The six, 3 to 4-month-old masked palm civets were provided by the special breeding base in Shaoguan (Guangdong, China) and kept under controlled conditions in an animal facility in South China Agricultural University. Prior to T. gondii infection, animal sera were screened via ELISA for feline panleukopenia virus, feline immunodeficiency virus, and feline coronavirus. All serological tests showed negative results.

The animals were separated into the experimental group (n=5) and control group (n=1). T. gondii ME49 was acquired from the South China Agricultural University’s Parasitology Lab (Guangdong, China). Kunming mice were intraperitoneally injected with ME49 tachyzoites and 10 days later, brains were harvested for cyst preparation. The cysts were enumerated under the microscope using a hemacytometer and diluted to 10³ cysts per microliter in PBS. The experimental group was inoculated with 1mL of 10³ cysts via intraperitoneal injection, and the control group was inoculated with 1 mL of sterile PBS via the same infection route. A modified agglutination test (MAT) was performed to confirm T. gondii infection in the masked palm civets. After 90 days, brain tissues were collected aseptically from the experimental and control groups and stored at -80°C.

DNA was extracted from the collected brain tissues of masked palm civets, T. gondii infection was detected targeting B1 gene (194bp) and repeated sequence Rep-529 (529bp) using primers (F: 5’-TCTTTAAAGCGTTCGTGTC-3’, R: 5’-GGAACTGCATCCGTTCATGAG-3’) and (F: 5’-CGCTGCAGGGAGGAAGACGAAAGTTG-3’, R: 5’-CGCTGCAGACACAGTGCATCTGGATT-3’), respectively.

For protein extraction, 100mg of brain tissue was taken and grinded in liquid nitrogen. A RIPA lysis buffer was added (50mM Tris, 150 mM NaCl, 1% TritonX-100, 1%SDS) and incubated at 4°C for 30mins. The mixture was centrifuged at 4°C, at 15,000 rpm for 15min. Supernatant was taken and stored at -80°C.

One hundred milligram of brain tissue was pulverized in liquid nitrogen. Tissue proteins were then isolated using lysate buffer and radioimmunoprecipitation (RIPA). The suspension was sonicated 10 times at 20 watts and centrifuged at 12000 rpm for 20 min at 4°C.

One hundred microgram of protein was digested with trypsin (trypsin to protein ratio 1:100) at 37°C for 4 hours and centrifuged at 12000 rpm. After centrifugation, the protein was solubilized using 0.5 M tetraethylammonium bromide and labeled with 6-plex iTRAQ reagent. Robust cation exchange chromatography was analyzed using a Prominence Nexera UHPLC LC-30A System. Four milliliters of the iTRAQ-labeled peptide mixture were added to buffer A (25 mM NaH2PO4 in 25% ACN, pH 2.7) and transferred to a 4.6×250 mm Ultremex substantial cation exchange column (Phenomenex) and eluted with different concentrations. The elution flow rate was set at 2 mL/min. The elution process is monitored using 214 nm of UV light.

Three brain samples per animal (3 × 5 samples from the experimental animals and 3 samples from the control) were obtained and subjected to LC-MS/MS. The mixture of iTRAQ-labeled peptides was resuspended in buffer A (5% Acetonitrile, 0.1% Formic acid), and the concentration of the peptides was adjusted to 0.5 μg/μL. Approximately ten microliter of the resuspension was added via autosampler to a 2 cm UHPLC LC-30A System. The assay samples were loaded at 8 μL/min for 4min and then analyzed with a gradient concentration of buffer B (90% Acetonitrile, 0.2% Formic acid) at a rate of 250 NL/min for 35 min. The gradient elution program using buffer B was as follows: 0%-5% for 5min, 5%-15% for 10min, 15%-35% for 60min, 35%-80% for 45min. The peptide fragments were detected at 214 nm of absorbance during the elution process.

Data acquisition and analysis were performed with the TripleTOF 6600 system (AB SCIEX, Concord, ON) and the Nanospray III (AB SCIEX, Concord, ON). Peptide data was obtained with an iron spray voltage of 2.5 kV, a dry gas pressure of 30 psi, a spray gas pressure of 15 psi, and a surface temperature of 150°C. The mass spectrometry was performed at a resolution of over 30,000 FWHM with a time-of-flight mass spectrometry (TOFMS).

Differential protein analysis was performed by searching the Uniprot database with Mascot software. The quantification of differentially expressed proteins (DEPs) was calculated using the algorithm included in the Mascot software. The criteria for differential proteins screening were p-value < 0.05 and multiple of difference more than 1.5 times.

GO and KEGG analysis of differential proteins was performed using R language package (Goseq) and KO-BAS3.0 to understand the biological activities involving proteins with significant alteration of expression (Chloe et al., 2022). The software employs authoritative databases in bioinformatic research to create gene symbols, functional annotation of proteins, and pathway enrichment information.

We evaluated the mRNA levels of DEPs using a combined proteomic and transcriptomic approach. The transcriptome data is previously published and available on NCBI (NCBI accession no: PRJNA760987) (Yuan et al., 2022).

The iTRAQ data was validated using western blot. Four upregulated (REEP3, REEP4, TEP1, and EEPD1) and one downregulated (GRB2) DEPs were chosen for validation. The protein samples were separated on a 10% SDS-PAGE at 120 volt and transferred to a nylon membrane at 45 mA for 120 minutes. The proteins were incubated with primary antibodies including Rabbit anti-REEP3 (Invitrogen, 1:150), Rabbit anti-REEP4 (Invitrogen, 1:150), Rabbit anti-TEP1 (Invitrogen, 1:150), Rabbit anti-EEPD1 (Invitrogen, 1:150), Rabbit anti-GRB2 (Invitrogen, 1:150). HRP-conjugated goat anti-rabbit IgG was used as a secondary antibody at 1:500. The membrane was visualized using a diaminobenzidine (DAB) substrate solution, and the image was analyzed using western blot detection system.

In addition, immunohistochemistry was performed to confirm the iTRAQ data. Tissue was fixed in paraffin and sectioned to 4 μm for IHC examination. After incubation with methanol followed by 0.75% of hydrogen peroxide, sections were incubated with primary antibody at 4°C overnight, secondary antibody at 37°C for 40 min. Anti-body probing followed the same procedure as shown in western blotting.

TIMER 2.0 (http://timer.cistrome.org) is a web platform used for systemic analysis of immune infiltrates in various cancer types. ssGSEA (single-sample gene set enrichment analysis) calculates the enrichment score which represents the degree of up or downregulation of genes in a particular gene set. Taking advantage of the ssGSEA and TIMER, we investigated the association between the protein expression patterns of REEP3, REEP4, TEP1, and EEPD1 and immune infiltration in GBM.

The proteins related to RREP3, REEP4, TEP1 and EEPD1 were selected using the STRING (http://string-db.org) database, and the degree of correlation among the proteins was quantified by the Pearson’s correlation coefficient. We used Cytoscape software to build a protein PPI network for visualization. The connection (edge) between nodes represents the protein-protein interaction. The degree of protein correlation was sorted according to the Pearson’s correlation coefficient from large to small, and the top 100 proteins with the highest correlation with RREP3, REEP4, TEP1 and EEPD1 proteins were screened out.

To better understand the underlying mechanisms of carcinogenesis, The Cancer Genome Atlas (TCGA) database was utilized to annotate cancer-related genes and proteins, including mutations, copy number variants, mRNA expression(Tomczak et al., 2015).

We downloaded the RNA-seq data of 23 tumor projects from the TCGA database (https://portal.gdc.cancer.gov) and extracted the data in TPM format. The total number of samples collected was 11124 (Tumor =7260, Normal =3864). We used the R language package (ggplot2 [3.3.6], stats [4.2.1], car [3.1-0]) for statistical analysis and data visualization.

The RNA-Seq data (TPM) of 1846 GBM clinical cases was downloaded from the TCGA database. R language is used for data analysis and processing. Using GEPIA (Gene Expression Profiling Interactive Analysis) dataset (http://gepia.cancer-pku.cn/), we compared the transcript and protein levels in normal brain tissue and GBM tumor in 370 samples (Tang et al., 2017).

Sixty days post T. gondii infection, the experimental group of masked palm civets began to show clinical signs including loss of appetite and lethargy, the control group remained in good health. The clinical symptoms lasted for 30 days. Ninety days post infection, masked palm civets in experimental and control groups were euthanized and dissected. Brain tissues were removed and stored at -80°C. A 194 bp fragment of B1 gene and a 529bp repeated sequence gene were successfully amplified by PCR, confirming the presence of T. gondii tissue cysts in brain tissues.

We used the iTRAQ-based quantitative proteomic method to identify the proteins and their expression levels in the brain of civet cats infected with T. gondii. Of the 377,063 total spectra, 252,916 were identified, among which 14,497 were matched to existing peptides in the database and 5447 were unique peptides ( Supplementary Figure 1A ). A total of 268 DEPs were detected and screened in the brain tissues of masked palm civets, of which 111 were up-regulated, and 157 were down-regulated ( Supplementary Figure 1B ). The differential proteins volcano plot ( Supplementary Figure 1C ) showed the upregulated and downregulated DEPs. As shown in SDS-PAGE, differences in the amount and variety of total, isolated proteins were not observed ( Supplementary Figure 1D ).

To identify the biological functions and regulatory pathways the differentially expressed proteins are enriched in, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were carried out using R language package (Goseq) and KO-BAS3.0.

GO analysis was based on three categories, namely biological process, cellular component, and molecular function ( Figure 1A ). The 268 DEPs were mapped to 30 pathways via KEGG enrichment analysis ( Figure 1B ). In addition, The GO ( Supplementary Figure 2 ) and pathway ( Supplementary Figure 3 ) analysis were conducted separately on the 111 upregulated and 157 downregulated DEPs.

These include immune regulatory and antitumor pathways, namely, NF-Kappa B signaling pathway, PPAR signaling pathway, Th17 cell differentiation, chemokine signaling pathway, TNF signaling pathway, Notch signaling pathway (Bazzoni and Bentivegna, 2019), MAPK signaling pathway(Lee et al., 2020), p53 signaling pathway (Marei et al., 2021), PI3K-Akt signaling pathway(Tewari et al., 2022), T cell receptor signaling pathway, Toll-like receptor signaling pathway.

Based on log2 fold change, the top five of the upregulated (REEP3, REEP4, TEP1, EEPD1, MYD88) and downregulated (CTLA4, PTPRC, CXCR4, CX3CR1, and GRB2) proteins were compiled in Table 1 .

To validate the iTRAQ data, we performed western blot. The top four most significantly upregulated proteins (REEP3, REEP4, TEP1 and EEPD1) and one significantly downregulated protein (GRB2) were selected ( Table 1 ). As shown in Figure 2 , REEP3, REEP4, TEP1, and EEPD1 were upregulated, while GRB2 was downregulated, consistent with the iTRAQ data.

To understand the expression of different proteins in the brain of masked palm civet infected with T. gondii at the transcriptional and protein levels, we performed proteomics and transcriptomics. Among the 268 differentially expressed proteins and 2808 differentially expressed genes, twenty-four showed consistent results ( Figure 3A ). The regulation patterns of these 24 genes were shown in Figure 3B . Despite variations in the expression amount, the two omics showed consistent expression patterns of these genes ( Figure 3B ). Noteworthily, REEP3, REEP4, TEP1, and EEPD1 were upregulated in both methods.

The twenty-four genes of interest were therefore subjected to functional annotation in categories of biological process, cellular component, and molecular function ( Figure 3C ). Gene Ontology analysis showed that in the biological process category, these 24 genes were involved in signal transduction, B cell activation, response to tumor necrosis factor, and other similar processes. These genes were enriched in cellular components such as PCNA-p21 complex, Toll-like receptor 1, and cyclin D1-CDK4 complex ( Figure 3C ). In terms of molecular function, these genes were involved in Schmidt-Lanterman incisure, MAP kinase activity, Toll-like receptor binding, among others ( Figure 3C ).

KEGG pathway enrichment analysis showed that the 24 genes were involved in pathways such as small-cell lung cancer, breast cancer, Toll-like receptor signaling pathway ( Figure 3D ). In addition to the cancer pathways identified, the 24 genes were mapped to the following signaling pathways, namely PI3K-Akt, Wnt, P53, Notch, and NF - Kappa B signaling pathway which were shown to be the critical pathways (Hellmark and Segelmark, 2014; Xie et al., 2019).

We used immunohistochemistry to verify the expression levels of REEP3, REEP4, TEP1, and EEPD1 in the brain of masked palm civets infected with T. gondii. The results showed that the expression of REEP3, REEP4, TEP1, and EEPD1 proteins were higher in T. gondii infected brain tissue than normal tissue ( Figure 4 ). This finding confirmed that T. gondii infection altered the protein expression of REEP3, REEP4, TEP1, and EEPD1 in the brain of masked palm civets.

We used immune infiltration analysis to identify respective immune cells that were associated with the expression regulation of REEP3, REEP4, TEP1, and EEPD1. Data was analyzed using TIMER 2.0 databases (Li et al., 2017; Ru et al., 2019). The immune cells that were associated with the up and down regulation of these four proteins were shown in Figure 5 . The upregulation of REEP3 was associated with T helper cell, T gamma delta (Tgd) cell, and T central memory (Tcm) cell. The downregulation of REEP3 was linked to CD56bright NK cell, Mast cell, and T follicular helper (Tfh) cell. The top three immune cells that were associated with REEP4 upregulation included macrophage, neutrophil, and eosinophil. In the case of TEP1, the T helper cell, macrophage, and Tgd were as-sociated with its upregulation. Tgd cell, plasmacytoid DC (pDC) cell, and CD8 T cell were related to the upregulation of EEPD1. Our finding suggests that the elevated expression of REEP3, REEP4, TEP1, and EEPD1 modulates immune cell infiltration in the brain microenvironment.

To better understand the regulatory mechanism of GBM, we examined the neighboring proteins that were closely associated with REEP3, REEP4, TEP1 and EEPD1 in the protein networks. As shown in Supplementary Figure 4 , the STRING database was utilized to screen the respective, top 100 proteins with close relation to REEP3, REEP4, TEP1, and EEPD1. The four groups of related proteins were analyzed collectively and a total of 44 proteins were found to be linked to all four proteins ( Figure 6A ). GO annotation provided functional annotation for the 44 proteins regarding biological processes, cellular components, and molecular functions ( Figure 6B ). The 44 proteins were mapped to the biological processes including cell division, chromatid segregation, and cell cycle phase transition. These proteins were implicated in cellular components such as spindle, microtubule, and kinetochore, and molecular function of receptor ligand activity, microtubule binding, and tubulin binding.

KEGG enrichment analysis identified the pathways involving the 44 proteins. As shown in Figure 6C , the majority of these proteins were significantly enriched in immunoregulation and cancer pathways. These included PI3K-Akt signaling pathway, cytokine-cytokine receptor interaction, Wnt signaling pathway, AMPK signaling pathway, breast cancer, prostate cancer, gastric cancer. This confirmed that REEP3, REEP4, TEP1, and EEPD1 and their network proteins were involved in immunoregulatory functions and cancer development.

To evaluate REEP3, REEP4, TEP1 and EEPD1 mRNA levels in multiple tumor and normal samples, we downloaded and compiled data from 11124 samples (Tumor =7260, Normal =3864) from the TCGA database. As shown in Figure 7 , REEP3, REEP4, TEP1 and EEPD1 was differentially expressed in 14, 17, 14, 13 tumor types, respectively. REEP4 was upregulated in all 17 types of tumor as compared to normal tissue while the expression level of the rest of the efforts is tumor-specific. All four effectors were upregulated in CHOL (cholangiocarcinoma), GBM (glioblastoma), LIHC (liver hepatocellular carcinoma), and STAD (stomach adenocarcinoma).

Taking advantage of Gene Expression Profiling Interactive Analysis (GEPIA) and The Cancer Genome Atlas Program (TCGA), we compared mRNA expression of REEP3, REEP4, TEP1, and EEPD1 between GBM tumor and normal brain tissue ( Supplementary Figure 5A ). The result showed that the expression of REEP3, REEP4, TEP1, and EEPD1 was significantly higher in GBM tumor tissue ( Supplementary Figure 5A ).

Next, we obtained 1846 GBM samples from the TCGA database. Of which, 1157 samples were from normal brain tissues and 689 were from GBM tumor tissues. As shown in ( Supplementary Figure 5B ), the REEP3, REEP4, TEP1, and EEPD1 had significantly higher expression in GBM tumor tissues compared to normal tissues.