All protocols were reviewed and approved by the Animal Research Ethics Committee of Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences. The experiments were performed in a strict accordance with the Animal Ethics Procedures and Guidelines of the People's Republic of China. All efforts were made to minimize suffering of the animals.

Six to eight week-old female BALB/c mice and 10 week-old female kitten were obtained from Laboratory Animal Center of Lanzhou Veterinary Research Institute. Type II Prugniuad (PRU) strain of T. gondii was maintained in our laboratory by passaging the cysts in mice as described previously (Zhou et al., 2016).

Proteins were extracted from at least two biological repeats of each of the life-cycle stages, namely tachyzoites (~10⁸), cysts (~10⁷) and oocysts (~10⁷). Briefly, each biological sample was homogenized with 5 volumes glass sand and lysed in 200 μl radioimmunoprecipitation assay (RIPA) buffer (20 mM Tris-HCl pH 7.5, 150 mM sodium chloride, 2 mM EDTA, 1% DOC, 1% Triton X-100) containing phenylmethylsulfonyl fluoride (PMSF). Then, samples were sonicated (2%, 1 s ON and 1 s OFF cycle, 5 times, 8 repeats) and the supernatant containing total soluble proteins was collected after 20-min centrifugation at 13,400 g. The concentration of the protein was determined by the protein quantitative kit (QuantiProTM BCA Assay Kit, Sigma).

For protein digestion, 100 μg protein of each sample was reduced, alkylated, and then precipitated by the methanol/chloroform precipitation method. Firstly, 50 mM Reducing Reagent (final concentration, 8-plex iTRAQ kit, AB Sciex, USA) were added into each sample followed by incubation at 60°C for 1 hr. Then, the sample was mixed with 200 mM Cysteine-Blocking Reagent (final concentration, 8-plex iTRAQ kit, AB Sciex, USA) and held at ambient temperature for 10 min. Following rinsing the 10 KD ultrafiltration cartridge with 70% ethanol and deionized water, the protein solution was poured into ultrafiltration cartridge and centrifuged at 13,400 g for 20 min. Solution at the bottom of ultrafiltration cartridge was discarded and the ultrafiltration cartridge was centrifuged with 100 μl 0.25M TEAB (triethyl ammonium bicarbonate) three times at 13,400 g for 20 min. The protein pellets were reconstituted in 50 μl of 6 M urea/50 mM TEAB with sonication (Possenti et al., 2013) and digested in 2% trypsin overnight (Promega). Then, the digested peptides were dried and reconstituted in 0.5 M TEAB. Then, peptides were labeled according to the instructions of 8-plex iTRAQ kit (Applied Biosystems/MDS Sciex, Foster City, CA). Label reagent mixed with 150 μl isopropyl alcohol was mixed with samples and incubated for 2 hr at ambient temperature. The reaction was terminated with deionized water and samples were thoroughly mixed. Following vacuum centrifugation, dried samples were preserved at −80°C until use. Proteins of sporulated oocysts were labeled with 115 and 116, whereas the proteins of the cysts and tachyzoites were labeled with 117, 118, and 119, 121, respectively.

Strong cation exchange (SCX) fractionation chromatography was performed using the high performance liquid chromatography (HPLC) system (Phenomenex columns; Gemini-NX 3u C18 110A; 150^*2.00 mM). The iTRAQ-labeled peptides were separated by a linear gradient formed by mobile phase A (20 mM HCOONH4, pH 10) and mobile phase B (20 mM HCOONH4, 80% ACN, pH 10). The flow of peptides elution was set to a constant rate of 200 μl/min. A total of 24 fractions were collected by a linear gradient (1 collection every 1 min, 100 min) and were acidified with trifluoroacetic acid (50%). The fractions were vacuum-dried for further analysis with LC-MS. Fractions (96 μg) were dissolved in buffer A [0.1% formic acid, 2% acetonitrile (ACN)] and pelleted at 13,447 g for 20 min. The supernatant was loaded onto analytical columns and was identified with an online Q Exactive system (Thermal Scientific). Component of mobile phase A and mobile phase B of LC-MS was formic acid (0.1%) and 80% ACN containing 0.1% formic acid, respectively. The flow rate of analytical columns was set at 350 nl/min and the peptides were analyzed within 65 min 3-step gradient (80% ACN in 0.1% formic acid from 4 to 50% over 45 min, 50 to 90% over 5 min and kept at 90% for 5 min). The parameters of the first grade MS included scan ranges from 350 to 1,800 m/z at a resolution of 70,000 with a maximum injection time of 40 ms. The second grade MS spectra were acquired in a resolution of 17,500 with 60 ms maximum injection time and the 20 top precursors for each MS cycle were selected.

The raw MS data was transformed into mascot generic format (.mgf) files with Proteome Discoverer™ 1.4 and the data file was used to query T. gondii ME49 strain database (http://www.toxodb.org/common/downloads/release-10.0/TgondiiME49/fasta/data/), which contains 8322 protein sequences. The ProteinPilot™ Software 4.5 (AB SCIEX) was used to further identify and quantity proteins. To filter the results, we employed false discovery rate (FDR) of less than 0.01 for identification and the confidence level of 95% or unused confidence score larger than 1.3 for quantification. For differentially expressed proteins (DEPs), those with |log₂ fold change|>1 were deemed upregulated or downregulated proteins, respectively.

Functional classification of the DEPs was performed according to Gene Ontology (GO) annotation and enrichment (http://www.geneontology.org) analysis. The DEPs were classified into three categories, namely molecular function, biological process and cellular component. The Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://www.kegg.jp/kegg/) was used to predict molecular function, biological processes and significant DEPs pathways. Protein-protein interactions (PPI) were predicted in the Search Tool of the Retrieval of Interaction Genes/Proteins (STRING) database and the interaction network was illustrated by Cytoscape software. The PPI whose combined score was >0.9 were subjected to further interaction network analysis.

iTRAQ was used to identify the proteomic differences among life-cycle stages of T. gondii, namely sporulated oocyst, tachyzoite, bradyzoites-containing cyst. Representative SCX chromatogram of the results is shown in Figure S1. A total of 6,285 proteins were identified from 53,335 peptides, which were matched with 76,534 spectra at a false discovery rate of 1%. Multiple proteins were identified, among which, 1667, 1697, and 1791, proteins were identified in O vs. T (oocysts vs. tachyzoites), T vs. C (tachyzoites vs. cysts) and C vs. O (cysts vs. oocysts), respectively after filtration (CV ≤ 0.5) (Tables S1–S3). Using 99% confidence level (CV≤0.5), 1602, 1625, and 1719 proteins were identified in O vs. T, T vs. C, and C vs. O, respectively (Tables S4–S6). Compared with analysis using 95% confidence level, 65 proteins, 72 proteins and 72 proteins were lost in O vs. T, T vs. C, and C vs. O. Repeatability analysis based on the Coefficient of variation (CV) showed that 79.3, 81.8, and 90.7% proteins can be covered in the total identified proteins in O vs. T, T vs. C, and C vs. O, respectively, when the CV was less than 0.5 (Figure 1).

In brief, 875 proteins, 656 proteins, and 538 proteins were defined as differentially expressed protein (DEPs) in O vs. T, T vs. C and C vs. O, respectively (|log₂ fold change|>1 and P < 0.05) (Tables S7–S9). Among DEPs in each group, 801 proteins and 74 proteins were upregulated or downregulated in oocyst compared with tachyzoite stage. There were 146 upregulated proteins and 510 downregulated proteins in T vs. C. Further, 180 proteins and 358 proteins had higher or lower expression level in cyst than in oocyst. Numbers of DEPs in different level are presented in Figure 2. Results of the hierarchical clustering analysis of the identified proteins are shown in Figure 3.

The DEPs were subjected to functional classification by Gene Ontology (GO) analysis. We identified 3371, 2846, and 1998 GO terms in O vs. T, T vs. C and C vs. O, respectively. Among these GO terms, there were 1708 molecular function terms, 874 biological process terms and 789 cellular component terms in O vs. T. GO terms in T vs. C included 1392 molecular function terms, 832 biological process terms, 622 cellular component terms and GO terms in C vs. O included 1083 molecular function terms, 634 biological process terms and 281 cellular component terms. Those GO terms were classified and enriched in order to investigate the properties of upregulated and downregulated proteins in each group. The most 40 enriched GO terms in each group and their destination to three main GO categories are presented in Figure 4.

As is shown in Figure 4A, for O vs. T, the top five enriched GO terms for upregulated proteins within molecular function were ion binding, oxidoreductase activity, kinase activity, peptidase activity and structural constituent of ribosome. The top five enriched GO terms for downregulated proteins were kinase activity, DNA binding, helicase activity, RNA binding, and oxidoreductase activity. Translation, small molecule metabolic process, catabolic process, cellular nitrogen compound metabolic process and transport were the five top terms for upregulated proteins under biological process. The top five enriched GO terms for downregulated proteins in biological process were cellular nitrogen compound metabolic process, biosynthetic process, small molecule metabolic process, cellular protein modification process, and DNA metabolic process. The most abundant GO terms for upregulated proteins under cellular component were intracellular, followed by protein complex, cytoplasm, ribosome and nucleus. Protein complex, nucleus, chromosome, cytoplasm and intracellular were the five top terms for downregulated proteins.

With regard to GO enrichment for upregulated proteins in T vs. C, ion binding, biosynthetic process, extracellular region were the most enriched terms in molecular function, biological process and cellular component respectively. For downregulated proteins, ion binding, structural constituent of ribosome, RNA binding were the top three terms under molecular function and translation, cellular nitrogen compound metabolic process, transport were the most three enriched terms under biological process. The top five terms within cellular component included intracellular, cytoplasm, protein complex, ribosome, nucleus (Figure 4B).

Among molecular function terms in C vs. O, ion binding, oxidoreductase activity, ligase activity, kinase activity and DNA binding were the mostly prevalent terms for upregulated proteins. The most five prominent terms for downregulated proteins were ion binding, oxidoreductase activity, kinase activity, peptidase activity, and lyase activity. Biosynthetic process and small molecule metabolic process occupied top one position in biological process for upregulated proteins and downregulated proteins respectively. The mostly enrichment terms within cellular component for upregulated proteins or downregulated proteins were cytoplasm and intracellular respectively (Figure 4C).

To identify the biological pathways operating during T. gondii development, we mapped the DEPs in each group to reference pathways contained in the KEGG pathway database. Among DEPs identified in each group, 401 DEPs, 654 DEPs, 536 DEPs had a KEGG Orthology (KO) ID and could be mapped to 55 pathways, 55 pathways, 61 pathways in O vs. T, T vs. C, and C vs. O, respectively. Of the 55 enriched pathways in group O vs. T, 42 pathways were upregulated and no pathway was downregulated. The top 5 enriched upregulated pathways were metabolic pathways, biosynthesis of secondary metabolites, biosynthesis of antibiotics, carbon metabolism, and ribosome (Figure 5A).

There were 5 upregulated pathways and 13 downregulated pathways in T vs. C. Ribosome, fatty acid metabolism, biosynthesis of unsaturated fatty acids, alpha-Linolenic acid metabolism, galactose metabolism were upregulated pathways and the top 5 enriched downregulated pathways included metabolic pathways, carbon metabolism, aminoacyl-tRNA biosynthesis, RNA transport, and protein processing in endoplasmic reticulum (Figure 5B).

In C vs. O, 2 and 25 pathways were up- and down-regulated, respectively. The up-regulated pathways included aminoacyl-tRNA biosynthesis, pyruvate metabolism and the five most enriched downregulated pathways were metabolic pathways, biosynthesis of secondary metabolites, biosynthesis of antibiotics, carbon metabolism, and biosynthesis of amino acids (Figure 5C).

To study the expression pattern of virulence factors across the life-cycle stages of T. gondii, the expression levels of 23 virulence factors, which simultaneously appear in all groups were analyzed. None of these virulence factors were downregulated and 20 factors were upregulated in oocyst compared with tachyzoite except RON4, PHIL1, and RON5 (Figure 6A). In T vs. C, while five virulence factors (SPATR, PP2C, VP1, ROP5, AMA1) were upregulated in tachyzoite vs. cyst, the number of downregulated virulence factors was 10 (RON5, ROP16, RON2, GRA4, MIC3, ROP2A, GRA1, GRA6, GRA7, MIC6) (Figure 6B). The outcome of comparison between C and O was similar to O vs. T. No virulence factor exhibited higher expression level in cyst and 11 virulence factors were upregulated in the oocysts (GRA7, ROM4, MIC3, MIC6, PP2C, ROP18, VP1, AMA1, MIC2, SPATR, ROP5) (Figure 6C).

The differences of ribosomal protein expression among the different life-cycle stages showed upregulation of 33 and 46 ribosomal proteins in oocyst and in cyst, compared to tachyzoite. Six ribosomal proteins had higher expression level in cysts compared with oocyst (Table 1).

The protein-protein interactions (PPI) whose combined score was >0.9 were used to build network using Cytoscape tool in each group. As shown in Figures 7, 8, there were 310 nodes and 1501 edges in the PPI network of O vs. T, and 273 nodes and 1769 edges in the PPI network of T vs. C. The PPI network of C vs. O contained 152 nodes and 308 edges (Figure 9).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.