Proteomics profiling in primary tumors of metastatic and non-metastatic breast cancers

Analysis of different proteomics profile in primary tumors of metastatic and non-metastatic breast cancers.

Sample Processing Protocol
Small pieces of 67NR and 66cl4 tumors were thawed briefly in lysis buffer and homogenized using 1.4 mm ceramic beads (Precellys, 03961-1-103) in reinforced tubes (KT03961-1-403.2) for 4 cycles á 40 sec homogenization, 2 min break. Lysis buffer: 8 M urea (Merck Millipore, #1084870500) with 0.5 % CHAPS, 100 mM DTT, (Sigma, #646563), 1x Complete® protease inhibitor (Roche, #1187350001) and 2x phosphatase inhibitor cocktail II (Sigma, #P5726) and III (Sigma, #P0044). Homogenized tissue samples were shaken before centrifugation (15 000 g, 20 min, 4°C). For cells, the cells were gorwn in medium with various concentrations of arginine ( 400µM, 4µM and 0µM for 24 and 48 h). The cells were harvested in 8 M urea lysis buffer. The samples were shaken (15 min in cold room) before centrifugation (15 000 g, 20 min, 4°C). Protein concentration was measured at 595 nm using BioRad protein assay dye reagent (Bio-Rad, #500-0006).15µg of each were added to 130µl 100mM ammonium bicarbonate. Proteins were reduced and alkylated with DTT (12mM) for 30min at 55°C. Samples were further alkylated with iodoacetamide (36mM) for 30min at room temperature and dark. Proteins were digested with 250ng trypsin at 37°C overnight and further acidified in acetic acid (0.5%) and desalted using Oasis HLB C18 solid phase extraction according to manufacturer’s instructions. After elution of peptides from C18, the samples were dried in speedvac and further dissolved in 18 µl 0.1% formic acid and LC-MS/MS were performed on a timsTOF Pro (Bruker Daltonics) connected to a nanoElute (Bruker Daltonics) HPLC. Peptides were separated using a Bruker15 (75µm*15cm) column with running buffers A (0.1% formic acid) and B (0.1% formic acid in acetonitrile) with a gradient from 0% B to 37%B for 100min. The timsTof instrument was operated in the DDA PASEF mode with 10 PASEF scans per acquisition cycle and accumulation and ramp times of 100 millisecond each. The ‘target value’ was set to 20,000 and dynamic exclusion was activated and set to 0.4 min. The quadrupole isolation width was set to 2 Th for m/z < 700 and 3 Th for m/z > 800.

Data Processing Protocol
Proteins were quantified by processing MS data using MaxQuant v.2.0.3.0 [1]. The open workflow provided in FragPipe [2] was used to inspect the raw files to determine optimal search criteria and accordingly search parameters were set as follows: enzyme specified as trypsin with maximum two missed cleavages allowed; deamidation of asparagine/glutamine, oxidation of methionine, and protein N-terminal acetylation as variable modifications; precursor and fragment mass tolerance was set to 20 parts per million (PPM). These were imported in MaxQuant which uses m/z and retention time (RT) values to align each run against each other sample with a minute window match-between-run function and 20 mins overall sliding window using a clustering-based technique. These were further queried against the mouse proteome including isoforms downloaded from Uniprot [3] in 2021 along with MaxQuant’s internal contaminants database using Andromeda built into MaxQuant. Both protein and peptide identifications false discovery rate (FDR) was set to 1%, only unique peptides with high confidence were used for final protein group identification. Peak abundances were extracted by integrating the area under the peak curve. Each protein group abundance was normalized by the total abundance of all identified peptides for each run and protein by calculated median summing all unique and razor peptide-ion abundances for each protein using label-free quantification (LFQ) algorithm [4] with minimum peptides ≥ 1. LFQ values for all samples were log-transformed with base 2. A correlation heatmap using R package pheatmap [5] was created using these transformed LFQ values and an outlier was removed. The rest of the values representing each condition were subjected to two-sided Student’s t-Tests [6] as implemented in R[7] in order to check the consistency of change. The amount of change was estimated by subtracting the median of these values representing each group (log2 median change). Directionality of the change is encoded within the sign of log2 median change whereby a negative sign reflecting decreased and a positive sign reflecting the increased expression of the respective protein group. Further, to estimate the false-discovery rate (FDR), the T-test p-values were corrected using the Benjamini-Hochberg procedure [8]. Differentially expressed (DE) protein groups were identified at FDR<0.1 and absolute log2 median change >1.5. The DE quantified only in one group were checked if their coefficient-of-variation of log2medianchange was within 5%. The Uniprot accession IDs of these DE were mapped to a volcano-plot using R package ggplot2 [9] . Volcano plots represented in the figures were drawn using the EnhancedVolcano R package (v 1.0.1), and the cut off was set to log2 median change ±1.5 and the corrected T-test p-value of < 0.05. Log2 median change is represented as log2FoldChange in the manuscript. To evaluate common biological functions of results of DEGs, functional enrichment analyses of all significantly highly expressed genes and all significantly low expressed genes were performed to identify the biological processes involved. The bioconductor package clusterProfiler [5] was used to conduct gene ontology (GO) functional enrichment analyses for biological process (BP), for DEGs from different groups applicable. The plot was visualized using the ggplot2 package (v3.2.1). 1. Tyanova, S., T. Temu, and J. Cox, The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat Protoc, 2016. 11(12): p. 2301-2319. 2. Geiszler, D.J., et al., PTM-Shepherd: Analysis and Summarization of Post-Translational and Chemical Modifications From Open Search Results. Mol Cell Proteomics, 2021. 20: p. 100018. 3. UniProt. UniProtKB - H3BJL3 (H3BJL3_MOUSE). 2021 October 2021]; Available from: (https://www.uniprot.org/proteomes/UP000000589. 4. Cox, J., et al., Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol Cell Proteomics, 2014. 13(9): p. 2513-26. 5. pheatmap: Pretty Heatmaps. Available from: https://cran.r-project.org/web/packages/pheatmap/index.html. 6. Student, The Probable Error of a Mean. Biometrika, 1908. 6(1): p. 1-25. 7. project, R. The R Project for Statistical Computing. Available from: https://www.r-project.org/. 8. Hochberg, Y.B.a.Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. 1995; Available from: https://www.jstor.org/stable/2346101. 9. Wickham, H. ggplot2: Elegant Graphics for Data Analysis. 2009; Available from: https://www.springer.com/gp/book/9780387981413.