The host response to COVID-19 pathophysiology over the first few days of infection remains largely unclear, especially the mechanisms in the blood compartment. We report on a longitudinal proteomic analysis of acute-phase COVID-19 patients, for which we used blood plasma, multiple reaction monitoring with internal standards, and data-independent acquisition. We measured samples on admission for 49 patients, of which 21 had additional samples on days 2, 4, 7, and 14 after admission. We also measured 30 externally obtained samples from healthy individuals for comparison at baseline. The 31 proteins differentiated in abundance between acute COVID-19 patients and healthy controls belonged to acute inflammatory response, complement activation, regulation of inflammatory response, and regulation of protein activation cascade. The longitudinal analysis showed distinct profiles revealing increased levels of multiple lipid-associated functions, a rapid decrease followed by recovery for complement activation, humoral immune response, and acute inflammatory response-related proteins, and level fluctuation in the regulation of smooth muscle cell proliferation, secretory mechanisms, and platelet degranulation. Three proteins were differentiated between survivors and nonsurvivors. Finally, increased levels of fructose-bisphosphate aldolase B were determined in patients with exposure to angiotensin receptor blockers versus decreased levels in those exposed to angiotensin-converting enzyme inhibitors. Data are available via ProteomeXchange PXD029437.

During tumorigenesis, DNA mutations in protein coding sequences can alter amino acid sequences which can change the structures of proteins. While the 3D structure of mutated proteins has been studied with atomic resolution, the precise impact of somatic mutations on the 3D proteome during malignant transformation remains unknown because methods to reveal in vivo protein structures in high throughput are limited. Here, we measured the accessibility of the lysine ε-amine for chemical modification across proteomes using covalent protein painting (CPP) to indirectly determine alterations in the 3D proteome. CPP is a novel, high-throughput quantitative mass spectrometric method that surveyed a total of 8052 lysine sites across the 60 cell lines of the well-studied anticancer cell line panel (NCI60). Overall, 5.2 structural alterations differentiated any cancer cell line from the other 59. Structural aberrations in 98 effector proteins correlated with the selected presence of 90 commonly mutated proteins in the NCI60 cell line panel, suggesting that different tumor genotypes reshape a limited set of effector proteins. We searched our dataset for druggable conformational aberrations and identified 49 changes in the cancer conformational landscape that correlated with the growth inhibition profiles of 300 drug candidates out of 50,000 small molecules. We found that alterations in heat shock proteins are key predictors of anticancer drug efficacy, which implies that the proteostasis network may have a general but hitherto unrecognized role in maintaining malignancy. Individual lysine sites may serve as biomarkers to guide drug selection or may be directly targeted for anticancer drug development.

Proteomics is a data-rich science with complex experimental designs and an intricate measurement process. To obtain insights from the large data sets produced, statistical methods, including machine learning, are routinely applied. For a quantity of interest, many of these approaches only produce a point estimate, such as a mean, leaving little room for more nuanced interpretations. By contrast, Bayesian statistics allows quantification of uncertainty through the use of probability distributions. These probability distributions enable scientists to ask complex questions of their proteomics data. Bayesian statistics also offers a modular framework for data analysis by making dependencies between data and parameters explicit. Hence, specifying complex hierarchies of parameter dependencies is straightforward in the Bayesian framework. This allows us to use a statistical methodology which equals, rather than neglects, the sophistication of experimental design and instrumentation present in proteomics. Here, we review Bayesian methods applied to proteomics, demonstrating their potential power, alongside the challenges posed by adopting this new statistical framework. To illustrate our review, we give a walk-through of the development of a Bayesian model for dynamic organic orthogonal phase-separation (OOPS) data.

Blood derivatives are the biofluids of choice for metabolomic clinical studies since blood can be collected with low invasiveness and is rich in biological information. However, the choice of the blood collection tubes has an undeniable impact on the plasma and serum metabolic content. Here, we compared the metabolomic and lipoprotein profiles of blood samples collected at the same time and place from six healthy volunteers but using different collection tubes (each enrolled volunteer provided multiple blood samples at a distance of a few weeks/months): citrate plasma, EDTA plasma, and serum tubes. All samples were analyzed via nuclear magnetic resonance spectroscopy. Several metabolites showed statistically significant alterations among the three blood matrices, and also metabolites' correlations were shown to be affected. The effects of blood collection tubes on the lipoproteins' profiles are relevant too, but less marked. Overcoming the issue associated with different blood collection tubes is pivotal to scale metabolomics and lipoprotein analysis at the level of epidemiological studies based on samples from multicenter cohorts. We propose a statistical solution, based on regression, that is shown to be efficient in reducing the alterations induced by the different collection tubes for both the metabolomic and lipoprotein profiles.

Protein post-translational modifications (PTMs) enable cells to rapidly change in response to biological stimuli. With hundreds of different PTMs, understanding these control mechanisms is complex. To date, efforts have focused on investigating the effect of a single PTM on protein function. Yet, many proteins contain multiple PTMs. Moreover, one PTM can alter the prevalence of another, a phenomenon termed PTM crosstalk. Understanding PTM crosstalk is critical; however, its detection is challenging since PTMs occur substoichiometrically. Here, we develop an enrichment-free, label-free proteomics method that utilizes high-field asymmetric ion mobility spectrometry (FAIMS) to enhance the detection of PTM crosstalk. We show that by searching for multiple combinations of dynamic PTMs on peptide sequences, a 6-fold increase in candidate PTM crosstalk sites is identified compared with that of standard liquid chromatography-tandem mass spectrometry (LC-MS/MS) workflows. Additionally, by cycling through FAIMS compensation voltages within a single LC-FAIMS-MS/MS run, we show that our LC-FAIMS-MS/MS workflow can increase multi-PTM-containing peptide identifications without additional increases in run times. With 159 novel candidate crosstalk sites identified, we envisage LC-FAIMS-MS/MS to play an important role in expanding the repertoire of multi-PTM identifications. Moreover, it is only by detecting PTM crosstalk that we can "see" the full picture of how proteins are regulated.

Fatty liver disease progresses through stages of fat accumulation and inflammation to nonalcoholic steatohepatitis (NASH), fibrosis and cirrhosis, and eventually hepatocellular carcinoma (HCC). Currently available diagnostic tools for HCC lack sensitivity and specificity. In this study, we investigated the use of circulating serum glycoproteins to identify a panel of potential prognostic markers that may be indicative of progression from the healthy state to NASH and further to HCC. Serum samples were processed and analyzed using a novel high-throughput glycoproteomics platform. Our initial dataset contained healthy, NASH, and HCC serum samples. We analyzed 413 glycopeptides, representing 57 abundant serum proteins, and compared among the three phenotypes. We studied the normalized abundance of common glycoforms and found 40 glycopeptides with statistically significant differences in abundances in NASH and HCC compared to controls. Summary level relative abundances of core-fucosylated, sialylated, and branched glycans containing glycopeptides were higher in NASH and HCC as compared to controls. We replicated some of our findings in an independent set of samples of individuals with benign liver conditions and HCC. Our results may be of value in the management of liver diseases. Data generated in this work can be downloaded from MassIVE (https://massive.ucsd.edu) with identifier MSV000088809.

Filter-aided sample preparation (FASP) is widely used in bottom-up proteomics for tryptic digestion. However, the sample recovery yield of this method is limited by the amount of the starting material. While ∼100 ng of digested protein is sufficient for thorough protein identification, proteomic information gets lost with a protein content <10 μg due to incomplete peptide recovery from the filter. We developed and optimized a flexible well-plate μFASP device and protocol that is suitable for an ∼1 μg protein sample. In 1 μg of HeLa digest, we identified 1295 ± 10 proteins with μFASP followed by analysis with liquid chromatography-mass spectrometry. In contrast, only 524 ± 5 proteins were identified with the standard FASP protocol, while 1395 ± 4 proteins were identified in 20 μg after standard FASP as a benchmark. Furthermore, we conducted a combined peptidomic and proteomic study of single pancreatic islets with well-plate μFASP. Here, we separated neuropeptides and digested the remaining on-filter proteins for bottom-up proteomic analysis. Our results indicate inter-islet heterogeneity for the expression of proteins involved in glucose catabolism, pancreatic hormone processing, and secreted peptide hormones. We consider our method to provide a useful tool for proteomic characterization of samples where the biological material is scarce. All proteomic data are available under DOI: 10.6019/PXD029039.

Machine learning is increasingly applied in proteomics and metabolomics to predict molecular structure, function, and physicochemical properties, including behavior in chromatography, ion mobility, and tandem mass spectrometry. These must be described in sufficient detail to apply or evaluate the performance of trained models. Here we look at and interpret the recently published and general DOME (Data, Optimization, Model, Evaluation) recommendations for conducting and reporting on machine learning in the specific context of proteomics and metabolomics.

It is important for the proteomics community to have a standardized manner to represent all possible variations of a protein or peptide primary sequence, including natural, chemically induced, and artifactual modifications. The Human Proteome Organization Proteomics Standards Initiative in collaboration with several members of the Consortium for Top-Down Proteomics (CTDP) has developed a standard notation called ProForma 2.0, which is a substantial extension of the original ProForma notation developed by the CTDP. ProForma 2.0 aims to unify the representation of proteoforms and peptidoforms. ProForma 2.0 supports use cases needed for bottom-up and middle-/top-down proteomics approaches and allows the encoding of highly modified proteins and peptides using a human- and machine-readable string. ProForma 2.0 can be used to represent protein modifications in a specified or ambiguous location, designated by mass shifts, chemical formulas, or controlled vocabulary terms, including cross-links (natural and chemical) and atomic isotopes. Notational conventions are based on public controlled vocabularies and ontologies. The most up-to-date full specification document and information about software implementations are available at http://psidev.info/proforma.

Human milk is a dynamic biofluid, and its detailed composition receives increasing attention. While most studies focus on changes over time or differences between maternal characteristics, interindividual variation receives little attention. Nevertheless, a comprehensive insight into this can help interpret human milk studies and help human milk banks provide targeted milk for recipients. This study aimed to map interindividual variation in the human milk proteome, peptidome, and metabolome and to investigate possible explanations for this variation. A set of 286 milk samples was collected from 29 mothers in the third month postpartum. Samples were pooled per mother, and proteins, peptides, and metabolites were analyzed. A substantial coefficient of variation (>100%) was observed for 4.6% and 36.2% of the proteins and peptides, respectively. In addition, using weighted correlation network analysis (WGCNA), 5 protein and 11 peptide clusters were obtained, showing distinct characteristics. With this, several associations were found between the different data sets and with specific sample characteristics. This study provides insight into the dynamics of human milk protein, peptide, and metabolite composition. In addition, it will support future studies that evaluate the effect size of a parameter of interest by enabling a comparison with natural variability.

Aberrant expression of certain glycosphingolipids (GSLs) is associated with the differentiation of acute myeloid leukemia (AML) cells. However, the expression patterns of GSLs in AML are still poorly explored because of their complexity, the presence of multiple isomeric structures, and tedious analytical procedures. In this study, we performed an in-depth GSL glycan analysis of 19 AML cell lines using porous graphitized carbon liquid chromatography-mass spectrometry revealing strikingly different GSL glycan profiles between the various AML cell lines. The cell lines of the M6 subtype showed a high expression of gangliosides with α2,3-sialylation and Neu5Gc, while the M2 and M5 subtypes were characterized by high expression of (neo)lacto-series glycans and Lewis A/X antigens. Integrated analysis of glycomics and available transcriptomics data revealed the association of GSL glycan abundances with the transcriptomics expression of certain glycosyltransferases (GTs) and transcription factors (TFs). In addition, correlations were found between specific GTs and TFs. Our data reveal TFs GATA2, GATA1, and RUNX1 as candidate inducers of the expression of gangliosides and sialylation via regulation of the GTs ST3GAL2 and ST8SIA1. In conclusion, we show that GSL glycan expression levels are associated with hematopoietic AML classifications and TF and GT gene expression. Further research is needed to dissect the regulation of GSL expression and its role in hematopoiesis and associated malignancies.

Extracellular vesicles (EVs) mediate communication in physiological and pathological conditions. In the pathogenesis of type 2 diabetes, inter-organ communication plays an important role in its progress and metabolic surgery leads to its remission. Moreover, gut dysbiosis is emerging as a diabetogenic factor. However, it remains unclear how the gut senses metabolic alterations and whether this is transmitted to other tissues via EVs. Using a diet-induced prediabetic mouse model, we observed that protein packaging in gut-derived EVs (GDE), specifically the small intestine, is altered in prediabetes. Proteins related to lipid metabolism and to oxidative stress management were more abundant in prediabetic GDE compared to healthy controls. On the other hand, proteins related to glycolytic activity, as well as those responsible for the degradation of polyubiquitinated composites, were depleted in prediabetic GDE. Together, our findings show that protein packaging in GDE is markedly modified during prediabetes pathogenesis, thus suggesting that prediabetic alterations in the small intestine are translated into modified GDE proteomes, which are dispersed into the circulation where they can interact with and influence the metabolic status of other tissues. This study highlights the importance of the small intestine as a tissue that propagates prediabetic metabolic dysfunction throughout the body and the importance of GDE as the messengers. Data are available via ProteomeXchange with identifier PXD028338.

The shell disorder models have predicted that SARS-CoV-2 is of a specific but peculiar evolutionary nature. All coronaviruses (CoVs) closely related to SARS-CoV-2 have been found to have the hardest outer shells (M protein) among CoVs. This hard shell (low M percentage of intrinsic disorder (PID)) is associated with burrowing animals, for example, pangolins, and is believed to be responsible for the high contagiousness of SARS-CoV-2 because it will be more resistant to antimicrobial enzymes found in saliva/mucus. Incoming clinical and experimental data do support this along with a prediction based on another aspect of the shell (N, inner shell) disorder models that SARS-CoV-1 is more virulent than SARS-CoV-2 because SARS-CoV-2 produces fewer virus copies in vital organs even if large amounts of infections particles are shed orally and nasally. A phylogenetic study using M reveals a closer relationship of SARS-CoV to pangolin-CoVs than the bat-RaTG13 found in Yunnan, China. Previous studies may have been confused by recombinations that were poorly handled. The shell disorder models suggest that a pangolin-CoV strain may have entered the human population in 2017 or before as an attenuated virus, which could explain why SARS-CoV is found to be highly adapted to humans.

As novel liquid chromatography-mass spectrometry (LC-MS) technologies for proteomics offer a substantial increase in LC-MS runs per day, robust and reproducible sample preparation emerges as a new bottleneck for throughput. We introduce a novel strategy for positive-pressure 96-well filter-aided sample preparation (PF96) on a commercial positive-pressure solid-phase extraction device. PF96 allows for a five-fold increase in throughput in conjunction with extraordinary reproducibility with Pearson product-moment correlations on the protein level of r = 0.9993, as demonstrated for mouse heart tissue lysate in 40 technical replicates. The targeted quantification of 16 peptides in the presence of stable-isotope-labeled reference peptides confirms that PF96 variance is barely assessable against technical variation from nanoLC-MS instrumentation. We further demonstrate that protein loads of 36-60 μg result in optimal peptide recovery, but lower amounts ≥3 μg can also be processed reproducibly. In summary, the reproducibility, simplicity, and economy of time provide PF96 a promising future in biomedical and clinical research.

For more than a decade, DNA and histone methylations have been the focus of extensive work, although their relationship with methyl group metabolism was overlooked. Recently, it has emerged that epigenetic methylations are influenced by methyl donor nutrient availability, cellular levels of S-adenosyl-methionine (SAM), and cytoplasmic methyltransferase activities. SAM-dependent methyltransferases methylate a wide range of targets, from small molecules to proteins and nucleic acids. However, few investigations of the global methylome of tumors have been performed. Here, untargeted NMR metabolomics of two mouse tumor models labeled with [13C-methyl]methionine were used to search for the NMR-visible set of cellular methyl acceptors denoted the global methylome. Tumor models were B16 melanoma cell cultures and B16 melanoma tumors, which may be considered as two stages of B16 tumor development. Based on 2D 1H-13C NMR spectra and orthogonal partial least squares discriminant analysis of spectra, our study revealed markedly different global methylomes for melanoma models. The methylome of B16 melanoma cell cultures was dominated by histone methylations, whereas that of B16 melanoma tumors was dominated by cytoplasmic small-molecule methylations. Overall, the technique gave access to the non-DNA methylome. Comparison of tumor models also exhibiting differential expression of aerobic glycolysis provided clues to a methyl metabolism shift during tumor progression.