Cell lines culture and membrane proteins separation
Two human gastric adenocarcinoma cell lines including MKN-45, AGS and one immortalized normal gastric epithelial cell line GES-1 were purchased from American Type Culture Collection (Manassas, VA, USA). These cells were incubated in RPMI 1640 medium (Gibco, Waltham, MA, USA), which were routinely supplemented with 10% fetal bovine serum (FBS; Gibco, Waltham, MA, USA) and 1% penicillin/streptomycin (Gibco, Waltham, MA, USA). Culture medium was refreshed and the cells were passaged every 3 days.
The plasma membrane fraction was isolated using a Mem-PER Plus Membrane Protein Extraction Kit (89842, Thermo Fisher Scientific) using the manufacturers recommended protocol. Prepared purified membrane protein samples with equal volumes were loaded for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis essentially followed the method described by Li et al[7]. Protein bands were visualized using Coomassie brilliant blue staining.
DIA proteomics analysis
The protein quantification and differentially expressed membrane proteins detection between tumor and normal cell lines were performed using data-independent acquisition (DIA) mass spectrometry at the Beijing Genomics Institution (BGI). In brief, a fraction of samples originating from the same cell lines were aliquoted and pooled for the generation of the spectral library. Each protein was quantified by average its unique peptides intensity[8]. For details, see Supplementary Methods.
Validation of DEPs in transcriptional profiling
Level-3 RNA-sequencing data (HTSeq-Counts and HTSeq-FPKM-UQ) of GC patients was downloaded from the UCSC Xena browser (https://xenabrowser.net/). Read count tables mapped on hg38 human genome with GENCODE v22 as gene annotation, were imported into the three different R packages (edgeR[9], Limma[10] and DESeq2[11]) to identify differentially expressed transcripts. To investigate the transcriptional and posttranscriptional activity, the expression level of identified cell surface proteins was confirmed in TCGA dataset. Besides, a Human microarray dataset (GSE13911) was obtained from the Gene Expression Omnibus (GEO database, https://www.ncbi.nlm.nih.gov/geo/). Microarray signals were normalized using a standard Robust Multichip Average (RMA) algorithm.
Immunohistochemical staining
This study enrolled 66 histologically diagnosed gastric adenocarcinoma patients from January 2019 to December 2020 at Nanjing Drum Tower Hospital. The use of these clinical samples was approved by the Ethics Committee of the Nanjing Drum Tower Hospital (The Affiliated Hospital of Nanjing University School). The GC tissues and the corresponding noncancerous mucosal tissues were fixed in 10% formalin, embedded in paraffin, and then cut into sections. Immunohistochemistry (IHC) on sections which were deparaffinized in xylene and rehydrated using graded alcohols was incubated with primary LRP1B antibody (1:100 dilution, Abcam, Cambridge, UK), followed by secondary antibodies IgG H&L (HRP; 1:200 dilution, Abcam, Cambridge, UK). The overall staining score was calculated as staining intensity score multiplied by the ratio of positively stained cells. The staining intensity was graded in four segments: 0 (no staining), 1 (light yellow), 2 (yellow brown) and 3(brownish-yellow). The number of positive cells was divided into four grades: 0 (< 5%), 1 (5–30%), 2 (31–70%), and 3 (71-100% of positive cells). A final score of 0 to 2 was defined it as negative staining, and 3 to 9 represented positive staining.
Bioinformatics analysis
Principal component analysis (PCA) on the measurements of membrane proteins expressions was carried out using the package “FactoMineR”[12]. Gene Ontology (GO) function and Gene Set Enrichment Analysis (GSEA) were performed using the package “ClusterProfiler”[13]. A GO enrichment analysis associating the DEPs to known biological categories is likely to reveal their functional roles. The value of log2 (Fold Change) calculated by DESeq2 package was used as ranking metric for GSEA. The canonical pathways sub-collection of the C2 collection in the Molecular Signatures Database was used as the reference gene sets.
Statistical analysis
The Mann-Whitney test (for continuous variables) and Pearson's χ2 test (for categorical variables) were used to examine differences of low-density lipoprotein receptor-related protein 1b (LRP1B) expression between GC and normal tissues. All statistical analysis was two-sided and P < 0.05 was defined as statistically significant. Statistical analysis was conducting on R programming language v.3.6.3.