Open Search-Based Proteomics Reveals Widespread Tryptophan Modifications Associated with Hypoxia in Lung Cancer

Chen, Jinfeng; Zhang, Lei; Sun, Zhao; Li, Hongyi; Li, Jingyi; Xue, Xinli; Zhu, Qingqing; Dong, Bowen; Wang, Yuanyuan; Yang, Yang; Dong, Yongqiang; Guo, Guangyu; Jiang, Hongqiang; Zhang, An; Zhang, Guoqing; Hou, Zhichao; Li, Xiangnan; Yang, Jing-Hua

doi:https://doi.org/10.1155/2022/2590198

Oxidative Medicine and Cellular Longevity

On this page

Abstract Introduction Materials and Methods Results and Discussion Conclusions Data Availability Conflicts of Interest Authors’ Contributions Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2022 | Article ID 2590198 | https://doi.org/10.1155/2022/2590198

Open Search-Based Proteomics Reveals Widespread Tryptophan Modifications Associated with Hypoxia in Lung Cancer

Jinfeng Chen,¹Lei Zhang,¹Zhao Sun,²Hongyi Li,²Jingyi Li,¹Xinli Xue,¹Qingqing Zhu,¹Bowen Dong,¹Yuanyuan Wang,¹Yang Yang,¹Yongqiang Dong,²and Guangyu Guo¹ et al.

Academic Editor: C. López-Alarcón

Received02 Sept 2021

Accepted08 Mar 2022

Published30 Apr 2022

Abstract

The tryptophan residue has a large hydrophobic surface that plays a unique role in the folded protein conformation and functions. Tryptophan modifications are presumably to be readily detected in proteins due to the vulnerability of the indole structure to electrophilic attacks. In this study, we report a systematic identification of sequence variations at tryptophan, termed tryptophan variants, from the proteome of patients with nonsmall cell lung cancer (NSCLC). Using shotgun proteomics and a modified open search algorithm, 25 tryptophan variants on 2481 sites in over 858 proteins were identified. Among these, 6 tryptophan variants are previously identified, 15 are newly annotated, and 4 are still unknown, most of which are involved in the cascade of oxidation in the blood microparticle. Remarkably, Trp313 of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was up-oxidized whereas Trp16 and Trp38 of hemoglobin (HBB) were down-oxidized in NCSLC tissues. The results were further supported by an independent cohort of 103 lung adenocarcinoma samples, reflecting a negative feedback and potential detoxification mechanism against tumor glycolysis and hypoxia. Overall, the study reports a quick approach to explore tryptophan variants at the proteomic scale. Our findings highlight the predominant role of tryptophan oxidation in regulating the redox balance of cancer cells and its potential role as prognostic biomarker for patients with NSCLC.

1. Introduction

Proteins are the main functional biomacromolecules in all organisms, which can be modified at the amino acid (AA) side chains, including by addition, subtraction, or alteration of chemical groups [1]. Modified proteins play key roles in many important cellular functions and regulatory processes [1, 2]. To date, more than 1500 various types of posttranslational modifications (PTMs) have been identified, including acetylation, methylation, glycosylation, phosphorylation, and ubiquitylation [3, 4]. The nucleophilic and high reactivity amino acids, cysteine (Cys) and lysine (Lys), are the two most attractive residues to the electrophilic small molecules [5, 6]. A global map of 3,000 covalently liganded cysteines in primary human T cells have been presented, which can impair T cell activation by distinct mechanisms involving the direct functional perturbation and/or degradation of proteins [5]. A new lysine modification, lactylation, was recently identified and shown to act as an endogenous “lactate clock” in bacterially challenged M1 macrophages to turn on gene expression to promote homeostasis [6].

Recently, tryptophan (Trp) has inspired the interest of researchers because of its unique reactivity and specificity. Tryptophan is the rarest amino acid in eukaryotes, accounting for only approximately 1.13% of all amino acids [7]. The average solvent-exposed value of tryptophan residue, calculated for globular proteins, is 0.19, which is much smaller than that of other charged amino acid residues, such as glutamate (0.50) [8, 9]. Therefore, tryptophan has a greater probability of being buried inside the proteins compared with other AA residues [10]. Nevertheless, tryptophan exists in almost all proteins [7]. Additionally, tryptophan residues tend to participate in the recognition of molecules interacting with enzymes and proteins [11–14]. Furthermore, tryptophan affords exclusive site selectivity owing to the presence of a unique indole ring, an electron-rich aromatic ring, possessing distinct chemical reactivity [12]. Since the 1960s, diffuse occurrence of tryptophan residues in heme- and/or metal-coordination sites of proteins has detected widespread metal-catalyzed oxidations [15–17]. On the other hand, the occurrence of accessible tryptophan residues in proteins present in close proximity to reactive oxygen species (ROS) promotes their selective oxidative modifications [18, 19]. Besides, tryptophan residue can also be nitrated [19, 20] or mutated to another amino acid, such as arginine (Arg) [21], glycine (Gly) [22]. Substitution of Arg for Trp at position 37 of hemoglobin greatly reduced its cooperativity and oxygen affinity [21], whereas the mutation of Trp to Gly resulted in a greater oxygen affinity than normal [22]. Collectively, these findings suggest the great importance of identifying tryptophan variants.

Shotgun proteomics, in which proteins are enzymatically digested and analyzed via liquid chromatography–high resolution mass spectrometry (LC-HRMS), is one of the most widely used approaches to analyze PTMs [23, 24]. Unlike the restricted search engines, the open search engines enlarge the search space with more types of modifications to retrieve more peptide candidates [23, 25]. To develop a comprehensive list of tryptophan modifications, we carried out a shotgun-based open search proteomic of 139 human tissues and primary cells. Our analysis not only provides the most comprehensive map of the tryptophan modifications to date, spanning nearly the full depth of the tryptophan modifications, but also highlights the predominant role of oxidative modifications in hemoglobin (HBB) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), mainly associated with tumor hypoxia. An independent cohort was further utilized to evaluate the potential prognostic value of the biomarkers of these oxidative tryptophan modifications, which reflect the hypoxia and metabolic dysregulation involved in lung tumor.

2. Materials and Methods

2.1. Sample Preparation of NSCLC Tissue for LC-MS/MS

Histologically-confirmed human nonsmall cell lung cancer (NSCLC) tissues and paired noncancerous adjacent tissues from 27 patients ( samples) were received from the tissue band of the First Affiliated Hospital of Zhengzhou University, Zhengzhou, China. The protocols for analysis of the samples were approved by the Institutional Research Ethics Review Boards. The tumor and adjacent lung tissues ( samples) were ground in liquid nitrogen and homogenized with Dounce in ice-cold RIPA buffer. The soluble proteins were collected by centrifugation, and their concentration was measured using Bio-Rad BCA protein assay kit. Approximately, 100 μg of the soluble proteins was precipitated with -80°C acetone and resuspended in 10 mM dithiothreitol and alkylated with 50 mM iodoacetamide. Proteins were digested with sequencing grade trypsin at an enzyme-to-substrate ratio of 1 : 100 overnight at 37°C. The peptides were desalted with the C18 column (Sigma-Aldrich) and fractionated using high-pH reversed-phase chromatography [26]. Briefly, approximately 50 μg of the desalted peptides was resuspended in 160 μL ammonia water () and loaded on the column filled with C18 reverse-phase medium (Agela Technologies). The column was eluted with increasing concentrations of acetonitrile (6−50%) in ammonia water. A total of 10 fractions were collected, lyophilized, and stored at −80°C for further use.

2.2. LC-MS/MS-Based Analysis of Peptides

For label-free proteomic analysis, trypsin-digested peptides were dissolved with 0.1% formic acid and analyzed by EASY-nLC1000 LTQ-Orbitrap Elite mass spectrometer (Thermo Fisher Scientific). The reversed-phase C18 column (Reprosil 3 μm, ) was eluted with a 180 min linear gradient of 5−35% acetonitrile at a flow rate of 250 nL/min (solvent A: 0.1% formic acid in water; solvent B: 0.1% formic acid in acetonitrile). The flex nanospray was used in positive mode, and the spray voltage was set to 2.30 kV using stainless steel emitters. The temperature of the transfer capillary was 250°C. The HCD collision energy was 27.5 V, and the MS spectra from 350 to 1800 m/z were acquired in the DDA mode with a resolution of 60,000. The 30 most intense ions were selected for MS/MS in the CID mode using ion trap. Normalized collision energies were 35% with 10 ms. The maximum precursor ion injection time was 100 ms. The repeat count was 2, and the dynamic exclusion duration was 90 s. The minimal ion count threshold was 1000. The acquired raw data (NSCLC proteomic data, Figure 1(a)) for 10 fractions of each sample were then analyzed by a wildcard open search using the Byonic software (v3.8) [25].

(a)

(b)

Figure 1

Algorithms for comprehensive identification of Trp variants by wildcard search-based proteomics. (a). NSCLC samples were digested, fractionated, and analyzed on the high-resolution and high-accuracy Orbitrap mass analyser. The human proteomics data (dataset identifier PXD000561, “A draft map of the human proteome” doi:10.1038/nature13302) were downloaded from the Proteome Xchange Consortium (http://proteomecentral.proteomexchange.org). Tandem mass spectrometry data were searched against the human UniProt database (only reviewed entries, human 20350 entries, proteome UP000005640, downloaded April 2020) using Byonic wilcard search algorithm. The obtained nonzero delta mass was clustered (Gaussian mixture model); then the tryptophan modifications with high confident peptide spectrum matches were filtered. (b). Workflow for the tryptophan modification annotation. The 62 preknown Trp variants are shown in Supplementary Table 2.

2.3. Identification of the Tryptophan Variants

To compare with normal tissues and cells, a total of 39 HRMS proteomics data (human proteomic data, Figure 1(a)) from 22 histologically normal human cell and tissue types were download from the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository with the dataset identifier PXD000561 [27]. Both the NSCLC proteomic data [28] and the human proteomic data [27] were analyzed against the UniProt database (20350 entries, human proteome UP000005640, April 2020) by a wildcard open search using the Byonic software (v3.8) [25]. The tolerance was set to 10 ppm for precursor ions, and 0.6 Da for the fragment ions. Two missed cleavages were allowed for trypsin digestions. The delta masses between coding and the observed amino acids were set from -150 to 500 Da and filtered using the criteria: . To determine the tryptophan variants, the delta masses were analyzed by the Bayesian Information Criterion (BIC) [29] followed by Gaussian mixture model [30] with 1 Da intervals from -150 to 500 Da. The clustered peaks with the expected delta masses and sufficient values of the goodness-of-fit () were identified as the qualified mass differences between the coding and the observed amino acids, which were considered different modifications or amino acid variations. A list of peptides containing the confident delta masses at the tryptophan residues were identified and termed tryptophan variants (Supplementary Table 1).

2.4. Annotation of the Tryptophan Variants

Based on the Unimod database (http://www.unimod.org/) and reference [31], a tryptophan modification database (62 preknown Trp variants, Supplementary Table 2) was first constructed for the annotation. We initially mapped the 25 tryptophan variants against the tryptophan modification database, resulting in the identification of six known modifications (, Table 1). The unmatched delta mass was further annotated according to biological cascade reactions. Accordingly, the obtained 1178 delta mass (19 unknown delta mass minus 62 preknown delta mass) was mapped against the 1514 entries of protein modifications (April 2021) in the Unimod database (). The resulted delta mass pair for each unknown Trp variant was reviewed, and the reasonable reactions were finally chosen to annotate the unknown Trp variants.

2.5. Gene Ontology (GO) and Network-Based Analyses

Latest GO database (https://www.ebi.ac.uk/QuickGO/) was used for GO enrichment analysis. For functional enrichment analysis, lists of interesting proteins were submitted for enrichment analysis using the UniProt database (20350 entries, human proteome UP000005640, April 2020) as a background dataset. All terms displayed had a . ClueGO, a Cytoscape plug-in, was used to decipher functionally grouped pathway annotation networks of the tryptophan-modified proteins [32]. For network-based analyses, protein lists were submitted to the STRING database (http://www.string-db.org/) [33]. The network was exported into a simple tabular output then imported into Cytoscape software (http://cytoscape.org/) for visualization [34].

2.6. Statistical Analysis

The R framework (version 4.0.3) was used to perform all statistical analyses of the bioinformatics data. Spearman correlation matrix was generated using “corr.test” function within the package “psych” (2.0.12) and visualized using the “ggcorrplot” function within the package “ggcorrplot” (0.1.3) and “pheatmap” (1.0.12). GO annotations were performed using the R package “clusterProfiler” (3.14.3) with the “enrichGO” function [35].

2.7. Survival Analysis

Kaplan-Meier survival curves (log-rank test) were used for overall survival (OS) or disease-free survival (DFS) of the proteomic subtypes and patients with Trp313 oxidation in GAPDH, Trp16, and Trp38 oxidations in HBB. Prior to the log-rank test, survminer 0.4.9 R package was applied to determine the optimal cut-off level for the selected samples. DFS or OS curves were then calculated (Kaplan-Meier analysis and log-rank test) based on the optimal cut-off level.

2.8. Molecular Docking

Molecular docking was conducted using MOE v2018.0101 [36]. The 3D structures of the proteins GAPDH and HBB were downloaded from the RCSB Protein Data Bank with PDB ID of 6M61 [37] and 1CBL [38], respectively. The 313W site of GAPDH and W16 and W38 sites of HBB were modified, respectively, and then, the energy of the modified structure was minimized to obtain the modified 3D protein structures. Prior to docking, the force field of AMBER10: EHT, and the implicit solvation model of Reaction Field (R-field) were selected. MOE-Dock was used for molecular docking simulations of the proteins with compounds.

The “induced fit” protocol was selected, wherein the side chains of the binding site in the receptor were allowed to move according to ligand conformations, and a constraint was applied on their positions. The weight used for tethering the side chain atoms to their original positions was 10. Firstly, all docked poses were ranked by London dG scoring function; then, force field refinement was applied on the top 30 poses followed by a rescoring of GBVI/WSA dG scoring function. The conformation with the lowest binding free energy was finally identified as the best probable binding mode.

3. Results and Discussion

3.1. Widespread Tryptophan Variants Identified by a Modified Open Search Approach

The tryptophan moiety in proteins is presumably reactive to many cellular metabolites that would generate a variety of tryptophan derivatives essential for protein conformation and functions. In this study, the shotgun proteomics and a modified open search approach were adapted to systematically identify any possible variations of tryptophan in the human proteome (termed tryptophan variants) in patients with nonsmall cell lung cancer (NSCLC). Briefly, total proteins of the tumor and adjacent tissues were digested with trypsin, fractionated, and analyzed with high-resolution LC-HRMS. Mass differences between the coding and observed amino acids were identified by a wildcard search using Byonic [25] against human UniProt database (Figure 1(a)). This generated a total of 2,153,367 nonzero delta masses accumulatively from the downloaded human proteomic data and the NSCLC proteomic data. To identify tryptophan variants, these delta masses were subsequently grouped with multivariate clustering followed by Gaussian regression [39], resulting in 25 unique delta mass clusters at tryptophan from 11,769 high-confident MS/MS spectra spread over 2481 tryptophan sites on 858 proteins (Supplementary Figure 1, Supplementary Table 1). These clustered delta masses of tryptophan were reasonably caused by different chemical modifications or structural alterations, reflecting the variants of tryptophan at the protein level.

3.2. Tryptophan Variants Largely due to Electrophilic Attacks of the Indole Moiety

As the accuracy of the clustered delta masses was greatly increased by clustering and regression (), it enabled more confident annotation simply based on the molecular weights and structure of the putative tryptophan variants (Figure 1(b)). By searching the database for known tryptophan variants (Supplementary Table 2), 6 of the 25 delta mass clusters matched the previously reported tryptophan modifications (Table 1, Figure 2, Supplementary Figure 2), including monooxidation (W+15.9949, 1) and di-oxidation (W+31.9899, 2) of tryptophan, tryptophan->kynurenine substitution (W+3.9953, 3), tryptophan->hydroxy-kynurenine substitution (W+19.9903, 4), tryptophan->Leu/Ile substitution (W-72.9961, 14), and dithiothreitol adduct (W+209.0179, 21). Herein, monooxidation includes keto and hydroxyl modifications of tryptophan, and di-oxidation includes quinonyl and dihydroxyl modifications of tryptophan. Additionally, dithiothreitol adduct was the artefact product derived from the dithiothreitol during the sample preparation.

The unknown delta mass clusters were assumed to be the products of stepwise reactions of the indole sidechain (Figure 1(b)). Because of the electron enrichment by oxidation, the HO-moiety indole moiety was found more vulnerable to attacks by electrophilic metabolites, leading to multiple modification productions of tryptophan (Figure 2). In this line, the clustered delta masses of W+115.9754 (5), W+354.1708 (6), W+368.1698 (7), and W+432.2146 (8) matched several acyl-modifications of oxidized tryptophan. In brief, dihydroxyl-tryptophan (2) might react with docosahexenoic acid (DHA), an essential compound for brain function [40], to generate docosahexenoyl-tryptophandione, suggesting the occurrence of W+354.1708 (6) in the proteome. Similarly, dihydroxyl-tryptophan might react with lactic acid, 10-hydroxydecanoic acid, and 3-oxocholic acid to generate lactoyl-tryptophandione (W+115.9754, 5), histidinyl-decanoyl-tryptophandione (W+368.1698, 7), and 3-oxocholoyl-tryptophandione (W+432.2146, 8), respectively. Notably, more-step reactions might occur after the mono-/trioxidized tryptophan in the human proteome. These included nitro-spermidinyl-tryptophan (W+246.1373, 10) and nitro-methyl-tryptophan (W+92.0277, 9) following monooxidation and hydroxyl-OPC6-tryptophan following tri-oxidation (W+312.1545, 11). Thus, these data agreed with our hypothesis that the oxidation products of tryptophan are the essential intermediates of many unknown tryptophan variants, and the dihydroxyl-tryptophan is apparently more vulnerable to further modifications. Following the same line, the nucleophilic nitrogen of the indole moiety is another potential site for electrophilic attacks. A good example was the clustered delta mass W+493.2657 (Supplementary Figure 2(A), 1(2)), which may be attributable to glycosylation of tryptophan by hexosamine followed by its reaction with levuglandinyl-lactam to contribute to the observed delta mass. Note that these two small molecules have been shown to be important metabolites for the hexosamine and levuglandin pathways [41]. Additionally, certain unknown delta mass clusters may involve amino acid substitutions prior to modifications. For instance, replacement of the tryptophan with histidine, aspartate, cysteine, and tyrosine, followed by histidine oxidation, aspartate benzoylation, cysteine thiobutyroylation, and tyrosine formylation was shown to be responsible for the tryptophan variants of W-59.0506 (15), W+32.9748 (18), W-13.0308 (17), and W+4.9792 (16), respectively. For the same reason, the tryptophan variants of W+88.9942 (19) and W+67.0058 (20) matched multiple reactions after Trp to Thr substitution (Supplementary Figure 2(B)).

Overall, the total frequency of all identified modifications was 11725 (Table 1); among these, 93.16% of tryptophan variants (total frequency: 10923) had resulted from primary and secondary modifications by reactive species, 2.74% (total frequency: 321) was due to amino acid substitutions, 0.32% (total frequency: 37) was due to direct chemical reaction with in vivo metabolites, 3.14% (total frequency: 368) was due to unknown reasons (Supplementary Figure 3A), and 0.65% (total frequency: 76) represented the artefact product derived from dithiothreitol during sample preparation. Thus, reactive species-induced modifications (Supplementary Figure 3(B)) were the most common nature of tryptophan variants in human proteomes. Collectively, our data suggested that the mechanisms for tryptophan variants were largely due to the vulnerability of the indole moiety to electrophilic attacks, which was significantly enhanced by single, double, or triple oxidization of indole (Table 1). Additionally, presubstitution of tryptophan by other amino acids followed by further reactions was another mechanism.

3.3. Tryptophan Variants Dominantly Found in Hemoglobins in Blood Microparticle

The Trp-modified proteins were grouped into 14 KEGG pathways (Figure 3(a), Supplementary Table 3) on bioinfomatics analysis using ClueGO of Cytoscape [32]. Most Trp oxidations were clustered in the glycolysis-related pathways (Group 11, Figure 3(b)) but not mitochondrion oxidative phosphorylation-related pathways (Groups 12 and 14), although mitochondria are the major source of cellular ROS in eukaryotic cells. The result was consistent with a previous study in which Cys, Pro, and Tyr, instead of Trp, were found to be predominantly oxidized in mitochondrial proteins [42]. Moreover, most Trp oxidations were clustered in the infection-related pathways, such as vibrio cholerae infection (Group 6), salmonella infection (Group 7), and antigen processing and presentation (Group 8). These results are highly consistent with the fact that ROS production is an important mechanism to counteract bacterial infections [43].

(a)

(b)

Figure 3

Proteins with tryptophan variants were largely clustered in cytoplasmic glycolysis pathway. (a) Functionally grouped KEGG pathways of the modified-tryptophan-containing proteins with terms as nodes linked based on their kappa score level (≥0.4), where the label of the most significant term per group is shown in color and bold. The node size represents the term enrichment significance. Functionally related groups show partial overlap. (b) Heatmap depicting the relative expression levels of the identified modifications in CC groups.

To understand the overall cellular distribution, the proteins with tryptophan polymorphisms were analyzed for their cellular component enrichment. Our data demonstrated that the proteins with tryptophan variants predominantly spanned over blood microparticle and ficolin-1-rich granule (Figure 4(a), Supplementary Table 4). Of note, the iron-containing and oxygen-carrying hemoglobins (HBB and HBG2) within the blood microparticle and ficolin-1-rich granules were the proteins with the highest modification rate (Figure 4(b), >135 per site) and the most abundant tryptophan oxidation (Supplementary Figure 3(C)). For instance, the high frequencies of modifications at Trp16 and Trp38 of HBB (P68871, Supplementary Table 1) and at Trp16, Trp38, and Trp131 of HBG2 (P69892) were presumably due to self-oxidation cascade [44]. Hemoglobin consists of two α- and two β-subunits containing the heme groups, a reduced ferrous ion in the porphyrin ring [44]. As an efficient O₂ carrier, the heme group coordinates with four O₂ molecules reversibly without exchanging electrons. Therefore, the indole moiety of the tryptophan residues in hemoglobin is reasonably vulnerable to self-oxidization by the reduced ferrous ions surrounding the heme-coordination sites [44]. The results indicate that Trp is prone to oxidize due to metacatalysis.

(a)

(b)

(c)

(d)

Figure 4

Proteins with tryptophan variants were largely clustered in blood microparticle. (a) Right panel displays the cellular components (CC) from GO enrichment analysis of the modified-tryptophan-containing proteins; left panel shows the frequencies per site of each CC term. (b) Network diagram of the modified-tryptophan-containing proteins mapped to the CC categories. Diamonds represent proteins; octagons represent cellular components. The size of the nodes represents the frequencies per site. (c) Heatmap depicting the Spearman correlation of tryptophan modifications in P68871 (HBB). (d) Color bar represents the relative frequencies of differentially expressed oxidation modification at the 16 W and 38 W sites of P68871 (HBB); the graph shows the overall structure of heme-core in P68871 (PDB: 1HBB). The linear distance between the 15(16)/37(38) Trp sites and heme is shown.

Next, the frequencies of tryptophan variants at different sites of hemoglobins (HBB and HBG2) were analyzed across the human proteome using Spearman correlation. The results demonstrated that the cascade of tryptophan oxidations such as W+3.9953, W+15.9949, W+19.9903, and W+31.9899 were significantly correlated (Figure 4(c) and Supplementary Figure 3(D)). The total oxidation frequencies varied at different tryptophan sites of HBB and were congruent with the sequence of monooxidation, di-oxidation, kynurenine, and hydroxyl-kynurenine (Figure 4(d), left panel), with the highest rates near the heme-coordination site. For instance, HBB at 16 W was detected with higher oxidative rate than HBB at 38 W, apparently due to the distance with the heme group (Figure 4(d), right panel). HBG2 at 16 W and 38 W was more sensitive to oxidation than HBG2 at 131 W for the similar reason (Supplementary Figure 3(E)).

Collectively, tryptophan variants were closely associated with the function of the heme proteins in the blood microparticle and were associated with anti-infection and energy metabolism pathways. Our data consistently suggested that the chemical properties or variants of tryptophan in the heme proteins were mainly regulated by self-catalyzed oxidation. Generally, our data suggested that the self-catalyzed oxidation was prone to target the tryptophan residues in the close proximity of the metal-coordination sites of HBB and HBG2 [14, 18, 19, 44]. The results improved our understanding of the diversity of distribution of Trp modifications, leading to the discovery of many Trp modifications with distinguished localization in specific subtypes of pathways.

3.4. Tryptophan Variants Closely Associated with Reactive Species in Lung Cancer

Reactive species are the byproducts of cellular metabolism known to oxidize proteins and cause cellular oxidative stress [31]. Oxidative stress triggers the antioxidant signaling pathways to eliminate reactive species [45, 46]. We propose that tryptophan variants are associated with oxidative stress induced by reactive species in lung cancer. Indeed, in our study, over 93% of the tryptophan variants was due to reactive species-derived modifications (Supplementary Figure 3(A)) and the majority of the modified proteins were associated with oxidation reactions of hemoglobin in lung cancer (Figure 4(b)). A total of 110 modified tryptophan sites were identified on 53 proteins predominantly in blood microparticle of patients with NSCLC (Supplementary Figure 4(A)). The most commonly modified tryptophan proteins were functionally associated with oxidoreductases, peroxidases, and antioxidant activities (Supplementary Figure 4(A)). Network-based analyses [33, 34] revealed that the occurrence of tryptophan variants in GAPDH, SOD2, HSPA8, ACTB, and HBB was closely connected with the antioxidant network (Figure 5(a)).

(a)

(b)

Figure 5

Tryptophan variants associated with antioxidants prone to oxidative stress in NSCLC. (a). Network diagram depicting the interaction of the modified-tryptophan-containing proteins in NSCLC samples that are mapped to the molecular function (MF) and cellular component (CC) categories. Diamonds represent protein, circulars represent MF, and hexagons represent CC. The color of the node corresponds to the fold change (log2) of the modification in each protein site between tumors and the adjacent normal tissues (NATs); red indicates higher expression and blue indicates lower expression in the tumor. The size of the node corresponds to the degree of correlation. (b). Heatmap depicting the modified levels of differentially identified modifications of each protein in lung tumor and adjacent normal tissues. Color bar from blue to red represents the fold change (log2) of modified level from increasing to decreasing. , , .

Consistent with this, the tryptophan modifications at GAPDH at 313 W were found significantly upregulated in tumors compared to that in the adjacent normal tissues (Figure 5(b)). This is consistent with previous studies in which the 313 W of GAPDH was found to be oxidized by the cytosolic ROS during metabolic and oxidative stress [45, 47–49]. The ROS level is typically upregulated in the tumor microenvironment [50]. However, the specific oxidation at 16 W and 38 W of HBB (HBB at 16 W, HBB at 38 W) was slightly reduced in lung cancer (Figure 5(b)), probably because HBB at 16 W and HBB at 38 W were self-oxidized near the heme-coordination site (Figure 4(d)). As GAPDH is typically upregulated in cancers as one of the critical enzymes for glycolysis, it catalyzed the oxidative phosphorylation of glyceraldehyde-3-phosphate (G3P) to generate 1,3-diphospho-glycerate (1,3-DPG), which were subsequently changed to 2,3-diphosphoglycerate (2,3-DPG) in the presence of the enzyme diphosphoglycerate mutase (DPGM) [51]. Upregulation of 2,3-DPG in cancers reduces the oxygen affinity leading to release of oxygen from HBB [51]. This explained our data that, while ROS enhanced oxidation at GAPDH at 313 W, the increase of 2,3-DPG suppressed self-oxidation at HBB 16 W and HBB 38 W, suggesting a negative regulation for the hypoxia environment in lung cancer (Figure 6(a)). Nevertheless, ROS-induced GAPDH at 313 W+15.9949 was localized within the catalytic domain, so that it would suppress the GAPDH activity (Figure 6(a)). The docking simulation studies finally confirmed these speculations (Supplementary Figure 5). The docking scores showed that the binding affinity of GAPDH at 313 W for NAD⁺ was weaker than that of wild type GAPDH (Supplementary Figure 5(A) and 5(B)), which would suppress the conversion of G3P into 1,3-DPG by GAPDH. In addition, the oxidized HBB-W16/W38 weakened the interactions between HBB and 2,3-bisphosphoglycerate (Supplementary Figure 5(C) and (D)), which enhanced the oxygen affinity and retained the oxygen from HBB (Figure 6(a)). In this context, our data suggested that tryptophan variants provided the negative regulation pathways for the aerobic glycolysis and hypoxia, two typical phenotypes for lung cancer. Finally, these data are consistent with the previously reported upregulation of the glycolytic enzymes in lung adenocarcinoma (LUAD) at the proteomic level [52]. In LUAD proteomic data [52], the expressions of PFKP, PFKL, ALDOA, ALDOC, TPI1, GAPDH, and PFKFB2 were significantly upregulated in tumor samples (Supplementary Figure 4(B)).

(a)

(b)

(c)

(d)

Figure 6

Tryptophan variants associated with the prognosis of NSCLC. (a). Schematic illustration of the mechanism underlying the regulation of various oxidative states of HBB and GAPDH. CDB3: cytoplasmic domain of band 3; 1,3-DPG: glycerate-1,3-diphosphate; 2,3-DPG: glycerate-2,3-diphosphate. (b). Relative expression levels of differentially identified modifications of HBB and GAPDH in tumor samples and adjacent normal tissues (NAT) in an independent cohort of 103 LUAD proteomic dataset (Xu et al., 2020, Cell 182, 245–261). (c). Association of three tryptophan oxidations with overall survival (OS) in an independent cohort of 103 LUAD proteomic dataset (Xu et al., 2020, Cell 182, 245–261), ( value from log-rank test). (d). Association of three tryptophan oxidations with disease-free survival (DFS) in an independent cohort of 103 LUAD proteomic dataset (Xu et al., 2020, Cell 182, 245–261), ( value from log-rank test).

3.5. Tryptophan Variants as Biomarkers for the Prognosis of Lung Cancer

As most tryptophan variants were associated with reactive species in lung cancer, we reasoned whether the specific tryptophan modifications of GAPDH at 313 W oxidation, HBB at 16 W oxidation, and HBB at 38 W oxidations could be developed as prognostic biomarkers. To this end, 103 lung adenocarcinoma samples from an independent cohort were examined [52]. The results further confirmed that oxidation of GAPDH at 313 W was significantly upregulated, while oxidation of HBB at 16 W and HBB at 38 W was downregulated in tumor samples (Figure 6(b)). Moreover, patients with higher levels of GAPDH at 313 W oxidation and lower levels of HBB at 16 W+15.9949 and HBB at 38 W+15.9949 had better prognostic outcomes (Figures 6(c)and 6(d), log-rank test, ). Therefore, the divergent oxidative tryptophan levels of GAPDH and HBB, mainly attributed to the aerobic glycolysis and hypoxia in lung tumor, presented a biomarker panel with potential prognostic relevance with respect to monitoring tumor progression, evaluating therapeutic response, and developing clinic treatments.

While hypoxia is known to induce ROS accumulation and lead to a diverse array of dysregulated signals, including aerobic glycolysis, oxidative stress, apoptosis, and eventually cell death, it also induces antioxidant measures to counteract ROS-induced damage [50]. Since the observed Trp oxidations are widely and naturally occurring and originate from ROS, it can be assumed that the biological impact of hypoxia may in part be mediated via oxidation of cellular proteins. Tryptophan variants are presumed as a negative feedback mechanism to counteract the glycolysis and hypoxia in lung cancer. On one the hand, the upregulation of ROS-induced GAPDH at 313 W oxidation may reduce the GAPDH activity and subsequently suppress GAPDH-mediated glycolytic pathways resulting in downregulation of self-oxidation of HBB. On the other hand, the downregulation of self-oxidation of HBB at 16 W and HBB at 38 W increases the oxygen affinity of HBB and leads to the suppression of ROS-induced hypoxia in cancers. Moreover, tumor hypoxia is an independent marker of poor prognosis for cancer [50, 53, 54], and targeting hypoxia is important for therapeutic outcome. Therefore, tryptophan variants, predominantly oxidative modifications at GAPDH at 313 W, HBB at 16 W, and HBB at 38 W represent a new avenue as potential therapeutic targets and prognostic biomarkers for patients with NSCLC.

4. Conclusions

Widespread tryptophan variations or tryptophan variants were identified across the proteome of patients with nonsmall cell lung cancer (NSCLC). These included 25 different tryptophan variants mapped to 2481 sites on 858 proteins, mostly involved in the cascade of oxidation in the blood microparticle, predominantly associated with reactive species in response to anaerobic glycolysis and hypoxia. The divergent Trp oxidation in HBB and GAPDH reflects and regulates the hypoxia and metabolic dysregulation involved in cancer cells, which are of potential prognostic relevance in the context of NSCLC. Overall, the results improve our understanding of the widespread tryptophan variations and highlight the predominant role of tryptophan oxidation at GAPDH at 313 W, HBB at 16 W, and HBB at 38 W in regulating redox balance of cancer cells and their potential role as prognostic biomarkers in patients with NSCLC.

Data Availability

All data used to support the findings of this study are included within the article and the supplementary information file.

Conflicts of Interest

The authors declare no conflict of interest.

Authors’ Contributions

Jing-Hua Yang and Jinfeng Chen contributed to the conceptualization; Jinfeng Chen and Lei Zhang contributed to the methodology; Hongyi Li and Zhao Sun contributed to the software; Jingyi Li, Qingqing Zhu and Hongqiang Jiang contributed to the validation; Jinfeng Chen contributed to the formal analysis; Xinli Xue, Yuanyuan Wang, and Bowen Dong contributed to the investigation; Guoqing Zhang, Zhichao Hou, and Xiangnan Li contributed to the resources; Jinfeng Chen and Lei Zhang contributed to the data curation; Jinfeng Chen contributed to the writing-original draft preparation; Jing-Hua Yang, Yang Yang, and Yongqiang Dong contributed to the writing-review and editing; Jinfeng Chen, An Zhang, and Guangyu Guo contributed to the visualization; Jing-Hua Yang contributed supervision; and Jinfeng Chen, Yongqiang Dong, Yang Yang, Lei Zhang, Xiangnan Li, and Jing-Hua Yang contributed to the funding acquisition of the study. All authors have read and agreed to the published version of the manuscript.

Acknowledgments

This research was funded by the National Natural Science Foundation of China [grant numbers 81903773, 81900521, 32070157, 31900502, and 32070623], Department of Science and Technology of Henan Province, China [grant number 201400210500], Medical Science and Technology Program of Henan Province, China [grant number 201802140], and Health Commission of Henan Province, China [grant number LHGJ20210338].

Supplementary Materials

Supplementary Figure 1: peptide spectrum matches (PSMs) of the identified 25 Trp variants. Supplementary Figure 2: proposed pathways of chemical reaction with in vivo metabolites (A) and tryptophan substitutions (B). The red-colored structures indicate the potential new modifications at tryptophan residue; the structures in grey color show the intermediates of the tryptophan modification pathway; the structures in black color show the preidentified tryptophan modifications. Supplementary Figure 3: proteins with tryptophan variants were largely clustered in blood microparticle, related to Figure 4. (a) Relative frequencies of each tryptophan modification group in the dataset. (b) Relative frequencies of delta mass clusters in the dataset. (c) Relative frequencies of each protein in the cellular components of blood microparticle, ficolin-1-rich granule lumen, and ficolin-1-rich granule. D. Heatmap depicting the correlation of tryptophan modifications in P69892 (HBG2). E. Color bar represents the relative frequency of differentially expressed oxidation modification at the 16 W, 38 W, and 131 W sites of P69892 (HBG2); the graph shows the overall structure of heme-core in P69892 (PDB: 4MQK). The linear distances of the 16 W, 38 W, and 131 W sites from the heme group are shown. Supplementary Figure 4: tryptophan variants associated with antioxidants prone to oxidative stress in NSCLC. A. Gene Ontology enrichment analysis of the modified-tryptophan-containing proteins in NSCLC, related to Figure 5(a); B. Relative expression levels of glycolytic enzymes in tumor samples and adjacent normal tissues in an independent cohort of 103 LUAD proteomic dataset (Xu et al., 2020, Cell 182, 245–261), related to Figure 6. Supplementary Figure 5: molecular docking of GAPDH (PubChem CID: 6 M61) with NAD⁺ and HBB (PubChem CID: 1CBL) with 2,3-diphosphoglycerate before and after oxidation, respectively. A. The 3D binding mode of NAD⁺ with GAPDH-wt; B. The 3D binding mode of NAD⁺ with GAPDH-wt. NAD⁺ is colored in yellow. The surrounding residues in the binding pocket are colored in light blue. The backbone of the receptor is depicted as light blue cartoon. C. The 3D binding mode of 2,3-diphosphoglycerate with HBB-wt. D. The 3D binding mode of 2,3-diphosphoglycerate with HBB-W16/W38. The 2,3-diphosphoglycerate is colored in yellow. The surrounding residues in the binding pocket are colored in light blue. The backbone and residue of HBB-wt alpha chain is colored in light blue, and the backbone and residue of HBB-wt beta chain is colored in white. The hydrogen bonds are depicted as yellow dashed lines. The salt bridges are depicted as magenta dashed lines. Supplementary Tables. Supplementary Table 1: the dataset of tryptophan modification. Supplementary Table 2: the preidentified Trp modifications was constructed based on the Unimod database (http://www.unimod.org/) and references [Bachi A, et al., Chem Rev 2013, 113 : 596-698.], related to Figure 1(b). Supplementary Table 3: the dataset of functionally grouped KEGG pathways of the modified-tryptophan-containing proteins. Supplementary Table 4: the dataset of cellular components of the modified-tryptophan-containing proteins and the corresponding frequencies per site, related to Figure 4. (Supplementary Materials)

References

P. Lappalainen, “Protein modification fine-tunes the cell’s force producers,” Nature, vol. 565, no. 7739, pp. 297-298, 2019.
View at: Publisher Site | Google Scholar
F. Carubbi, A. Alunno, R. Gerli, and R. Giacomelli, “Post-translational modifications of proteins: novel insights in the autoimmune response in rheumatoid arthritis,” Cell, vol. 8, no. 7, p. 657, 2019.
View at: Publisher Site | Google Scholar
H. Huang, B. R. Sabari, B. A. Garcia, C. D. Allis, and Y. Zhao, “SnapShot: Histone Modifications,” Cell, vol. 159, no. 2, 2014.
View at: Publisher Site | Google Scholar
K. Y. Huang, M. G. Su, H. J. Kao et al., “dbPTM 2016: 10-year anniversary of a resource for post-translational modification of proteins,” Nucleic Acids Research, vol. 44, no. D1, pp. D435–D446, 2016.
View at: Publisher Site | Google Scholar
E. V. Vinogradova, X. Zhang, D. Remillard et al., “An activity-guided map of electrophile-cysteine interactions in primary human t cells,” Cell, vol. 182, no. 4, pp. 1009–1026.e29, 2020.
View at: Publisher Site | Google Scholar
D. Zhang, Z. Tang, H. Huang et al., “Metabolic regulation of gene expression by histone lactylation,” Nature, vol. 574, no. 7779, pp. 575–580, 2019.
View at: Publisher Site | Google Scholar
D. Gilis, S. Massar, N. J. Cerf, and M. Rooman, “Optimality of the genetic code with respect to protein stability and amino-acid frequencies,” Genome Biology, vol. 2, no. 11, pp. research0049.1–research004912, 2001.
View at: Publisher Site | Google Scholar
O. Motohisa and O. Tastsuo, “Characteristic thermodynamic properties of hydrated water for 20 amino acid residues in globular Proteins1,” Journal of Biochemistry, vol. 104, no. 3, pp. 433–439, 1988.
View at: Publisher Site | Google Scholar
F. Yamakura and K. Ikeda, “Modification of tryptophan and tryptophan residues in proteins by reactive nitrogen species,” Nitric Oxide, vol. 14, no. 2, pp. 152–161, 2006.
View at: Publisher Site | Google Scholar
M. Moller and A. Denicola, “Protein tryptophan accessibility studied by fluorescence quenching,” Biochemistry and Molecular Biology Education, vol. 30, no. 3, pp. 173–178, 2002.
View at: Google Scholar
S. Madabushi, A. K. Gross, A. Philippi, E. C. Meng, T. G. Wensel, and O. Lichtarge, “Evolutionary trace of G protein-coupled receptors reveals clusters of residues that determine global and class-specific functions,” The Journal of Biological Chemistry, vol. 279, no. 9, pp. 8126–8132, 2004.
View at: Publisher Site | Google Scholar
J. J. Huntley, W. Fast, S. J. Benkovic, P. E. Wright, and H. J. Dyson, “Role of a solvent-exposed tryptophan in the recognition and binding of antibiotic substrates for a metallo-beta-lactamase,” Protein Science, vol. 12, no. 7, pp. 1368–1375, 2003.
View at: Publisher Site | Google Scholar
Y. Xie and J. B. Cohen, “Contributions of _Torpedo_ nicotinic acetylcholine receptor γTrp-55 and δTrp-57 to agonist and competitive antagonist function,” The Journal of Biological Chemistry, vol. 276, no. 4, pp. 2417–2426, 2001.
View at: Publisher Site | Google Scholar
A. Previero, M. A. Coletti-Previero, and P. Jolles, “Localization of non-essential tryptophan residues for the biological activity of lysozymet,” Journal of Molecular Biology, vol. 24, no. 2, pp. 261–268, 1967.
View at: Publisher Site | Google Scholar
S. Ozaki, I. Hara, T. Matsui, and Y. Watanabe, “Molecular engineering of myoglobin: the improvement of oxidation activity by replacing Phe-43 with tryptophan,” Biochemistry, vol. 40, no. 4, pp. 1044–1052, 2001.
View at: Publisher Site | Google Scholar
T. D. Pfister, T. Ohki, T. Ueno et al., “Monooxygenation of an aromatic ring by F43W/H64D/V68I myoglobin mutant and hydrogen peroxide: myoglobin mutants as a model for P450 hydroxylation chemistry,” The Journal of Biological Chemistry, vol. 280, no. 13, pp. 12858–12866, 2005.
View at: Publisher Site | Google Scholar
M. Ehrenshaft, L. J. Deterding, and R. P. Mason, “Tripping up Trp: modification of protein tryptophan residues by reactive oxygen species, modes of detection, and biological consequences,” Free Radical Biology & Medicine, vol. 89, pp. 220–228, 2015.
View at: Publisher Site | Google Scholar
A. R. Correia, S. Y. Ow, P. C. Wright, and C. M. Gomes, “The conserved Trp155 in human frataxin as a hotspot for oxidative stress related chemical modifications,” Biochemical and Biophysical Research Communications, vol. 390, no. 3, pp. 1007–1011, 2009.
View at: Publisher Site | Google Scholar
F. Yamakura, T. Matsumoto, K. Ikeda et al., “Nitrated and oxidized products of a single tryptophan residue in human Cu,Zn-superoxide dismutase treated with either peroxynitrite-carbon dioxide or myeloperoxidase-hydrogen peroxide-nitrite,” Journal of Biochemistry, vol. 138, no. 1, pp. 57–69, 2005.
View at: Publisher Site | Google Scholar
S. M. Vaz, F. M. Prado, P. Di Mascio, and O. Augusto, “Oxidation and nitration of ribonuclease and lysozyme by peroxynitrite and myeloperoxidase,” Archives of Biochemistry and Biophysics, vol. 484, no. 2, pp. 127–133, 2009.
View at: Publisher Site | Google Scholar
S. K. Jeffrey, H. R. Paul, A. C. David, and A. Arthur, “High-resolution x-ray study of deoxyhemoglobin rothschild 37.beta. Trp .fwdarw. Arg: a mutation that creates an intersubunit chloride-binding site,” Biochemistry, vol. 31, no. 16, pp. 4111–4121, 1992.
View at: Publisher Site | Google Scholar
M. C. Owen, P. A. Ockelford, and R. M. G. Wells, “HB Howick [β37(C3)TRP→GLY]: a new high oxygen affinity variant of the α1β2 contact,” Hemoglobin, vol. 17, no. 6, pp. 513–521, 1993.
View at: Publisher Site | Google Scholar
A. T. Kong, F. V. Leprevost, D. M. Avtonomov, D. Mellacheruvu, and A. I. Nesvizhskii, “MSFragger: ultrafast and comprehensive peptide identification in mass spectrometry-based proteomics,” Nature Methods, vol. 14, no. 5, pp. 513–520, 2017.
View at: Publisher Site | Google Scholar
R. Aebersold and M. Mann, “Mass-spectrometric exploration of proteome structure and function,” Nature, vol. 537, no. 7620, pp. 347–355, 2016.
View at: Publisher Site | Google Scholar
M. Bern, Y. J. Kil, and C. Becker, “Byonic: advanced peptide and protein identification software,” Current Protocols in Bioinformatics, vol. 40, 2012.
View at: Publisher Site | Google Scholar
N. A. Kulak, G. Pichler, I. Paron, N. Nagaraj, and M. Mann, “Minimal, encapsulated proteomic-sample processing applied to copy-number estimation in eukaryotic cells,” Nature Methods, vol. 11, no. 3, pp. 319–324, 2014.
View at: Publisher Site | Google Scholar
M. S. Kim, S. M. Pinto, D. Getnet et al., “A draft map of the human proteome,” Nature, vol. 509, no. 7502, pp. 575–581, 2014.
View at: Publisher Site | Google Scholar
Y. Duan, J. Li, F. Wang et al., “Protein modifications throughout the lung cancer proteome unravel the cancer- specific regulation of glycolysis,” Cell Reports, vol. 37, no. 12, pp. 110137–110137, 2021.
View at: Publisher Site | Google Scholar
C. Fraley, A. E. Raftery, T. B. Murphy, and L. Scrucca, Mclust Version 4 for R: Normal Mixture Modeling for Model-Based Clustering, Classification, and Density Estimation, Department of Statistics University of Washington, Seattle, USA, 2012.
C. Fraley and A. E. Raftery, “How many clusters? Which clustering method? Answers via model-based cluster analysis,” The Computer Journal, vol. 41, no. 8, pp. 578–588, 1998.
View at: Publisher Site | Google Scholar
A. Bachi, I. Dalle-Donne, and A. Scaloni, “Redox proteomics: chemical principles, methodological approaches and biological/biomedical promises,” Chemical Reviews, vol. 113, no. 1, pp. 596–698, 2013.
View at: Publisher Site | Google Scholar
G. Bindea, B. Mlecnik, H. Hackl et al., “ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks,” Bioinformatics, vol. 25, no. 8, pp. 1091–1093, 2009.
View at: Publisher Site | Google Scholar
D. Szklarczyk, A. L. Gable, D. Lyon et al., “STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets,” Nucleic Acids Research, vol. 47, no. D1, pp. D607–D613, 2019.
View at: Publisher Site | Google Scholar
P. Shannon, A. Markiel, O. Ozier et al., “Cytoscape: a software environment for integrated models of biomolecular interaction networks,” Genome Research, vol. 13, no. 11, pp. 2498–2504, 2003.
View at: Publisher Site | Google Scholar
G. Yu, L. G. Wang, Y. Han, and Q. Y. He, “clusterProfiler: an R package for comparing biological themes among gene clusters,” OMICS, vol. 16, no. 5, pp. 284–287, 2012.
View at: Publisher Site | Google Scholar
Molecular Operating Environment (MOE) CCGI, 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7, 2014.
Y. Yan, X. Zang, C. S. Jamieson et al., “Biosynthesis of the fungal glyceraldehyde-3-phosphate dehydrogenase inhibitor heptelidic acid and mechanism of self-resistance,” Chemical Science, vol. 11, no. 35, pp. 9554–9562, 2020.
View at: Publisher Site | Google Scholar
G. E. Borgstahl, P. H. Rogers, and A. Arnone, “The 19 Å structure of deoxyβ₄ hemoglobin: analysis of the partitioning of quaternary-associated and ligand-induced changes in tertiary structure,” Journal of Molecular Biology, vol. 236, no. 3, pp. 831–843, 1994.
View at: Publisher Site | Google Scholar
J. O. Eriksson, M. Rezeli, M. Hefner, G. Marko-Varga, and P. Horvatovich, “Clusterwise peak detection and filtering based on spatial distribution to efficiently mine mass spectrometry imaging data,” Analytical Chemistry, vol. 91, no. 18, pp. 11888–11896, 2019.
View at: Publisher Site | Google Scholar
F. Gomez-Pinilla and E. Tyagi, “Diet and cognition: interplay between cell metabolism and neuronal plasticity,” Current Opinion in Clinical Nutrition and Metabolic Care, vol. 16, no. 6, pp. 726–733, 2013.
View at: Publisher Site | Google Scholar
J. Kim, H. M. Lee, F. Cai et al., “The hexosamine biosynthesis pathway is a targetable liability in _KRAS_ / _LKB1_ mutant lung cancer,” Nature Metabolism, vol. 2, no. 12, pp. 1401–1412, 2020.
View at: Publisher Site | Google Scholar
N. Bagwan, E. Bonzon-Kulichenko, E. Calvo et al., “Comprehensive quantification of the modified proteome reveals oxidative heart damage in mitochondrial heteroplasmy,” Cell Reports, vol. 23, no. 12, pp. 3685–3697.e4, 2018.
View at: Publisher Site | Google Scholar
H. D. Bismuth, G. Brasseur, B. Ezraty, and L. Aussel, “Bacterial genetic approach to the study of reactive oxygen species production in galleria mellonella during salmonella infection,” Frontiers in Cellular and Infection Microbiology, vol. 11, pp. 1–8, 2021.
View at: Publisher Site | Google Scholar
W. Xiang, V. Weisbach, H. Sticht et al., “Oxidative stress-induced posttranslational modifications of human hemoglobin in erythrocytes,” Archives of Biochemistry and Biophysics, vol. 529, no. 1, pp. 34–44, 2013.
View at: Publisher Site | Google Scholar
S. Jawon, J. Jaeho, M. K. Young, H. Narae, P. Eunok, and L. Kong-Joo, “Strategy for comprehensive identification of post-translational modifications in cellular proteins, including low abundant modifications: application to glyceraldehyde-3-phosphate dehydrogenase,” Journal of Proteome Research, vol. 7, no. 2, pp. 587–602, 2008.
View at: Publisher Site | Google Scholar
A. M. Rostila, S. L. Anttila, M. M. Lalowski et al., “Reactive oxygen species-regulating proteins peroxiredoxin 2 and thioredoxin, and glyceraldehyde-3-phosphate dehydrogenase are differentially abundant in induced sputum from smokers with lung cancer or asbestos exposure,” European Journal of Cancer Prevention, vol. 29, no. 3, pp. 238–247, 2020.
View at: Publisher Site | Google Scholar
A. L. Samson, A. S. Knaupp, I. Kass et al., “Oxidation of an exposed methionine instigates the aggregation of glyceraldehyde-3-phosphate dehydrogenase,” The Journal of Biological Chemistry, vol. 289, no. 39, pp. 26922–26936, 2014.
View at: Publisher Site | Google Scholar
M. A. Tossounian, B. Zhang, and I. Gout, “The writers, readers, and erasers in redox regulation of GAPDH,” Antioxidants, vol. 9, no. 12, pp. 1288–1310, 2020.
View at: Publisher Site | Google Scholar
J. L. Jenkins and J. J. Tanner, “High-resolution structure of human D-glyceraldehyde-3-phosphate dehydrogenase,” Acta Crystallographica. Section D, Biological Crystallography, vol. 62, no. 3, pp. 290–301, 2006.
View at: Publisher Site | Google Scholar
C. T. Lee, M. K. Boss, and M. W. Dewhirst, “Imaging tumor hypoxia to advance radiation oncology,” Antioxidants & Redox Signaling, vol. 21, no. 2, pp. 313–337, 2014.
View at: Publisher Site | Google Scholar
H. Mairbaurl and R. E. Weber, “Oxygen transport by hemoglobin,” Comprehensive Physiology, vol. 2, no. 2, pp. 1463–1489, 2012.
View at: Publisher Site | Google Scholar
J. Y. Xu, C. Zhang, X. Wang et al., “Integrative proteomic characterization of human lung adenocarcinoma,” Cell, vol. 182, no. 1, pp. 245–261.e17, 2020.
View at: Publisher Site | Google Scholar
S. Matsumoto, S. Kishimoto, K. Saito et al., “Metabolic and physiologic imaging biomarkers of the tumor microenvironment predict treatment outcome with radiation or a hypoxia-activated prodrug in mice,” Cancer Research, vol. 78, no. 14, pp. 3783–3792, 2018.
View at: Publisher Site | Google Scholar
G. L. Semenza, “Oxygen sensing, hypoxia-inducible factors, and disease pathophysiology,” Annual Review of Pathology, vol. 9, no. 1, pp. 47–71, 2014.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Jinfeng Chen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

462

Downloads

516

Citations