Abundant Plasma Proteins Research Articles

P0156 Plasma immune protein profiles in therapy-naïve ulcerative colitis relate to extent of affected tissue and density of intestinal infiltration, differentiating patients with discreet disease courses

Abstract Background Ulcerative colitis (UC) is a chronic intestinal inflammation driven by inflammatory memory T helper cell (ThM) reactivation. Heterogeneity in clinical disease and variability in response to treatment require new strategies targeting the patient’s underlying immune disease. As inflammation in UC only affects the intestinal mucosa, it is debated whether peripheral blood immune responses could reliably reflect ongoing intestinal immune disease. Methods To unequivocally establish this, we performed a uniquely deep immune characterization of therapy-naïve pediatric UC patients (n=34) and controls (n=55) by measuring 92 plasma immune proteins, circulating ThM subpopulations, intestinal mRNA expression, and intestinal immunohistochemistry. Extensive clinical follow-up was used to relate immune responses to clinical disease severity and therapy response. Results At diagnosis, 23 plasma proteins were differentially abundant in UC compared to controls. Increased protein concentrations mainly related to the IL-17 pathway (IL-17, IL-8, CCL20), tissue remodelling (MMP-10, MMP1, HGF), acute phase response (IL-6, OSM) and chemokinesis (CXCL9, CXCL10, CCL3). Interestingly, intestinal mRNA expression of highly abundant plasma proteins similarly increased, whereas no proportional changes in circulating ThM were detected. Unbiased hierarchical clustering of plasma immune profiles yielded two clusters of UC patients; UC_A with significantly increased concentrations of 17 proteins, many of which Th17/neutrophil-related, versus UC_B with lower concentrations. Notable increases in UC_A were: IL- 17A; OSM, a neutrophil product previously associated with anti-TNF therapy resistance; CXCL9 and CXCL11, IFN-g-induced chemokines. Strikingly, this UC_A immune profile strongly related to increased size of the affected intestinal surface and degree of infiltrating IL-17A+ cells and number of neutrophils in the biopsies. Clinically, UC_A patients had higher clinico-pathological parameters, more severe intestinal disease at endoscopy and on histology , and required earlier treatment intensification during 1.5 year follow-up, compared to UC_B patients. Conclusion These data firmly establish that, even though inflammation in UC only affects the intestinal mucosa, peripheral blood immune protein responses reliably reflect the degree of ongoing intestinal immune disease in UC. Implementing immune profiling in routine clinical characterization at diagnosis improves identification of patients who will develop a severe disease course allowing tailored treatment with earlier start of more intensive immunomodulatory treatment to treat to target and avoid tissue damage.

The Analysis of Plasma Proteomics for Luminal A Breast Cancer.

Breast cancer is the prevailing malignancy among women, exhibiting a discernible escalation in incidence within our nation; hormone receptor-positive (HR+) human epidermal growth factor receptor 2-negative (HER2-) breast cancer is the most common subtype. In this study, we aimed to search for a non-invasive, specific, blood-based biomarker for the early detection of luminal A breast cancer through proteomic studies. To explore new potential plasma biomarkers, we applied data-independent acquisition (DIA), a technique combining liquid chromatography and tandem mass spectrometry, to quantify breast cancer-associated plasma protein abundance from a small number of plasma samples in 10 patients with luminal A breast cancer, 10 patients with benign breast tumors, and 10 healthy controls. The proteomes of 30 participants in all cohorts were analyzed using the DIA method, and a total of 517 proteins and 3584 peptides were quantified. We found that there were significant differences in plasma protein expression profiles between breast cancer patients and non-breast cancer patients, and breast cancer was mainly related to lipid metabolism pathways. Finally, the optimal protein combinations for the diagnosis of breast cancer were PON3, IGLV3-10, and IGHV3-73 through multi-model analysis, which had a high prediction accuracy for breast cancer (AUC = 0.92), and the model could also distinguish breast cancer from HC (AUC = 0.92) and breast cancer from benign breast tumor (AUC = 0.91). The study revealed proteomic signatures of patients with luminal A breast cancer, identified multiple differential proteins, and identified three plasma proteins as potential diagnostic biomarkers for breast cancer. It provides a reference for the screening of biomarkers for breast cancer.

Small molecule modulation of protein corona for deep plasma proteome profiling

The protein corona formed on nanoparticles (NPs) has potential as a valuable diagnostic tool for improving plasma proteome coverage. Here, we show that spiking small molecules, including metabolites, lipids, vitamins, and nutrients into plasma can induce diverse protein corona patterns on otherwise identical NPs, significantly enhancing the depth of plasma proteome profiling. The protein coronas on polystyrene NPs when exposed to plasma treated with an array of small molecules allows for the detection of 1793 proteins marking an 8.25-fold increase in the number of quantified proteins compared to plasma alone (218 proteins) and a 2.63-fold increase relative to the untreated protein corona (681 proteins). Furthermore, we discovered that adding 1000 µg/ml phosphatidylcholine could singularly enable the detection of 897 proteins. At this specific concentration, phosphatidylcholine selectively depletes the four most abundant plasma proteins, including albumin, thus reducing the dynamic range of plasma proteome and enabling the detection of proteins with lower abundance. Employing an optimized data-independent acquisition approach, the inclusion of phosphatidylcholine leads to the detection of 1436 proteins in a single plasma sample. Our molecular dynamics results reveal that phosphatidylcholine interacts with albumin via hydrophobic interactions, H-bonds, and water bridges. The addition of phosphatidylcholine also enables the detection of 337 additional proteoforms compared to untreated protein corona using a top-down proteomics approach. Given the critical role of plasma proteomics in biomarker discovery and disease monitoring, we anticipate the widespread adoption of this methodology for the identification and clinical translation of biomarkers.

Identification of fibrinogen as a plasma protein binding partner for lecanemab biosimilar IgG.

Recombinant monoclonal therapeutic antibodies like lecanemab, which target amyloid beta in Alzheimer's disease, offer a promising approach for modifying the disease progression. Due to its relatively short half-life, lecanemab administered as a bi-monthly infusion (typically 10 mg/kg) has a relatively brief half-life. Interaction with abundant plasma proteins binder in the bloodstream can affect pharmacokinetics of drugs, including their half-life. In this study, we investigated potential plasma protein binding (PPB) interaction to lecanemab using lecanemab biosimilar. Lecanemab biosimilar used in this study was based on publicly available sequences. ELISA and western blotting were used to assess lecanemab biosimilar immunoreactivity in the fractions of human plasma obtained through size exclusion chromatography. The binding of lecanemab biosimilar to candidate plasma binders was confirmed by western blotting, ELISA, and surface plasmon resonance analysis. Using a combination of equilibrium dialysis, ELISA, and western blotting in human plasma, we first describe the presence of likely PPB partners to lecanemab biosimilar and then identify fibrinogen as one of them. Utilizing surface plasmon resonance, we confirmed that lecanemab biosimilar does bind to fibrinogen, although with lower affinity than to monomeric amyloid beta. In the context of lecanemab therapy, these results imply that fibrinogen levels could impact the levels of free antibodies in the bloodstream and that fibrinogen might serve as a reservoir for lecanemab. More broadly, these results indicate that PPB may be an important consideration when clinically utilizing therapeutic antibodies in neurodegenerative disease.

Carrier-Guided Proteome Analysis in a High Protein Background: An Improved Approach to Host Cell Protein Identification.

Many shotgun proteomics experiments are negatively influenced by highly abundant proteins, such as those measuring residual host cell proteins (HCP) amidst highly abundant recombinant biotherapeutic or plasma proteins amidst albumin and immunoglobulins. While western blotting and ELISAs can reveal the presence of specific low abundance proteins from highly abundant background proteins, mass spectrometry approaches are required to define the low abundance protein composition in these scenarios. The challenge in detecting low abundance proteins in a high protein background by standard shotgun approaches is that spectra are often not triggered on their peptides in data dependent acquisition methods but rather on the highly abundant background peptides. Here, we use tandem mass tags (TMT) to introduce a carrier proteome approach to enhance the detection of proteins, such as from residual host cell proteomes amidst a highly abundant background. Using a mixture of bovine serum albumin (BSA) and E. coli as a mock high background/low abundance target protein formulation, we demonstrate proof-of-principle experiments allowing the improved detection of target proteins amidst a high protein background. While we observed significant coisolation interference, we mitigated it by using a spike-in interference detection TMT channel. Finally, we use the approach to identify 300 residual E. coli proteins from a protein A pulldown of a human IgG antibody, demonstrating that it may be applicable to analysis of HCPs in biotherapeutic protein formulations.

Proteomic Evidence for Amyloidogenic Cross-Seeding in Fibrinaloid Microclots

In classical amyloidoses, amyloid fibres form through the nucleation and accretion of protein monomers, with protofibrils and fibrils exhibiting a cross-β motif of parallel or antiparallel β-sheets oriented perpendicular to the fibre direction. These protofibrils and fibrils can intertwine to form mature amyloid fibres. Similar phenomena can occur in blood from individuals with circulating inflammatory molecules (and also some originating from viruses and bacteria). Such pathological clotting can result in an anomalous amyloid form termed fibrinaloid microclots. Previous proteomic analyses of these microclots have shown the presence of non-fibrin(ogen) proteins, suggesting a more complex mechanism than simple entrapment. We thus provide evidence against such a simple entrapment model, noting that clot pores are too large and centrifugation would have removed weakly bound proteins. Instead, we explore whether co-aggregation into amyloid fibres may involve axial (multiple proteins within the same fibril), lateral (single-protein fibrils contributing to a fibre), or both types of integration. Our analysis of proteomic data from fibrinaloid microclots in different diseases shows no significant quantitative overlap with the normal plasma proteome and no correlation between plasma protein abundance and their presence in fibrinaloid microclots. Notably, abundant plasma proteins like α-2-macroglobulin, fibronectin, and transthyretin are absent from microclots, while less abundant proteins such as adiponectin, periostin, and von Willebrand factor are well represented. Using bioinformatic tools, including AmyloGram and AnuPP, we found that proteins entrapped in fibrinaloid microclots exhibit high amyloidogenic tendencies, suggesting their integration as cross-β elements into amyloid structures. This integration likely contributes to the microclots’ resistance to proteolysis. Our findings underscore the role of cross-seeding in fibrinaloid microclot formation and highlight the need for further investigation into their structural properties and implications in thrombotic and amyloid diseases. These insights provide a foundation for developing novel diagnostic and therapeutic strategies targeting amyloidogenic cross-seeding in blood clotting disorders.

Trypsin Digestion Conditions of Human Plasma for Observation of Peptides and Proteins from Tandem Mass Spectrometry.

Previous meta-analysis indicated that plasma or serum proteome groups using various experimental conditions detected different peptides from the same plasma proteins, which is strong evidence for the veracity of blood fluid LC-ESI-MS/MS but also evidences that the trypsin digestion step is a key source of variation in plasma proteomics. Agreement between different digestion conditions and MS/MS algorithms may serve as an independent confirmation of the validity of the LC-ESI-MS/MS analysis of plasma peptides. Plasma contains a high percentage of albumin held together by multiple disulfide bonds; hence, reduction and/or alkylation may greatly enhance the digestion efficiency of albumin. Plasma proteins were precipitated in 90% acetonitrile, collected over quaternary amine resin, and eluted in NaCl prior to digestion treatments. To determine the effect of trypsin digestion methods, the plasma proteins were digested in 600 mM urea and 5% acetonitrile with trypsin alone, or reduced with 2 mM DTT followed by trypsin, or DTT followed by 15 mM iodoacetamide and then trypsin. The resulting peptides were analyzed by LC-ESI-MS/MS with a linear quadrupole ion trap (LIT). The MS/MS spectra were directly fit to peptides by the X!TANDEM and SEQUEST algorithms. Blank noise injections served as the analytical control, and 30 million random MS/MS served as the statistical control. Digesting human plasma with DTT reduction, or reduction and alkylation, resulted in a dramatic increase in the number and observation frequency of albumin peptides. In contrast, digestion with trypsin alone suppressed the observation of albumin, and instead, many low abundance plasma and cellular proteins showed higher observation frequency. Digestion with trypsin alone increased the observation frequency of APOC1, ACAN, ATRN, CPB2, GP2, GPX3, HBA1, PAPD5, PKD1, and many cellular proteins. After correction against noise and random controls, SEQUEST showed good agreement with the true positive plasma proteins identified by X!TANDEM and resulted in an R-squared of 0.5238 with an F-statistic of 10,930 on 9,935 protein gene symbols with a p-value < 2.2e-16. Digestion of plasma with trypsin alone avoids the complete digestion of albumin and permits the enhanced detection of some other cellular proteins from plasma. Different digestion approaches were complimentary and together resulted in a more comprehensive plasma proteome. The protein FDR q-values, the modest effect of background and Monte Carlo correction, and the significant STRING analysis were all consistent with the high fidelity of the rigorous X!TANDEM algorithm. In contrast, SEQUEST required significant correction against noise and statistical controls and selection of high cross correlation (XCorr) scores to show good agreement with X!TANDEM. There was qualitative and quantitative agreement between plasma proteins digested without alkylation from the orbital ion trap (OIT) versus the LIT instrument that showed highly significant regression against the X!TANDEM OIT monoisotopic results, those from heavy isotopes and other masses from X!TANDEM, and with those from MaxQuant. There was significant qualitative and quantitative agreement between the complementary digestion conditions consistent with the good fidelity of plasma analysis by LC-ESI-MS/MS with a sensitive linear ion trap.

Trypsin Partially Cleaves Apolipoprotein A-I (ApoA-I) Precursor into Mature ApoA-I Hindering the Quantification of Naturally Occurring ApoA-I Proteoforms by Liquid Chromatography in Multiple Reaction Monitoring Mode Mass Spectrometry (LC-MRM-MS).

Apolipoprotein A-I (ApoA-I), one of the most abundant proteins in plasma and the major protein component of high-density lipoprotein (HDL), is naturally found in several proteoforms; two of them are ProApoA-I and mature ApoA-I. These two proteoforms of ApoA-I coexist in biological samples and differ only in their N-terminal end. Virtually, the only way to differentiate them is by detecting the proteoform-specific N-terminal proteolytic peptides (RHFWQQDEPPQSPWDR and DEPPQSPWDR, respectively) using liquid chromatography in multiple reaction monitoring mode mass spectrometry (LC-MRM-MS). We have developed a bottom-up LC-MRM-MS method to simultaneously detect proApoA-I and mature ApoA-I. To test the specificity of the method, we digested with trypsin purified mature ApoA-I and recombinant proApoA-I. As expected, only the N-term peptide corresponding to the mature ApoA-I proteoform (DEPPQSPWDR) was detected when digesting mature ApoA-I. However, the digestion of the proApoA-I produced not only the N-terminal peptide corresponding to proApoA-I (RHFWQQDEPPQSPWDR) but also the N-terminal tryptic peptide corresponding to mature ApoA-I (DEPPQSPWDR). This effect was produced by standard and high-specificity trypsin as well as by the Arg-C enzyme in a self-limited manner (approximately 10% of the total). The synthetic proApo-I peptide is not cleaved by trypsin, suggesting that the here reported effect is dependent on protein conformation. The effect is not negligible, as it can be detected by LC-MRM-MS, and correction calculations should be applied to accurately quantify proApoA-I and mature ApoA-I in biological samples where these two proteoforms may coexist.

Deep Plasma Proteome Profiling by Modulating Single Nanoparticle Protein Corona with Small Molecules.

The protein corona, a dynamic biomolecular layer that forms on nanoparticle (NP) surfaces upon exposure to biological fluids is emerging as a valuable diagnostic tool for improving plasma proteome coverage analyzed by liquid chromatography-mass spectrometry (LC-MS/MS). Here, we show that spiking small molecules, including metabolites, lipids, vitamins, and nutrients (namely, glucose, triglyceride, diglycerol, phosphatidylcholine, phosphatidylethanolamine, L-α-phosphatidylinositol, inosine 5'-monophosphate, and B complex), into plasma can induce diverse protein corona patterns on otherwise identical NPs, significantly enhancing the depth of plasma proteome profiling. The protein coronas on polystyrene NPs when exposed to plasma treated with an array of small molecules (n=10) allowed for detection of 1793 proteins marking an 8.25-fold increase in the number of quantified proteins compared to plasma alone (218 proteins) and a 2.63-fold increase relative to the untreated protein corona (681 proteins). Furthermore, we discovered that adding 1000 μg/ml phosphatidylcholine could singularly enable the detection of 897 proteins. At this specific concentration, phosphatidylcholine selectively depleted the four most abundant plasma proteins, including albumin, thus reducing the dynamic range of plasma proteome and enabling the detection of proteins with lower abundance. By employing an optimized data-independent acquisition (DIA) approach, the inclusion of phosphatidylcholine led to the detection of 1436 proteins in a single plasma sample. Our molecular dynamic results revealed that phosphatidylcholine interacts with albumin via hydrophobic interactions, h-bonds, and water-bridges. Addition of phosphatidylcholine also enabled the detection of 337 additional proteoforms compared to untreated protein corona using a top-down proteomics approach. These significant achievements are made utilizing only a single NP type and one small molecule to analyze a single plasma sample, setting a new standard in plasma proteome profiling. Given the critical role of plasma proteomics in biomarker discovery and disease monitoring, we anticipate widespread adoption of this methodology for identification and clinical translation of proteomic biomarkers into FDA approved diagnostics.

Optimized workflow of EV enrichment from human plasma samples for downstream mass spectrometry analysis

To improve the prognosis of bladder and prostate cancer, highly specific and sensitive biomarkers are needed for early detection, prognosis prediction, and therapeutic stratification. Extracellular vesicles (EV) from plasma could fill this gap due to their potential to serve as cancer biomarkers. However, the enrichment of EV is a major challenge, because the highly abundant plasma proteins are interfering with analytical downstream applications like mass spectrometry (MS). Therefore, the purity requirements of the EV samples must be carefully considered when selecting or developing a suitable EV enrichment method. The aim of this study was to compare a self-designed EV enrichment method based on density cushion centrifugation (DCC) combined with size exclusion chromatography (SEC) and concentration (method 1) with the exoRNeasy midi kit from Qiagen (method 2) and with unprocessed plasma. Furthermore, the single steps of method 1 were evaluated for their effectiveness to enrich EV from plasma. The results showed that the EV samples enriched with method 1 contained the highest levels of EV and exosome markers with simultaneously low levels of highly abundant plasma proteins. In summary, the combination of DCC, SEC and concentration proved to be a promising approach to discover EV-based biomarkers from plasma of cancer patients.

Optimized AF4 combined with density cushion ultracentrifugation enables profiling of high-purity human blood extracellular vesicles.

Extracellular vesicles (EVs) have emerged as a promising tool for clinical liquid biopsy. However, the identification of EVs derived from blood samples is hindered by the presence of abundant plasma proteins, which impairs the downstream biochemical analysis of EV-associated proteins and nucleic acids. Here, we employed optimized asymmetric flow field-flow fractionation (AF4) combined with density cushion ultracentrifugation (UC) to obtain high-purity and intact EVs with very low lipoprotein contamination from human plasma and serum. Further proteomic analysis revealed more than 1000 EV-associated proteins, a large proportion of which has not been previously reported. Specifically, we found that cell-line-derived EV markers are incompatible with the identification of plasma-EVs and proposed that the proteins MYCT1, TSPAN14, MPIG6B and MYADM, as well as the traditional EV markers CD63 and CD147, are plasma-EV markers. Benefiting from the high-purity of EVs, we conducted comprehensive miRNA profiling of plasma EVs and nanosized particles (NPs), as well as compared plasma- and serum-derived EVs, which provides a valuable resource for the EV research community. Overall, our findings provide a comprehensive assessment of human blood EVs as a basis for clinical biopsy applications.

Comparing the Proteomic Profiles of Extracellular Vesicles Isolated using Different Methods from Long-term Stored Plasma Samples

BackgroundThe lack of standardized protocols for isolating extracellular vesicles (EVs), especially from biobank-stored blood plasma, translates to limitations for the study of new biomarkers. This study examines whether a combination of current isolation methods could enhance the specificity and purity of isolated EVs for diagnosis and personalized medicine purposes.ResultsEVs were isolated from healthy human plasma stored for one year by ultracentrifugation (UC), size exclusion chromatography (SEC), or SEC and UC combined (SEC + UC). The EV isolates were then characterized by transmission electron microscopy imaging, nanoparticle tracking analysis (NTA) and western blotting. Proteomic procedures were used to analyze protein contents. The presence of EV markers in all isolates was confirmed by western blotting yet this analysis revealed higher albumin expression in EVs-UC, suggesting plasma protein contamination. Proteomic analysis identified 542 proteins, SEC + UC yielding the most complex proteome at 364 proteins. Through gene ontology enrichment, we observed differences in the cellular components of EVs and plasma in that SEC + UC isolates featured higher proportions of EV proteins than those derived from the other two methods. Analysis of proteins unique to each isolation method served to identify 181 unique proteins for the combined approach, including those normally appearing in low concentrations in plasma. This indicates that with this combined method, it is possible to detect less abundant plasma proteins by proteomics in the resultant isolates.ConclusionsOur findings reveal that the SEC + UC approach yields highly pure and diverse EVs suitable for comprehensive proteomic analysis with applications for the detection of new biomarkers in biobank-stored plasma samples.

Proteomics of human platelet lysates and insight from animal studies on platelet protein diffusion to hippocampus upon intranasal administration.

Human platelet lysates (HPLs) from allogeneic platelet concentrates (PCs) are biomaterials, which are rich in various trophic factors, increasingly used in regenerative medicine and biotherapy. Understanding how preparation methods influence the HPL protein profile, biological function, and clinical outcomes is crucial. Our study sheds light on the proteomes and functionality of different HPLs, with the aim of advancing their scientifically grounded clinical applications. To achieve this, PCs suspended in plasma underwent three distinct processing methods, resulting in seven HPL types. We used three characterization techniques: label-free proteomics and tandem mass tag (TMT)-based quantitative proteomics, both before and after the immunodepletion of abundant plasma proteins. Bioinformatic tools assessed the proteome, and western blotting validated our quantitative proteomics data. Subsequent pre-clinical studies with fluorescent labeling and label-free proteomics were used as a proof of concept for brain diffusion. Our findings revealed 1441 proteins detected using the label-free method, 952 proteins from the TMT experiment before and after depletion, and 1114 proteins from the subsequent TMT experiment on depleted HPLs. Most detected proteins were cytoplasmic, playing key roles in catalysis, hemostasis, and immune responses. Notably, the processing methodologies significantly influenced HPL compositions, their canonical pathways, and, consequently, their functionality. Each HPL exhibited specific abundant proteins, providing valuable insight for tailored clinical applications. Immunoblotting results for selected proteins corroborated our quantitative proteomics data. The diffusion and differential effects to the hippocampus of a neuroprotective HPL administered intranasally to mice were demonstrated. This proteomics study advances our understanding of HPLs, suggesting ways to standardize and customize their production for better clinical efficacy in regenerative medicine and biotherapy. Proteomic analyses also offered objective evidence that HPPL, upon intranasal delivery, not only effectively diffuses to the hippocampus but also alters protein expression in mice, bolstering its potential as a treatment for memory impairments.

High-efficient depletion and separation of histidine-rich proteins via Cu2+-chelated porous polymer microspheres

Depletion and separation of histidine-rich proteins from complicated biosamples are crucial for various downstream applications in proteome research and clinical diagnosis. Herein, porous polymer microspheres coated with polyacrylic acid (SPSDVB-PAA) were fabricated through double emulsion interfacial polymerization technique and followed by immobilization of Cu2+ ions on the surface of SPSDVB-PAA. The as-prepared SPSDVB-PAA-Cu with uniform size and nanoscale pore structure enabled coordination interaction of Cu2+ with histidine residues in his-rich proteins, resulting in high-performance adsorption. As metal affinity adsorbent, the SPSDVB-PAA-Cu exhibited favorable selectivity for adsorbing hemoglobin (Hb) and human serum albumin (HSA) with the maximum adsorption capacities of 152.2 and 100.7 mg g−1. Furthermore, the polymer microspheres were used to isolate histidine-rich proteins from human whole blood and plasma, underscoring their effectiveness. The liquid chromatography tandem mass spectrometry (LC-MS/MS) results indicated that the content of 14 most abundant proteins in human plasma was depleted from 81.6 % to 30.7 % and low-abundance proteins were enriched from 18.4 % to 69.3 % after treatment with SPSDVB-PAA-Cu, illustrating potential application of SPSDVB-PAA-Cu in proteomic research.

Longitudinal plasma proteomic analysis of 1117 hospitalized patients with COVID-19 identifies features associated with severity and outcomes.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is characterized by highly heterogeneous manifestations ranging from asymptomatic cases to death for still incompletely understood reasons. As part of the IMmunoPhenotyping Assessment in a COVID-19 Cohort study, we mapped the plasma proteomes of 1117 hospitalized patients with COVID-19 from 15 hospitals across the United States. Up to six samples were collected within ~28 days of hospitalization resulting in one of the largest COVID-19 plasma proteomics cohorts with 2934 samples. Using perchloric acid to deplete the most abundant plasma proteins allowed for detecting 2910 proteins. Our findings show that increased levels of neutrophil extracellular trap and heart damage markers are associated with fatal outcomes. Our analysis also identified prognostic biomarkers for worsening severity and death. Our comprehensive longitudinal plasma proteomics study, involving 1117 participants and 2934 samples, allowed for testing the generalizability of the findings of many previous COVID-19 plasma proteomics studies using much smaller cohorts.

Proteomic changes in the human cerebrovasculature in Alzheimer's disease and related tauopathies linked to peripheral biomarkers in plasma and cerebrospinal fluid.

Cerebrovascular dysfunction is a pathological hallmark of Alzheimer's disease (AD). Nevertheless, detecting cerebrovascular changes within bulk tissues has limited our ability to characterize proteomic alterations from less abundant cell types. We conducted quantitative proteomics on bulk brain tissues and isolated cerebrovasculature from the same individuals, encompassing control (N=28), progressive supranuclear palsy (PSP) (N=18), and AD (N=21) cases. Protein co-expression network analysis identified unique cerebrovascular modules significantly correlated with amyloid plaques, cerebrovascular amyloid angiopathy (CAA), and/or tau pathology. The protein products within AD genetic risk loci were concentrated within cerebrovascular modules. The overlap between differentially abundant proteins in AD cerebrospinal fluid (CSF) and plasma with cerebrovascular network highlighted a significant increase of matrisome proteins, SMOC1 and SMOC2, in CSF, plasma, and brain. These findings enhance our understanding of cerebrovascular deficits in AD, shedding light on potential biomarkers associated with CAA and vascular dysfunction in neurodegenerative diseases.

Effect of the 35 nm and 70 nm Size Exclusion Chromatography (SEC) Column and Plasma Storage Time on Separated Extracellular Vesicles.

The technical difficulty of separating extracellular vesicles (EVs) from plasma proteins in human blood presents a significant hurdle in EV research, particularly during nano ultra-high-performance liquid chromatography-tandem mass spectrometric (UHPLC-MS/MS) analysis, where detecting "vesicular" proteins among abundant plasma proteins is challenging. Standardisation is a pressing issue in EV research, prompting collaborative global efforts to address it. While the MISEV guidelines offer valuable recommendations, unanswered questions remain, particularly regarding sample storage. We compared size exclusion chromatography (SEC) columns with pore sizes of 35 nm and 70 nm to identify fractions with minimal contaminating proteins and the highest concentration of small EVs (sEVs). Following column selection, we explored potential differences in the quality and quantity of sEVs isolated from platelet-free plasma (PFP) after long-term storage at -80 °C (>2.5 years) compared to freshly drawn blood. Our methodologically rigorous study indicates that prolonged storage, under correct storage and processing conditions, does not compromise sEV quality. Both columns effectively isolated vesicles, with the 70 nm column exhibiting a higher abundance of "vesicular" proteins. We propose a relatively rapid and moderately efficient protocol for obtaining a comparatively pure sEV fraction from plasma, facilitating sEV processing in clinical trials.

Characterization of effective, simple, and low-cost precipitation methods for depleting abundant plasma proteins to enhance the depth and breadth of plasma proteomics.

Plasma is an abundant source of proteins and potential biomarkers to aid in the detection, diagnosis, and prognosis of human diseases. These proteins are often present at low levels in the blood and difficult to identify and measure due to the large dynamic range of proteins. The goal of this work was to characterize and compare various protein precipitation methods related to how they affect the depth and breadth of plasma proteomic studies. Abundant protein precipitation with perchloric acid (PerCA) can increase protein identifications and depth of plasma proteomic studies. Three acid- and four solvent-based precipitation methods were evaluated. All methods tested provided excellent plasma proteomic coverage (>600 identified protein groups) and detected protein in the low pg/mL range. Functional enrichment analysis revealed subtle differences within and larger changes between the precipitant groups. Methanol-based precipitation outperformed the other methods based on identifications and reproducibility. The methods' performance was verified using eight lung cancer patient samples, where >700 protein groups were measured and proteins with an estimated plasma concentration of ∼10pg/mL were detected. Various protein precipitation agents are amenable to extending the depth and breadth of plasma proteomes. These data can guide investigators to implement inexpensive, high-throughput methods for their plasma proteomic workflows.

Agnostic identification of plasma biomarkers for postpartum hemorrhage risk

Postpartum hemorrhage is difficult to predict, is associated with significant maternal morbidity, and is the leading cause of maternal mortality worldwide. The identification of maternal biomarkers that can predict increased postpartum hemorrhage risk would enhance clinical care and may uncover mechanisms that lead to postpartum hemorrhage. This retrospective case-control study employed agnostic proteomic profiling of maternal plasma samples to identify differentially abundant proteins in controls and postpartum hemorrhage cases. Maternal plasma samples were procured from a cohort of >60,000 participants in a single institution's perinatal repository. Postpartum hemorrhage was defined as a decrease in hematocrit of ≥10% or receipt of transfusion within 24 hours after delivery. Postpartum hemorrhage cases (n=30) were matched by maternal age and delivery mode (vaginal or cesarean) with controls (n=56). Mass spectrometry was used to identify differentially abundant proteins using integrated peptide peak areas. Statistically significant differences between groups were defined as P<.05 after controlling for multiple comparisons. By study design, cases and controls did not differ in race, ethnicity, gestational age at delivery, blood type, or predelivery platelet count. Cases had slightly but significantly lower predelivery and postdelivery hematocrit and hemoglobin. Mass spectrometry detected 1140 proteins, including 77 proteins for which relative abundance differed significantly between cases and controls (fold change >1.15, P<.05). Of these differentially abundant plasma proteins, most had likely liver or placental origins. Gene ontology term analysis mapped to protein clusters involved in responses to wound healing, stress response, and host immune defense. Significantly differentially abundant proteins with the highest fold change (prostaglandin D2 synthase, periostin, and several serine protease inhibitors) did not correlate with predelivery hematocrit or hemoglobin but identified postpartum hemorrhage cases with logistic regression modeling revealing good-to-excellent area under the operator receiver characteristic curves (0.802-0.874). Incorporating predelivery hemoglobin with these candidate proteins further improved the identification of postpartum hemorrhage cases. Agnostic analysis of maternal plasma samples identified differentially abundant proteins in controls and postpartum hemorrhage cases. Several of these proteins are known to participate in biologically plausible pathways for postpartum hemorrhage risk and have potential value for predicting postpartum hemorrhage. These findings identify candidate protein biomarkers for future validation and mechanistic studies.

Abstract 1854: Plasma proteomic biomarkers identify non-responders and reveal biological insights about the tumor microenvironment in melanoma patients after PD1 blockade

Abstract Most patients treated with immune checkpoint blockade (ICB) do not have durable treatment responses, stressing a critical need to identify early non-invasive biomarkers of response. Circulating biomarkers provide easy access for serial monitoring and can provide insight on the mechanisms of response to ICB. We performed plasma proteomics (~3000 proteins) on 250 metastatic melanoma patients before and on ICB treatment and performed analysis to identify response- and time point-associated peripheral protein biomarkers. We next generated patient-matched peripheral blood lymphocyte (PBL) surface proteomic data and tumor sample bulk transcriptomic data to build a multimodal dataset. We leveraged these data to train machine learning models for unsupervised analysis and to classify treatment responses and learned features predictive of survival and patient endotypes. Using univariate linear models to predict protein abundance based on time point, response and the interaction term for time point and response, we identified 343 time point-associated proteins, 141 response-associated proteins, and one interaction-associated protein. We further investigated response-related proteins by fitting multivariate logistic regression and cox proportional hazards regression models to further investigate the ability of these proteins to predict response, progression-free survival (PFS), and overall survival (OS) and identify those proteins that are most important in predicting response. We calculated covarying modules of proteins that were found to be either response-associated or time point-associated to discover underlying biological networks at different treatment time points and disease response states. Additionally, we leveraged these modules to endotype patients and identified trends in protein modules across patient subsets, including modules related to biological processes such as immune infiltration, apoptosis, metaplasia and cell adhesion. To understand the contribution of different compartments on plasma proteomic abundance, we studied correlations of plasma protein abundance with PBL surface protein abundance and tumor bulk mRNA-sequencing expression. Finally, we integrated our data modalities and fit various statistical models to evaluate the relative importance of plasma protein biomarkers in predicting response compared to PBL surface proteins and tumor bulk mRNA-sequencing expression. Together, these data represent one of the deepest peripheral biomarker studies using paired samples in melanoma patients treated with anti-PD1 therapy. Citation Format: Samuel J. Wright, Deepika Yeramosu, Milan Parikh, Marijana Rucevic, Ngan Nguyen, Russell W. Jenkins, Keith T. Flaherty, Nir Hacohen, Genevieve M. Boland, Arnav Mehta. Plasma proteomic biomarkers identify non-responders and reveal biological insights about the tumor microenvironment in melanoma patients after PD1 blockade [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1854.