With an aging population, there has been significant progress in the discovery and measurement of bone turnover biomarkers since the 2000s, especially for monitoring skeletal diseases like osteoporosis. Multiple markers derived from type I collagen, such as CTX, NTX, PINP, and ICTP, have been developed. Extensive efforts have been devoted to characterizing these molecules; however, their complex crosslinked structures have posed significant analytical challenges, and to date, these biomarkers remain poorly characterized. Previous attempts at characterization involved gel-based separation methods and MALDI-TOF analysis on collagen peptides directly extracted from bone. However, using bone powder, while rich in collagen, does not represent the true structure of the peptides in the biofluids. In this study, our goal was to characterize plasma and serum CTX for subsequent LC-MS/MS method development. We extracted and characterized type I collagen peptides directly from human plasma and serum using a proteomics workflow that integrates preparative liquid chromatography, affinity chromatography, and high-resolution mass spectrometry. Subsequently, we successfully identified numerous CTX species, providing valuable insights into the characterization of these crucial biomarkers.

Single-cell proteomics (SCP) has advanced significantly in recent years, with new tools that have been specifically designed for preparation and analysis of single cells now commercially available to researchers. The field is sufficiently mature to be broadly accessible to any lab capable of isolating single cells and performing bulk-scale proteomic analyses. In this review, we highlight recent work in the SCP field that has significantly lowered the barrier to entry, thus providing a practical guide for those who are newly entering the SCP field. We outline the fundamental principles and report multiple paths to accomplish the key steps of a successful SCP experiment including sample preparation, separation and mass spectrometry data acquisition and analysis. We recommend that researchers start with a label-free SCP workflow, as achieving high-quality and quantitatively accurate results is more straightforward than label-based multiplexed strategies. By leveraging these accessible means, researchers can confidently perform SCP experiments and make meaningful discoveries at the single-cell level.

LC-MS intact mass analysis and LC-MS/MS peptide mapping are foundational assays for developing biologic drugs and other commercial protein products. Certain PTM types, such as truncation and oxidation, increase the difficulty of precise proteoform characterization owing to inherent limitations in peptide and intact protein analyses. Top-down MS (TDMS) can resolve this ambiguity via fragmentation of specific proteoforms. We optimized our existing flow-programmed denaturing online buffer exchange ((fp)dOBE) approach to improve ESI sensitivity and increase TDMS sampling time for industrial applications. Using bovine alpha-lactalbumin (αLac), we tested data-dependent (DDA) and targeted strategies with 14 different MS/MS scan types featuring combinations of collisional- and electron-based fragmentation as well as proton transfer charge reduction. This large dataset was processed using a new software platform, named TDAcquireX, that improves proteoform characterization through TDMS data aggregation. (fp)dOBE-based DDA-TDMS analysis readily identified truncated proteoforms. For targeted TDMS, we used Sliding Window fragment ion deconvolution to analyze composite proteoform (cPrSM) results. This strategy facilitates probability-based noise filtering of individual fragments, simultaneously increasing matched fragments while decreasing total fragment masses. Using this strategy, we characterized oxidation positional isomers on αLac, finding ETD fragmentation uniquely provided accurate relative occupancy ratios by oxidation-specific challenges.

Single-cell proteomics (SCP) aims to characterize the proteome of individual cells, providing insights into complex biological systems. It reveals subtle differences in distinct cellular populations that bulk proteome analysis might overlook, which is essential for understanding disease mechanisms and developing targeted therapies. Mass spectrometry (MS) methods in SCP allow the identification and quantification of thousands of proteins from individual cells and this review highlights the role of data-independent acquisition MS (DIA-MS) in SCP. One major hurdle in SCP is the limited material in single-cell samples, but DIA-based techniques offer multiple potential solutions for their analysis. Utilizing wide precursor isolation windows to fragment multiple peptides simultaneously, DIA-based methods improve sensitivity, quantitative accuracy, and reproducibility at a cost in data analysis complexity. DIA methods can also be combined with sample multiplexing methods to increase the sample throughput, currently a key limitation in SCP. Challenges remain for interpreting sample multiplexed data from DIA-based SCP experiments, particularly with regards to isobaric tagging methods. Even still, we believe that DIA-based SCP approaches will play a major role in our understanding of systems biology.

Targeted proteomics, which includes parallel reaction monitoring (PRM), is typically utilized for more precise detection and quantitation of key proteins and/or pathways derived from complex discovery proteomics datasets. Initial discovery-based analysis using data independent acquisition (DIA) can obtain deep proteome coverage with low data missingness while targeted PRM assays can provide additional benefits in further eliminating missing data and optimizing measurement precision. However, PRM method development from bioinformatic predictions can be tedious and time-consuming because of the DIA output complexity. We address this limitation with a Python script that rapidly generates a PRM method for the TIMS-TOF platform using DIA data and a user-defined target list. To evaluate the script, DIA data generated from HeLa cell lysate (200 ng, 45-minute gradient method) as well as canonical pathway information from Ingenuity Pathway Analysis was utilized to generated a pathway-driven PRM method. Subsequent PRM analysis of targets within the example pathway, regulation of apoptosis, resulted in improved chromatographic data and enhanced quantitation precision (100% peptides below 10% CV with a median CV of 2.9%, n=3 technical replicates). The script is freely available at https://github.com/StevensOmicsLab/PRM-script and provides a framework that can be adapted to multiple DDA/DIA data outputs and instrument-specific PRM method types.

Sequencing the tyrosine phosphoproteome using MS-based proteomics is challenging due to the low abundance of tyrosine phosphorylation in cells, a challenge compounded in scarce samples like primary cells or clinical samples. The broad-spectrum optimization of selective triggering (BOOST) method was recently developed to increase phosphotyrosine sequencing in low protein input samples by leveraging tandem mass tags (TMT), phosphotyrosine enrichment, and a phosphotyrosine-loaded carrier channel. Here, we demonstrate the viability of BOOST in T cell receptor (TCR)-stimulated primary murine T cells by benchmarking the accuracy and precision of the BOOST method and discerning significant alterations in the phosphoproteome associated with receptor stimulation. Using 1 milligram of protein input (about 20 million cells) and BOOST, we identify and precisely quantify more than 2,000 unique pY sites compared to about 300 unique pY sites in non-BOOST control samples. We show that although replicate variation increases when using the BOOST method, BOOST does not jeopardise quantitative precision or the ability to determine statistical significance for peptides measured in triplicate. Many pY previously uncharacterised sites on important T cell signalling proteins are quantified using BOOST, and we identify new TCR responsive pY sites observable only with BOOST. Finally, we determine that the phase-spectrum deconvolution method on Orbitrap instruments can impair pY quantitation in BOOST experiments.

Alcohol consumption and high-fat diets often coincide in Western society, exerting negative synergistic effects on the liver. While many studies have demonstrated the impact of ALD and NAFLD on organ protein expression, none have offered a comprehensive view of the dysregulation at the level of the membrane proteome. In this study, we utilize peptidisc and solvent precipitation (SP4) methods to isolate and compare the membrane protein content of the liver with its unique biological functions. Using mice treated with a high-fat diet and ethanol in drinking water, we identified 1,563 liver proteins, with 46% predicted to have a transmembrane segment. Among these, 106 integral membrane proteins are dysregulated compared to the untreated sample. Gene ontology analysis reveals several dysregulated membrane processes associated with lipid metabolism, cell adhesion, xenobiotic processing, and mitochondrial membrane formation. Pathways related to cholesterol and bile acid transport are also mutually affected, suggesting an adaptive mechanism to counter the steatosis of the liver model. Our peptidisc-based membrane proteome profiling thus emerges as an effective way to gain insights into the role of the transmembrane proteome in disease development, warranting further in-depth analysis of the individual effect of the identified dysregulated membrane proteins.

Ectodomain shedding, which is the proteolytic release of transmembrane proteins from the cell surface, is crucial for cell-to-cell communication and other biological processes. The metalloproteinase ADAM17 mediates ectodomain shedding of over 50 transmembrane proteins ranging from cytokines and growth factors, such as TNF and EGFR ligands, to signaling receptors and adhesion molecules. Yet, the ADAM17 sheddome is only partly defined and biological functions of the protease have not been fully characterized. Some ADAM17 substrates (e.g. HB-EGF) are known to bind to heparan sulfate proteoglycans (HSPG), and we hypothesised that such substrates would be under-represented in traditional secretome analyses, due to their binding to cell surface or pericellular HSPGs. Thus, to identify novel HSPG-binding ADAM17 substrates, we developed a proteomic workflow that involves addition of heparin to solubilize HSPG-binding proteins from the cell layer, thereby allowing their mass spectrometry detection by heparin-secretome (HEP-SEC) analysis. Applying this methodology to murine embryonic fibroblasts stimulated with an ADAM17 activator enabled us to identify 47 transmembrane proteins that were shed in response to ADAM17 activation. This included known HSPG-binding ADAM17 substrates (i.e. HB-EGF, CX3CL1) and 17 novel HSPG-binding putative ADAM17 substrates. Two of these, MHC-I and IL1RL1, were validated as ADAM17 substrates by immunoblotting.

Botryllus schlosseri, is a model marine invertebrate for studying immunity, regeneration, and stress-induced evolution. Conditions for validating its predicted proteome were optimized using nanoElute® 2 deep-coverage LCMS, revealing up to 4,930 protein groups and 20,984 unique peptides per sample. Spectral libraries were generated and filtered to remove interferences, low-quality transitions, and only retain proteins with >3 unique peptides. The resulting DIA assay library enabled label-free quantitation of 3,426 protein groups represented by 22,593 unique peptides. Quantitative comparisons of a laboratory-raised with two field-collected populations revealed (1) a more unique proteome in the laboratory-raised population, and (2) proteins with high/low individual variabilities in each population. DNA repair/replication, ion transport, and intracellular signaling processes were unique in laboratory-cultured colonies. Spliceosome and Wnt signaling proteins were the least variable (highly functionally constrained) in all populations. In conclusion, we present the first colonial tunicate’s deep quantitative proteome analysis, identifying functional protein clusters associated with laboratory conditions, different habitats, and strong versus relaxed abundance constraints. These results empower research on B. schlosseri with proteomics resources and enable quantitative molecular phenotyping of changes associated with transfer from in situ to ex situ and from in vivo to in vitro culture conditions.

Predicting protein function from protein sequence, structure, interaction, and other relevant information is important for generating hypotheses for biological experiments and studying biological systems, and therefore has been a major challenge in protein bioinformatics. Numerous computational methods had been developed to advance protein function prediction gradually in the last two decades. Particularly, in the recent years, leveraging the revolutionary advances in artificial intelligence (AI), more and more deep learning methods have been developed to improve protein function prediction at a faster pace. Here, we provide an in-depth review of the recent developments of deep learning methods for protein function prediction. We summarize the significant advances in the field, identify several remaining major challenges to be tackled, and suggest some potential directions to explore. The data sources and evaluation metrics widely used in protein function prediction are also discussed to assist the machine learning, AI, and bioinformatics communities to develop more cutting-edge methods to advance protein function prediction.

We here present a chatbot assistant infrastructure (https://www.ebi.ac.uk/pride/chatbot/) that simplifies user interactions with the PRIDE database, the most popular proteomics data repository. Our system utilizes two advanced Large Language Models (LLM), llama2-13b and chatglm2-6b, and includes a web service API (Application Programming Interface), web interface, and sophisticated algorithms. We have developed a novel approach to construct vector-based representations for enabling the LLM responses, featuring a curated version and a comprehensive database of relevant links and paragraphs for each generated response. An important part of the framework is a benchmark component based on an Elo-ranking system, providing a scalable method for evaluating not only the performance of llama2-13b and chatglm2-6b but also, of any other available and future open-source LLMs. Throughout the benchmarking process, the PRIDE documentation for external users was refined to enhance the clarity and efficacy in addressing user queries. Importantly, while our infrastructure is exemplified through its application in the PRIDE database context, the modular and adaptable nature of our approach positions it as a valuable tool for improving user experiences across a spectrum of bioinformatics and proteomics tools and resources, among other domains. The integration of advanced LLMs, innovative vector-based construction, the benchmarking framework, and optimized documentation collectively form a robust and transferable chatbot assistant infrastructure.

State-of-the-art mass spectrometers combined with modern bioinformatics algorithms for peptide-to-spectrum matching (PSM) with robust statistical scoring allow for more variable features (i.e., post-translational modifications) being reliably identified from (tandem-) mass spectrometry data, often without the need for biochemical enrichment. Semi-specific proteome searches, that enforces a theoretical enzymatic digestion to solely the N- or C-terminal end, allow to identify native protein termini or those arising from endogenous proteolytic activity (also referred to ‘neo-N-termini’ analysis or ‘N-terminomics’. Nevertheless, deriving biological meaning from these search outputs can be challenging in terms of data mining and analysis. Thus, we introduce Fragterminomics, a data analysis approach for the (1) annotation of peptides according to their enzymatic cleavage specificity, (2) differential abundance and enrichment analysis of N-terminal sequence patterns, (3) visualization of neo-N-termini location, and (4) mapping neo-N-termini to known protein processing features. We illustrate the use of Fragterminomics by applying it to tandem mass tag (TMT)-based proteomics data of a mouse model of polycystic kidney disease and assess the semi-specific searches for biological interpretation of cleavage events and the variable contribution of proteolytic products to general protein abundance. The Fragterminomics approach and example data are available as an R package at https://github.com/MiguelCos/Fragterminomics.

The tooth is one of the ideal models for developmental study, involving in epithelial-mesenchymal transition and cell differentiation. The essential factors and pathways identified in tooth development will help understand the natural development process and the malformations of mineralized tissues such as skeleton. The time-dependent proteomic changes were investigated by healthy human molars proteomics of embryonic stages from the cap-to-early bell stage. A total of 713 differentially expressed proteins (DEPs) with five temporal expression patterns were filtered. 24 potential driver proteins of tooth development were screened by weighted gene co-expression network analysis (WGCNA) including CHID1, RAP1GDS1, HAPLN3, AKAP12, WLS, GSS, DDAH1, CLSTN1, AFM, RBP1, AGO1, SET, HMGB2, HMGB1, ANP32A, SPON1, FREM1, C8B, PRPS2, FCHO2, PPP1R12A, GPALPP1, U2AF2 and RCC2. The hub proteins in different temporal expression patterns were extracted. And the potential cell resources and the temporal expression patterns at transcriptomic level were explored using single cell RNA-sequencing (scRNA-seq). This study provides invaluable resources for the mechanistic studies of human embryonic epithelial and mesenchymal cell differentiation and tooth development.

This pilot experiment examines if a loss in muscle proteostasis occurs in people with obesity and whether endurance exercise positively influences either the abundance profile or turnover rate of proteins in this population. Men with (n = 3) or without (n = 4) obesity were recruited and underwent a 14-d measurement protocol of daily deuterium oxide (D2O) consumption and serial biopsies of vastus lateralis muscle. Men with obesity then completed 10-weeks of high-intensity interval training (HIIT), encompassing 3 sessions per week of cycle ergometer exercise with 1 min intervals at 100 % maximum aerobic power interspersed by 1 min recovery periods. The number of intervals per session progressed from 4 to 8, and during weeks 8-10 the 14-d measurement protocol was repeated. Proteomic analysis detected 352 differences (p < 0.05, false discovery rate < 5%) in protein abundance and 19 (p < 0.05) differences in protein turnover, including components of the ubiquitin-proteasome system. HIIT altered the abundance of 53 proteins and increased the turnover rate of 22 proteins (p < 0.05) and tended to benefit proteostasis by increasing muscle protein turnover rates. Obesity and insulin resistance are associated with compromised muscle proteostasis, which may be partially restored by endurance exercise.

The mzIdentML file format, originally developed by the Proteomics Standards Initiative in 2011, is the open XML data standard for peptide and protein identification results coming from mass spectrometry. We present mzIdentML version 1.3.0, which introduces new functionality and support for additional use cases. First of all, a new mechanism for encoding identifications based on multiple spectra. Furthermore, the main mzIdentML specification document can now be supplemented by extension documents which provide further guidance for encoding specific use cases for different proteomics subfields. One extension document has been added, covering additional use cases for the encoding of crosslinked peptide identifications. The ability to add extension documents facilitates keeping the mzIdentML standard up to date with advances in the proteomics field, without having to change the main specification document. The crosslinking extension document provides further explanation of the crosslinking use cases already supported in mzIdentML version 1.2.0, and provides support for encoding additional scenarios that are critical to reflect developments in the crosslinking field and facilitate its integration in structural biology. These are: (i) support for cleavable crosslinkers, (ii) support for internally linked peptides, (iii) support for noncovalently associated peptides, and (iv) improved support for encoding scores and the corresponding thresholds.

Seeds are an important part of plants, ensuring the continuation of plants’ life and providing nutrient reserves for humans and animals. Seed development is controlled by the interplay of several physiological processes. We applied label-free proteomics to round and wrinkled peas using seeds sampled at five growth stages (4 days after anthesis (DAA), 7DAA, 12DAA, 15DAA, and maturity). Phenotypic results indicated that wrinkled peas had lower starch concentration compared to round peas (29.5% vs. 46.6-55.1%). A total of 4,126 high confident proteins were detected, with 22–26% shared across all sampling times within an entry. Early seed growth stages were characterized by more unique proteins compared to maturity. Two-way ANOVA revealed 1,685 proteins significantly different among samples, of which 722 proteins were characterized into 29 functional classes. The four major classes (comprising over 50 proteins) were protein biosynthesis, protein homeostasis, enzymes, and carbohydrate metabolism. Of the two types of comparisons (time-point and entry-wise), time-point comparisons yielded more differentially abundance proteins (596 proteins in total). Different protein classes exhibited different patterns of change during seed development. For example, cell division related proteins were abundant early in seed development, whereas storage proteins were abundant later in seed development (especially after 12DAA). Compared to the round pea entries, the wrinkled entry had significantly lower abundance of starch branching enzymes, a protein involved in the biosynthesis of amylopectin in starch. In conclusion, the results of this study provide valuable information to improve our understanding of seed development and form the basis for further studies.

Immunotherapy harnesses neoantigens encoded within the human genome, but their therapeutic potential is hampered by low expression, which may be controlled by the Nonsense-Mediated Decay (NMD) pathway. This study investigates the impact of UPF1-knockdown on the expression of non-canonical/mutant proteins, employing proteogenomic to explore UPF1 role within the NMD pathway. Additionally, we conducted a comprehensive pan-cancer analysis of UPF1 expression and evaluated UPF1 expression in Triple-Negative Breast Cancer (TNBC) tissue in-vivo. Our findings reveal that UPF1-knockdown leads to increased transcription of non-canonical/mutant proteins, particularly those originating from retained-introns, pseudogenes, long non-coding RNAs, and unannotated transcript biotypes. Moreover, our analysis demonstrates elevated UPF1 expression in various cancer types, with notably heightened protein levels in patient-derived TNBC tumours compared to adjacent tissues. This study elucidates UPF1 role in mitigating transcriptional noise by degrading transcripts encoding non-canonical/mutant proteins. Intriguingly, we observe an upregulation of the NMD pathway in cancer, potentially acting as a “neoantigen-masking” mechanism that suppresses non-canonical/mutant protein expression. Targeting this mechanism may reveal a new spectrum of neoantigens accessible to the antigen presentation pathway. Our novel findings provide a strong foundation for the development of therapeutic strategies aimed at targeting UPF1 or modulating the NMD pathway.

Endometrial cancer is the most prevalent gynaecological cancer globally. Its association with obesity and metabolic diseases is a key aetiology, increasingly among younger females. Early diagnosis and improved treatment decisions are crucial for these women whose outcomes could be improved by discovering new biomarkers. We took a new approach to extracellular vesicle (EV) biomarker discovery - profiling the proteome of enriched EVs isolated directly from frozen biobanked endometrial cancers. Nine tissue pools, each generating collagenase-digested tissue and matched small EVs, were analysed using label-free proteomics. Three clinical subgroups: Endometrioid low BMI (body mass index), Endometrioid high BMI, and Serous, irrespective of BMI, were compared to identify shared secreted proteins, proteins associated with histological subtype, and proteins related to BMI. EVs were enriched for common EV markers and large secreted proteins. Cell lysates were enriched in mitochondrial and blood proteins. EV protein profiles were most different between the high BMI subgroup and the others, highlighting a significant influence of comorbidities on the intra-tumoural EV secretome. Proteins differentially abundant between subgroups in tissues were strikingly not also differential in the matched EVs. This work has identified secreted proteins implicated in the complex pathophysiology of endometrial cancer and pinpointed candidate biomarkers for diagnosis.