MAOA suppresses the growth of gastric cancer by interacting with NDRG1 and regulating the Warburg effect through the PI3K/AKT/mTOR pathway

Previous studies have indicated that neurotransmitters play important roles in the occurrence and development of gastric cancer. MAOA is an important catecholamine neurotransmitter-degrading enzyme involved in the degradation of norepinephrine, epinephrine and serotonin. To find a potential therapeutic target for the treatment of gastric cancer, the biological functions of MAOA and the underlying mechanism in gastric cancer need to be explored. The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO) datasets, Kaplan‒Meier (KM) plotter were used to identify the differentially expressed genes, which mainly involved the degradation and synthesis enzymes of neurotransmitters in gastric cancer. We also investigated the expression pattern of MAOA in human and mouse tissues and cell lines by immunohistochemistry and Western blotting analysis. Western blotting, quantitative real-time PCR, enzyme-linked immunosorbent assay (ELISA) and a Seahorse experiment were used to identify the molecular mechanism of cancer cell glycolysis. MAOA expression and patient survival were analysed in the Ren Ji cohort, and univariate and multivariate analyses were performed based on the clinicopathological characteristics of the above samples. MAOA expression was significantly downregulated in gastric cancer tissue and associated with poor patient prognosis. Moreover, the expression level of MAOA in gastric cancer tissue had a close negative correlation with the SUXmax value of PET-CT in patients. MAOA suppressed tumour growth and glycolysis and promoted cancer cell apoptosis. We also reported that MAOA can interact with NDRG1 and regulate glycolysis through suppression of the PI3K/Akt/mTOR pathway. MAOA expression may serve as an independent prognostic factor in gastric cancer patients. MAOA attenuated glycolysis and inhibited the progression of gastric cancer through the PI3K/Akt/mTOR pathway. Loss of function or downregulation of MAOA can facilitate gastric cancer progression. Overexpression of MAOA and inhibition of the PI3K/Akt/mTOR pathway may provide a potential method for gastric cancer treatment in clinical therapeutic regimens.


Introduction
Gastric cancer remains an important cancer worldwide with over one million new cases in 2020 and an estimated 769,000 deaths (equal to one in every 13 deaths globally), ranking fifth for incidence and fourth for mortality globally [1].Gastric cancer is a major cancer with high mortality and morbidity, causing a considerable health burden worldwide.The majority of gastric cancer cases are diagnosed in East Asia, Eastern Europe and South America [2].Although systemic chemotherapy, radiotherapy, surgery, immunotherapy, and targeted therapy have proven efficacy in gastric adenocarcinoma [3], some advanced gastric cancer patients do not benefit significantly from the above treatment approaches.Therefore, further exploration of the characteristics of gastric cancer is required.
It has been reported that peripheral nerves take control of the tumour microenvironment to achieve tumour innervation by secreting neurotransmitters [4][5][6].The gastrointestinal tract is rich in nerve fibers, and its activity is regulated by neurotransmitters, such as the sympathetic and vagus nerves [7,8].The body's sympathetic nerves continue to be overexcited, and stress can cause gastrointestinal ulcers, which may further progress to tumours [9,10].Therefore, it is of interest and meaningful to study the role of neurotransmitters such as norepinephrine (NE) and epinephrine (E) in gastric cancer.
The major NE/E degrading enzyme is monoamine oxidase A (MAOA).In addition to neurons and astrocytes, MAOA is also expressed in the digestive tract, liver and placenta.MAOA is located on the X chromosome and acts as a monoamine oxidase to catalyse the degradation of catecholamine neurotransmitters.Previous studies have shown that its gene mutation inactivation is associated with violent behaviour [11][12][13][14].The main function of MAOA is to degrade neurotransmitters such as E, NE and dopamine and to regulate the level of neurotransmitters in the body or in the tumour microenvironment.In hepatocellular carcinoma and prostate cancer, MAOA was reported to be closely related to cancer progression [15].However, MAOA has rarely been reported in previous studies of stomach adenocarcinoma.Therefore, its function and mechanism in gastric cancer need to be further explored.
In this study, we investigated the biological functions of MAOA in gastric cancer.In vitro and in vivo experiments demonstrated the inhibitory effects of MAOA on the processing of gastric cancer growth.Moreover, MAOA was found to play an important role in regulating the PI3K/ Akt/mTOR pathway by interacting with NDRG1, which is upstream of glycolysis.The inhibitory effects of MAOA-NDRG1 on the PI3K/Akt/mTOR pathway led to suppression of the Warburg effect, thus inhibiting the growth of gastric cancer.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
The thermocycling conditions were as follows: 60 °C for 34 s and 95 °C for 15 s for 40 cycles.Total RNA was extracted from gastric cancer tissues or gastric cancer cell lines using TRIzol reagent (Thermo Fisher Scientific, Inc.) and reverse transcribed using a PrimeScript RT-PCR kit (Takara Bio., Inc., Otsu, Japan) according to the manufacturer's protocols.RT-qPCR was performed with SYBR Premix Ex Taq (Takara Bio., Inc.) using a 7500 Real-time PCR system (Thermo Fisher Scientific, Inc.).The 2-ΔΔCq method (15) was used to quantify relative gene expression, which was normalized to β-actin [16,18].The primers used in this study can be found in Supplementary Table 1.

Western blotting
Total protein was extracted using total protein extraction buffer (Beyotime Institute of Biotechnology, Haimen, China), and the protein concentration was measured using a bicinchoninic acid protein assay kit (Thermo Fisher Scientific, Inc.).Proteins (30 μg per lane) were separated using 10% SDS-PAGE and transferred onto a nitrocellulose membrane.Following blocking with 1% BSA at room temperature for 1 h, the membrane was probed with MAOA (1:1000, Proteintech,

TCGA database, GEO database and K-M plotter analysis
For

ECAR and OCR measurements in a Seahorse experiment
In vitro cell metabolic alterations were monitored with the Seahorse XF96 Flux Analyser (Seahorse Bioscience) according to the manufacturer's instructions.BGC-823 or SGC-7901 cells were seeded on an XF96-well plate, followed by PBS and specific treatment for 24 h.For assessment of the real-time glycolytic rate (ECAR), an indicator of net proton loss during glycolysis, cells were incubated with unbuffered medium followed by a sequential injection of 10 mM glucose, 1 mM oligomycin (Sigma-Aldrich), and 80 mM 2-deoxyglucose (2-DG, Sigma-Aldrich, D8375).Mitochondrial respiration (OCR) was assessed using sequential injections of 1 mM oligomycin, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP, Sigma-Aldrich, C2920), and 2 mM antimycin A and rotenone (Sigma-Aldrich).To achieve maximal OCR, antimycin A and rotenone were used to inhibit mitochondrial respiration by blocking complex III (ubiquinone: cytochrome b-c complex).Both ECAR and OCR measurements were normalized to total protein content and reported as pmole/min for OCR and mph/min for ECAR [16,19].

Glucose uptake and lactate production measurement
Briefly, cells were seeded onto 6-well plate culture dishes and starved for 24 h after they had adhered.Then, the cells were treated with the indicated antagonists for 2 h, followed by specific treatment for an additional 24 h.The culture medium was collected for the measurement of glucose and lactate concentrations, and cells were harvested for protein lysates.The glucose uptake assay was performed using a glucose assay kit (Sigma-Aldrich).Glucose consumption was calculated by deducting the measured glucose concentration in the medium from the original glucose concentration.Lactate production in the medium was detected using the Lactate Assay Kit (BioVision, Mountain View, CA) according to the manufacturer's instructions.The results were normalized on the basis of the total protein concentration of each sample [16,19].

Co-Immunoprecipitation
For immunoprecipitation assays, BGC-823 cells were lysed in WB and IP lysis buffer (NCM Biotech, P70100, China), and cell lysates were incubated with Pierce Anti-HA Magnetic Beads (Thermo Scientific, 88,836, USA) overnight at 4 °C.Then, beads were washed three times in TBST, after which the immunoprecipitated protein complexes were resuspended in SDS-PAGE sample loading buffer followed by western blot analysis with indicated antibody.

Patients and tissue microarray (TMA) construction
This was a retrospective analysis of 587 patients with primary GC who underwent gastrectomy at the Department of Gastrointestinal Surgery, Ren Ji Hospital, ( Formalin-fixed paraffin-embedded (FFPE) tissue blocks were collected from the Pathology Department of Ren Ji Hospital.TNM staging was performed based on the American Joint Committee on Cancer (8th Edition) staging system.For each case, the diagnosis was confirmed by two senior pathologists through a review of H&E-stained slides.
Representative FFPE blocks were selected for construction of the TMA using a tissue arrayer of 5-μm thickness.

Statistical analysis
The statistical analysis was performed using SPSS 22.0 software (IBM, USA).All of the data are expressed as the mean ± standard deviation.Student's t test or one-way ANOVA was used to compare the means of two or three groups.The correlation between MAOA expression and clinicopathological characteristics was calculated using the chi-square test or Fisher's exact test.A Kaplan-Meier survival curve was used to evaluate overall survival, and the log-rank test was used to compare differences between curves.A P value of < 0.05 was considered significant (*P < 0.05, **P < 0.01, and ***P < 0.001).

The expression of MAOA is downregulated in gastric cancer tissues and closely related to poor prognosis of patients
To identify the potential biological functions of NE/E-related synthetase and degrading enzymes in gastric cancer (Fig. 1A), we used the TCGA databases to analyse the expression levels of dopamine β-hydroxylase (DBH), monoamine oxidase A (MAOA), monoamine oxidase B (MAOB) and catechol-O-methyltransferase (COMT).The mRNA expression levels of MAOA and MAOB were significantly downregulated in tumour tissues compared with normal tissues (Fig. 1B and 1C).Moreover, we also observed the expression levels of MAOA, MAOB, DBH and COMT in normal and tumour tissues of gastric cancer in GSE13911, GSE13861 and GSE19826.The downregulatory result in all GEO datasets was only observed for MAOA, which suggests the important role of MAOA in the progression of gastric cancer (Supplementary Fig. 1).To further determine the expression pattern of MAOA and MAOB in gastric cancer tissue and normal tissue, we analysed at the mRNA and protein levels 15 tumour-normal paired patient tissues from patients who received surgery in Renji Hospital.
We found MAOA to be obviously downregulated in gastric cancer tissues compared with normal tissues.For MAOB, no significant difference was found (Fig. 1D).
To explore the prognostic value of MAOA expression, we analysed the overall survival (OS) of gastric cancer patients for MAOA expression on the K-M plotter website.The results showed that patients with high expression of MAOA had significantly longer OS times than those with low expression of MAOA.Meanwhile, the expression of MAOB was also closely related to the OS of gastric cancer patients (Fig. 1E).The expression of MAOA or MAOB was further examined in normal and tumour tissues of gastric cancer by immunohistochemistry.It was found that the expression of MAOA was downregulated in gastric cancer tissues, while the expression of MAOB had no changes between normal and gastric cancer tissues (Fig. 1F).

MAOA expression in gastric cancer tissues is associated with improved prognosis and acts as an independent prognostic factor in gastric cancer
After MAOA expression was explored in the TCGA database, we further investigated its expression using a gastric cancer tissue microarray (TMA) containing 587 gastric cancer patients.According to the intensity and area of MAOApositive staining in the above TMA, 256 cases showed high MAOA expression, while 331 cases showed low MAOA expression.In the high expression cases, MAOA expression was found to be localized in the cytoplasm and extracellular area of gastric cancer tissues (Fig. 2A).Next, we analysed the OS of the 587 cases of gastric cancer patients, and observed that patients with high MAOA expression had an improved survival time compared with those in the low MAOA expression group (p = 0.0193) (Fig. 2B).Some of the gastric cancer patients underwent PET-CT examination before radical gastrectomy, and we analysed the relationship between the SUXmax value of the gastric cancer tissue and the mRNA expression level of MAOA.We found that MAOA expression had a moderate negative correlation with the SUXmax value of PET-CT (R 2 = 0.4838).At the protein level, based on MAOA expression in Western blotting, the high and low expression groups showed evident differences in the PET-CT SUVmax value (P = 0.000) (Fig. 2C).Two representative cases of gastric cancer patients with high and low expression of MAOA underwent PET-CT examination before surgery.Case 1 with low expression of MAOA showed high glucose uptake, while case 2 with high expression of MAOA displayed low glucose uptake (Fig. 2D).To decipher the significance of MAOA expression in gastric cancer samples, we applied multiple-factor analysis.We concluded that MAOA can act as an independent prognostic factor to predict patient survival (Fig. 2E).Collectively, these results suggest that high expression of MAOA in gastric cancer patients is linked to an improved prognosis and may also act as an independent prognostic factor for gastric cancer patients.

MAOA can inhibit gastric cancer cell growth in vitro and in vivo
To elucidate the biological functions of MAOA in gastric cancer, we first investigated cell proliferation and MAOA expression in gastric cancer cell lines.MGC-803 and SGC-7901 had relatively high expression, and BGC-823 and HGC-27 had relatively low expression of MAOA of the 8 gastric cancer cell lines (Supplementary Fig. 2).A CCK-8 assay was performed to test the function of MAOA on cell proliferation.Knockdown of MAOA gene expression in MGC-803 and SGC-7901 cell lines resulted in acceleration of cancer cell proliferation.In the two relatively low MAOA expression cell lines BGC-823 and HGC-27, overexpression of the MAOA gene resulted in cell proliferation inhibition (Fig. 3A and 3C).To further examine the cell proliferation function, colony formation assays were performed.Silencing of MAOA gene expression increased the colony formation of cancer cells, while the colony formation of cells decreased in the MAOA-overexpressing group (Fig. 3B and 3D).Next, we investigated the effect of MAOA on cell apoptosis.When MAOA expression in MGC-803 and SGC-7901 cells was silenced, the proportion of apoptotic cells was significantly increased (Supplementary Fig. 3A).
For BGC-823 and HGC-27 cell lines with low expression of MAOA, overexpression of MAOA resulted in a marked reduction in cell apoptosis (Supplementary Fig. 3B).
After gastric cancer cell proliferation was detected in vitro, the biological functions of MAOA were also examined in a nude mouse model in vivo.MAOA-overexpressing and control Vector BGC-823 cells were subcutaneously injected into nude mice.In the above subcutaneous xenograft model, gastric cancer tumour growth in the BGC-823-MAOA group was significantly reduced compared with that in the BGC-823-Vector group, as shown by the tumour volume growth curve (Fig. 3E and 3F).The final subcutaneous tumour weight was significantly reduced in the MAOA overexpression group compared with the control group (Fig. 3G).All of the above results indicate that MAOA can significantly inhibit tumour growth in a nude mouse subcutaneous xenograft model.Subsequently, we performed immunohistochemistry to investigate the differential expression of MAOA in the overexpression group and the control group (Fig. 3H).We stained for Ki-67 in the MAOA overexpression group and vector group, and the number of Ki-67-positive cells was obviously decreased in the MAOA overexpression group (Fig. 3I).Staining for another proliferation marker, proliferating cell nuclear antigen (PCNA), in the two groups also showed a marked decrease in the MAOA overexpression group (Fig. 3J).Therefore, these experiments suggest that MAOA can profoundly inhibit the growth of gastric cancer in vitro and in vivo.

MAOA suppresses gastric cancer cell growth by modulating the Warburg effect
To uncover the molecular mechanism by which MAOA inhibits gastric cancer progression, we annotated the signalling pathways of cellular functions (HALLMARK) in the TCGA cohort using GESA software.The mTOR, c-MYC and glycolysis pathways were significantly activated in the MAOA-low expression group (Fig. 4A).Next, we analysed all the pathways in the HALLMARK classification, and the cancer-related or cell proliferation-related pathways were relatively activated in the MAOA low expression group, while in the MAOA high expression group, the tumour suppressing and metabolism-related signalling pathways were activated (Fig. 4B).
To explore the function of MAOA in cancer cell glycolysis, we compared the mRNA expression of glycolysisrelated genes between the MAOA overexpression group and the vector group.In BGC-823 cells, the mRNA expression levels of PDK1, PFKL, LDHA, ENO1 and PKM2 were obviously decreased in the MAOA overexpression group compared with the vector group (Fig. 4C).In another cell line, HGC-27, the mRNA expression of HK2, PDK1, PFKL, LDHA, ALDOA and PKM2 also showed a downwards trend in the MAOA overexpression group compared with that in the vector group (Fig. 4D).In summary, MAOA overexpression in BGC-823 and HGC-27 cancer cells clearly inhibited the mRNA expression levels of glycolysis-related enzymes.The results for PDK1, PFKL, LDHA and PKM2 were further confirmed at the protein expression level by Western blotting (Fig. 4I).
Next, a Seahorse analysis was performed to detect the effects of MAOA on cancer cell glycolysis.When MAOA was overexpressed in BGC-823 cell line, the extracellular acidification rate (ECAR) and the oxygen consumption rate (OCAR) of the cancer cells were significantly inhibited compared with those in the control vector group (Fig. 4E  and F).Meanwhile, the ECAR and OCAR of SGC-7901 cells were significantly increased using siRNA of MAOA (Supplementary Fig. 4A and 4B).The glycolysis and glycolytic capacity of cancer cells were significantly inhibited in the MAOA overexpression group in the cancer cell line BGC-823.Moreover, ATP production and maximal respiration were also suppressed in the MAOA overexpression group (Fig. 4G and H).The Warburg effect, or glycolysis capacity, determines glucose uptake and lactate production.
Therefore, the glucose uptake and lactate production capacity were also examined between the MAOA overexpression group and vector group of gastric cancer cells.The relative glucose uptake was significantly reduced in the MAOA overexpression group compared with that in the vector group for both the BGC-823 and HGC-27 cell lines.The lactate production capacity was also obviously reduced in the MAOA overexpression group (Fig. 4J).Thus, MAOA may inhibit the Warburg effect in gastric cancer.

MAOA interacts with NDRG1, and analysis of the correlation between them in gastric cancer
The protein interaction prediction website (GeneMANIA) and previous related studies suggest that MAOA may interact with NDRG1 and regulate the malignant biological behaviour of tumours (Fig. 5A).Therefore, we performed coimmunoprecipitation to further examine the interaction between MAOA and NDRG1 in the BGC-823 cell line (Fig. 5B).Then, we analysed the relationship between MAOA and NDRG1 at the mRNA and protein levels.There was a significant positive correlation between MAOA and NDRG1 at the mRNA level when analysing 30 gastric cancer patient tissues (R 2 = 0.5136) (Fig. 5D).In 587 gastric cancer tissue microarrays, the positive correlation of MAOA and NDRG1 at the protein expression level was further confirmed (p = 0.000).Patients with high MAOA expression typically also had high NDRG1 expression.Similarly, patients with low MAOA expression typically also had low NDRG1 expression (Fig. 5C).Next, we analysed the mRNA expression level of NDRG1 and the SUVmax value for the 30 patients who completed the PET-CT examination before the operation.We found that NDRG1 mRNA expression and the SUVmax value were negatively correlated (R 2 = 0.4108) (Fig. 5E).To explore MAOA and NDRG1 colocalization, immunohistochemistry of MAOA and NDRG1 in a tumour microarray was performed.MAOA and NDRG1 expression showed obvious colocalization (Fig. 5F).Next, we analysed the relationship between NDRG1 expression and patient survival in 587 gastric cancer patients.We found that patients with high NDRG1 expression had no obvious improvement in survival (p = 0.1290) (Fig. 5G).

The MAOA-NDRG1 axis suppresses gastric cancer cell growth by regulating the PI3K/AKT/ mTOR pathway
From the above analysis, we learned that MAOA is closely related to glycolysis and may hold a tumour suppression gene function.To demonstrate whether the antioncogene and anti-glycolysis roles of MAOA in gastric cancer are dependent on the metabolism-related signalling pathway, we examined the protein expression of PI3K/AKT/mTOR.When MAOA was overexpressed in BGC-823 and HGC-27 cell lines, the phosphorylation levels of PI3K, AKT and mTOR were significantly decreased compared with those in the vector group.The total protein levels of PI3K, AKT and mTOR were not obviously different between the MAOA overexpression group and the vector group (Fig. 6A).When the MAOA-specific inhibitor clorgyline was used to treat MAOA-overexpressing and control gastric cancer cells, the inhibitory effects of MAOA overexpression on the phosphorylation levels of PI3K, AKT and mTOR were significantly abrogated.However, clorgyline could only inhibit the enzyme activity of MAOA and had no effect on its expression level (Fig. 6A).Furthermore, we found that overexpression of MAOA could increase the expression level of NDRG1.Administration of clorgyline had no effect on the expression levels of either MAOA or NDRG1 (Fig. 6B).
To investigate the role of the MAOA-NDRG1 axis in the growth of gastric cancer cells, we used NDRG1 shRNA to treat MAOA-overexpressing and control BGC-823 cells.Western blotting analysis showed that NDRG1 knockdown rescued the inhibitory effect of MAOA overexpression on the phosphorylation level of mTOR (Fig. 6C).Using CCK8 and colony formation assays, we found that knockdown of NDRG1 could rescue the inhibitory effect of MAOA overexpression on the proliferation and colony formation of gastric cancer cells (Fig. 6D-G).Taken together, we conclude that the MAOA-NDRG1 axis can regulate the Warburg effect through the PI3K-AKT-mTOR pathway.
In Fig. 6H, we show a schematic diagram summarizing our results and the function of MAOA in gastric cancer.MAOA acts as a neurotransmitter-degrading enzyme and can degrade NE and E to prevent the combination of neurotransmitters with their receptors, such as ADRB2 and ADRA1A, thereby further inhibiting classical neurotransmitter signalling pathway function.MAOA can interact with NDRG1 and inhibit the Warburg effect in cancer cells by suppressing PI3K/AKT/mTOR pathway activation, thus restraining cancer cell proliferation and colony formation.In summary, the MAOA-NDRG1 axis can inhibit the cancer cell Warburg effect by suppressing the PI3K/AKT/mTOR pathway to exert its tumour suppressor gene role (Fig. 6H).

Discussion
Neurotransmitters released from peripheral and autonomic nerves play a very wide spectrum of roles in the signalling from the cells of the nervous system to target cells through binding to their respective receptors.In the tumour microenvironment, neurotransmitters can affect immune cells, endothelial cells and tumour cells to promote tumour progression.In our previous research, we confirmed that MAOA can affect the canonical neurotransmitter signalling pathway and indirectly activate the EGFR signalling pathway to inhibit the progression of hepatocellular carcinoma [15].In this study, we found that MAOA can restrain glycolysis by inhibiting the PI3K/AKT/mTOR signalling pathway, thereby inhibiting the occurrence and progression of gastric cancer.
Previous studies on MAOA and tumours have mainly focused on prostate cancer [20,21] and hepatocellular carcinoma, while reports on gastrointestinal tumours such as gastric cancer are relatively rare.We only found one related article on MAOA in gastric cancer, but it did not explore the molecular mechanism of MAOA on the occurrence and progression of gastric cancer or combine MAOA expression in gastric cancer tissues to analyse its related functions clinically [22].Additionally, MAOA was mentioned to regulate the expression of the glycolysis rate-limiting enzymes HK2 and PDH, but the specific mechanism was not elaborated in detail.Moreover, the authors only discussed the effect of MAOA on glycolysis and the effect on ROS in general, and did not discuss the specific effect and the role of MAOA on these processes.Therefore, it remained necessary to conduct in-depth research and elucidate the mechanisms by which MAOA affects glycolysis and gastric cancer progression.
First, in our study, we found that MAOA can affect the occurrence and progression of gastric cancer by regulating the Warburg effect of tumour cells in addition to the classical neurotransmitter signalling pathway.The Warburg effect is a special metabolic pattern of tumour cells in the tumour microenvironment [23,24].It can quickly obtain energy and metabolites through glycolysis under aerobic conditions, providing material and energy support for the malignant biological behaviour of tumour cells.We found that the effect of MAOA on the Warburg effect was mainly regulated by the upstream PI3K/AKT/mTOR pathway, which is an important pathway for cell metabolism.Previous studies have indicated that the PI3K/AKT/mTOR pathway is involved in various cellular physiological activities, such as glucose metabolism, and is closely related to the proliferation, division and metabolism of tumour cells [25][26][27].In addition, one previous study of trastuzumab, a humanized antibody targeting HER2 in gastric cancer, also described a close relationship between the Warburg effect and the PI3K/ AKT/mTOR pathway [28].
Moreover, we observed an interaction between MAOA and NDRG1, and their binding can affect the downstream PI3K/AKT/mTOR pathway.In the case of MAOA overexpression and associated increase of its activity, MAOA may be able to interact with NDRG1 by recruiting more proteins of NDRG1 to the surrounding environment, thereby affecting its downstream signaling pathways.NDRG1 is a member of the N-myc downregulated gene family, which belongs to the alpha/beta hydrolase superfamily.The protein encoded by this gene is a cytoplasmic protein involved in stress responses, hormone responses, cell growth, and differentiation [29,30].NDRG1 acts as a tumour suppressor in many cancer types.In colorectal cancer, NDRG1 inhibits EMT, migration and invasion through the interaction and promotion of caveolin-1 ubiquitylation [31].Suppression of NDRG1 transcription can promote colorectal cancer growth and metastasis [32].In breast cancer, NDRG1 expression is also known to be an independent prognostic factor in inflammatory breast cancer, and NDRG1 is widely described as a metastasis suppressor in breast cancer [33].In fact, previous studies have demonstrated that NDRG1 is closely associated with the PI3K/AKT/mTOR signalling pathway.A study on chemotherapy resistance in gliomas identified NDRG1 as a unique hypoxia-, steroid-, and mTORC2/SGK1-regulated molecule that may be developed as a predictive biomarker for response to treatment [34].Other studies in pancreatic cancer and prostate cancer also demonstrated a regulatory relationship between NDRG1 and PI3K/AKT/mTOR.NDRG1 is able to inhibit the Ras and PI3K pathways while upregulating NEDD4L, SMAD4, and PTEN expression in pancreatic cancer [35].In prostate cancer, NDRG1 has an important role in mediating the tumour-suppressive effects of Dp44mT via selective targeting of the PI3K/AKT, TGF-b and ERK pathways [36].In a study of endometrial cancer, the NDRG1/PTEN/AKT axis was found to regulate endometrial cancer growth and metastasis [37].In an angiogenesis-related study of human placental cells, the authors demonstrated that NDRG1 regulates angiogenesis in vitro and suggested the possibility that NDRG1 inhibits angiogenesis through suppression of PI3K/ AKT signalling [38].In our study, we analysed the correlation between NDRG1 and MAOA and found that there was a positive correlation at the mRNA level, and that there was a negative correlation between the mRNA level of NDRG1 and the SUVmax value of PET-CT examination in gastric cancer patients.However, after analysing the expression of NDRG1 and the prognosis of gastric cancer patients, we found no significant difference between the high and low expression groups, suggesting that NDRG1 cannot be used as an independent indicator to predict the survival prognosis of gastric cancer patients.
In addition, we found that MAOA expression is closely related to the survival and prognosis of gastric cancer patients.This is a similar conclusion as our previous study in hepatocellular carcinoma demonstrating the tumour suppressive role of MAOA [15].In prostate cancer, the role and mechanism of MAOA are different from that of digestive tumours [39].Moreover, MAOA can also be used as an independent prognostic factor to predict the survival and prognosis of gastric cancer patients clinically.Future studies should further expand the sample number and combine multicentre and multiunit data to support this conclusion.To explore the relationship between MAOA and the glucose metabolism of gastric cancer patients, we analysed gastric cancer patients who underwent PET-CT examination before surgery and found that MAOA was related to the SUVmax value of PET-CT at both the RNA expression level and protein expression level.Gastric cancer patients with high metabolism often have low or missing expression of MAOA.Therefore, the expression of MAOA can reflect the activity of glucose metabolism in gastric cancer patients to a certain extent.This further supports the close relationship between MAOA and metabolism, especially glucose metabolism.It also suggests that the SUVmax value of gastric cancer patients undergoing PET-CT examination in the future can be preliminarily judged by the expression of MAOA, though the specific and exact relationship requires further study.

Conclusions
In the tumour microenvironment, tumour cells rapidly obtain energy ATP and metabolites, such as pyruvate and lactate, through a unique metabolic pathway.The Warburg effect, namely, aerobic glycolysis, is the basis for the rapid proliferation, invasion and migration of tumour cells and other malignant behaviours.It provides energy, substrate and raw materials, which are necessary for cancer cell division.MAOA can degrade NE/E and attenuate its effect on its receptors ADRB2/ADRA1A and the classical neurotransmitter signalling pathway, thus ultimately exerting a tumour-suppressing effect in the tumour microenvironment.Importantly, we show that MAOA can interact with NDRG1, inhibiting the activity of the downstream PI3K/AKT/mTOR pathway, thereby attenuating the Warburg effect of gastric cancer cells and ultimately inhibiting the proliferation of tumour cells and their malignant biological behaviours.This study may provide new ideas for future research on the roles of neurotransmitters and their degrading enzymes and synthetases in gastric cancer.Future studies based on these results may lead to the development of specific target drugs and may provide new strategies for the treatment of gastric cancer patients, especially in terms of activated glucose metabolism.

Fig. 1
Fig. 1 MAOA is downregulated in gastric cancer tissues and closely related to poor prognosis of patients.A NE/E-related synthetase and degrading enzyme.B The relative mRNA expression of DBH, MAOA, MAOB and COMT in gastric cancer and normal tissues as analysed using the TCGA database.C The relative mRNA expression of DBH, MAOA, MAOB and COMT in paired gastric cancer and normal tissues as analysed using the TCGA database.D The relative

Fig. 2
Fig.2MAOA is associated with improved prognosis and acts as an independent prognostic factor in gastric cancer.A Immunohistochemical staining of MAOA in gastric cancer tissues.Scale bars: 100 μm, 50 μm.B The OS of gastric cancer patients for MAOA (P = 0.0193) expression based on a 587-case gastric cancer tissue microarray.C The correlation between MAOA expression and the SUXmax value of PET-CT of gastric cancer patients at the mRNA level (upper) (R 2 = 0.4838) and protein level (lower) (P = 0.000).D Two representative cases of gastric cancer patients with high and low expression of MAOA who underwent PET-CT examination.Scale bars: 100 μm.E Multiple-factor analysis of MAOA in gastric cancer tissue microarrays ◂

Fig. 3
Fig. 3 MAOA suppresses gastric cancer cell growth in vitro and in vivo.A Cell viability assay of MGC-803 and SGC-7901 cells administered MAOA or NC siRNA.B Colony formation assay of MGC-803 and SGC-7901 cells administered MAOA or NC siRNA.C Cell viability assay of BGC-823 and HGC-27 cells administered MAOA or vector lentivirus.D Colony formation assay of BGC-823 and HGC-27 cells administered MAOA or vector lentivirus.E The subcutaneous tumours of mice injected with MAOA-overexpressing or vector BGC-823 cells.F and G The tumour growth (F) and weight (G) of mice injected with MAOA-overexpressing or vector BGC-823 cells.H Immunohistochemical staining of MAOA in the tumours of mice injected with MAOA-overexpressing or vector BGC-823 cells.Scale bars: 100 μm, 50 μm.I and J Immunohistochemical staining of Ki67 (I) and PCNA (J) in the tumours of mice injected with MAOA-overexpressing or vector BGC-823 cells.Scale bars: 100 μm, 50 μm.**: P < 0.01 ◂

Fig. 4
Fig. 4 MAOA suppresses the Warburg effect in gastric cancer cells.A GSEA of mTOR, c-MYC and the glycolysis pathway from HALL-MARK.B Pathway enrichment analysis in the MAOA low and high expression groups.For the above analysis, the top 100 cases with the highest mRNA expression and 100 cases with the lowest mRNA expression of MAOA were compared in the TCGA database.C and D The mRNA expression of glycolysis-related genes in MAOA-overexpressing and vector BGC-823 (C) and HGC-27 cells (D).E and

Fig. 5 ◂Fig. 6
Fig.5 The interaction and correlation analysis between MAOA and NDRG1 in gastric cancer.A The protein interaction prediction of MAOA and other proteins from GeneMANIA.B Coimmunoprecipitation of MAOA (HA is added at the C-terminus and N-terminus) and NDRG1 in BGC-823 cells.C The correlation between MAOA and NDRG1 in 587 gastric cancer tissue microarrays.D The correlation between MAOA and NDRG1 at the mRNA level by analysing 30 gastric cancer patient tissues (R 2 = 0.5136).E The correlation between NDRG1 expression and the SUXmax value of PET-CT of gastric cancer patients at the mRNA level (R 2 = 0.4108).F Immunohistochemical staining of NDRG1 in gastric cancer tissues.Scale bars: 100 μm, 50 μm.G The OS of gastric cancer patients for NDRG1 (P = 0.1290) expression based on a 587-case gastric cancer tissue microarray.*: P < 0.05; **: P < 0.01 ◂