Abstract
Background/Aim: N-glycans are potential serum biomarkers due to their aberrant structure and abundance alteration during disease progression. Few studies have been associated with relative quantitative N-glycans profiling during different gastric disease stages. In this study, we conducted an investigation on the profiling of N-glycans in patients with gastric disease, as well as in healthy controls. Materials and Methods: In this study, the porous graphitization carbon chromatography-high resolution Fourier transform mass spectrometry (PGC-FTMS) method was applied to assess comprehensive N-glycans profiling in patients at different stages of gastric disease, including gastritis, atrophic gastritis, gastric ulcer, gastric polyps, and gastric cancer. Results: A total of 45 N-glycans (relative abundance >0.1%) were detected, and 9 N-glycans were found to be potential biomarkers for gastric disease detection. Along with the progression of gastric disease, the abundance of sialylated N-glycans increased, while that of core-fucosylated N-glycans decreased. Multivariate statistical analysis demonstrated that N-glycans profiling between gastritis and healthy controls had significant differences. The characteristic N-glycans distinguished gastric cancer from healthy controls, which had strong clinical diagnostic value. Conclusion: The relative quantitative profile of N-glycans in different gastric disease stages was revealed and serum N-glycans are proposed for distinguishing gastric disease stages in clinical application.
- Gastric disease
- N-glycans
- serum
- porous graphitic carbon liquid chromatography-mass spectrometry
- biomarker
Gastric disease is a predominate digestive tract disease. Due to social competition, life rhythm speed, and the intense mental pressure, the incidence rate of digestive diseases is increasing year by year, resulting in individuals exhibiting sub-healthy gastric conditions. Gastric disorder with high prevalence and chronic persistence is associated with a variety of risk factors, including smoking, obesity, high-salt diet, and infection with gram-negative H. pylori (1-3). However, gastric disease generally presents no specific symptoms in the initial stages and is thus typically diagnosed in advanced stages (4). Gastric cancer is one of the most prevalent cancers worldwide, with over 1,000,000 new cases in 2020 and 796,000 deaths (2). The incidence and mortality rate are the highest in East Asian countries, such as China, Japan, and South Korea, making gastric cancer the fifth most common cancer and the fourth leading cause of cancer-related death (5). Gastric cancer is primarily discovered in locally advanced or metastatic stages, rendering an extremely poor prognosis (6). Biochemical indicators often change in gastric disease, such as tissue sirtuin 1 expression, structural alterations and dysfunction of E-cadherin during gastric cancer progression and abnormal expression of lncRNA and microRNAs in gastric cancer (7-9). Until recently, several attempts have been made to discover biomarkers for gastric disease, including CEA, CA19-9, and CA72-4, which lack sensitivity and are not useful in the detection of premalignant gastric cancer lesions (10-12). Of particular concern is that no reliable serological markers are available for early diagnosis, monitoring, and prognosis of patients. Thus, it is imperative to develop sensitive, specific, economical, and non-invasive biomarkers for gastric disease to improve the survival and prognosis of patients.
Glycosylation is the basis of protein modification, that can alter the function of proteins and play a key role in several pathophysiological processes of cancer, including cell adhesion, migration, interaction with extracellular matrix, immune surveillance, cell signaling, and cell metabolism (13-15). Given the crucial function of glycosylation in tumor biology at the onset and during disease progression, glycans could be a source for development of new non-invasive biomarkers (16, 17). Glycans are covalent assemblies of oligosaccharides and polysaccharides, which can usually be divided into N-glycans and O-glycans. N-glycans are post-translational modifications through linkage of asparagine (Asn) and peptide chain. Serological glycomic profiling is an emerging non-invasive screening tool for discovering potential markers in the diagnosis of cancer. N-glycans have been closely linked to various pathological conditions and have been recognized as potential biomarkers in some diseases, including colorectal cancer, breast cancer, thyroid cancer, and diabetes (18-21). Relevant studies on N-glycans in patients with gastric cancer have also confirmed that N-glycans change during the course of cancer, with several studies suggesting that high mannose N-glycans, complex glycans, and galactosylated di-antennary glycans decreased in gastric cancer, whereas non-galactose forms of di-antennary, three-antenna, and four-antenna fucosylated N-glycans increased (22, 23). In addition, the low core-fucosylation of N-glycans has been confirmed in gastric cancer tissues (24). Previous studies have demonstrated that the level of sialyl Lewis X epitopes present on triantennary glycans in gastric cancer increased compared to healthy controls (25). However, these studies focused on tissues or had complicated pretreatment, which restricted invasion or required relatively large amounts of biological samples. With the advancement of high-throughput mass spectrometry platforms (MS) and liquid chromatography separation technologies (LC), LC-MS has become an indispensable and unparalleled tool in providing definitive structural information on glycans for biomarker discovery (26-28).
In our study, we performed an accurate and systematic analysis of the serum N-glycans in patients with gastric disease based the porous graphitization carbon chromatography-high resolution Fourier transform mass spectrometry (PGC-FTMS) approach. The main purpose of our work was to understand the changes in N-glycans between patients with gastric disease and healthy individuals and identify differences related to disease stages.
Materials and Methods
Chemicals and reagents. Anthranilic acid (2-AA), dimethyl sulfoxide (DMSO), sodium cyanoborohydride (NaBH3CN), ammonium bicarbonate, and glacial acetic acid were purchased from Sigma Aldrich (St. Louis, MO, USA). PNGase F digestion agents (P0709S, glycerol-free) were purchased from New England Biolabs (NEB, USA). Sodium dodecyl sulfate (SDS), dithiothreitol (DTT), and nonidet P 40 (NP-40) were purchased from Solarbio (Beijing, PR China). Acetonitrile (ACN) was purchased from Merck (Darmstadt, Germany), and other reagents were analytical grade.
Serum preparation. The serum samples of healthy volunteers and patients at different stages of gastric disease were donated by the Affiliated Hospital of Qingdao University (Qingdao, PR China) with written informed consent of all patients following ethical standards of Qingdao University. We collected a total of 36 healthy control donors (NC), 21 gastritis (GT), 9 atrophic gastritis (AG), 5 gastric ulcer (GU), 9 gastric polyps (GP), and 28 gastric cancer (GC) samples with histological verification. Venous blood was collected into a vacuette pipet and left to coagulate at room temperature for 30 min. The tube was subsequently centrifuged at 3,000 rpm for 15 min, and the separated serum was aliquoted, transported, and stored at −80°C until use.
N-glycan release and derivative. Denaturing buffer (5% SDS, 0.4 mol/L DTT) was added to 20 μl serum, and the samples were incubated at 100°C for 5 min. Then, 0.5 mol/L sodium phosphate (pH 7.0) and 10% NP-40 were added to neutralize the denaturing reagent when it cooled to room temperature, which was then shaken for 10 s before being incubated with PNGase F (500 units) for 10 h at 37°C. N-glycans were subsequently released from macromolecular proteins. Following alcohol precipitation, the precipitate was removed by centrifugation at 10,000 rpm. The supernatant was collected and lyophilized, and the lyophilized sample was labeled by 2-AA. The glycans were then mixed with 20 μl labeling solution [0.32 mol/l 2-AA and 1.0 mol/l of NaCNBH3 dissolved in 1.0 ml of 4% methanol/2% sodium acetate-boric acid solution (w/v)]. All labeled solutions were freshly prepared. Derivatization was then performed by incubation at 65°C for 2 h without light. Finally, the labeled N-glycans were analyzed by PGC-FTMS.
PGC-FTMS analysis. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis was performed with an LTQ Orbitrap XL MS (Thermo Scientific, Waltham, MA, USA) using the electrospray ionization (ESI)-negative mode. The spray voltage of the MS spectra was set to 3 kV, with a capillary voltage of −41 V, a lens voltage of −120 V, a capillary temperature of 275°C, a sheath gas flow rate of 8 L/min, and a mass range of 280-3,000 m/z. An Agilent 1290 UPLC (Agilent Technologies, Wilmington, DE, USA) system was utilized for separation on a porous graphitic carbon column (PGC, 150 mm×2.1 mm, 3 μm, Thermo Scientific) at 30°C with a gradient elution program at a flow rate of 120 μl/min. The mobile phases consisted of (A) 10 mmol/l ammonium bicarbonate and (B) acetonitrile. The gradient elution program was as follows: 12-30% B (10-40 min) and 30-75% B (40-52 min).
Statistical analysis. LC-MS/MS data were generated by Xcalibur, Decon2LS, GlycResoft, and GlycoWorkBench software. The candidate N-glycan residue compositions were obtained by given limited conditions in the Generate Composition Hypothesis of GlycResoft as follows: Hex (C6H10O5, 3-10), HexNAc (C8H13NO5, 2-9), Fuc (C6H10O4, 0-6), NeuAc (C11H17NO8, 0-6), and 2-AA (C7H7NO, 1). The combination relationship was set as HexNAc+Hex ≥5, HexNAc+Hex-Fuc ≥4, and HexNAc+Hex-NeuAc ≥5 to ensure core pentasaccharide structure, the rationality of the fucose site, and the sialic acid site, respectively. Fragment ions were then assigned using GlycoWorkBench. Finally, the accurate structures of the glycans were obtained by the MS and MS2 spectra and the matching results generated by GlycoWorkBench. The relative abundance of N-glycans was calculated by dividing the individual N-glycan abundance by the total N-glycans abundance.
All the reported t-tests with two-tailed distribution were generated using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA) to evaluate the difference of N-glycan expression in patients with gastric disease and healthy controls. The glycomic data with statistical significance (p<0.05) were assessed by the receiver operator characteristics (ROC) test using SPSS v. 22.0 (IBM Corporation, Armonk, NY, USA). Heatmap and OPLS-DA model were used to realize the visualization of multiple information.
Results
Separation of serum N-glycans by LC-MS/MS. Chromatograms of serum N-glycans were analyzed using GPC-FTMS. The retention time of N-glycans could be analyzed in 22 min, and the amounts of N-glycans were calculated based on the areas of their mass chromatograms. N-glycans are different from peptides or proteins because they contain same pentasaccharide core sequences and are synthesized by a well-defined pathway. Thus, mass spectrometry information can provide the structure of N-glycan, including monosaccharide composition, sequence and linkage. Raw Liquid chromatography–mass spectrometry (LC-MS/MS) data was deconvoluted by Decon2LS, then N-glycan structure was matched and screened using GlycResoft and GlycoWorkBench. Extracted ion chromatograms (EICs) represent the signals from all ion species associated with a single compound. Therefore, each EIC peak can be attributed to a single distinct compound (Figure 1). Serum N-glycan analysis identified 45 types of 2-AA labeled-N-glycans (relative abundance >0.1%). In the healthy control group, 56.64% sialylated N-glycans (C/H Sia), 17.08% fucosylated N-glycans (C/H Fuc), 23.06% sialylated and fucosylated N-glycans (C/H Sia & Fuc), 2.65% high mannose N-glycans (high man), and 0.15% complex/hybrid N-glycans (C/H) were detected. Significant differences in the N-glycans profiling between the healthy controls and patients at different stages of gastric disease were shown in Figure 2a.
Extracted ion chromatograms of N-glycans found in healthy controls. Colors denote different N-glycan classes: high man (green); C/H (blue); C/H Fuc (red); C/H Sia (purple); and C/H Sia & Fuc (orange). Structures shown are merely putative assignments based on previously published structural characterization of common N-glycans in human serum (29).
Comparative analysis of serum N-glycans profiling. (a) The relative abundance of N-glycans between healthy controls and patients with gastric disease. (b-f) The ROC curve of N-glycans with potential diagnostic value in different gastric disease stages. *p<0.05, **p<0.01, ***p<0.001. NC: Normal control; GT: gastritis; AG: atrophic gastritis; GU: gastric ulcer; GP: gastric polyps; GC: gastric cancer.
N-glycans differentiate patients at different stages of gastric disease. LC-MS mass spectra of N-glycans profiling of patients with gastric disease and healthy controls were preprocessed for difference comparison. ROC curve and t-tests were used to investigate N-glycans that had statistically significant expression differences among five kinds of gastric disease when compared to the healthy control group. Lower levels of fucosylated N-glycans and higher levels of sialylated N-glycans were found with progression of gastric disease, relative to healthy control donors, as shown in Figure 2a. C/H fucosylated N-glycans of disease groups were significantly reduced by a 0.62-0.89-fold change, whereas C/H sialylated N-glycans were more clearly increased 1.07-1.10 fold in advanced disease groups. ROC analysis of N-glycans between patients with gastric disease and healthy controls assessed the predictive ability of the biomarker candidates. The glycan-modification, with area under the curve (AUC) greater than 0.7, showed potential for clinical application. When compared to the NC group, high mannose type N-glycans indicated relative sensitivity in GT, AG, and GU, AUC was 0.7128, 0.8438, and 0.7188, respectively (Figure 2b). AG and GU demonstrated excellent diagnostic accuracy in complex/hybrid N-glycans, with AUC greater than 0.9 (Figure 2c). The AUC of fucosylated or sialylated N-glycans was greater than 0.7, indicating strong diagnostic value in more severe gastric disease (Figure 2d, e). While fucosylated and sialylated N-glycans with AUC less than 0.7 in different stages gastric disease exhibited low sensitivity and specificity, which showed weak clinical diagnostic value (Figure 2f).
Forty-five N-glycans were identified using GlycResoft software, and t-tests were used to analyze the difference of N-glycans between patients at different stages of gastric disease and healthy controls. Multiple N-glycans with potential diagnostic value for different stages of gastric disease were listed. The levels of core fucosylated and monosialylated/core fucosylated N-glycans decreasing significantly with disease progression (Figure 3). Non-fucosylated mono- and di-sialylated biantennary N-glycans demonstrated an increasing trend in the box plot, with N-glycan expression levels subsequently increasing during disease progression (Figure 4). These results indicated increasing levels of sialylated N-glycans to be a common molecular feature in gastric cancer, which was consistent with previous studies (30-32).
The dot plot of decreased relative abundance of N-glycan expression in patients at different stages of gastric disease and healthy controls. *p<0.05, **p<0.01, ***p<0.001. NC: Normal control; GT: gastritis; AG: atrophic gastritis; GU: gastric ulcer; GP: gastric polyps; GC: gastric cancer.
Dot plot of increased relative abundance of N-glycan expression in patients at different stages of gastric disease and healthy controls. *p<0.05, **p<0.01, ***p<0.001. NC: Normal control; GT: gastritis; GC: gastric cancer.
Multivariate statistical analysis of potential N-glycan biomarkers. Orthogonal partial least squares discriminant analysis (OPLS-DA) is a common statistical tool for biomarker discovery and predictive model training (33). The OPLS-DA model quality is assessed by its ability to fit (R2) and predict (Q2) variance of the data. OPLS-DA model was used to preliminarily distinguish patients at different stages of gastric disease and healthy controls, Q2 value was 0.287, as shown in Figure 5a. Through the OPLS-DA model, we determined that GC and NC are well distinguished, while the separation between GT and NC was slightly weaker. Other disease groups did not exhibit strong separation, which might be affected by a smaller number of samples.
Orthogonal partial least squares discriminant analysis (OPLS-DA) model of N-glycans of different gastric disease groups versus healthy controls using LC-MS data. (a) OPLS-DA score plot of different gastric disease groups and healthy controls. (b) OPLS-DA score plot of gastritis group (blue) and healthy controls (green). (c) OPLS-DA score plot of atrophic gastritis group (red) and healthy controls (green). (d) OPLS-DA score plot of gastric cancer group (purple) and healthy controls (green). (e) VIP map of N-glycan prediction. (f) Validation of model proved based on 200-permutation of the six components. Intercepts: R2=0.218 (green triangles) and Q2=−0.591 (blue squares). NC: Normal control; GT: gastritis; AG: atrophic gastritis; GU: gastric ulcer; GP: gastric polyps; GC: gastric cancer.
Preliminary statistical analysis based on OPLS-DA model indicated that there would be a certain degree of separation between different groups. Further, the OPLS-DA model was applied to analyze two distinct groups. Differences between the healthy controls and gastritis and atrophic gastritis and gastric cancer groups were shown in Figure 5b-d. We determined that this separation became more obvious with disease progression. For the healthy controls and gastritis group, the OPLS-DA model showed strong separation [R2Y (cum) was 94.4% and predictive ability Q2 (cum) was 85.8%] (Figure 5b). The OPLS-DA model also showed clear separation ability [R2Y (cum) was 96.0%] and prediction ability [Q2 (cum) was 79.2%] in the healthy controls and atrophic gastritis group (Figure 5c). We observed that in the early and middle stages of gastric disease, the difference between N-glycans was even more pronounced. The OPLS-DA model demonstrated stronger separation ability [R2Y (cum) was 93.5%] and prediction ability [Q2 (cum) was 88.5%] in the healthy controls and gastric cancer group (Figure 5d). By calculating the Variable Importance for the Projection (VIP), the influence intensity and explanatory ability of N-glycan on sample classification were evaluated to assist in the screening of biomarkers. The results were shown in Figure 5e, N-glycans with VIP value greater than one were relatively important in the progression of gastric disease, which had good predictive ability. As shown in Figure 5f, characteristic N-glycans were found between the gastric cancer group and the healthy controls, with R2 greater than 0.2, which had potential clinical diagnostic value.
According to color order rule, the relative abundance of N-glycan was converted to a heatmap (Figure 6a). We observed the serum N-glycan expression profiling in patients at different stages of gastric disease and the healthy controls intuitively. In sialylated N-glycans, H5N4S2 and H5N4S1 were upregulated gradually, and H6N5S3 decreased first and then increased. Fucosylated N-glycans containing H5N4F1 and H4N4F1 showed a downward trend. N-glycans profiling indicated that gastric disease progression would be accompanied by N-glycan changes, and obvious differences in the expression of N-glycans could become disease markers in later clinical research.
N-glycan cluster analysis and correlation between different stages of gastric disease and healthy controls. (a) Heatmap of the expression levels of N-glycans in the serum from different gastric disease patients. Red: fluorescence signal upregulation. Green: signal downregulation. Black: no clear link. (b) Pearson correlation analysis of healthy controls, gastritis group, and gastric cancer group. NC: Normal control; GT: gastritis; AG: atrophic gastritis; GU: gastric ulcer; GP: gastric polyps; GC: gastric cancer.
In addition, Pearson regression analysis was applied to analyze the correlation of N-glycan content among groups (Figure 6b). The data indicated that the correlation coefficient between the two sialic acid modified N-glycans was 0.45, including H4N3S1 and H5N4S1, demonstrating a moderate correlation between the healthy controls and the gastritis group. The correlation coefficient between H4N4F1 and H4N4F1S1 in the gastric cancer group and the gastritis group was 0.57, which is also a moderate correlation. N-glycan correlation can provide the basis for a stronger diagnostic model for future studies on potential combination therapies.
Based on multivariate statistical analysis and t-tests, 9 N-glycans were screened out from 45 detected N-glycans were listed in Table I, with significant changes in the progression of gastric disease and clinical diagnostic value. These N-glycans were mainly core-fucosylated and sialylated N-glycans. Characteristic N-glycans not only showed significant differences in gastric cancer, but also indicated abnormal expression in premalignant gastric cancer lesions.
Characteristic variation of N-glycans in gastric disease groups.
Detailed analysis of potential N-glycan biomarkers by LC-MS/MS. Following preliminary informatization of N-glycans to obtain qualitative and quantitative results. Based on first-order MS, we used Xcalibur software to extract tandem mass spectrometry (MS/MS) spectra and analyzed the spectra to confirm fine structure of N-glycans.
NeuAc is a major component of human serum N-glycans. The primary method of binding is to connect non-reductive Gal residues through α2-3 or GalNAc residues through α2-6. As shown in Figure 7a, the m/z 306 ion corresponding to 0,4A2α–CO2 only appeared in the spectrum of NeuAc2-6-linked lactose, while the m/z 408 fragment ion can be used as diagnostic ion of α2-3, these ions readily differentiate the 2-3 and 2-6 linkages (34-36).
The structure of N-glycans was analyzed by tandem mass spectrometry (MS/MS). (a) MS/MS spectra of monosialylated N-glycans. (b) MS/MS spectra of fucosylated N-glycans.
Fucose in N-glycans is primarily linked to the labeled GlcNAc to form core fucose types, or to the branched GlcNAc. In the case of non-derivative labeling, the core fucose will exhibit 368, 350, 571, or 553 characteristic ions, with no single fracture of GlcNAc in the core pentasaccharides and no 2,4 trans-ring fracture of the end GlcNAc (37, 38). In Figure 7b, the presence of fragment ions m/z 690.21 and m/z 487.31 indicated that the fucose substitution site was GlcNAc at the reductive end.
Discussion
N-glycan alterations have been reported in serum proteins and plasma of patients with gastric cancer. However, alterations in serum N-glycans in different gastric disease stages have been rarely reported. Recent advances in MS and bioinformatics have accelerated glycomics research and have provided a new paradigm for cancer biomarker discovery. In this study, we performed serum N-glycan analysis using GPC-FTMS and successfully detected abnormal expression of N-glycans in patients with gastric disease. Using this serum N-glycan profiling assay, a panel of serum N-glycans were successfully identified as potential markers for detecting gastric disease and cancer. Several N-glycans with potential diagnostic values were significantly correlated with the serological markers during occurrence and progression of gastric disease.
In our study, core fucosylated N-glycans decreased, demonstrating that this glycol-subclass is more likely to be a biomarker, particularly for the early detection of gastric disease. In previous studies, core fucose N-glycan residues decreased in both tissues and serum from patients with gastric cancer, resulting from the lower expression of fucosyltransferase. The core-fucosylation process was catalyzed by the only FUT 8 fucosyltransferase, and the expression level of FUT 8 protein exhibited a downward trend in gastric cancer (39, 40). Upregulation of core fucosylation can inhibit proliferation of human gastric cancer cells. Core fucose decreases in gastric cancer can be attributed to core-fucosylation being downregulated in gastric cancer (41, 42). We also reported a similar decrease in gastric cancer as well as early gastric disease. Core-fucosylation N-glycan H5N4F1, for example, was decreased in AG patients, with AUC score of 0.94. In recent studies, a comparative analysis of exosome N-glycans from the urine of healthy controls and patients with gastric cancer demonstrated that the content of fucosylated N-glycans exhibited a decreasing trend, while the content of sialic acid type N-glycans significantly increased (43). The same trend also appeared in patients with prostate cancer and lung cancer (44-47).
Further analysis showed an increase in the level of sialylated N-glycan species, particularly in gastric cancer. Previous research indicated that sialyl Lewis X increased in the total serum glycome (48, 49). IgG, haptoglobin, transferrin, and alpha1-acid glycoprotein may be protein biomarkers with core-fucosylated agalactosyl biantennary glycans and sialylation increasing (50, 51). H5N4S2 which was the most abundant N-glycan in human beings, increased significantly compared to healthy controls. This clear difference may help reveal the mechanism for the occurrence and development of gastric cancer.
We identified several glyco-subclasses on the basis of their characteristic structures and showed that N-glycan profiling of healthy controls and patients with gastric cancer could be clearly distinguished. To assess novel biomarkers with clinical diagnostic value, an OPLS-DA model was used to examine the N-glycan isolation in patients with gastric disease and healthy controls. A strong separation and prediction ability were consistent with previous OPLS-DA models. R2 and Q2 values should be more than 0.2 clinically, with Q2 being 0.287 in this study. These results indicated that the diagnostic models based on N-glycan markers were valuable and non-invasive alternatives for identifying gastric cancer. The method used partial least squares regression to establish a relationship model between metabolite expression and sample category and to realize the prediction of sample category.
Conclusion
In this study, we performed high-throughput and comprehensive PGC-FTMS for semi-quantitative N-glycan expression profiling of serum N-glycan analysis. OPLS-DA statistical model combined with ROC curve analysis was used to judge the diagnostic ability. According to the establishment of the OPLS-DA model among groups, the separation degree and predictive ability of each group were analyzed. Using multivariate statistical analysis of the characteristics of N-glycan expression profiles, we provide a basis for the glycomics study of advanced gastric diseases. The analysis of N-glycans in patients at various stages of gastric disease not only helps in understanding the significant role of N-glycans in disease progression but also facilitates the discovery of potential biomarkers. These findings have the potential to contribute to the diagnosis and treatment of gastric diseases.
Footnotes
Authors’ Contributions
Conceptualization, G.L. and G.Y; methodology, X.J., and Q.L.; validation, W.Z. and C.W.; formal analysis, X.J., W.Z. and Q.H; investigation, G.L.; resources, J.Z. and X.L; data curation, X.J. and W.Z.; writing—original draft, X.J., W.Z. and Q.H.; writing—review & editing, G.L. and H.J.; visualization, X.J. and Q.H; supervision, G.L.; project administration, G.L.; funding acquisition, G.L. H.J. and G.Y.; All Authors have read and agreed to the published version of the manuscript.
Funding
This research was funded by the National Natural Science Foundation of China (31600646, 81802777), Key Scientific and Technological Projects of Shandong Province (2021ZDSYS22, 2021KJ012), Qingdao Marine Science and Technology Center (2022QNLM030003-2), Qingdao Basic and Applied Research Project (18-2-2-25-jch), the Fundamental Research Funds for the Central Universities (201762002), and the Taishan Scholar Climbing Project (TSPD20210304).
Conflicts of Interest
The Authors declare that they have no competing interests.
- Received August 24, 2023.
- Revision received October 24, 2023.
- Accepted October 30, 2023.
- Copyright © 2024, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).