Data Source and Online Analysis Tool
We acquired samples with sufficient clinical information from the TCGA database (https://portal.gdc.cancer.gov/). We also downloaded data samples from the GSE10810 and GSE86166 datasets of the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). R(3.6.3 version) was utilized to process the original data. Other online tools used comprise TISIDB (http://cis.hku.hk/TISIDB/index.php), Kaplan-Meier Plotter (http://kmplot.com/analysis/), UALCAN (http://ualcan.path.uab.edu/), GEPIA2 (http://gepia2.cancer-pku.cn/), MethSurv (https://biit.cs.ut.ee/methsurv/), and LinkedOmics (http://www.linkedomics.org/login.php).
Differentially Expression
We downloaded data from the TCGA database and proceeded with Logistics regression analysis of the correlation between PIGR and clinicopathological characteristics in BRCA. The Wilcoxon rank sum test was adopted to compare different PIGR expression levels between tumors and corresponding normal tissues across 23 human malignancies. Then we analyzed the expression of PIGR in the TCGA-BRCA dataset. We further verified the results in the GSE10810 dataset. An ROC curve was generated based on the data from the TCGA-BRCA dataset to assess the accuracy in identifying BRCA tissues.
Pathological Samples Collection and IHC Staining
We collected 18 samples of paraffin-embedded BRCA tissues and corresponding paracancerous normal tissues from the Pathology Department of Qilu Hospital (Qingdao). All informed consents have been obtained from patients or families. Following the procedures explained in the instructions, immunohistochemistry (IHC) staining was performed. Tissues were fixed in 10% formalin, paraffin-embedded, 4~6 μm thinly sliced, and spread out on slides. After the deparaffinization, rehydration, and microwave antigen retrieval, slides were incubated with Anti-PIGR (Atlas Antibodies, Rabbit #HPA012012) antibody at 1:500 dilution at 4°C overnight. The slides were incubated with secondary antibody at room temperature for 30 min and stained with DAB substrate, followed by hematoxylin counterstaining.
MCF7 BRCA Cell Culture and Transfection
We observed theproliferation, clone formation, and migration of MCF7 BRCA cells with or without down-regulated PIGR expression. MCF7 cell lines were derived from the preservation in the laboratory. We cultured the MCF7 cell lines in Dulbecco's Modified Eagle's medium containing 10% fetal bovine serum and 1% penicillin and streptomycin. shRNA was purchased from Genechem (Shanghai, China) with the sequence as follows: Negative Control, NC: (5'-TTCTCCGAACGTGTCACGT-3'), shPIGR: (5 '- CCCAACTATACAGGAAGAATA - 3'). Anti-PIGR antibodies were purchased from Atlas Antibodies (HPA012012). Transfection of shRNA was performed according to the instructions of Genechem (Shanghai, China).
CCK8 (Cell Counting Kit 8) Proliferation Assay and Clone Formation Assay
We used western blotting to examine the MCF7 cell lines constructed. Then we seeded cells in 96-well plates following 4×103 cells/well concentrations. We repeated the seeding three times as a group and added 10ul CCK8 reagent (CK04, Dojindo CO. Ltd., Kumamoto, Japan) to each well every 24 hours. After incubating for 1-4 hours, the absorbance at OD450nm was measured by a microplate reader. To study cell clone formation, we seeded MCF7 cell lines in six-well plates following 1×103 cells/well concentrations and cultured them for two weeks. The cells were washed three times with PBS at the end of the culture. Then the cells were fixed, stained, dried, and counted in succession.
Transwell Migration Assay
We performed the transwell migration assay using a transwell package containing 24-well plates, 6.5mm diameter inserts, and 8.0μm pore size polycarbonate membrane (Corning Incorporated costar, Kennebunk, ME). Following the guidelines for use, 4.0 ×104 cells were resuspended in 200ul serum-free DMEM. The cells were seeded in the upper transwell chambers. Then, a medium containing 20% FBS was put in the lower transwell chambers. We scraped off cells on the upper surface with a cotton swab after 24 hours of culturing. Cells on the lower surface of the chambers were fixed with methanol and stained with crystal violet. Last, we counted the number of cells under a microscope.
Prognostic Value
The Log-rank test checked the relationship between PIGR and the prognosis of BRCA patients. We used TISIDB to explore the associations between PIGR and patients across 30 human malignancies. Kaplan-Meier curves of OS or RFS between PIGR low- and high-expression groups were carried out based on data from the GSE86166 dataset (group cutoff value: 0.94) and the Kaplan-Meier plotter database by the Log Rank test. Prognostic factors of BRCA were shown in the forest plot of the multivariate Cox regression analysis. We regarded factors with HR >1 and p-value <0.05 as risk factors and factors with HR <1 and p-value <0.05 as protective factors. A nomogram was formed to predict the probability of 1-, 3-, and 5-year OS of BRCA patients. We formed calibration curves to evaluate the prediction efficacy of the nomogram.
DNA Methylation Analysis
To explore possible mechanisms of PIGR low expression in BRCA, we studied the different levels of DNA promoter methylation in BRCA and normal breast tissues via the UALCAN website. We observed the DNA methylation status in the PIGR gene and studied the prognostic value of the CpG islands' methylation status via the MethSurv database.
Co-expression Networks and Gene Set Enrichment Analysis
We explored PIGR co-expressed genes via LinkedOmics. A survival map of these genes was formed via the GEPIA2. Then, we performed the Gene Ontology (GO) biological processes analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to explore the possible functional mechanisms of PIGR co-expression genes joined. We proceeded with a gene set enrichment analysis (GSEA) method on the LinkedOmics data platform.
Immune Cell Infiltration Analysis and Immune-Related Signatures Analysis
We studied the relationship between PIGR expression and infiltration levels of 24 immune cell types in BRCA based on data from TCGA. The effect of PIGR on the infiltration of seven typical immune cells was explored. Data were processed by R(3.6.3 version). We further investigated the relationship between PIGR expression and the abundance of immune molecules in pan-cancer on the TISIDB data platform. These molecules comprise tumor-infiltrating lymphocytes (TILs), immunomodulators (including immune inhibitors, immunostimulators, and major histocompatibility complex (MHC) molecules), chemokines, and receptors.