Journal of Proteome Research最新文献

Thermal proteome profiling (TPP) and its higher-throughputderivative,the proteome integral solubility alteration (PISA) assay, measurechanges in protein thermal stability upon ligand binding or otherperturbations and have been widely adopted in drug discovery and biomedicalresearch. Though the PISA workflow is straightforward, key parameters,including detergent concentration, methods for removing denaturedaggregates, and temperature range selection, vary across studies andcan markedly influence assay outcomes. Yet these factors have notbeen systematically evaluated, limiting rational experimental designand data interpretation. Here, through a combined use of TPP, PISA,tandem mass tag (TMT)-based multiplexing, and computational simulation,we systematically characterize these parameters based on the meltingbehavior of ∼9000 proteins. We find that reducing detergentconcentration elevates apparent Tm by 1.5–2 °C proteome-wide,and aggregate removal by filtration versus centrifugation furtheralters measurements. We leverage these observations to characterizehow these parameters shape PISA and then apply selected conditionsto identify the aminopeptidase NPEPPS as a previously uncharacterizedbinding partner of angiotensin II, a key vasoactive peptide hormonein blood pressure regulation. Together, this work provides a generalframework for assay design and data interpretation and extends theutility of PISA beyond small molecules to dissecting peptide–proteininteractions, an increasingly important modality in drug discovery.

Precision-cut liver slices (PCLSs) are an ex vivo liver model that allows isolation of extracellular vesicles(EVs),which are nano-sized particles involved in intercellular communication.However, EV recovery from PCLSs is hindered by adhesion to the extracellularmatrix and macromolecules such as extracellular DNA. Enzymatic treatments,such as collagenase (COL) and the endonuclease benzonase (BEN), canaid their release, but potential effects on the tissue are unclear.Here, the effect of these enzymes separately, as well as combined,was analyzed to determine their effects on morphology, protein content,and the proteome of the PCLS. The enzymes showed minimal effects oncell viability and protein content after two different incubationtimes, whereas morphological analysis confirmed collagen degradationby COL and COL(BEN), along with decreased ribosomal and secretorysecretory protein levels. BEN decreased the levels of proteins relatedto protein synthesis and degradation. Increased levels of proteinsrelated to detoxification and energy production were observed uponaddition of COL, BEN, and COL(BEN). This study emphasizes the potentialand risks for the use of collagenase and benzonase for EV isolationfrom the PCLS.

Melanomais a highly aggressive skin cancer with strong metastaticpotential and frequent chemoresistance. Aberrant N-glycosylation plays a critical role in tumor progression and drugresistance; however, cisplatin-induced alterations in melanoma surface N-glycans remain poorly understood. Thus, this study aimedto perform comprehensive structural and quantitative profiling ofsurface N-glycans in G-361 (low metastatic potential),A375P (moderate), and SK-MEL-28 (high) melanoma cells following cisplatintreatment using LC–MS/MS. A total of 25, 39, and 28 surface N-glycans were identified in untreated G-361, A375P, andSK-MEL-28 cells, respectively, whereas 40, 37, and 40 N-glycans were detected after cisplatin treatment. Absolute quantificationrevealed a marked increase in total N-glycan abundancein G-361 (3.2 to 8.4 pmol) and A375P (5.7 to 10.9 pmol), whereas SK-MEL-28maintained homeostasis (3.9 to 4.0 pmol). Surface N-glycans were predominantly high-mannose and complex-type glycanswith extensive sialylation and fucosylation. Cisplatin induced distinctcell line-specific remodeling: G-361 accumulated high-mannose glycanswith reduced Neu5Gc-containing sialylation; A375P showed increasedglycan branching and Neu5Ac-containing sialylation; and SK-MEL-28maintained a stable glycan distribution but exhibited increased Neu5Gc-containingsialylation. These results demonstrate adaptive N-glycosylation responses to cisplatin in melanoma cells with distinctmetastatic potentials and provide a glycomic framework for understandingcisplatin-associated glycan alterations.

Accurate characterization of host cell proteins (HCPs)is criticalfor biologics development, yet analytical outcomes are strongly influencedby the digestion strategy and chromatographic design. Here, we systematicallyevaluated digestion parameters, including digestion mode, enzyme ratio,and incubation time, together with LC-MS gradient architecture usinga CHO null harvest as a controlled model system to isolate workflow-dependentanalytical effects from biological variability. Specifically, thestudy was designed to determine how these variables jointly influenceHCP detection and the physicochemical and biological characteristicsof the detected protein populations. By comparing reducing and nonreducingdigestion approaches across multiple trypsin ratios and incubationtimes, we found that under low-enzyme conditions, the primary constrainton HCP recovery is insufficient proteolysis rather than chromatographicundersampling. Reconfiguration of the gradient architecture alteredpeptide sampling efficiency and MS/MS utilization, enabling controlledcomparison of digestion workflows under equivalent acquisition constraints.While higher enzyme ratios increased overall HCP identifications,subcellular localization analysis revealed that low-enzyme nativedigestion preferentially detected secreted and extracellular proteinswith limited access to intracellular species. Extension of nativedigestion partially restored intracellular protein detection but didnot fully recapitulate the coverage achieved by higher enzyme workflows.These findings demonstrate that digestion accessibility and chromatographicspacing impose workflow-dependent effects that shape both the depthand biological origin of detected HCPs, underscoring the need to considerthe digestion strategy and chromatographic design together in HCPcharacterization workflows.

Unequivocal identification and reproducible quantitationof proteinsare key components of research and development activity in drug discovery.In some cases, the target protein might be a protein highly conservedacross species; antibody-based assays are not specific enough to discriminatebetween orthologues, thus forcing drug developers to use multipleanalytical platforms across preclinical models. Here, we present amodular mass spectrometry-based assay for the analysis of orthologousproteins across species in the context of translational science usingataxin-2 (ATXN2) as an example. Instead of relying on recombinantproteins, we applied rational peptide design to target human, mouse,human/pig/non-human primates (NHP), or pan-species regions of thefull-length protein. Specificity and sensitivity of the method wereproven in wild-type and transgenic models. Dose–response upontreatment with different doses of an adeno-associated virus (AAV)-deliveringmicroRNA (miRNA) targeting ATXN2 showed that knock-down at the transcriptlevel was consistent with downregulation at the protein level, confirmingthe assay suitability for target engagement evaluation. Furthermore,the data presented here indicate that this method is readily transferableto minipigs, ensuring analytical continuity across preclinical studies.More broadly, this work provides a translational blueprint that canbe adapted to other orthologues, reducing cost and timeline in drugdevelopment.

Helicobacter pylori (H. pylori) induces gastric precancerous lesions (GPL),but their underlying biomarkers remain poorly characterized. To establisha GPL model, BALB/c mice were inoculated with H. pylori via gavage. Immunofluorescence confirmed successful modeling bydetecting H. pylori distribution andaberrant expression of precancerous lesion markers. Serum-based label-freeproteomics and bioinformatic analyses were performed to screen keyproteins. Data are available via ProteomeXchange (PXD076808). Westernblot and immunofluorescence were used to verify key proteins and markersof DNA damage, fibrosis, and epithelial mesenchymal transition (EMT).The results revealed high H. pylori load in gastric mucosa with abnormal precancerous markers. Proteomicanalysis identified 103 differentially expressed proteins and fivecore proteins as potential biomarkers, all overexpressed in gastriccancer. Col1a1, a fibrosis marker, was identified as a key upregulatedprotein and validated by Western blot and immunofluorescence. Col1a1colocalized with the DNA damage marker γ-H2AX in gastric mucosa. H. pylori infection triggered DNA damage, fibrosis,and EMT in GPL. In conclusion, H. pylori induced DNA damage, fibrosis, and EMT in gastric mucosa, acceleratingprecancerous lesion progression, with Col1a1 playing a significantrole.

Gastric cancer is driven by aberrantkinase signalingthat promotesuncontrolled proliferation and metabolic adaptations. Although calcium/calmodulin-dependentprotein kinase 2 (CAMKK2) is overexpressed in gastric cancer, itsrole in regulating tumor-associated metabolic programs remains unclear.Here, we employed an integrated multiomics approach, centered on untargetedmetabolomics, to investigate the effects of CAMKK2 inhibition in gastriccancer cells. Pharmacological inhibition of CAMKK2 using STO-609 resultedin G1/S phase cell cycle arrest, multinucleation, and coordinatedmetabolic, proteomic, and phosphoproteomic alterations. Metabolomicprofiling revealed extensive metabolic reprogramming characterizedby the depletion of purine and pyrimidine intermediates required forDNA synthesis. Pathway analysis demonstrated the suppression of nucleotide,lipid, and central carbon metabolism, indicating impaired biosyntheticcapacity. Integration with proteomic and phosphoproteomic datasetsrevealed coordinated downregulation of DNA replication machinery andattenuation of signaling pathways involved in cell cycle progression.Collectively, these findings suggest that CAMKK2 contributes to thecoordination of metabolic and signaling programs associated with DNAreplication and proliferation in gastric cancer. CAMKK2 inhibitionwas associated with the coordinated depletion of nucleotide metabolitesand disruption of replication-associated signaling, consistent withimpaired DNA replication and a replication-stress-like phenotype.These findings highlight CAMKK2 as a potential therapeutic targetfor exploiting metabolic vulnerabilities in gastric cancer.

Ionizingradiation (IR) induces oxidative stress andlipid peroxidationwith ferroptosis contributing to radiation-induced cancer cell death.Peroxiredoxin 6 (PRDX6), a multifunctional enzyme involved in phospholipidrepair and redox regulation, has been implicated in ferroptosis regulation,but its role in the radiosensitivity of cancer cells remains unclear.Here, we show that PRDX6 depletion enhances radiosensitivity in A549lung cancer cells. PRDX6-silenced A549 cells exhibited elevated lipidperoxidation and labile iron levels, indicating a ferroptosis-pronestate that was amplified by radiation. To identify the underlyingdrivers of ferroptosis sensitivity in the PRDX6 background, we performedglobal proteomic profiling of PRDX6-silenced A549 cells, with andwithout radiation. Basal PRDX6-deficient cells displayed increasedlevels of ACSL4 and DHCR7. ACSL4 activates peroxidation-prone polyunsaturatedfatty acids for incorporation into membrane phospholipids, and DHCR7depletes the antiferroptotic moiety, 7-dehydrocholesterol. MPST, anantioxidant protein, was downregulated. These changes were accentuatedupon IR exposure with further suppression of GINS4, a negative regulatorof ferroptosis. Functional rescue by oleic acid suggested the roleof ACSL4 in the observed phenotypes. Radiosensitization via ferroptosisinduction was additionally demonstrated in another human lung cancercell line under the PRDX6-silenced state. Collectively, PRDX6 lossremodels the cellular proteome to establish a presensitized redoxand lipid compositional state, which favors ferroptosis-driven radiosensitivity.

Blood-basedproteomics may complement existing MCED approachesby capturing tumor-secreted proteins and systemic host responses.We reviewed prospective and multicancer studies published from 2020to 2025 that used Olink, SomaScan, or mass spectrometry and comparedevidence from case–control and prospective designs. Candidateproteins were classified as established or novel and evaluated usingpathway and cancer-hallmark enrichment analyses, GEPIA3.0 Cox analyses,TCGA Kaplan–Meier survival curves, and TPCPA/RPPA protein abundancedata. Across platforms, established markers, including CEACAM5, WFDC2/HE4,and GDF15 were complemented by novel candidates converging on a matrix–immune–secretoryaxis. This axis included MMP12 and ADAM8, CD74 and CXCL13, TGFB1,and CDCP1. Secretory and matrix-associated candidates were linkedto poorer survival in several epithelial cancers, whereas immune-associatedproteins showed context-dependent effects. Tumor-type-specific proteinelevation supported their detectability, and interaction analysisorganized the candidates into biologically coherent modules. Thesefindings support an MCED framework combining higher-risk secretoryand matrix markers with immune-context indicators to improve sensitivityand biological interpretation. Prospective validation and down-selectioninto scalable targeted assays are required before population-levelapplication.

Coronary calcificationis the final stage in coronary heart disease,associated with significantly increased morbidity and mortality. Themolecular pathology of calcification and especially events triggeringcalcification are poorly understood. Metagenomic and molecular microbialtechniques have identified the presence of oral biofilm-producingand dental calculus-forming bacteria in atherosclerotic plaques, suggestingthat they may be involved in calcification. To enable studying thisat the proteome level, we compared four different protein extractionmethods to find a robust and reliable method for the identificationand quantification of human and bacterial proteins from 10 calcifiedarterial plaques. We identified 3 913 human proteins and 24 bacterialproteins. These findings provide further support for the presenceof bacterial proteins in calcified areas of coronary artery plaques,with Fusobacteria being the most abundant bacterial protein source.Comparisons of the extraction methods showed that the urea-based extractionmethod was most optimal due to its simplicity and reproducible identificationof a high number of proteins, which makes it suitable for furtherlarge-scale analysis of coronary artery plaque samples.