasmid co-expressing the reporter and target genes (immune mapped protein 1 of E. maxima, EmIMP1) mediated by P2A sequence in a single expression cassette plasmid [Figure 1A; Table S1 in Supplementary Material; (17)]. We obtained the stably transfected parasite (Et-EmIMP1) expressing the reporter gene after an exhaustive selection in vivo (Figure 1B; Table S2 in Supplementary Material) and confirmed that the exogenous plasmid was integrated into the Eth_scaff 16 locus of the Et-EmIMP1 genome (Figure 1C). We also analyzed the oocysts shedding pattern and total oocyst output of the chickens after inoculation with Et-EmIMP1 and its wild type to reveal the reproduction of the parasites affected by the transgenic manipulation. We found similar oocyst shedding patterns as well as total oocyst output for Et-EmIMP1 and its wild type (Figures 1D,E). The peak corresponding to the maximum oocyst output for the transgenic manipulation control [EtER, a transgenic E. tenella line described previously (17)] showed a 24-h delay respect to the wild type (WT) and Et-EmIMP1 (Figure 1D). These data demonstrated that we obtained a pure transgenic Eimeria line to investigate its immunogenicity.

The schizogony stages are considered to be the most immunogenic of the diverse endogenous developmental stages undergone by eimerian coccidian (10). To observe the EmIMP1 expression pattern during the complex life cycle of the transgenic parasite, we first confirmed that the co-expressing reporter gene was continuously expressed in the diverse endogenous developmental stages of Et-EmIMP1 as sporozoite, trophozoite, schizont, and gametocyte, but not in the unsporulated oocyst stage (Figure 2A). These results suggested that EmIMP1 was expressed in the whole life cycle of Et-EmIMP1, as both EYFP and EmIMP1 were controlled by one set of regulatory elements (17). To further confirm the expression pattern, we conducted an immunoblot assay. Consistent with the expression of EYFP, EmIMP1 was highly expressed in the diverse endogenous developmental stages of the transgenic parasite, except in the unsporulated oocyst stage (Figure 2B).

Homologs of the IMP1 gene can be readily identified in eimerian and non-eimerian apicomplexan parasites, and these may also be candidate protective antigens (12, 26, 27). The IMP1 of E. tenella is an immunodominant antigen that elicits partial protective immunity against E. tenella infection (28). To determine whether the exogenous EmIMP1 with such high level of expression would interfere with the endogenous EtIMP1, we analyzed the expression and distribution of EmIMP1 and endogenous EtIMP1 for the transgenic parasite. We found that EmIMP1 was expressed in the whole sporozoites of Et-EmIMP1 except in the refractile body (Figure 2C). The endogenous EtIMP1 was expressed mainly on the cell surface of Et-EmIMP1 sporozoites (Figure 2C), as reported in previous experiments with transient transfection of E. tenella sporozoites with EtIMP1 tagged with red fluorescent protein (28). In general, EmIMP1 was continuously highly expressed during the whole life cycle of Et-EmIMP1 and did not interfere with the endogenous EtIMP1.

Most Eimeria parasites are immunogenic pathogens that elicit systemic and mucous immunity against parental parasite reinfection (29). With few exceptions, transgenic manipulation of Eimeria parasites expressing molecules not relevant for the immune response did not affect their immunogenicity (25, 30). To evaluate the immunogenicity of our transgenic population expressing high levels of immune molecules compared with its parental strain, we first investigated the cell-mediated immune response stimulated by Et-EmIMP1. This response is revealed by the ratio of parasites’ antigen-specific IFN-γ secreting cells in PBMCs after immunization, as they play a major role in protecting the host from Eimeria parasite reinfection (31). We found no significant difference between Et-EmIMP1, its wild-type immunized birds or the transgenic control (EtER) for the E. tenella oocyst antigens (Et Ag)-specific in IFN-γ secreting cells in PBMCs (Figures 3A,B). Humoral immunity also contributes to provide additional immunity against Eimeria parasites infection, especially against E. maxima infection (32). As expected for a cell-mediated immune response, we found that Et-EmIMP1 did not reduce antibody production after immunization (Figure 3C).

To assess the protective efficacy of Et-EmIMP1 as an avian coccidiosis vaccine strain against its parental parasites infection, we infected Et-EmIMP1 immunized birds by oral inoculation with wild-type E. tenella. We found there was no oocyst output neither in Et-EmIMP1 nor in its wild-type immunized birds when the challenge dose was 10,000 oocysts (data not shown). When the challenge dose was 50,000 E. tenella oocysts, the oocyst output was significantly reduced compared to naïve chickens (Figure 3D). With reduced oocyst output, both Et-EmIMP1 and wild-type parasite immunized birds gained much more body weight (Figure 3E). Taken together, these results clearly show that Et-EmIMP1 maintained the immunogenicity of its wild-type parasite and indicate that a coccidiosis vaccine formulation will protect chickens against E. tenella infection.

Vaccination with live parasites (attenuated or un-attenuated strains) is the most powerful way to control coccidiosis in chickens (5, 6, 10). Commercial vaccines against coccidiosis based on live parasites contain multiple species, as there is no cross-protection stimulated by a single parasite. The results presented above showed that Et-EmIMP1 maintained its paternal strain’s immunogenicity protecting chickens from E. tenella infection. As next step, we evaluated the EmIMP1 and E. maxima-specific immune responses elicited by Et-EmIMP1 as well as the protective efficacy against E. maxima infection in chickens. We found that Et-EmIMP1 elicited foreign antigen-specific cellular and humoral immune responses revealed by the number of EmIMP1-specific IFN-γ secreting cells (Figure S1A in Supplementary Material) and antibody production (Figure S1B in Supplementary Material). Immunities against the whole parasite antigens of E. maxima were also detected in Et-EmIMP1 immunized chickens, compared to both naïve and WT parasite immunized birds (Figures 4A–C). To test the protective efficacy of these immunities against E. maxima infection, we infected the chickens immunized with or without Et-EmIMP1 or its wild-type parasite with 50 E. maxima oocysts at 35 dpi and analyzed the E. maxima oocyst production. We found significantly reduced oocyst production of Et-EmIMP1 immunized birds’ respect to the wild-type and transfection control groups (Figure 4D). Our results also demonstrated the well-established paradigm indicating that there is no cross-protective immunity between different Eimeria species, as the same oocyst output was detected in the naïve and wild-type E. tenella immunized birds after E. maxima infection (Figure 4D).

Floor pen test is an effective intermediate step in the process of translating an anticoccidial vaccine from the laboratory to commercial trials. We vaccinated 1-week-old SPF chickens with 200 Et-EmIMP1 oocysts or its wild type. The chickens were fed in the cages with litter of chopped straw. Half of the chickens of each group were removed to new cages and infected with 50 E. maxima oocysts at 14 and 28 dpi, respectively. We found that protective immunity against E. maxima infection was established as early as 14 dpi (Table 1). The immunity was enhanced after self-boosting induced by the offspring oocysts in the litter, as the oocyst production after E. maxima infection of Et-EmIMP1 immunized birds was nearly half of the naïve and wild-type parasite immunized birds at 14 dpi and was reduced to nearly 1/3 of the original at 28 dpi (Table 1).

Immune mapped protein 1 has been identified as immunoprotective antigen from other apicomplexan parasites, such as E. tenella, Toxoplasma, and Neospora (26–28). We tested for the possible cross-protective immunity elicited by the overexpressed EmIMP1 against other Eimeria spp., as it is known that several epitopes are conserved among the seven Eimeria spp. in chickens (Figure S2 in Supplementary Material). We infected the birds with Et-EmIMP1 or wild-type parasite immunization with 50 E. acervulina or E. mitis oocysts and found that at 28 dpi there was no significant difference in the E. acervulina (Figure 5A) and E. mitis (Figure 5B) production between immunized and naïve birds. These results demonstrated that protective immunity elicited by EmIMP1 is species faithful and only protected chickens against E. maxima infection but not against the other six species.