Molecular & Cellular Proteomics最新文献

Although the plasma membrane (PM) is among the most biologically important and therapeutically targeted cellular compartments, it is among the most challenging to faithfully capture using proteomic approaches. The quality of quantitative surfaceomics data depends heavily on the effectiveness of the cell surface enrichment used during sample preparation. Enrichment improves sensitivity for low abundance PM proteins and ensures that the changes detected reflect PM expression changes rather than whole cell changes. Cell surface biotinylation with PM-impermeable, amine-reactive reagents is a facile, accessible, and unbiased approach to enrich PM proteins. However, it results in unexpectedly high contamination with intracellular proteins, reducing its utility. We report that biotinylating human cells with amine-reactive reagents intracellularly labels a small but reproducible population of nonviable cells. Although these dead cells represent only 5 ± 2% of the total, we find that in T cell preparations the dead cells account for 90% of labelled proteins. Depleting Annexin V positive dead T cells postlabelling removes ∼99% of the intracellularly labelled cells, resulting in markedly improved PM identifications, peptide counts, and intensity-based absolute quantification intensities. Correspondingly, we found substantial depletion of intracellular proteins, particularly of nuclear origin. Overall, the cumulative intensity of PM proteins increased from 4% to 55.8% with dead cell depletion. Finally, we demonstrate that immature ER/Golgi glycoforms of CD11a and CD18 are selectively removed by dead-cell depletion. We conclude that high intracellular labelling of nonviable cells is the major source of intracellular protein contaminants in amine-reactive surface enrichment methods and can be reduced by dead-cell depletion postlabelling, improving both the sensitivity and accuracy of PM proteomics.

Post-translational modifications (PTMs) play a central role in cellular regulation and are implicated in numerous diseases. Database searching remains the standard for identifying modified peptides from tandem mass spectra but is hindered by the combinatorial expansion of modification types and sites. De novo peptide sequencing offers an attractive alternative, yet existing methods remain limited to unmodified peptides or a narrow set of PTMs. Here, we curated a large dataset of spectra from endogenous and synthetic peptides from ProteomeTools spanning 19 biologically relevant amino acid-PTM combinations, covering phosphorylation, acetylation, and ubiquitination. We used this dataset to develop Modanovo, an extension of the Casanovo transformer architecture for de novo peptide sequencing. Modanovo achieved robust performance across these amino acid-PTM combinations (median area under the precision-coverage curve 0.92), while maintaining performance on unmodified peptides (0.93), nearly identical to Casanovo (0.94). The model outperformed π-PrimeNovo-PTM and InstaNovo-P and showed increased precision and complementarity to the database search tool MSFragger. Robustness was confirmed across independent datasets, particularly at peptide lengths frequently represented in the curated dataset. Applied to a phosphoproteomics dataset from monkeypox virus-infected cells, Modanovo recovered numerous confident peptides not reported by database search, including new viral phosphosites supported by spectral evidence, thereby demonstrating its complementarity to database-driven identification approaches. These results establish Modanovo as a broadly applicable model for comprehensive de novo sequencing of both modified and unmodified peptides.

Investigating multiple protein post-translational modifications (PTMs) is critical for unraveling the complexities of protein regulation and the dynamic interplay among PTMs, a growing focus in proteomics. However, simultaneous analysis of diverse PTMs remains a significant technical challenge, as existing workflows struggle to balance throughput, sensitivity, and reproducibility, particularly when sample amounts are limited. To address these limitations, we present MoSAIC, a multi-PTM workflow integrating coenrichment strategies, multiplexing, fractionation, hybrid data acquisition, and unified data analysis, optimized for clinically relevant biological samples. This approach targets phosphorylation, glycosylation, acetylation, and ubiquitination, enabling comprehensive interrogation of these modifications simultaneously. Compared with the traditional Clinical Proteomic Tumor Analysis Consortium workflow, MoSAIC doubles PTM coverage (four versus two PTMs) while maintaining the same instrument time (24 mass spectrometry runs), achieving increased identifications of PTM-modified peptides. By leveraging fractionation and tandem mass tag labeling, we achieved concurrent identification and quantification of PTM-specific peptides from the same sample, enhancing throughput and data consistency. This robust workflow addresses key limitations in multi-PTM proteomics, providing a cost-effective and efficient platform to advance biological and clinical research.

Post-translational modifications (PTMs) are pivotal in cellular regulations, and their crosstalk is related to various diseases such as cancer. Given the prevalence of PTM crosstalk within close amino acid ranges, identifying peptides with multiple PTMs is essential. However, this task is an NP-hard combinatorial problem with exponential complexity, posing significant challenges for existing analysis methods. Here, we introduce PIPI-C (PTM-Invariant Peptide Identification with a Combinatorial model), a novel search engine that addresses this challenge through a mixed integer linear programming (MILP) model, thereby overcoming the limitations of existing approaches that struggle with high-order PTM combinations. Rigorous validation across diverse datasets confirms PIPI-C's superior performance in detecting PTM combinations. When applied to over 72 million mass spectra of three human cancers-lung squamous cell carcinoma (LSCC), colorectal adenocarcinoma (COAD), and glioblastoma (GBM)-PIPI-C reveals significantly upregulated PTM combinations. In LSCC, 50% of 860 upregulated unique PTM site patterns (UPSPs) (when comparing cancer vs. normal samples) carried at least two PTMs, including literature-supported crosstalks such as di-methylation with trifluoroleucine substitution and amidation with proline-to-valine substitution. Similar findings in COAD and GBM highlight PIPI-C's utility in uncovering cancer-relevant PTM combination landscapes. Overall, PIPI-C provides a robust mathematical framework for decoding complex PTM patterns, advancing our understanding of PTM-driven cellular processes in diseases.

Translational errors (TEs) result in a mismatch between mRNA codons and the amino acids (AAs) of the corresponding protein. Unlike DNA mutations or RNA editing, where nucleotide sequences can be used to infer AA substitutions, TEs can only be detected at the protein level. Although high-throughput mass spectrometry (MS) proteomics offers the potential to resolve peptide sequences and could theoretically be used to identify TEs, the feasibility of current MS data analysis approaches for this application remains uncertain. Here, we utilize patient-derived xenograft proteomics data, which include both human and mouse peptides with identifiable cross-species AA variations, as a ground truth for benchmarking TE identification methods. By using high-confidence mouse peptides as surrogates for "TE-containing" peptides, we show that current open search approaches can achieve >65% overall sensitivity and >70% overall precision for high-quality samples. The intersection of different search strategies significantly enhances precision, albeit at the expense of reduced sensitivity. Notably, the evaluation metrics vary significantly across individual AA substitutions, suggesting that caution is warranted when detecting or interpreting specific AA substitutions. Moreover, closed searches targeting predefined AA changes exhibit poor precision, with post-translational modification mislocalization identified as a key bottleneck for this application. Overall, our study provides a first-of-its-kind benchmark for MS-based TE discovery and offers guidance for optimizing MS search strategies.

We present the first integrated transcriptomic and proteomic profiling of the iridocorneal region in the spontaneous murine glaucoma model DBA/2J and DBA/2J-Gpnmb⁺/Sj controls to define molecular changes associated with ocular hypertension and glaucoma. Using RNA sequencing and label-free quantitative proteomics, we identified over 20,000 transcripts and 8500 proteins, creating a comprehensive molecular atlas of glaucoma-related alterations in DBA/2J mice. Principal component and differential expression analyses revealed distinct genotype-specific molecular signatures. In DBA/2J mice, upregulated genes were enriched in pathways related to extracellular matrix remodeling, collagen organization, TGF-β signaling, and inflammation. Proteomic data confirmed increased levels of complement components, antigen presentation proteins, and autophagy markers. Integrated analyses identified 29 genes upregulated at both transcript and protein levels, primarily involved in extracellular matrix structure and immune regulation. Downregulated genes were associated with melanocyte differentiation and pigment-organelle function, including Pmel, a gene implicated in pigmentary glaucoma. Cross-referencing with human genome-wide association studies data revealed overlap with glaucoma-associated genes (LTBP2, LOXL1, COL11A1, VCAM1), alongside reduced expression of Angpt and Lmx1b, linked to ocular hypertension. Together, these findings support the existence of an immune-fibrotic feed-forward loop and implicate collagen-elastic fiber dysfunction as a central mechanism in glaucoma pathogenesis.

Membrane proteins (MPs) are vital to cellular signaling, metabolism, and disease pathology, yet remain underrepresented in proteomics. To address this, several independent workflows have been developed to enable the profiling of the membrane proteome; however, the relative advantages and limitations of each method remain poorly defined. Here, we systematically compare four classical solid-phase membrane proteomic workflows (SP3, SP4, FASP, S-Trap) and three membrane mimetic strategies (Peptidisc, nanodisc, and SMALP copolymer) for mass spectrometry-based membrane proteome profiling, using healthy (LFD) and obese (HFD) mouse liver tissue. We found that the solid-phase methods yield higher total protein identifications, while the membrane mimetic systems enrich MPs. SMALP copolymer displays intermediate characteristics between the solid-phase and membrane mimetic workflows. Peptidisc and nanodisc stand out for their enrichment of MPs, although Peptidisc shows better enrichment of plasma membrane integral MPs, particularly those with 11+ transmembrane segments. In the context of HFD-induced liver proteome remodeling, the Peptidisc workflow outperformed the other six methods by capturing the highest number of differentially expressed MPs and demonstrating the lowest standard deviation of MP-level dysregulation. Collectively, this comparative analysis highlights the trade-offs between depth of proteome coverage and MP enrichment across workflows, underscoring the importance of method selection based on total protein counts, MP enrichment, and the precise detection of MP-level dysregulation.

Cell-cell communications are widely explored to understand tissue homeostasis and diseases. Numerous computational tools have been developed to infer cellular interactions from transcriptomic or proteomic expression data. However, proteins often carry post-translational modifications (PTMs) that can induce conformational switches and alter their functional properties. A key challenge remains to incorporate PTM data in the inference and analysis of cellular interactions. Here, we propose an extension of our previously published tool BulkSignalR to integrate PTM information in ligand-receptor interactions and downstream pathways predictions. This new functionality is compatible with bulk and single-cell data, and it supports all types of PTMs. Based on two illustrative datasets, we show that this new feature provides deeper insights into biological pathway regulation, and that PTM integration helps reducing false positive results occasionally produced by standard approaches.

Mass spectrometry (MS) is the method of choice for high-throughput identification of immunopeptides, which are generated by intracellular proteases, unlike proteomics peptides that are typically derived from trypsin-digested proteins. Therefore, the searching space for immunopeptides is not limited by proteolytic specificity, requiring more sophisticated software algorithms to handle the increased complexity. Despite the widespread use of MS in immunopeptidomics, there is a lack of systematic evaluation of data processing software, making it challenging to identify the optimal solution. In this study, we provide a comprehensive benchmarking of the most widespread/used data-dependent acquisition (DDA)-based software platforms for immunopeptidomics: MaxQuant, FragPipe, PEAKS and MHCquant. The evaluation was conducted using data obtained from the JY cell line using the Thunder-DDA-PASEF method. We assessed each software's ability to identify immunopeptides and compared their identification confidence. Additionally, we examined potential biases in the results and tested the impact of database size on immunopeptide identification efficiency. Our findings demonstrate that all software platforms successfully identify the most prominent subset of immunopeptides with 1% false discovery rate (FDR) control, achieving medium to high identification confidence correlations. The largest number of immunopeptides were identified using the commercial PEAKS software, which is closely followed by FragPipe, making it a viable non-commercial alternative. However, we observed that larger database sizes negatively impacted the performance of some software platforms more than others. These results provide valuable insights into the strengths and limitations of current MS data processing tools for immunopeptidomics, supporting the immunopeptidomics/MS community in determining the right choice of software.

Alzheimer's disease (AD) is an age-associated neurodegenerative disorder characterized by amyloid plaques, tau hyperphosphorylation, and synaptic dysfunction. Most available cellular AD models lack aging features, limiting their ability to recapitulate key pathological mechanisms. Here we applied high-resolution mass spectrometry-based multiplexed proteomics and phosphoproteomics in a discovery setting to characterize an accelerated AD (acAD) model that combines amyloid precursor protein (APP) and presenilin (PSEN) mutations with progerin, an aging-associated Lamin A mutant that accelerates aging. Across four phenotypes (control, progerin, classic AD, and acAD), we identified 8279 proteins, quantified 6081 proteins, and detected phosphorylation dynamics. Relative to the classic model, acAD exhibited broader proteome remodeling, including amplified downregulation of synaptic and cytoskeletal proteins, upregulation of transcription and translation machinery, and pathway-level changes in neuronal signaling, mitochondrial dynamics, and neuroinflammation. Phosphoproteome analysis revealed widespread changes in RNA-binding and cytoskeletal proteins, aligning with recent data from two murine AD models. These findings show that acAD captures canonical AD phenotypes while uniquely modeling age-related inflammation and phosphorylation, providing a resource to accelerate studies of proteome-level mechanisms of AD progression and to inform strategies targeting cytoskeletal and inflammatory pathways.