Isobaric Tagging and Data Independent Acquisition as Complementary Strategies for Proteome Profiling on an Orbitrap Astral Mass Spectrometer

Abstract

Comprehensive global proteome profiling that is amenable to high throughput processing will broaden our understanding of complex biological systems. Here, we evaluated two leading mass spectrometry techniques, Data Independent Acquisition (DIA) and Tandem Mass Tagging (TMT), for extensive protein abundance profiling. DIA provides label-free quantification with a broad dynamic range, while TMT enables multiplexed analysis using isobaric tags for efficient cross-sample comparisons. We analyzed 18 samples, including four cell lines (IHCF, HCT116, HeLa, MCF7) under standard growth conditions, in addition to IHCF treated with two H₂O₂ concentrations, all in triplicate. Experiments were conducted on an Orbitrap Astral mass spectrometer, employing Field Asymmetric Ion Mobility Spectrometry (FAIMS). Despite utilizing different acquisition strategies, both the DIA and TMT approaches achieved comparable proteome depth and quantitative consistency, with each method quantifying over 10,000 proteins across all samples, with slightly more protein-level precision for the TMT strategy. Relative abundance correlation analysis showed strong agreement at both peptide and protein levels. Our findings highlight the complementary strengths of DIA and TMT for high-coverage proteomic studies, providing flexibility in method selection based on specific experimental needs.

Keywords: Astral; DIA; FAIMS; IHCF; TMTpro.

Figure 1:. Workflow overview for processing 18 samples by DIA and TMT.

Tryptic peptides were prepared from proteins extracted from reduced and alkylated cell lysates. Peptide pools were then aliquoted with 0.5 μg of each sample to be analyzed by DIA once without FAIMS and again with FAIMS (single CV of −35V). Aliquots (25 μg) from the same sample were labeled with TMTpro, pooled, and pre-fractionated with basic pH reversed-phase chromatography. The subsequent 24 concatenated super-fractions were analyzed by LC-FAIMS-MS/MS.

Figure 2:. Comparison of DIA datasets acquired with no FAIMS and with one FAIMS compensation voltage (CV=−35V).

A) Count of the total peptides quantified in each sample for data acquired without (“NoFAIMS”) and with FAIMS (“CV-35”). B) Venn diagram depicting the overlap in peptide sequences quantified by both DIA acquisition methods. C) Count of the proteins quantified in each sample for each DIA acquisition method. D) Venn diagram illustrating the protein-level overlap between DIA acquisition methods. E) Peptide level completeness for each DIA acquisition method. “Completeness” requires a measurement for each replicate for a given peptide. If no measurement is made for an entire set of triplicate samples, it does not count against completion, as it is assumed that the given protein is not expressed in that cell line. F) Protein level completeness for each DIA acquisition method using the same criteria as in (E). G) Box-and-whiskers plot depicting the number of data points for a given chromatographic peak. H) Correlation plot of the average fold change in abundance between IHCF and MCF7 cell lines for the two DIA acquisition methods at the peptide level. I) Corresponding correlation plot of the average fold change in abundance between IHCF and MCF7 cell lines at the protein level.

Figure 3:. Comparison of DIA and TMT datasets acquired on an Orbitrap Astral mass spectrometer.

A) Bar graph depicting the number of quantified proteins in all three datasets. The stacked bars for the DIA datasets include proteins in which all triplicates of a given condition had measurements (or all had no measurements) for a given protein (bottom portion, solid) and those with missing values (top portion, slashes). B) Box-and-whiskers plot illustrating the number of peptides with unique sequences quantified per protein. C) An UpSet plot showing the peptide overlap across the three data acquisition methods. Correlation plot of the fold change in abundance between IHCF and MCF7 cell lines at the peptide level for D) the TMT and DIA: NoFAIMS datasets, and E) the TMT and DIA: CV-35 datasets. Box-and-whiskers plot illustrating the distribution of the coefficient of variation (relative standard deviation, RSD) for triplicate measurements of each condition (or cell line) at the F) peptide and G) protein level for all three datasets.

Figure 4:. Quantitative assessment of the three data acquisition methods.

Principal components analysis (PCA) plotting the first two principal components for the A) DIA: No FAIMS (left), DIA:CV-35 (center), and TMT (right) datasets. Example protein abundance profiles for B) Collagen alpha-1(I) chain (COL1A1, P02452), C) Sortilin (SORT1, Q99523), D) Proline-rich protein 5-like (PRR5L, Q6MZQ0) and E) Receptor-type tyrosine-protein phosphatase gamma (PTPRG, P23470).