Introduction

Coccidiosis, a parasitic disease caused by Eimeria spp. infection, is responsible for major economic loss to the poultry industry (Williams 1999). Currently, prophylactic control with infeed anticoccidials is the common practice. However, long-term prophylactic drug usage not only promotes the development of drug resistance and adds great cost to the poultry industry, it also creates concerns in the public about the chemical residues in food (Allen and Fetterer 2002). Unattenuated live vaccines or attenuated live vaccines have been available for the control of coccidiosis in the poultry industry (Irfan Anwar et al. 2008). However, these unattenuated live vaccines are limited to a certain degree by the pathogenicity of the parasites used. Attenuated live vaccines had the possibility reverting back to a pathogenic form (Sharman et al. 2010). Consequently, research efforts have been invested in the development of anticoccidial subunit vaccines composed of protective antigens as an alternative to live vaccines. The identification of protective antigens is vital in the development of subunit vaccines. So efforts continue to be directed towards the finding of novel vaccine targets (Li et al. 2006).

Apicomplexan pathogens replicate exclusively within the confines of a host cell. Invasion into the host cells requires an array of specialized parasite molecules, many of which have long been considered to have potential as targets of drug or vaccine-based therapies (Muhammad et al. 2010; Dowse et al. 2008; O'Donnell and Blackman 2005). Among the targets that have been studied, rhomboid proteases seem promising.

Members of the rhomboid protease family have been identified and studied in protozoa, such as Toxoplasma gondii, Plasmodium spp., Eimeria tenella, Entamoeba histolytica, and Cryptosporidium parvum (Baker et al. 2006; Li et al. 2006; Trasarti et al. 2007; Brossier et al. 2008; Sheiner et al. 2008). Recent studies have suggested that proteases played critical roles in the process of apicomplexan parasites invasion. Rhomboid proteases were found in Toxoplasma and Plasmodium to cleave cell adhesins that were essential for invasion (Urban and Freeman 2003; Kim 2004; Zhou et al. 2004; Brossier et al. 2005; Carruthers and Blackman 2005; Dowse et al. 2008; Baker et al. 2006; Freeman 2008). Although rhomboid proteins are related to the apicomplexan protozoa invasion process, the protective efficacy as vaccine antigens against parasitic diseases remains unclear.

In prior works, we have cloned a rhomboid-like cDNA sequence from E. tenella with an open reading frame of 774 bp. Recombinant proteins of this novel cDNA were successfully expressed in Escherichia coli (Li et al. 2006). Sequence analysis revealed seven transmembrane domains and a rhomboid domain within the protein sequence (Li et al. 2006; Zheng et al. 2011). It has been reported that recombinant E. tenella rhomboid-like proteins expressed by recombinant fowlpox virus (rFPV) and recombinant Mycobacterium bovis BCG (rBCG) strains could elicit immune responses and provide partial protection against E. tenella challenge (Yang et al. 2008; Wang et al. 2009).

In the present study, specific humoral and cellular immune responses were evaluated in chicken immunized with a recombinant subunit vaccine of E. tenella rhomboid protein. The protective efficacy of this vaccine against challenge with E. tenella sporulated oocysts was also being examined.

Materials and methods

Production of recombinant rhomboid protein

For recombinant protein expression in E. coli, Rosetta host bacterial cells transformed with p-ETRH01 plasmid were induced with 1 mM IPTG for 4 h. Inclusion body from the induced bacteria was purified and refolded with previously reported procedure (Carrio et al. 2000). The recombinant rhomboid protein fractions were visualized on 10% SDS–acrylamide gels stained with Coomassie brilliant blue and on Western blots probed with horseradish peroxidase-conjugated anti-His monoclonal antibody (1:2,000; Tiangen), and stored at −20°C.

Immunization and challenge infection

One-day-old male Leghorn broilers obtained from commercial breeders (Changchun, China) were reared in a coccidian-free environment in wire cages. At day 7, the chickens were randomly assigned to three groups (20 birds in each group). Birds in the experimental group were immunized intramuscularly with 100 μg recombinant rhomboid protein mixed with equal volume of Freund's complete adjuvant. These birds were injected again at days 7 and 21 after the first immunization with the same amount of antigen mixed with equal volume of Freund's incomplete adjuvant or protein only, respectively. Birds in the control group were injected three times with PBS. Birds were challenged with sporulated oocysts of E. tenella (Xinjiang strain) at a dose of 3 × 104 per bird 14 days after the final immunization.

Specific antibody responses induced by recombinant rhomboid protein in chicken

To assay for specific antibodies, serum samples were collected from 10 chickens in each group randomly at days 7, 14, 28, and 42 and examined by ELISA. Briefly, the microtiter plates were coated overnight at 4°C with crude antigens of E. tenella sporozoites (100 μg/ml) and blocked with 5% bovine serum albumin (BSA) in PBS at room temperature (RT) for 2 h. After washing with PBST three times, sera at 1:200 diluted in 1% BSA–PBS were added to corresponding wells and incubated at 37°C for 1 h. After washed with PBST for three times, the plates were incubated with HRP-labeled goat anti-chicken IgG antibody (Beijing Boisynthesis Biotechnology Co., Ltd.; 1:1,000) at 37°C for 1 h and washed again. Finally, substrate solution containing 15 μl 30% H2O2, 10 ml citrate–phosphate and 4 mg O-phenylenediamine was added (100 μl/well). The reaction was stopped by 2 M H2SO4 and the optical density (OD) value was read at 490 nm in a microplate reader.

Production of interleukin-2 and γ-interferon induced by recombinant rhomboid protein in chicken

Serum samples were collected from individual bird at days 7, 14, 28, and 42. Interleukin-2 (IL-2) and interferon-γ (IFN-γ) were measured with ELISA kits (BG Biotechnology Co., Ltd.) according to the manufacturer's instructions. The results were described as picograms of IL-2 or IFN-γ per 100 μl of samples.

Measurements of the numbers of CD4+ and CD8+ cells induced by recombinant rhomboid protein in chicken

For evaluation of cellular immunity, chickens were killed at day 14 after the boosting immunization and their spleens were removed aseptically. The splenocytes were washed twice and resuspended in PBS at a concentration of 107cells/ml and incubated with R-phycoerythrin-conjugated mouse anti-chickens CD4 antibody (0.1 mg/ml) and fluorescein-conjugated mouse anti-chickens CD8α antibody (0.5 mg/ml; Southern Biotech Associates, Inc.) for 40 min at RT in the dark. After two washes with PBS, cells were resuspended in a fluorescence preservative solution and the proportions of CD4+ and CD8+ cells were analyzed by flow cytometry.

Evaluation of protective efficacy immunization with recombinant rhomboid protein against E. tenella sporulated oocysts challenge in chicken

The protective efficacy of the recombinant rhomboid protein was measured according to the cecal lesion scores, oocyst output, and body weight gain (BWG). Cecal lesion scores were determined 7 days after chickens being challenged with the sporulated oocysts of E. tenella according to the method of Johnson and Reid (1970). For the calculation of oocyst output, feces from each group were collected separately between 7 and 9 days postchallenge and the numbers of oocysts per gram feces were calculated using Danforth's counting technique (Danforth 1998). The BWG of the chickens in each group were determined at the beginning and the end of the experiment. Percentage protection = (the number of oocysts from control chickens − the number of oocysts from vaccinated chickens)/the number of oocysts from control chickens × 100%.

Statistical analysis

Statistical analysis was performed by variance (ANOVA) and Duncan's multiple range tests using SPSS 16.0 software. Differences were considered to be statistically significant between two groups when p < 0.05.

Results

Expression of recombinant rhomboid protein

The recombinant rhomboid protein containing an NH2-terminal His6 epitope tag was expressed in E. coli, and the encoded protein was purified by Ni +2 chelate affinity chromatography. A protein band of approximately 30 kDa was observed on an SDS–acrylamide gel (Fig. 1a). A similar band profile was detected by western blotting using a monoclonal antibody against the His epitope tag (Fig. 1b). The final yield of the affinity purified protein was 4.0 mg/l.

Fig. 1
figure 1

Expression of recombinant rhomboid protein. a Inclusion body of recombinant rhomboid protein and control were resolved on 12% SDS–acrylamide gels stained with Coomassie brilliant blue. Lane 1 lysate sediment of E. coli expressing rhomboid, lane 2 lysate supernatant of E. coli expressing rhomboid, lane 3 lysate of E. coli not expressing rhomboid, lane M protein size markers. b Refolded His-tagged recombinant rhomboid protein was analyzed by western blotting with horseradish peroxidase-conjugated anti-His monoclonal antibody. Lane 1 refolded recombinant rhomboid protein, lane 2 lysate of E. coli not expressing rhomboid, lane M protein size markers

Full size image

Evaluation of humoral immunity

Specific antibody in chickens induced by recombinant rhomboid protein was detected by ELISA. No obvious difference in mean absorbance values was detected at days 14 and 28. However, a significant difference in mean absorbance values was observed at day 42 in chickens immunized with rhomboid protein. The data are shown in Fig. 2.

Fig. 2
figure 2

Antibody responses induced by recombinant rhomboid protein (rETRHO1). Chickens were immunized i.n. with recombinant rhomboid protein and PBS as control under the same condition. Sera were collected and analyzed by ELISA. A490 OD readings of sera diluted 1:200 were shown. Each bar represents the mean OD (±SE, n = 10), the asterisks represent significant increase of serum antibody titers when compared with those of the PBS control (p < 0.05)

Full size image

Evaluation of IL-2 and IFN-γ productions

After individual immunization with recombinant rhomboid protein, IL-2 and IFN-γ levels in the serum samples were examined using an ELISA assay. The average expression level of IL-2 was estimated to be 313.7 pg/100 μl at day 42, which was more than threefolds increase compared with the PBS group (96.2 pg/100 μl; p < 0.01, N = 10; as shown in Fig. 3). The average expression level of IFN-γ was estimated to be 260.2 pg/100 μl at day 42, which is significantly increased over the PBS group (98.7 pg/100 μl; p < 0.05; N = 10). The data are shown in Fig. 4.

Fig. 3
figure 3

IL-2 response induced by recombinant rhomboid protein. Each bar represents mean ± SD (N = 10 for birds rhomboid immunized; N = 6 for birds PBS immunized). The asterisks indicate significant increase in IL-2 expression level compared to the PBS group. **p < 0.01

Full size image
Fig. 4
figure 4

IFN-γ response induced by recombinant rhomboid protein. Each bar represents mean ± SD (N = 10 for birds rhomboid immunized; N = 6 for birds PBS immunized). The asterisks indicate significant increase in IFN-γ expression level compared to the PBS group. *p < 0.05

Full size image

Evaluation of the percentages of CD4+ and CD8+ lymphocytes

The percentages of CD4+ and CD8+ lymphocytes (CD4+ with 16.89 ± 2.00 and CD8+ with 25.20 ± 1.75) in chickens immunized with recombinant rhomboid protein were significantly higher than those in the control group (CD4+ with 9.90 ± 0.32 and CD8+ with 20.58 ± 0.66, p < 0.05). The data are shown in Table 1.

Table 1 Changes of CD4+ and CD8+ T cells from the chickens immunized with recombinant rhomboid antigen
Full size table

Protective efficacy of immunization with the recombinant rhomboid protein

Oocyst output, BWG, cecal lesion scores, and percentage protection are summarized in Table 2. Chickens immunized with recombinant rhomboid protein revealed a significant decrease in oocyst output when compared with chickens in the PBS group (p < 0.05). The oocyst output indicated that recombinant rhomboid protein provided a partial protection around 77.3%. The body weight increased significantly (p < 0.05), and the cecal lesion decreased significantly.

Table 2 Protective efficacy of recombinant rhomboid protein against E. tenella challenge
Full size table

Discussion

In the present study, the recombinant rhomboid antigen has been evaluated as a candidate for E. tenella subunit vaccine. Specific antibody responses as well as cell-mediated responses were detected in chicken immunized with recombinant rhomboid protein. Oocyst output, BWG and cecal lesion scores, and protective efficacy were also evaluated after challenged with E. tenella sporulated oocysts. The results demonstrated that recombinant rhomboid protein can provide partial protection against E. tenella challenge.

Rhomboid proteases belonged to a nearly ubiquitous family of intramembrane serine proteases with unique activity of being able to cleave within the first few residues of their substrates (Koonin et al. 2003). Rhomboid-1 from Drosophila melanogaster, the first member of this widespread protease family to be identified, promoted the cleavage of the membrane-anchored TGF-alpha-like growth factor Spitz, allowing it to activate the Drosophila EGF receptor (Urban et al. 2001). These proteins typically contained seven transmembrane domains and a catalytic triad within the membrane bilayer involving an asparagine, a histidine, and a serine residue that might be involved in the protease functions (Urban et al. 2001). Recent biochemical reconstitution and high-resolution crystal structures have provided proof that rhomboid proteins functioned as novel intramembrane proteases, with a serine protease-like catalytic apparatus embedded within the membrane bilayer, buried in a hydrophilic cavity formed by a protein ring (Ha 2009). Although the function of most rhomboids was not yet known, they have already been implicated in growth factor signaling, mitochondrial function, and protein translocation across membranes in bacteria, especially host cell invasion by apicomplexan parasites (Freeman 2008).

It has been a hot pursuit for a suitable candidate as target for E. tenella vaccine, especially proteins involved with host cell invasion. Among them the rhomboid protein might be a good choice. Experimental data from T. gondii have linked rhomboid protein with the invasion process. A spatially localized rhomboid protease cleaved cell surface adhesins that were essential for invasion by Toxoplasma (Brossier et al. 2005; Dowse et al. 2008). Rhomboid proteases preferentially cleaving different adhesions implicated in all invasive stages of malaria (Baker et al. 2006). PfROM4 cleaved a key adhesin erythrocyte-binding antigen 175 during malaria parasite invasion into erythrocytes (O'Donnell et al. 2006).

As for Eimeria rhomboid ETRH01, some typical features of rhomboid proteins were found in the sequence, such as seven transmembrane domains and a rhomboid domain which was a typical characteristic of the intramembranous rhomboid protease (Li et al. 2006). The recombinant protein of ETRH01 has been expressed in E. coli, fowlpox virus, M. bovis BCG, and yeast strain cdc25H (Li et al. 2006; Yang et al. 2008; Wang et al. 2009; Zheng et al. 2011). Immune responses and efficacy of recombinant proteins expressed in different hosts in protecting against homologous challenge have also been evaluated separately. These results suggest that the rhomboid was capable of eliciting humoral response and activating cell-mediated immunity and impart partial protection efficacy in birds.

In Eimeria infections, protective immunity is thought to rely on a strong cell-mediated response with antibodies supposedly playing a minor role. In the report, CD4+ and CD8+ T lymphocytes, IFN-γ and IL cytokines have also been evaluated. The numbers increase of CD4+ and CD8+ T lymphocytes of the chickens immunized with the recombinant rhomboid protein indicated that the recombinant rhomboid protein could activate cell-mediated immunity. Cytokines synthesized and secreted by leukocytes play important regulatory roles during the immune response to infection. In avian coccidiosis, it was clear that IFN-γ was produced by the host at sites of infection (Rothwell et al. 2000; Yun et al. 2000), and IFN-γ release has been used to screen for protective antigens against E. tenella infections (Breed et al. 1999). In this report, IFN-γ levels in the serum samples were significantly higher when compared with those of the PBS group (p < 0.05). Another important cytokine IL-2 level in the serum samples was also significantly higher when compared with those of the PBS group (p < 0.01). IL-2 was confirmed as an immunity-related cytokine enhancing protective immunity against coccidiosis (Lillehoj et al. 2005; Xu et al. 2008). However, under certain conditions, antibodies seem to be significant in protection in Eimeria infections (Constantinoiu et al. 2008). In the present study, specific IgG antibody response against E. tenella was generated in the chickens immunized with recombinant rhomboid protein expressed in E. coli. These results suggest that the rhomboid was capable of eliciting humoral response and activating cell-mediated immunity in birds.

The protection efficacies of recombinant rhomboid proteins expressed in different hosts in protecting against homologous challenge have also been evaluated separately. The rFPV-rhomboid elicited humoral immune response and stimulated proliferation of peripheral blood lymphocytes. Immunization with rFPV-rhomboid reduced oocyst shedding, resulting in a protection rate of 39.6%, 41.1%, or 41.7% at different doses of 102 PFU, 104 PFU, or 106 PFU of rFPV-rhomboid, respectively (Yang et al. 2008). Two rBCG strains carrying the E. tenella rhomboid gene delivered by extrachromosomal vector pMV261 and integrative vector pMV361 were evaluated for their ability to protect chickens against E. tenella challenge. Vaccination with rBCG pMV261-Rho produced a 56.04% protection rate (Wang et al. 2009). As shown from the results, the numbers of oocysts and cecal lesion were decreased significantly and the body weight also increased significantly in birds immunized with recombinant rhomboid protein when compared with the PBS group. The oocyst output indicated that the recombinant rhomboid protein gave a percentage protection of 77.3. From the present study, the recombinant E. tenella rhomboid protein could be a potential candidate for the development of a vaccine against E. tenella.