Abstract

   Metabolomics and 16S rDNA sequencing have shown great potential in
   elucidating complex mechanisms associated with diseases. Currently,
   there is little research on the omics of gastric cancer and it lacks
   effective biomarkers.

Objective

   Based on plasma metabolomics and 16S rDNA sequencing to evaluate the
   changes in metabolites and fecal microbiota of advanced gastric cancer.

Method

   Firstly, plasma metabolomics was used to screen for differential
   metabolites and metabolic pathways in gastric cancer. Then, 16S rDNA
   sequencing was performed on fecal samples to study the differential
   intestinal microbiota in gastric cancer patients. Finally, conduct a
   correlation analysis between them.

Result

   A total of 152 differential metabolites were identified, and we
   screened 10 of them. All metabolites were enriched into 42 differential
   metabolic pathways, of which 13 have P values less than 0.05. 16S rDNA
   sequencing showed significant differences in 4 microbial communities at
   the phylum level. There are significant differences in 23 communities
   at the genus level. We focus on Lactobacillales, Lactobacillus,
   Streptococcus, Veillonella, Bacilli and Megasphaera. Correlation
   analysis shows that the intestinal microbiota and plasma metabolites
   jointly affect the occurrence and development of gastric cancer.

Conclusion

   For the first time, we comprehensively used plasma metabolomics and 16S
   rDNA sequencing to reveal the changes and correlations between
   metabolites and intestinal microbiota in advanced gastric cancer. We
   have discovered new potential biomarkers for gastric cancer. This
   deepens our understanding of the physiological and pathological
   mechanisms of advanced gastric cancer and helps to improve the
   diagnosis and treatment of advanced gastric cancer.

   Keywords: Plasma metabolomics, 16S rDNA sequencing, Gastric cancer, Gut
   flora

1. Introduction

   Gastric cancer (GC) is a malignant tumor with high incidence rate and
   high mortality in China. At present, the treatment of advanced gastric
   cancer mainly relies on surgery, radiotherapy and chemotherapy, but its
   toxicity and side effects seriously affect the patients' quality of
   life [[41]1,[42]2]. And it is prone to recurrence and metastasis. This
   greatly affects the overall treatment effect of gastric cancer, and the
   prognosis of patients is extremely poor [[43]3,[44]4].

   Gastric cancer is a highly heterogeneous disease driven by multiple
   genetic mutations and epigenetic abnormalities. It involves changes in
   multiple genes, pathways and stages [[45]5]. Due to the lack of
   accurate and reliable biomarkers, the diagnosis and treatment of
   gastric cancer still have significant limitations. With the emergence
   of gene sequencing, biological analysis and big data analysis tools,
   finding its biomarkers is crucial for diagnosing diseases, predicting
   disease progression and treating diseases.

   Currently, metabolomics and 16S rDNA sequencing are becoming powerful
   tools for discovering biomarkers and revealing pathological processes.

   An increasing number of studies indicate that cancer cells promote
   survival and proliferation by altering multiple metabolic pathways. The
   reprogrammed cellular metabolism supports the occurrence and
   development of tumors [[46]6,[47]7]. We use metabolomics to detect
   cancer in order to discover its key biomarkers. Targeted regulation of
   metabolism is expected to become a powerful therapeutic strategy for
   cancer [[48]8,[49]9]. Metabolomics studies the endogenous metabolites
   of organisms and their relationship with influencing factors. It is a
   science about the types, quantities, and changes of metabolic
   substances in living organisms [[50]10].

   16S rDNA sequencing can detect microorganisms in specific environments,
   providing a variety of information such as sample species
   classification, species abundance and population structure. It is one
   of the most commonly used bacterial classification standards. In recent
   years, with the fast advances in high-throughput sequencing, the
   intestinal microbiota has gradually attracted people's attention. More
   and more evidence suggests that intestinal microbiota has local or
   remote regulatory functions on target organs. The balance of them is
   closely related to human health [[51]11,[52]12]. Disrupting their
   balance will lead to a series of diseases. Studying the diversity and
   composition of intestinal microbiota in gastric cancer patients
   provides a theoretical basis for disease research and treatment
   [[53]13].

   This study aims to analyze the differences in plasma metabolites and
   intestinal microbiota between gastric cancer patients and the normal
   population, in order to discover new targeted biomarkers for gastric
   cancer.

2. Materials and methods

2.1. Research object

   Referring to the diagnostic criteria for gastric cancer in the
   "Diagnosis and Treatment Guidelines for Common Malignant Tumors in
   China", the participants were 30 patients with stage IIIB-IV advanced
   gastric cancer who visited Yangzhou Traditional Chinese Medicine
   Hospital, Gaoyou Traditional Chinese Medicine Hospital, Yizheng
   Traditional Chinese Medicine Hospital and Baoying Traditional Chinese
   Medicine Hospital from July 2022 to December 2022 (referring to the
   staging criteria of the American Joint Commission on Cancer (AJCC) 8th
   edition). The study has been approved by the Ethics Committee of
   Yangzhou Traditional Chinese Medicine Hospital in Jiangsu
   Province((2022) Lun Shen No. (31)).

2.1.1. Inclusion criteria

     * (1)
       Age range from 18 to 75 years old, regardless of gender.
     * (2)
       Gastric cancer patients with clear pathological diagnosis.
     * (3)
       Gastric cancer patients with stages IIIB-IV.
     * (4)
       The expected survival period is more than 3 months.
     * (5)
       KPS score ≥60 points.
     * (6)
       Participants have good compliance and must sign an informed
       consent.

2.1.2. Exclusion criteria

     * (1)
       Individuals who have experienced significant traumatic damage such
       as surgical treatment in the past month.
     * (2)
       Individuals with severe bleeding or uncontrolled infection.
     * (3)
       Individuals with persistent purulent and chronic infected wounds.
     * (4)
       Individuals with uncontrolled diseases such as splenic
       hyperfunction, hyperthyroidism, connective tissue disease, and
       tuberculosis.
     * (5)
       Patients with severe cardiovascular and cerebrovascular
       complications, active hepatitis, and severe liver and kidney
       dysfunction.
     * (6)
       People who take drugs that affect the intestinal microbiota, such
       as microbial regulators and antibiotics.
     * (7)
       Individuals suffering from malabsorption syndrome and other
       diseases that affect gastrointestinal absorption, as well as eating
       difficulties.
     * (8)
       Individuals with or experiencing malignant tumors in other parts of
       the body (excluding fully treated cervical carcinoma in situ, skin
       basal or squamous cell carcinoma, or other tumors that have been
       cured surgically and have not recurred for at least 5 years).
     * (9)
       Women of childbearing age who have tested positive for urinary
       pregnancy, are breastfeeding, or have insufficient contraception.
     * (10)
       There are any unstable or potentially hazardous situations that may
       endanger patient safety and affect compliance with the study, such
       as those with severe mental illnesses such as schizophrenia.
     * (11)
       There are any circumstances that may affect the test results, such
       as someone having eaten before being sampled for blood.

2.2. Metabolomics

2.2.1. Chemicals and reagents

   All of the standards of targeted metabolites were obtained from
   Sigma-Aldrich (St. Louis, MO, USA), Steraloids Inc. (Newport, RI, USA)
   and TRC Chemicals (Toronto, ON, Canada). All the standards were
   accurately weighed and prepared in water, methanol, sodium hydroxide
   solution, or hydrochloric acid solution to obtain individual stock
   solution at a concentration of 5.0 mg/mL. Appropriate amount of each
   stock solution was mixed to create stock calibration solutions. Formic
   acid was of (Optima grade and obtained from Sigma-Aldrich (St. Louis,
   MO, USA). Methanol (Optima LC-MS), acetonitrile (Optima LC-MS), and
   isopropanol (Optima LC-MS) were purchased from Thermo-Fisher Scientific
   (FairLawn, NJ, USA). Ultrapure water was produced by a Mill-Q Reference
   system equipped with a LC-MS Pak filter (Millipore, Billerica, MA,
   USA).

2.2.2. Sample preparation

   Samples were thawed on ice-bath to diminish sample degradation. 20 μL
   of plasma was added to a 96-well plate. Then the plate was transferred
   to the Eppendorf epMotion Workstation(Eppendorf Inc., Humburg,
   Germany). 120 μL ice cold methanol with partial internal standards was
   automatically added to each sample and vortexed vigorously for 5 min.
   The plate was centrifuged at 4000g for 30 min (Allegra X-15R, Beckman
   Coulter, Inc., Indianapolis, IN, USA). Then the plate was returned back
   to the workstation. 30 μL of supernatant was transferred to a clean
   96-well plate, and 20 μL of freshly prepared derivative reagents was
   added to each well. The plate was sealed and the derivatization was
   carried out at 30 °C for 60 min. After derivatization, 330 μL of
   ice-cold 50 % methanol solution was added to dilute the sample. Then
   the plate was stored at −20 °C for 20 min and followed by 4000g
   centrifugation at 4 °C for 30 min 135 μL of supernatant was transferred
   to a new 96-well plate with 10 μL internal standards in each well.
   Serial dilutions of derivatized stock standards were added to the left
   wells. Finally the plate was sealed for LC-MS analysis.

2.2.3. Instrumentation

   An ultra-performance liquid chromatography coupled to tandem mass
   spectrometry (UPLC-MS/MS) system (ACQUITY UPLC-Xevo TQ-S, Waters Corp.,
   Milford, MA, USA) was used to quantitate a l l t a r g e t e d
   metabolite s in this project. The optimized instrument settings are
   briefly described below. The instrument performance optimization and
   routine maintenance were performed every week. ([54]Table 1, The
   optimized instrument settings)

Table 1.

   The optimized instrument settings.
   UPLC-MS/MS instrument settings
   UPLC
   Column ACQUITY UPLC BEH C18 1.7 μM VanGuard pre-column (2.1 × 5 mm) and
   ACQUITY UPLC BEH C18 1.7 μM analytical column (2.1 × 100 mm)
   Column Temp. (°C) 40
   Sample Manager Temp. (°C) 10
   Mobile Phases A = water with 0.1 % formic acid; and
   B = acetonitrile/IPA (70:30)
   Gradient Conditions 0–1 min (5 % B), 1–11min (5–78 % B), 11–13.5 min
   (78–95 % B), 13.5–14 min (95–100 % B), 14–16 min (100 % B),16–16.1 min
   (100-5% B),16.1–18 min (5 % B).
   Flow Rate (mL/min) 0.40
   Injection Vol. (μl) 5.0
   MASS SPECTROMETER
   Capillary (Kv) 1.5 (ESI+), 2.0 (ESI-)
   Source Temp (°C) 150
   Desolvation Temp (°C) 550
   Desolvation Gas Flow (L/Hr) 1000
   [55]Open in a new tab

2.2.4. Data analysis

2.2.4.1. Software

   The raw data files generated by UPLC-MS/MS were processed using the
   MassLynx software (v4.1, Waters, Milford, MA, USA) to perform peak
   integration, calibration, and quantitation for each metabolite. The
   iMAP platform (v1.0; MetaboProfile, Shanghai, China) was used for
   statistical analyses.

2.2.4.2. Quantitation

   Mass spectrometry-based quantitative metabolomics refers to the
   determination of the concentration of a substance in an unknown sample
   by comparing the unknown to a set of standard samples of known
   concentration (i.e., calibration curve). The calibration curve is a
   plot of how the analytical signal changes with the concentration of the
   analyte (the substance to be measured). For most analyses a plot of
   instrument response vs. concentration will show a linear relationship.
   This yields a model described by the equation y = ax + b, where y is
   the instrument response e.g., peak height or area, a represents the
   slope/sensitivity, and b is a constant that describes the background.
   The analyte concentration (x) of unknown samples may be calculated

   from this equation.

2.2.4.3. Multivariate statistical analyses

   Partial Least Square Discriminant Analysis (OPLS-DA) algorithm
   decomposes the raw data set into three parts: systemic variations,
   orthogonal/unrelated information, and residual. This leads to a model
   with a minimal number of predictive components defined by the number of
   degrees of freedom (k − 1 dimensions) between group variances (k
   classes). This partitioning of the X-data facilitates model
   interpretation and model prediction.

2.3. 16SrDNA sequencing

   Stool samples from all participants were freshly collected and frozen
   at −80 °C within 3 h after sampling.

2.3.1. DNA extraction, PCR amplification and sequencing

   Microbial DNA was extracted from fecal samples using an E.Z.N.A.® soil
   DNA Kit (Omega Bio-tek, Norcross, GA), according to the manufacturer ‘s
   protocols. The NanoDrop 2000 UV–vis spectrophotometer (Thermo
   Scientific, Wilmington, Delaware, United States) was used to determine
   the final DNA concentration and purity, and DNA quality was checked
   using 1 % agarose gel electrophoresis. The hypervariable regions of the
   bacterial 16S rRNA gene were amplified with the primers by a
   thermocycler PCR system (GeneAmp 9700, ABI, Foster, CA, United States).
   The PCR reactions were as follows: denaturation at 95 °C for 3 min, 27
   cycles of denaturation at 95 °C for 30s, annealing at 55 °C for 30s,
   elongation at 72 °C for 45 s, extension at 72 °C for 10 min, and ending
   at 4 °C. Each 20 μL reaction mixture contained 4 μL of 5 × TransStart
   FastPfu buffer, 2 μL of 2.5 mM dNTPs, 0.8μ L of each primer (5μ M),
   0.4μ L TransStart FastPfu DNA Polymerase, and 10 ng template DNA. PCR
   was performed in triplicate. The PCR product was extracted from 2 %
   agarose gel and purified using an AxyPrep DNA Gel Extraction Kit
   (Axygen Biosciences, Union City, CA), according to the manufacturer's
   instructions, and quantified using Qubit 4 (Thermo Fisher, United
   States). Purified amplicons were pooled in equimolar and paired-end
   sequenced on the Illumina MiSeq PE300 platform (Illumina, San Diego,
   United States) according to the standard protocols by Honsunbio
   Technology Co. Ltd (Shanghai, China).

2.3.2. Bioinformatics analysis

   Sequencing reads were demultiplexed, quality controlled by fastp
   (version 0.21.0), and merged by FLASH (version 1.2.7). Shortly, reads
   with adaptor sequences and low quality bases (quality score < Q20) were
   trimmed. Truncated reads shorter than 50 bp and reads containing
   ambiguous nucleotides were discarded. Subsequently, the paired-end
   reads were merged according to the minimum overlap of 10 bp with
   maximum mismatch ratio of 0.2 in the overlapping region. Only merged
   sequences were retained for downstream analyses. The UPARSE algorithm
   was used to cluster sequences with a 97 % similarity cutoff, while
   chimeric sequences were identified and removed. Next, the taxonomy of
   each OTU representative sequence was assigned by using RDP Classifier
   against the reference database SILVA138 with a minimum confidence score
   of 0.7. Rarefaction was performed in order to compare the abundance of
   OTUs across samples. Sequences demultiplexed were imported to QIIME2
   (version 2022.8). The DADA2 algorithm was used to quality filter and
   denoise sequences, while chimeric sequences were removed. Next, the
   taxonomy of each ASV representative sequence was assigned by using RDP
   Classifier against the reference database SILVA138 with a minimum
   confidence score of 0.7. The number of sequences from each sample was
   normalized to the lowest number of read counts by randomly selecting
   subsets of sequences.

2.3.3. Statistical analysis

   The R (version 4.1.3) was used to perform general statistical analysis
   and visualize results via packages vegan (v2.6–4), phyloseq (v1.38.0),
   tidyverse (v1.3.2), ggpubr (v0.5.0), ComplexHeatmap (v2.10.0) and
   corrplot (v0.92). Alpha diversity was estimated using the Sobs, ACE,
   Chao1, Shannon and Simpson indices. Principal coordinates analysis
   (PCoA) based on bray-curtis matrices with statistical significance
   determined by permutational multivariate analysis of variance
   (PERMANOVA) was conducted to assess the differences in beta diversity
   between groups. For comparing the relative abundance of different taxa
   between groups, linear discriminant analysis (LDA) effect size (LEfSe)
   method was performed with a p-value <0.05 for the Kruskal–Wallis test
   and a size-effect threshold of 2.0 on the logarithmic LDA score.
   Spearman's rank correlation analysis was used for correlation analysis.

3. Results

3.1. Metabolomics

   Metabolomics technology has successfully discovered a variety of
   cancer-related metabolite markers, which can reveal the differences in
   metabolites under pathological conditions and provide new insights for
   early diagnosis and treatment of diseases. Studies have pointed out
   that amino acid transport in many malignant tumor cells is
   significantly increased. For example, the metabolism of gastric cancer
   often requires a large consumption of serine and glutamine [[56]14].
   Metabolites that have been identified for distinguishing gastric cancer
   include isoleucine, lactic acid, glutamic acid, glutathione, etc. Their
   expression levels change with the progression of gastric cancer.
   Metabolic pathways such as glycolysis, tricarboxylic acid cycle, and
   glutamine catabolism in gastric cancer are also affected to varying
   degrees [[57]15]. Different syndrome types of the disease have
   different levels of energy metabolism and lipid metabolism. For
   example, He Nana et al. discovered 10 related biomarkers for gastric
   cancer with spleen deficiency and dampness excess [[58]16]. Currently,
   the study of gastric cancer with other syndrome type has also become a
   new research hotspot.

3.1.1. Metabolite classification

   We conducted an in-depth analysis of the metabolites present in all
   samples and detected 14 different categories of metabolites. Compared
   to the plasma metabolites in the normal group, the expression of 8
   metabolites in gastric cancer exhibits significant differences.These
   metabolites include benzenoids, imidazoles (p < 0.001), short-chain
   fatty acids, phenylpropanoic acids (p < 0.01), phenols, amino acids,
   benzoic acids, and carbohydrates (p < 0.05). Among them, the absolute
   value of the log2FC of benzenoids is 1.71. Notably, except for the
   decreased expression of amino acids in the gastric cancer patient
   group, the expression of other categories of metabolites has increased
   (see [59]Fig. 1A and B).

Fig. 1.

   [60]Fig. 1
   [61]Open in a new tab

   (A) Stacked bar chart showing the relative abundance of the median
   values of various metabolite types in each sample group; (B) Stacked
   bar chart displaying the relative abundance of various metabolite types
   in each individual sample.

3.1.2. Identification of differential metabolites

   First, we performed a multidimensional statistical analysis. The
   results of the principal component analysis showed that the
   contribution rate of the first principal component was 14.2 %, and the
   contribution rate of the second principal component was 9.9 % ([62]Fig.
   2A). The proximity of QC sample points in the PCA score plot indicates
   good stability of the instrument detection. In the heatmap of
   correlations between QC samples ([63]Fig. 2B), the color of each cell
   represents the correlation coefficient between a QC sample and other QC
   samples. A correlation coefficient closer to 1 indicates that the QC
   samples are more similar to each other. This indicates that the
   experimental detection data is more stable and the quality control is
   effective. Then, we utilized Orthogonal Partial Least
   Squares-Discriminant Analysis (OPLS-DA) for analysis. The horizontal
   coordinate P1 represents the first predictive principal component of
   the model, and the vertical coordinate O1 represents the first
   orthogonal component of the model. The percentages in parentheses
   represent the explanatory power of the principal components, and the
   points of different colors represent different sample groups ([64]Fig.
   2C). To prevent overfitting of the OPLS-DA model, we set a threshold
   with Q2Y > 0.2 in the OPLS-DA model, and the intercept of the Q2Y
   fitting curve on the Y-axis in the permutation test was less than 0.
   The permutation test results showed that R2Y = 0.847 and Q2Y = 0.74.
   This indicates that the model has good stability and predictive
   performance. ([65]Fig. 2D).

Fig. 2.

   [66]Fig. 2

   [67]Fig. 2
   [68]Open in a new tab

   (A) PCA 2D score plot; (B) Heatmap of correlations between QC samples;
   (C) OPLS-DA 2D score plot; (D) OPLS-DA permutation test results; (E)
   Venn diagram of differentially expressed metabolites screened by
   multidimensional and unidimensional methods; (F) Volcano plot of
   unidimensional metabolites; (G) Heatmap of potential biomarkers.

   Based on the normality and homogeneity of variance of the data, we
   further conducted a one-dimensional statistical analysis. The volcano
   plot of one-dimensional metabolites shows the fold change and p-value
   of each metabolite screened for differential metabolites ([69]Fig. 2F).
   The threshold for selecting differential metabolites is: (1) P < 0.05;
   (2) |log2FC| ≥ 0. The vertical dashed line represents the
   log-transformed FC threshold, with log2FC as the x-coordinate. The
   horizontal dashed line represents the P = 0.05 threshold, with the
   corresponding -logeP value as the y-coordinate. Points that satisfy
   both being above the horizontal dashed line and on both sides of the
   vertical dashed line will be highlighted. The red highlights on the
   right representing metabolites with increased concentration in the
   gastric cancer group. The blue highlights on the left representing
   metabolites with decreased concentration in the gastric cancer group.
   Gray points represent metabolites that do not meet the set threshold
   requirements. The results showed that compared with the healthy group,
   there were 152 differentially expressed plasma metabolites, of which 25
   metabolites were upregulated and 24 metabolites were downregulated
   ([70]Fig. 2F).

   After taking the intersection of the differentially expressed
   metabolites obtained from the aforementioned unidimensional and
   multidimensional statistical analyses (unidimensional test p < 0.05,
   multidimensional statistics VIP>1), we selected the biomarkers relevant
   to this study for subsequent analysis ([71]Fig. 2E). A total of 55
   differentially expressed metabolites were identified, with 6
   metabolites only present in the multidimensional results, 10
   metabolites only in the unidimensional results, and 39 metabolites are
   common to both ([72]Fig. 2E). The heatmap of potential biomarkers shows
   that these metabolites mainly including amino acids, benzene ring
   compounds, bile acids, carbohydrates, fatty acids, imidazoles, organic
   acids, phenolic compounds, and short-chain fatty acids ([73]Fig. 2G).

3.1.3. Screening of characteristic diagnostic metabolites based on random
forest algorithm and ROC curve analysis

   Through the random forest algorithm, 39 significant characteristic
   diagnostic metabolites were screened out ([74]Fig. 3B). The top 10
   metabolites include Sarcosine, Glycolic acid, Creatine, Phenylacetic
   acid, Phenylacetylglutamine, Succinic acid, Threonic acid,
   Methylmalonic acid, Histidine, and Imidazolepropionic acid. The area
   under the curve (AUC) of the ROC curve for the model using these 10
   characteristic metabolites is equal to 1 ([75]Fig. 3A). Among these AUC
   values, Sarcosine has the highest AUC of 0.949, while
   Imidazolepropionic acid has the lowest AUC of 0.795 ([76]Table 2).
   Indicating high accuracy of the model. Glycolic acid, Creatine, and
   Sarcosine are the top three characteristic metabolites ranked by
   MeanDecreaseGini. These related components are closely related to the
   occurrence and development of gastric cancer. Except for Phenylacetic
   acid and Phenylacetylglutamine, all of them are expressed at lower
   levels in gastric cancer patients ([77]Fig. 3C–L).

Fig. 3.

   [78]Fig. 3

   [79]Fig. 3
   [80]Open in a new tab

   (A) ROC curve for the 10 characteristic diagnostic metabolites; (B)
   Evaluation of metabolite importance by Random Forest; (C–L) Box plots
   for the top 10 ranked unidimensional differential metabolites.

Table 2.

   Basic information and ROC curve analysis results of the 10
   characteristic diagnostic metabolites.
   Metabolite class AUC MeanDecreaseGini CI1 CI1
   1 Sarcosine Amino Acids 0.949 3.675498931 0.901098103 0.996044754
   2 Glycolic acid Organic Acids 0.946 5.413611281 0.886586953 1
   3 Creatine Amino Acids 0.907 3.958991033 0.822227885 0.991105448
   4 Phenylacetic acid Benzenoids 0.898 2.447379694 0.812323993
   0.983866483
   5 Phenylacetylglutamine Amino Acids 0.896 1.772019824 0.817094156
   0.975286797
   6 Succinic acid Organic Acids 0.881 2.478942047 0.79367162 0.968233142
   7 Threonic acid Carbohydrates 0.868 1.309109285 0.774114708 0.961123387
   8 Methylmalonic acid Organic Acids 0.864 1.996615598 0.7690512
   0.958567847
   9 Histidine Amino Acids 0.827 0.753045002 0.721323646 0.932009687
   10 Imidazolepropionic acid Imidazoles 0.795 0.711454578 0.684481196
   0.905994994
   [81]Open in a new tab

3.1.4. Differential metabolite enrichment analysis results

   In the enrichment analysis bubble plot ([82]Fig. 4A), each circle
   corresponds to a metabolic pathway. The x-axis represents the degree of
   pathway impact. The size of the circle is related to the pathway's
   impact value, with larger impact values corresponding to larger
   circles. The y-axis represents the negative logarithm of the P-value
   obtained from the pathway enrichment analysis. The color change is
   positively correlated with the negative logarithm of the P-value of the
   pathway change. The results show that the differential metabolites
   identified in the KEGG database were enriched in 42 different metabolic
   pathways ([83]Fig. 4B and C), of which 13 pathways had a P-value less
   than 0.05. These pathways are: Aminoacyl-tRNA biosynthesis,
   Phenylalanine metabolism, Glycine, serine and threonine metabolism,
   Alanine, aspartate and glutamate metabolism (P < 0.001), Propanoate
   metabolism, Citrate cycle, Arginine and proline metabolism, Glyoxylate
   and dicarboxylate metabolism, D-Arginine and D-ornithine metabolism
   (P < 0.01), Nitrogen metabolism, Butanoate metabolism, Cyanoamino acid
   metabolism, and Cysteine and methionine metabolism (P < 0.05). The
   corresponding metabolites enriched in these 13 pathways are shown in
   [84]Table 3. Among them, the differential metabolites are mainly
   enriched in three pathways: Aminoacyl-tRNA biosynthesis, Phenylalanine
   metabolism, and Glycine, serine and threonine metabolism.

Fig. 4.

   [85]Fig. 4
   [86]Open in a new tab

   (A) Barplot of the top enriched pathways of metabolites in the KEGG
   database; (B) Pathway Enrichment Analysis Barplot Using
   Pathway-associated metabolite sets (Pathway-associated metabolite sets
   (SMPDB)); (C) Pathway Enrichment Analysis Barplot Using
   Pathway-associated metabolite sets (Predicted metabolite sets).

Table 3.

   Metabolites enriched in the 13 pathways.
   Metabolic Pathway Raw P Enriched_Compounds Impact
   1 Aminoacyl-tRNA biosynthesis 0.000000024451 Asparagine、Histidine
   Arginine、Serine
   Methionine、Valine、Alanine
   Lysine、Threonine
   Tryptophan 0.19356
   2 Phenylalanine metabolism 0.00002301 Phenylacetic acid
   Benzoic acid
   Phenylacetylglutamine
   Succinic acid、Fumaric acid p-Hydroxyphenylacetic acid 0.10909
   3 Glycine, serine and threonine metabolism 0.000033668 Serine、Glyceric
   acid
   Sarcosine、Threonine
   Creatine、Tryptophan 0.25424
   4 Alanine, aspartate and glutamate metabolism 0.00025673
   Asparagine、Alanine
   Fumaric acid、Succinic acid 0.08571
   5 Propanoate metabolism 0.0011373 Propionic acid
   Methylmalonic acid
   Succinic acid、Valine 0.10416
   6 Citrate cycle (TCA cycle) 0.0022845 Succinic acid、Malic acid
   Fumaric acid 0.125
   7 Arginine and proline metabolism 0.0034583 Ornithine、Arginine
   Creatine、Fumaric acid
   Sarcosine 0.17476
   8 Glyoxylate and dicarboxylate metabolism 0.0043267 Glycolic
   acid、Glyceric acid
   Malic acid、Succinic acid 0.13637
   9 D-Arginine and D-ornithine metabolism 0.0048485 Arginine、Ornithine
   0.14286
   10 Nitrogen metabolism 0.015341 Tryptophan、Asparagine
   Histidine 0.07692
   11 Butanoate metabolism 0.016431 Butyric acid、Succinic acid
   Fumaric acid 0.06
   12 Cyanoamino acid metabolism 0.019403 Asparagine、Serine 0.08333
   13 Cysteine and methionine metabolism 0.039754
   Serine、Methionine、Alanine 0.07408
   [87]Open in a new tab

4. 16S rDNA sequencing

   The intestine is the largest habitat for microbiota in the human body
   and serves as an essential "organ" in the host's immune and metabolic
   systems [[88]17]. Increasing evidence suggests that intestinal
   microbiota imbalance may affect the occurrence and progression of
   systemic diseases beyond the intestine, and even influence the
   effectiveness of immunotherapy [[89]18].We use fecal samples to study
   the composition of intestinal microbiota, which is convenient and
   non-invasive.

4.1. The intestinal microbial community structure at the phylum level

   At the phylum level, the relative abundance of intestinal microbiota is
   mainly composed of Actinobacteria, Proteobacteria, Bacteroidetes, and
   Firmicutes. Compared to the healthy group, patients with gastric cancer
   have a decreased expression of Firmicutes and Actinobacteria, while
   Bacteroidetes and Proteobacteria show an increased expression. Other
   bacterial groups also have higher contents and abundances ([90]Fig.
   5A).

Fig. 5.

   [91]Fig. 5
   [92]Open in a new tab

   (A) Composition of the microbial community structure at the phylum
   level; (B) Composition of the microbial community structure at the
   genus level; (C) Composition of the microbial community structure in
   each sample; (D) Bar plot of LEfSe analysis.

4.2. Intestinal microbial community structure at the genus level

   At the genus level, the intestinal microbiota is more diverse. The
   microbiota related to gastric cancer includes Lactobacillus,
   Streptococcus, and Escherichia coli. There expressions in the gastric
   cancer group are significantly higher than those in the healthy group.
   These genera dominate the intestinal microbiota in gastric cancer
   patients. In contrast, the expressions of Faecalibacterium,
   Phascolarctobacterium, and Bifidobacterium are significantly decreased
   ([93]Fig. 5B).

   To determine the microbiota types associated with advanced gastric
   cancer, we analyzed the microbiota using LEfSe and presented the
   results at the genus level ([94]Fig. 5C). Compared to the healthy
   group, 73 microbial genera showed significant differences between the
   gastric cancer group and the healthy group. Among them, 36 were
   enriched in the healthy group samples, while 37 were enriched in the
   gastric cancer group samples. In gastric cancer, Lactobacillus,
   Lactococcus, Bacillus, Streptococcus, Veillonella, and Megasphaera were
   the most enriched genera. While Clostridium, Faecalibacterium, and
   Romboutsia were more enriched in the healthy group samples. These
   significant changes in the microbial community structure in gastric
   cancer are associated with an increased risk of gastric cancer and may
   serve as potential diagnostic biomarkers for gastric cancer. They
   deserve our close attention.

5. Correlation between differential plasma metabolites and differential fecal
microbiota metabolites

   To further investigate the relationship between plasma metabolites and
   intestinal microbiota, we conducted a correlation analysis. Among the
   top ten differential metabolites, three plasma amino acids (creatine,
   sarcosine, and histidine), two organic acids (glycolic acid and
   succinic acid), and a carbohydrate (threonine) showed a negative
   correlation with Lactobacillus, Streptococcus, Veillonella,
   Megasphaera, and Actinobacteria. They showed a positive correlation
   with Faecalibacterium, Erysipelotrichaceae, Romboutsia, Ruminococcus,
   and Campylobacter. An organic acid (methylmalonic acid) and a plasma
   amino acid (phenylacetylglutamine), a phenyl derivative (phenylacetic
   acid), and an imidazole (imidazolpropionic acid) showed a positive
   correlation with Lactobacillus, Streptococcus, Veillonella,
   Megasphaera, and Actinobacteria. They showed a negative correlation
   with Faecalibacterium, Erysipelotrichaceae, Romboutsia, Ruminococcus,
   and Campylobacter ([95]Fig. 6). Besides the top ten metabolites, we
   also found that Lysine, Tryptophan, NorDCA, Methionine, Asparagine,
   Arginine, and Serine were closely related to intestinal microbiota.

Fig. 6.

   [96]Fig. 6
   [97]Open in a new tab

   (A) Heatmap of joint analysis of differential plasma metabolomic
   profiles and gut microbiota profiles.

   By comprehensively applying plasma metabolomics and 16S rDNA sequencing
   technology, we discovered significant changes in metabolic
   characteristics and intestinal microbiota structure in patients with
   advanced gastric cancer. There ia a significant correlation between the
   two. This not only helps us better understand the pathogenesis of
   gastric cancer but also provides new theoretical foundations for the
   diagnosis and treatment of gastric cancer.

6. Discussion

   Gastric cancer is a common malignant tumor with high morbidity and
   mortality rates. Especially in the case of advanced gastric cancer, the
   lack of effective biomarkers makes it difficult to detect and diagnose
   in the early stages, and the underlying pathogenic mechanisms remain
   unclear. These limitations underscore the importance of gaining new
   insights into the pathogenesis of advanced gastric cancer. Modern
   medicine has delved into the genetic and epigenetic levels in
   understanding gastric cancer. This suggests that there may be
   significant differences in biological characteristics, clinical
   manifestations, treatment response, and prognosis. With the advent of
   gene sequencing, bioanalytical technologies, and big data analysis
   tools, the treatment of gastric cancer has entered the era of precision
   medicine.

   In this study, we employed a combined approach of metabolomics and 16S
   rDNA sequencing to investigate the plasma metabolome and intestinal
   microbiota of 30 patients with stage IIIB-IV gastric cancer.

   Through metabolomics analysis, we identified a total of 152 relevant
   plasma differential metabolites, of which 8 metabolites showed
   significant differences. These metabolites play crucial roles in
   biological processes such as protein synthesis, energy metabolism, and
   signal transduction, which are associated with the proliferation,
   migration, and other activities of gastric cancer.

   In our study, several metabolites have been reported to be closely
   related to gastric cancer. For instance, alterations in fatty acid
   metabolism are associated with changes in the immune microenvironment
   of gastric cancer [[98]19]. They are one of the most important
   hallmarks of cancer metastasis initiation [[99]20]. Fatty acid-induced
   O-GlcNAcylation promotes the transcription and function of CD36 by
   directly modifying the S468 and T470 sites of CD36 through activation
   of the NF-κB pathway, thereby driving gastric cancer metastasis
   [[100]21]. Among the fatty acids that have been extensively studied,
   docosahexaenoic acid has been confirmed by both in vitro and in vivo
   studies to induce apoptosis [[101]22,[102]23], inhibit cell invasion
   [[103]24,[104]26], and suppress cell growth [[105]25] in gastric cancer
   cells. We identified fatty acid components such as Methylsuccinic acid,
   3−Hydroxyisovaleric acid, Palmitoleic acid, and Myristoleic acid et al.
   But their mechanisms of action remain unclear and could be the focus of
   future research.

   Methylmalonic acid (MMA) is one of the top 10 differential metabolites.
   It belongs to the category of organic acids. Elevated MMA is a good
   indicator of vitamin B12 deficiency, which is a common complication
   among patients after gastric resection. The study found that MMA levels
   were significantly increased in gastric cancer patients after gastric
   resection. These results suggest that additional assessment of MMA
   levels could help evaluate early changes in vitamin B12 levels in
   patients after gastrectomy and may even serve as a diagnostic marker
   for gastric cancer [[106]27].

   Amino acid are frequently found in tests as differential
   metabolitesand. It plays crucial roles in various biological processes
   such as protein synthesis, energy metabolism, and signal transduction.
   Current studies have indicated that L-arginine intervention promotes
   apoptosis. It can significantly increase the expression of
   apoptosis-promoting proteins p53, while significantly reduce the
   expression of anti-apoptotic proteins Bcl-2 and survivin
   [[107]28,[108]29]. Numerous studies have confirmed that arginine plays
   a vital role in immune function and intestinal mucosal protection. The
   combined use of arginine and glutamine in enteral and parenteral
   nutrition for patients after gastric cancer surgery has led to
   increased levels of albumin, serum protein, and transferrin, as well as
   decreased levels of C-reactive protein, C3, and C4 [[109]30]. Cellular
   immune indicators were also significantly improved, and both apoptosis
   and PCNA expression in tumor tissues were improved [[110]31,[111]32].
   Graziosi L et al. found that methionine can regulate the proliferation
   and migration of gastric tumor cells and reduce the production of
   peritoneal nodules [[112]33]. These amino acids were all detected in
   our study and ranked high in terms of their importance.

   Bile acids are conducive to the absorption of dietary fat and can also
   affect the process of gastric cancer [[113]34]. Currently, the
   differential bile acid components we detected include NorDCA and
   LCA−3S. There are many studies on bile acids at present. Bile acid
   reflux can damage gastric mucosa [[114]35], and high concentration of
   bile acids can induce apoptosis of gastric cancer cells [[115]36].
   Among them, hydrophobic bile acids such as chenodeoxycholic acid
   promote cell invasion through PKC activation and COX-2 induction in
   gastric cancer cells. Ursodeoxycholic acid weakens PGE2 synthesis by
   disturbing the bile acid pool, which affects tumor invasiveness
   [[116]37], and induces cell death according to the intracellular
   signaling environment [[117]38]. Secondary bile acids are typical
   carcinogenic bile acids [[118]40,[119]41]. For example, deoxycholic
   acid (DCA) has been shown to be related to alkaline gastritis
   [[120]39]. It can promote the expression of E-cadherin and inhibit the
   invasion and migration of gastric cancer cells by upregulating the
   expression of MUC2 mRNA and downregulating the expression of Snail and
   MMP9 [[121]42]. DCA can also induce apoptosis of gastric cancer cells
   through mitochondrial pathway, which is manifested by the increase of
   Bax/Bcl-2 ratio and the collapse of mitochondrial membrane potential
   [[122]43].

   Our research results show that glycolic acid, creatine and sarcosine
   are also closely related to the condition of gastric cancer. But we
   have read all the relevant literature, and found that there has been no
   detailed research on the mechanisms of these three metabolites in the
   literature. This provides us with new ideas for our future research. We
   believe that future research on the specific mechanisms of these
   metabolites will also be a new breakthrough point.

   Our research findings are consistent with most of the previous studies.
   This indicate the reliability of our results. Furthermore, compared
   with previous reports, we have uncovered more changes in plasma
   metabolites related to advanced gastric cancer, which will be conducive
   to screening for new biomarkers.

   In addition, metabolomics has also revealed the metabolic differences
   between gastric cancer cells and normal cells. Gastric cancer cells
   tend to have higher glycolytic activity and amino acid metabolic
   activity, which may be related to the rapid proliferation and increased
   energy demand of gastric cancer cells. Meanwhile, gastric cancer cells
   may also exhibit abnormal expression of certain specific metabolic
   pathways, which may be correlated with the malignancy, progression
   rate, and therapeutic response of gastric cancer.

   Among the identified differential metabolites in the KEGG database, 42
   differential metabolic pathways were enriched, with 13 pathways having
   a P-value less than 0.05. These metabolic pathways may ultimately lead
   to pathological changes. Further functional and clinical sample
   analysis of these metabolic pathways is needed to further prove their
   roles.

   The study of metabolomics provides us with a new perspective and tool
   to gain a deeper understanding of the mechanisms of gastric cancer
   development. It offers new strategies and methods for the diagnosis,
   treatment, and prevention of gastric cancer. By further investigating
   the relationship between metabolites and gastric cancer, we can better
   discover the biological characteristics of gastric cancer.

   The intestinal microbiota is the largest microecosystem in the human
   body. The ecological imbalance provides opportunities for the growth of
   pathogenic bacteria. The intestinal flora includes probiotics and
   conditional pathogens. The former mainly includes Lactobacillus and
   Bifidobacterium, while the latter includes Enterococcus, Escherichia,
   Shigella, Klebsiella, Streptococcus, and Streptococcus gastricus et al.
   [[123]44].

   Through 16SrDNA sequencing, we found that compared to the healthy
   group, the expression of Firmicutes and Actinobacteria was reduced in
   the gastric cancer group at the phylum level, while the expression of
   Bacteroidetes and Proteobacteria was increased. Bacteroidetes is an
   important clinical pathogen that contributes to anaerobic infections.
   It can break down polysaccharides in the intestine into short-chain
   fatty acids, increase the permeability of the intestinal mucosa,
   promote inflammation, and has a mutually reinforcing symbiotic
   relationship with Firmicutes. The imbalance in the ratio of Firmicutes
   to Bacteroidetes may indicate impaired intestinal absorption and
   inflammation, further promoting the occurrence of gastric cancer.

   The bacterial flora in the gastric cancer group exhibited a more
   diverse composition at the genus level. The expression of
   Lactobacillus, Streptococcus, and Escherichia coli is significantly
   higher than in the healthy group. While the expression of
   Faecalibacterium, Phascolarctobacterium, and Bifidobacterium was
   significantly reduced. There were significant differences in 73
   bacterial taxa between the two groups, with 37 taxa enriched in the
   gastric cancer group. We focused our attention on changes in
   Lactococcus, Lactobacillus, Bacillus, Streptococcus, Veillonella, and
   Megasphaera.

   Lactobacillus has the closest relationship with gastric cancer. It can
   block the invasion and colonization of pathogens such as bacteria,
   viruses, and fungi through competition. It can also maintain the
   integrity of the epithelial barrier, and promote the production of
   antibacterial compounds [[124]45,[125]46]. However, more than 200
   Lactobacillus-related infections have been found in immunocompromised
   individuals, such as cancer patients [[126]47,[127]48]. The increase in
   Lactobacillus abundance in gastric cancer is consistent with previous
   studies, showing a positive correlation between Lactobacillus content
   and disease severity in gastric cancer [[128]49,[129]50]. Lactobacillus
   proliferates in the stomach of gastric cancer patients and moves
   downward to the intestine [[130]51]. Dynamic monitoring of
   Lactobacillus levels has important clinical significance for predicting
   the development of gastric cancer. In addition, some studies have found
   that Lactococcus and Megasphaera are related to gastric cancer
   [[131]52], and the potential pathogenic bacterium Veillonella is
   significantly increased in gastric cancer [[132]53,[133]54,[134]55].
   This result is consistent with our findings. But the mechanism of
   action of Megasphaera and Veillonella is still unclear. Their
   relationship and mechanism of action with gastric cancer deserve
   further investigation.

   Streptococcus is an anaerobic bacterium. Previous studies have also
   shown that the abundance of Streptococcus in gastric cancer tumor
   tissues has increased [[135]56,[136]57]. There have been research
   reports using Streptococcus as a diagnostic marker. Zhou et al.
   diagnosed gastric cancer based on Streptococcus anginosus and
   Streptococcus constellatus in fecal samples, with an AUC of 0.91
   [[137]58]. Yu et al. explored Streptococcus as a marker for liver
   metastasis of gastric cancer, with an AUC of 0.651 [[138]59]. It is
   reported that the content of Streptococcus in colorectal cancer also
   increased significantly [[139]60,[140]61]. This indicate its potential
   as a diagnostic marker for gastrointestinal cancer. Our analysis also
   validates the above results.

   Shili Liu et al. found that the number of beneficial bacteria such as
   Faecalibacterium, Roseburia, and Lachnospira decreased in gastric
   cancer. The number of Escherichia and Streptococcaceae increased in
   gastric cancer patients [[141]62]. Previous studies have also reported
   significant changes in Roseburia and Escherichia among patients, which
   are related to the patient's disease status [[142]63]. These findings
   are fully consistent with our results. This suggests that Roseburia,
   Escherichia, Faecalibacterium, and Lachnospira could be used as
   biomarkers to distinguish gastric cancer from health. This demonstrate
   the potential application value of gut microbiota in the diagnosis of
   gastric cancer.

   The development of gastric cancer is associated with multiple risk
   factors. We conducted a correlation analysis of the components of these
   metabolic profiles and found that these differential metabolites are
   closely related to and mutually influence fecal microbiota.

   We discovered that Lactobacillus, Faecalibacterium, and Bacteroides may
   collectively alter the abundance of fatty acyls and glycerophospholipid
   metabolites. The increased relative abundance of differential
   metabolites in pathways such as aminoacyl-tRNA biosynthesis and
   nucleotide sugar metabolism may be due to the collective influence of
   Lactobacillus, Streptococcus, Empedobacter, and Faecalibacterium. They
   may also jointly be responsible for carbohydrate synthesis. This
   indicates that these microbiota affect the degradation and synthesis of
   most metabolites, and the composition of different metabolites also
   affects the living environment of the microbiota.

   These results indicate that there is a strong interaction between
   gastric cancer metabolites and intestinal microbiota, which jointly
   affect gastric cancer. For example, it has been reported that the aryl
   hydrocarbon receptor and Helicobacter pylori can interact with the
   tryptophan metabolite kynurenine to weaken the immune system, increase
   the production of inflammatory cytokines, and induce gastric cancer. In
   gastric macrophages, Helicobacter pylori inhibits the uptake of
   arginine, thus inhibiting the production of NO, reducing the killing of
   Helicobacter pylori and tumor angiogenesis [[143]64]. A study by Noto
   JM et al. confirmed that Helicobacter pylori infection reduces the
   level of specific bile acids and the accumulation of secondary bile
   acids, inhibits the activation of harmful signaling pathways, and
   increases the occurrence of gastric cancer [[144]65].

   The homeostasis of amino acids and other compounds provides metabolic
   precursors for microbial growth. It also ensure the homeostasis of
   various metabolic products such as amino acids in the host. For
   example, in our research results, creatine and sarcosine are among the
   top three in importance. They participate in various pathways such as
   glycine, serine, and threonine metabolism, connecting key metabolic
   networks. Studies have shown that L-threonine can generate SCFAs,
   acetate, butyrate, and propionate. It plays an important role in
   maintaining normal intestinal function and integrity [[145]66].

   It is possible that the microbiota participates in the degradation and
   synthesis of most of the differential metabolites in these categories.
   Plasma metabolome and intestinal microbiota profiling have great
   potential in the detection and treatment of gastric cancer. They may
   help understand its underlying mechanisms. However, there are currently
   few studies on the specific mechanisms of their interactions, which is
   a major focus of future research on the impact of metabolome and
   microbiota profiling on gastric cancer.

   The initial interpretation of these results suggests that differences
   in the microbiome and metabolome ultimately affect the occurrence and
   progression of gastric cancer. Many of the findings in this study are
   consistent with previously reported results, which confirms the
   reliability of our research results. In addition, our study reveals
   more plasma metabolites and changes in intestinal microbiota associated
   with advanced gastric cancer. The metabolome described in this study
   could serve as biomarkers for the diagnosis and treatment of gastric
   cancer or targets for its prevention. However, we must acknowledge some
   limitations in our study. First, the sample size is relatively small,
   and these potential new biomarkers need to be further validated in a
   larger number of patients and controls.

7. Conclusion

   In summary, our study is the first to comprehensively apply plasma
   metabolomics and 16SrDNA sequencing technology. The experiments
   revealed the metabolic and intestinal microbiota characteristics of
   advanced gastric cancer as well as their correlations, and discovered
   more signature differential metabolites.These findings provide new
   perspectives and ideas for gastric cancer research, offer new potential
   biomarkers for the diagnosis and monitoring of advanced gastric cancer,
   and provide new evidence for the diagnosis, treatment, and prognosis of
   gastric cancer. In the future, we will further explore the application
   value of metabolomics and microbiota in gastric cancer research, aiming
   to make greater contributions to improving the survival rate and
   quality of life of patients with gastric cancer.

CRediT authorship contribution statement

   Xinyi Feng: Writing – original draft, Software, Resources, Project
   administration, Methodology. Yu Zhang: Formal analysis, Data curation.
   Jun Feng: Validation, Software. Zhongjun Li: Resources. Zhi Zhang:
   Resources, Methodology. Lin Zhu: Supervision, Methodology,
   Investigation. Ruoyu Zhou: Validation, Supervision. Haibo Wang: Writing
   – review & editing, Project administration, Conceptualization. Xiaojun
   Dai: Resources, Project administration. Yanqing Liu: Writing – review &
   editing, Validation, Project administration, Formal analysis.

Ethics and consent declarations

   Ethical approval have been obtained before the research study commenced
   or data was collected, and informed consent obtained before the
   collection of data from participants.

For ethics approval

   This study was reviewed and approved by Medical Ethics Committee of
   Yangzhou Traditional Chinese Medicine Hospital with the approval
   number:(2022) Ethics Review No. (31); dated:2022.11.04.

For consent

   All participants/patients (or their proxies/legal guardians) provided
   written informed consent to participate in the study and for their data
   to be published.

Funding

   National Natural Science Foundation of China(82104946, 82274603).

   Natural Science Foundation of Jiangsu Province(BK20210817).

   The Traditional Chinese Medicine Science and Technology Development
   Project of Jiangsu Province (Project code: QN202008, MS2022119).

   The young scientific and technological talents uplift project of
   Jiangsu Association of Integrated Traditional Chinese and Western
   Medicine (JSZXTJ-2024-A05).

   The Qing Lan Project of Yangzhou University.

   The Postgraduate Research & Practice Innovation Program of Jiangsu
   Province (SJCX23_2047).

Declaration of competing interest

   The authors declare that the research was conducted in the absence of
   any commercial or financial relationships that could be construed as a
   potential conflict of interest.

References