E. granulosus vesicles and protoscolices were isolated from the livers of sheep naturally-infected with CE, which were obtained from a local commercial slaughterhouse. Two pathogen-free female New Zealand rabbits used for anti-Egp38 antibody preparation were purchased from the animal center of Xinjiang Medical University, housed in cages with a 12-hr light/dark cycle, and provided normal rabbit feed and water. All animals received humane care in compliance with the guidelines of the Medical Research Center, and animal procedures were approved by the Animal Care and User Committee and the Ethical Committee of the First Affiliated Hospital of Xinjiang Medical University, China (20081205-2).

Total RNA was isolated from E. granulosus protoscolices using TRIzol reagent (Invitrogen Co., Beverly, Massachusetts, USA) according to the manufacturer’s instructions. The cDNA was then synthesized from the purified RNA according to the instructions included with the reverse transcription kit (Invitrogen) and used as a template to amplify the Egp38 cDNA with the primers p38-5EC (5′-GTGCCCGATGTAAATGAGCG-3′) and p38-3EC (5′-CGCGTTGATTGGCGAGTAC-3′) [21]. The following thermal cycling conditions were used: 35 cycles of 30 sec at 94°C, 1 min at 53°C, and 1 min at 72°C. The amplified fragment was cloned into the pMD18-T vector with a TA cloning kit (Takara Co., Dalian, China) and verified by sequencing (Shanghai Sangon Co., Shanghai, China).

Amino acid sequences of the p38 MAPK homologous genes in E. multilocularis, Homo sapiens, Drosophila melanogaster, Caenorhabditis elegans, and Saccharomyces cerevisiae were downloaded from GenBank (https://www.ncbi.nlm.nih.gov/genbank/) and compared using DNAMAN 5.0. Phylogenetic trees were constructed with the software tool MegAlign for multiple sequence alignment. Catalytic domains were predicted with a simple modular architecture research tool (SMART; http://smart.embl-heidel-berg.de) [23]. GraphPad Prism software (GraphPad Software) version 5 was used for statistical analyses.

The 3 dimensional-structures of the Egp38 proteins were predicted by submitting the amino acid sequences to the SwissModel Automatic Modeling server (http://Swissmodel.expasy.org) [24]. The Egp38 homology model was calculated based on the crystal structure of the Cryptosporidium Ser/Thr protein kinase template (3eb0A). The EmMPK2 homology model was calculated based on the crystal structure of the human tau Ser/Thr protein kinase template (1j1bB). Structural comparisons and figure generation were carried out using DeepView/Swiss-Pdbviewer 4.01 [25].

Equal amounts of E. granulosus vesicles and protoscolices cDNA were employed for qRT-PCR to determine the expression of Egp38 using the primers p38-5q (5′-ACTCAACTACACCAGCAAGAAG-3′) and p38-3q (5′-TCCTACAGGCTCATCACTCTC-3′) or elp using the primers elp-5q (5′-CGGGAGGGTAATGACAAG-3′) and elp-3q (5′-GGAAGTAACCGTAGTGATGGAG-3′). The following thermal cycling conditions were employed: initiation with a 10 min denaturation at 95°C, followed by 40 cycles of 30 sec at 94°C and 60 sec at 60°C. The relative gene expression of Egp38 was calculated via the 2^−ΔΔCT method [26]; elp was used as a housekeeping gene [27]. In all cases, each PCR trial was performed in triplicate and repeated at least 3 times.

Egp38 was expressed in E. coli using the pET-28a system. Briefly, the entire Egp38 open reading frame (ORF) was PCR-amplified from the plasmid pMD18-T-Egp38 using the primers p38-F-28a (5′-GGAATTCATGCCCGATGTAAATGA-3′) and p38-R-28a (5′-GTCGACCGTTGATTGGCGAGTA-3′), employing the conditions described above, and cloned into the pET-28a vector with EcoR I and Sal I, resulting in an N-terminal in-frame fusion with a hex-histidine tag. The production of recombinant proteins after induction with 0.1 mM IPTG was carried out according to the pET user manual. The recombinant Egp38 proteins were purified from E. coli lysates under native conditions via affinity chromatography against the hex-histidine tag. After collecting recombinant E. coli by centrifugation (6,000 g, 20 min, 4°C) from 2 L of culture, the pellet was resuspended in 20 ml of lysis buffer (300 mM NaCl, 50 mM sodium-phosphate buffer at pH 8.0, 10 mM imidazole). Cell debris was removed by centrifugation, and the supernatant was transferred to 4 ml of a nickel bead slurry and equilibrated with lysis buffer. After overnight rotation at 4°C, beads were washed 3 times with 300 ml of washing buffer (300 mM NaCl, 50 mM sodium-phosphate buffer at pH 8.0, 20 mM imidazole). Recombinant protein was then extracted with 6 ml elution buffer I (300 mM NaCl, 50 mM sodium-phosphate buffer at pH 8.0, 250 mM imidazole) and 6 ml elution buffer II (300 mM NaCl, 50 mM sodium-phosphate buffer at pH 8.0, 500 mM imidazole). Protein-containing fractions were pooled and dialyzed overnight against PBS at 4°C.

To detect Egp38 in Echinococcus larval stages, we generated an anti-Egp38 antibody in rabbits by carrying out 4 rounds of immunization with recombinant Egp38. As the pTGpY phosphorylation motif is highly conserved in Egp38, a commercially available antibody for the phosphorylated (pTGpY) form of p38 MAPK (Promega, Mannheim, Germany) was used to detect phosphorylated Egp38. Rabbit anti-β-actin (Santa Cruz Biotechnology, Santa Cruz, California, USA) was used to detect β-actin of E. granulosus. For western blotting, vesicles and protoscolices lysates from CE sheep were produced, separated on 12% acrylamide gels, and electro-transferred to nitrocellulose membranes. Detection was performed with the above-mentioned antibodies according to the manufacturer’s instructions, using an anti-rabbit (whole molecule) IgG-peroxidase antibody as the secondary antibody (Sigma-Aldrich, Taufkirchen, Germany). For immunohistochemistry, vesicles and protoscolices from infected CE sheep were isolated and embedded in paraffin for sectioning and antibody staining. Immunohistochemistry was then performed with the phospho-specific Egp38 antibody (see above) according to a previously established protocol [28].

Freshly isolated E. granulosus protoscolices were washed in sterile PBS (Gibco-BRL, Eggenstein, Germany) and incubated in 96-well plates (approximately 400 protoscolices per well) in the presence of culture medium (89% Dulbecco’s minimal essential medium, 10% fetal calf serum, 1% penicillin-streptomycin solution). ML3403 was dissolved in dimethylsulphoxide (DMSO). Negative control groups did not contain DMSO or ML3403 (Calbiochem, Gibbstown, New Jersey, USA). The DMSO group which had 2 μl of DMSO per well is solvent control. The 7.5 μM, 15 μM, 30 μM, and 60 μM ML3403 groups contained 2 μl of the appropriate concentration of ML3403 per well to achieve the desired final concentrations. At each time point (0, 1, 2, 3, 4, and 5 days), 3 wells for each group were stained with eosin, and the viability was immediately assessed by counting under a microscope. Red stained protoscolices were dead, and protoscolies lacking color staining were alive. The survival percentage was calculated as the number of live protoscolices divided by the total number of live and dead protoscolices. All experiments were performed independently in quadruplicate.

In addition, freshly prepared protoscolices were treated with 60 μM ML3403 for 4 hr in vitro, to detect native or phosphorylated Egp38 by western blotting.

Freshly prepared protoscolices were acclimated via cultivation in culture medium (99% Dulbecco’s minimal essential medium, 1% penicillin-streptomycin solution) for 7 days, then treated with 1 nM TGF-β1 (R&D Systems, Minneapolis, Minneapolis, USA) [29] for 24 hr. At each time point (0 hr, 4 hr, 8 hr, and 24 hr), protoscolices were removed to detect native or phosphorylated Egp38 by western blotting.

The entire Egp38 ORF sequence comprised 1,107 bp and 368 codons (excluding the TGA stop codon). As shown in Fig. 1, the encoded protein (368 amino acids, 42 kDa) exhibited significant homology with p38 proteins in mammals, insects, nematodes, yeasts, and plants. The p38 homologue EmMPK2 from E. multilocularis demonstrated the highest level of homology (98.6% identity). Lower levels (43.0–61.9% identity) of homology were shared with p38 homologues from a wide variety of phylogenetically distinct animals, including mammals, insects, and nematodes. Thus, we designated the corresponding E. granulosus cDNA and the encoded 42 kDa protein as Egp38.

Based on EmMPK2 sequence analysis [21], the amino acid sequence of Egp38 exhibits many characteristics of p38 homologues. A Ser/Thr protein kinase catalytic domain between residues Tyr²⁵ and Phe³⁰⁹ of Egp38 was identified with the SMART software package [23]. Many conserved amino acid residues in eukaryotic protein kinases were also observed in EgP38 and at corresponding positions (Fig. 1). Furthermore, Egp38 contains a T-G-Y dual phosphorylation site in the activation loop of the kinase domain.

We investigated the phylogenetic relatedness of p38 homologue sequences. The constructed evolutionary tree revealed the close identity of p38 in E. granulosus and E. multilocularis to orthologs in Schistosoma japonicum and yeast, but divergence from p38 in nematodes, filarial parasites, amphibians, and mammals (Fig. 2).

The models for Egp38, EmMPK2, and p38α were very similar (Fig. 3A–C), particularly in terms of the catalytic domain, as illustrated by color-coding. Only minimal differences were observed between the N-terminal regions of Egp38 (Pro²²-Met¹¹⁰) and EmMPK2 (Asp²³-Met¹¹⁰). In contrast, the N- and C-terminal regions of p38α (Fig. 3C) were very different compared to Egp38 and EmMPK2; in particular, p38α included 2 longer helices (colored sky blue and red).

As shown in Fig. 4, there were no significant differences between the vesicles and protoscolices of E. granulosus, indicating the constitutive expression of Egp38 cDNA, during the larval stages. An anti-Egp38 antibody and an antibody for the phosphorylated (pTGpY) form of p38 MAPK were employed to detect Egp38 and its active form, respectively. Egp38 and its active form were both identified in Echinococcus vesicles and protoscolices lysates, and the proteins migrated at approximately 45 kDa.

As shown in Fig. 6A and D, both the germinal layer of the vesicle and the defined structures of the protoscolex reacted with rabbit-anti Egp38 immune sera, which detected total Egp38. In the vesicles, strong staining was observed in the cytoplasm and nuclei of the germinal layer cells. In the protoscolices, staining was present in the tissue surrounding the rostellum and in the suckers. As shown in Fig. 6B and E, localization of phosphorylated Egp38 was similar to that of the total Egp38 in the germinal layer of the vesicle and protoscolex. However, phosphorylated Egp38 was localized in the cytoplasm and nuclei of certain germinal layer cells in the vesicles.

Taken together, the western blotting and immunohistochemistry findings from parasite lysates indicated that Egp38 was constitutively present throughout the Echinococcus larval development stages.

To test the effects of p38 MAPK inhibitors on E. granulosus larvae, we employed ML3403 and examined the phosphorylation status of Egp38. After treatment with 60 μM ML3403 for 4 hr, the phospho-Egp38 band nearly disappeared, while the phospho-Egp38 band in the control group, with or without DMSO, was clearly present (Fig. 7B); thus, the p38 MAPK inhibitor effectively blocked Egp38 phosphorylation activity in the parasite.

On the 3rd day, the viability of the protoscolices treated with 30 and 60 μM ML3404 (Fig. 7A) began to decline. On the 5th day, the protoscolices viability of the 60 μM ML3404 treatment group declined to only 7.3% of the total (Fig. 7A). However, protoscolices survival in the normal group, the DMSO group, the 7.5 μM ML3404 group and the 15 μM ML3404 group remained unaffected (Fig. 7A, Supplementary Fig. S1).

To investigate whether human TGF-β1 exerts effects via Egp38, we treated E. granulosus protoscolices with human TGF- β1 for 24 hr in vitro. Human TGF-β1 activated the phosphorylation of Egp38 over time. At 24 hr, the phosphorylation levels of Egp38 were increased by approximately 4-folds (Fig. 8).