The Serotonergic and Cholinergic Signaling Pathways Drive Gastric Precancerous Lesions and Its Potential Mechanism

ABSTRACT

Gastric cancer (GC) is a leading lethal malignancy globally. Gastric precancerous lesion (GPL) represents a critical step in GC, but its neural regulatory mechanisms remain largely unclear. We combined bioinformatics, patch‐clamp recording, immunofluorescence, and chemogenetics to investigate serotonergic and cholinergic signaling in GPL. An N‐methyl‐N‐nitrosourea (MNU) ‐induced mouse model was used for in vivo validation. High 5‐HTR1D expression correlated with poor survival in gastric adenocarcinoma patients and was upregulated in AGS and MKN45 cells, as well as in gastric tissues of GPL mice. 5‐HT evoked non‐selective cation currents involving 5‐HTR1D and TRPV4 in gastric epithelial and cancer cells, as supported by both pharmacological inhibition and siRNA‐mediated TRPV4 knockdown. Blockade of 5‐HTR1D or TRPV4 attenuated intestinal metaplasia and dysplasia, and this improvement was accompanied by reduced AKT/mTOR activation in GPL mice. Increased gastric M3R/5‐HT co‐localization was associated with elevated acetylcholine and overactivated cholinergic neurons in the dorsal motor nucleus of the vagus (DMV). Chemogenetic inhibition of DMV cholinergic neurons decreased ChAT and 5‐HT levels and alleviated gastric pathology in GPL mice. These findings suggest that 5‐HT/5‐HTR1D signaling may promote TRPV4‐involved cation currents, contributing to mucosal pathology and intestinal metaplasia in GPL model mice, whereas DMV cholinergic neurons projecting to the stomach may facilitate gastric 5‐HT secretion and thereby potentially contribute to GPL progression. This work establishes a novel neuro‐epithelial network linking vagal cholinergic output to serotonergic signaling that may be involved in GPL progression, providing potential targets for prevention and intervention.

Vagal cholinergic neurons promote gastric 5‐HT secretion, which drives gastric precancerous lesions through 5‐HTR1D and TRPV4, highlighting these as candidate targets for early intervention.

Abbreviations

5‐HT

5‐hydroxytryptamine

5‐HTR

5‐hydroxytryptamine receptor

AAV

adeno‐associated virus

AB‐PAS

alcian blue‐periodic acid schiff

ACh

acetylcholine

AKT

protein kinase B

CaMKK

Ca²⁺/calmodulin‐dependent protein kinase kinase

ChAT

choline acetyltransferase

CNO

clozapine N‐oxide

DFI

disease‐free interval

DMV

dorsal motor nucleus of vagus

DSS

disease‐specific survival

ELISA

enzyme linked immunosorbent assay

FG

flouro‐gold

GC

gastric cancer

GPL

gastric precancerous lesions

IM

intestinal metaplasia

M3R

muscarinic acetylcholine receptor M3

MNU

N‐Nitroso‐N‐methylurea

mTOR

mammalian target of rapamycin

OS

overall survival

qRT‐PCR

quantitative real‐time PCR

siRNA

small interfering RNAs

TME

tumor microenvironment

TRPV4

transient receptor potential vanilloid type 4

VIP

vasoactive Intestinal peptide

VPAC1

vsoactive intestinal peptide receptor 1

1. Introduction

Gastric cancer (GC) remains a major global health threat and concern, ranking as the fifth most common malignancy and the third leading cause of cancer‐related death [1, 2]. The development of GC is a multi‐step process driven by complex interactions between genetic predisposition and environmental exposures. This carcinogenic cascade typically evolves from normal gastric mucosa to chronic gastritis, then to mucosal atrophy, intestinal metaplasia (IM), and dysplasia, before culminating in invasive carcinoma [3]. IM and dysplasia are collectively defined as gastric precancerous lesions (GPL), marking a pivotal phase with a high risk of malignant transformation. The severity of GPL is strongly correlated with the risk of subsequent GC, underscoring the clinical significance [4, 5, 6]. Consequently, the GPL stage represents a critical window for clinical intervention, and early effective management is essential to arrest progression to invasive cancer [7, 8].

5‐Hydroxytryptamine (5‐HT), a monoamine neurotransmitter synthesized from tryptophan, is predominantly found in the gastrointestinal tract, platelets, and the central nervous system. In the periphery, approximately 95% of 5‐HT is produced by enterochromaffin cells in the gut, where it regulates motility and secretory functions. There is growing evidence that dysregulated 5‐HT signaling, often mediated by the overexpression of its receptors (5‐HTRs), contributes to the proliferation and survival of various cancers, including colorectal, pancreatic, hepatocellular, breast, lung, and prostate cancers [9, 10, 11, 12, 13, 14]. Seven types of 5‐HTRs (5‐HTR1‐7), comprising a total of 14 subtypes, have been identified. Among the subtypes, 5‐HTR1D is reported to serve as an independent prognostic risk factor in GC. Silencing 5‐HTR1D or treating GC cell lines (AGS and MGC803) with the antagonist GR127935 suppressed DNA replication, arrested cell cycle, induced ferroptosis, and altered metabolic pathways, ultimately inhibiting proliferation, invasion, and migration [15]. Although HTR1F was also found to be upregulated in gastric cancer cells, its prognostic significance and functional role in gastric cancer remain largely unknown. Therefore, the present study focused on 5‐HTR1D. However, the potential role of 5‐HTR1D in gastric mucosal injury and the development of intestinal metaplasia during the GPL stage remains unexplored.

Acetylcholine (ACh), a key neurotransmitter in both the central and peripheral nervous systems, is essential for cognitive function, autonomic regulation, and neuromuscular transmission [16, 17, 18]. Its involvement extends to the pathophysiology of various conditions, including chronic inflammation, gastrointestinal disorders, and several cancers [19, 20, 21, 22, 23, 24]. For example, ACh can function as a growth factor, stimulating tumor cell proliferation in lung, breast, colon, and gastric cancers [25, 26, 27, 28]. Studies using ectopic transplantation models have demonstrated that the muscarinic ACh receptor M3 (M3R) antagonists, such as 4‐diphenylacetoxy‐N‐methylpiperidine methiodide and darifenacin, significantly reduce gastric tumor volume [29]. Vagotomy proved more effective than localized M3R blockade in a mouse model of GC [30], suggesting that additional pathways induced by ACh contribute to gastric carcinogenesis. It is known that the vagal nerve also modulates mucosal secretory cells. For instance, vagus nerve‐derived ACh stimulates gastrointestinal mast cells to release histamine, enhancing gastric acid secretion [31], and activates chromaffin cells to secrete 5‐HT, which promotes gastrointestinal motility [32]. Whether ACh‐induced 5‐HT release plays a role in GPL is currently unknown. Clarifying the mechanism could provide important insights into the cholinergic regulation of GC development.

Transient receptor potential (TRP) channels are increasingly recognized for their roles in various diseases, including cancers [33]. The TRP channels assemble as homo‐ or heterotetrameric complexes that permit Na⁺/Ca²⁺ transport across membranes, modulating cellular excitability and Ca²⁺‐dependent signaling cascades [34]. TRPV4 channel, in particular, has drawn significant interest due to its involvement in tumor initiation, progression, and metastasis, such as colorectal, lung, and gastric carcinoma [35, 36]. Particularly in gastric cancer, abnormal Ca²⁺ influx resulting from TRPV4 channel activation further fuels malignancy. Overexpression of the TRPV4 channel in GC patients is strongly associated with deeper invasion, lymph node metastasis, and reduced overall and disease‐free survival [37]. Excessive opening of the TRPV4 channel caused by Calcium‐sensing receptor and vasoactive intestinal peptide receptor 1 (VPAC1) increases Ca²⁺ entry and accelerates tumor progression in GC animal models [38, 39]. Moreover, dysregulation of Ca²⁺ signaling of the TRPV4 channel induced by the VIP/VPAC1 axis enhances malignant phenotypes in GC cells [38]. These observations lead us to propose that the TRPV4 channel also plays a critical role in GPL. Previous work indicates that the activation of 5‐HTR could modulate the function of the TRPV4 channel to mediate nociceptive signaling [40]. Nevertheless, it remains unclear whether the activation of 5‐HTR1D affects the activity of the TRPV4 channel in GPL, or whether their interaction contributes to precancerous development.

Building on these findings, the following hypothesis was proposed: In GPL progression, abnormal expressed 5‐HT, triggered by elevated vagus nerve‐derived Ach, activates 5‐HTR1D, which enhances Ca²⁺ influx through TRPV4 channels (Figure 1). This cascade drives gastric mucosal alterations, including intestinal metaplasia, a hallmark of precancerous lesions. This study employs various methods such as patch‐clamp recording, immunofluorescence chemical staining, and chemogenetics to analyze ion channel regulation, neurotransmitter signaling, and classical oncogenic signaling pathways during GPL. By delineating the crosstalk among 5‐HTR1D, TRPV4, and cholinergic signaling, our work provides a theoretical foundation and identifies potential new directions for the early intervention and treatment of GC.

FIGURE 1.

Association between 5‐HTR1D expression and survival in gastric cancer patients. (a) mRNA expression level of 5‐HTR1D was significantly elevated in gastric adenocarcinoma tissues compared to normal controls. (b–d) Kaplan–Meier survival analysis revealed that high 5‐HTR1D expression was significantly associated with shorter disease‐free interval (DFI; b), disease‐specific survival (DSS; c), and overall survival (OS; d) in gastric adenocarcinoma patients. ***p < 0.001 versus normal controls.

2. Materials and Methods

2.1. Animals

Male BALB/c mice (SPF grade, 4–5 weeks old, 18–25 g) were obtained from the Laboratory Animal Center of Qingdao University (Shandong, China). Mice were housed under a 12 h light/dark cycle at 22°C–24°C with 45% ± 5% humidity and ad libitum access to food and water. All animal procedures were conducted in accordance with Qingdao University's Guidelines for the Care and Use of Laboratory Animals. The procedures were approved by the Animal Ethics Committee of Qingdao University's Medical College (Approval Document Number: NO. 2‐23‐5‐1Balb/c9620250125148).

2.2. Bioinformatics Analysis

5‐HTR1D mRNA expression in gastric adenocarcinoma was retrieved from the TIMER database (http://timer.cistrome.org). Survival analyses were performed using the PanCanSurvPlot tool (https://smuonco.shinyapps.io/PanCanSurvPlot/) in the TCGA‐STAD cohort, assessing disease‐free interval (DFI), disease‐specific survival (DSS), and overall survival (OS). Patients were dichotomized into high‐ and low‐expression groups using the optimal cutoff determined by the platform.

2.3. Cell Culture

Human gastric cancer cell lines MKN45 and AGS, and the normal gastric epithelial cell line GES‐1, were obtained from the Chinese Academy of Sciences Cell Bank (Shanghai, China). Cells were cultured in DMEM‐high glucose medium supplemented with 10% fetal bovine serum, 100 IU/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified atmosphere containing 5% CO₂.

2.4. RNA Extraction and Quantitative Real‐Time PCR (qRT‐PCR)

Total RNA was isolated using RNAiso Plus reagent. cDNA was synthesized from 1 μg RNA with PrimeScript RT‐polymerase. Quantitative PCR was performed on a StepOnePlus Real‐Time PCR System using SYBR Premix Ex Taq II, with 50 ng cDNA as template. All reactions were run in triplicate, and GAPDH was used as the endogenous control. Relative mRNA levels of target genes were calculated via the 2 − ΔΔCt method. Primer sequences are listed in Table S1 [41, 42, 43].

2.5. siRNA Knockdown of TRPV4

TRPV4 expression was knocked down in AGS and MKN45 cells using small interfering RNAs (siRNAs, sequences listed in Table S2) via reverse transfection with Lipofectamine 3000 according to the manufacturer's instructions [44]. A non‐targeting siRNA was used as the negative control. Cells were cultured for 48–72 h before subsequent patch‐clamp recordings. Knockdown efficiency was confirmed by qRT‐PCR.

2.6. Electrophysiological Recordings

Whole‐cell patch‐clamp recordings were performed on GES‐1, MKN45, and AGS cells using an EPC 10 USB Double amplifier. Signals were digitized at 10 kHz and low‐pass filtered at 5 kHz. Cells were voltage‐clamped at 0 mV to inactivate voltage‐gated calcium and sodium channels, and a 100‐ms linear ramp protocol (−100 mV to 100 mV) was applied every 2 s. Extracellular solution contained (mM): 140 NaCl, 5 KCl, 2 CaCl₂, 2 MgCl₂, and 10 HEPES (pH 7.3). Pipette solution contained (mM): 140 CsCl, 5 EGTA, 3 Mg‐ATP, and 10 HEPES (pH 7.3). Osmolality was adjusted to approximately 300 mOsm/kg. 5 HT (30 μM), LY310762 (10 μM), and RN 1734 (10 μM) were applied via bath perfusion.

2.7. Establishment of the GPL Model and Grouping

Mice were randomly assigned to four groups: (1) control (distilled water + Vehicle); (2) model (MNU + Vehicle); (3) 5‐HTR1D antagonist (MNU + LY310762); and (4) TRPV4 antagonist (MNU + RN‐1734). The GPL model was induced by administering 120 ppm MNU (Cat. No. N425449, Aladdin, China) in drinking water as previously described (alternate day, light protected bottles, 20 weeks) [45, 46]. Starting at Week 10, three mice from the control and model groups were randomly euthanized for gastric tissue collection; hematoxylin and eosin (H&E) staining confirmed the presence of pathological changes. The remaining model mice were then randomly divided into the model, 5‐HTR1D antagonist, and TRPV4 antagonist groups. From week 10, the control and GPL model groups received 0.4 mL/day vehicle (≤ 5% DMSO in 0.9% NaCl) via intragastric administration. The 5‐HTR1D antagonist group was administered LY310762 (1 mg/kg/day in 0.4 mL vehicle; Cat. No. GC17138, Glpbio, USA) [47], and the TRPV4 antagonist group received RN‐1734 (10 μmol/L in 0.4 mL vehicle; Cat. No. HY‐19975, MedChemExpress, USA). All treatments were given intragastrically for 10 consecutive weeks. At the end of the study, all mice were sacrificed for collection of stomach, brain, and blood samples for further analysis.

2.8. H&E Staining

Gastric corpus tissue samples were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at a thickness of 3 μm. Sections were stained with H&E following standard protocols. Histopathological evaluation was performed under a light microscope. All slides were assessed in a blinded manner by two independent pathologists.

2.9. Alcian Blue‐Periodic Acid Schiff (AB‐PAS) Staining

AB‐PAS staining was employed to evaluate intestinal metaplasia (IM) in gastric tissues. This method distinguishes neutral mucins (magenta) in normal mucosa from acidic mucins (blue) characteristic of IM lesions. Morphological assessment was conducted under a microscope.

2.10. Western Blotting

Total protein was extracted using RIPA buffer and quantified with a BCA kit. Samples (20 μg) were resolved on 10% SDS‐PAGE gels, transferred to PVDF membranes, and blocked with 5% non‐fat milk in TBST. Membranes were incubated with primary antibodies at 4°C overnight, followed by HRP‐conjugated secondary antibodies for 1 h at room temperature. Signals were developed with an ECL system, and band densities were analyzed with ImageJ software relative to an internal reference.

2.11. ELISA

Gastric and dorsal motor nucleus of the vagus (DMV) tissue samples (0.04 g) were separately homogenized in 360 μL phosphate‐buffered saline (PBS) and centrifuged at 1,450 × g for 20 min at 4°C. Supernatants were collected and assayed for the concentrations of ChAT and 5‐HT using commercial ELISA kits according to the manufacturer's instructions.

2.12. Immunofluorescence Staining

Brain tissues were post‐fixed for 4 h and cryoprotected overnight in 30% sucrose. Coronal sections (10 μm thick) were cut using a cryostat. Sections were blocked for 2 h at room temperature with 4% goat serum and 0.5% Triton X‐100, followed by incubation with primary antibodies overnight at 4°C. For gastric cells, the primary antibodies were rabbit anti‐ChAT (Cat. No. ab181023, Abcam, UK) or rabbit anti‐5‐HT (Cat. No. 20080, Immunostar, USA). For neurons in the DMV, co‐staining was performed using rabbit anti‐ChAT and chicken anti‐c‐Fos (Cat. No. 226009, Synaptic Systems, Germany). The following day, sections were washed three times (5 min each) with PBS and incubated for 2 h at room temperature with appropriate secondary antibodies: goat anti‐rabbit Alexa Fluor 488 (Cat. No. 115–545‐003, Jackson ImmunoResearch, USA) and goat anti‐chicken Alexa Fluor 594 (Cat. No. ab150176, Abcam, UK). After three additional PBS washes, sections were mounted with antifade medium. Images were captured using a BX50 microscope equipped with a DP50 digital camera. For quantification, immunoreactive cells were counted in five distinct regions across five brain slices per mouse. The percentage of double‐labeled cells was calculated as (number of double‐labeled cells/total number of positive neurons) × 100% [48].

2.13. Fluoro‐Gold (FG) Retrograde Tracking

Mice were anesthetized with ketamine (100 mg/kg) and xylazine (20 mg/kg, i.p.). A 1.5‐cm midline abdominal incision was made to expose the stomach. FG (2%, 0.6 μL, Cat. No. 52–9400, Fluorochrome, USA) was injected into the stomach wall using a microinjector (maximum effective volume: 1 μL). The syringe was left in place for 10 min to allow diffusion. The incision was sutured and disinfected. Postoperatively, penicillin was administered for 3 days. After 7 days, stomach and brain tissues were harvested and fixed in 4% PFA. Brain sections (20 μm) were prepared following sucrose gradient cryoprotection for immunohistochemical analysis.

2.14. Adeno‐Associated Virus (AAV) Vector Injection

AAV vectors—rAAV‐ChAT‐hM4D(Gi)‐mCherry‐WPRES(−) and the control virus rAAV‐hSyn‐mCherry‐WPRE‐hGH pA (titer: 2 × 10¹², 0.5 μL, BrainVTA, China) were stereotaxically injected into the DMV of GPL model mice 10 weeks after modeling. The coordinates for the DMV were determined according to the mouse brain atlas of Paxinos and Franklin (2nd edition): AP, −7.34 mm; ML, ±0.3 mm; DV, −4.5 mm. Mice were randomly divided into the following four groups (n = 6) based on intraperitoneal injection of NS or Clozapine‐N‐oxide (CNO, MedChemExpress, USA), combined with DMV microinjection of either rAAV‐hSyn‐mCherry (mCherry) or rAAV‐ChAT‐hM4D(Gi)‐mCherry (hM4D): mCherry + NS, mCherry + CNO, hM4D + NS, and hM4D + CNO. Two weeks after viral expression, CNO (1 mg/kg dissolved in saline) was administered via intraperitoneal injection 5 days per week for 10 consecutive weeks to continuously inhibit hM4D‐positive ChAT neurons within the DMV [49]. At the end of the 20 weeks modeling period, gastric tissues were collected for H&E staining, AB‐PAS staining to assess gastric mucosal injury, and immunofluorescence staining to examine the expression of 5‐HT and ChAT in the gastric wall.

2.15. Histological Identification

After the experiment, mice were perfused and fixed with 4% paraformaldehyde. Coronal brain sections (50 μm thick) were prepared using a cryostat to verify the accuracy of the nucleus injection site. Data from mice with incorrect injection locations were excluded from subsequent statistical analyses.

2.16. Statistical Analysis

Data are expressed as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism version 8.0. One‐way analysis of variance followed by Bonferroni/Dunn correction for multiple comparisons was used. Paired t‐tests were applied where appropriate to evaluate differences between groups. A p‐value of less than 0.05 was considered statistically significant.

3. Results

3.1. The Association Between 5‐HTR1D Expression and Survival Duration in Patients With Gastric Adenocarcinoma

Employing the TIMER2.0 platform (http://timer.cistrome.org), we analyzed the mRNA expression level of 5‐HTR1D in gastric adenocarcinoma patients by integrating extensive gene expression data from a large cohort of gastric cancer samples. As depicted in Figure 1a, the expression level of 5‐HTR1D in patients with gastric adenocarcinoma was remarkably elevated (p < 0.001). To further investigate the clinical relevance of 5‐HTR1D, we assessed its prognostic impact using the PanCanSurvPlot tool. In the TCGA‐STAD cohort, high 5‐HTR1D expression was significantly associated with shorter DFI (p < 0.05; Figure 1b), DSS (p < 0.05; Figure 1c), and OS (p < 0.05; Figure 1d). Collectively, these data indicate that elevated 5‐HTR1D expression is associated with poor prognosis in GC.

3.2. The mRNA Expression of 5‐HTR1D Was Increased in MKN45 and AGS Gastric Cancer Cell Lines

The mRNA expression of nine 5‐HTR subtypes in GES‐1, MKN45 and AGS was detected by qRT‐PCR (Figure 2a–i). The results showed that compared with GES‐1 cells, the mRNA expression of 5‐HTR1D (2.335 ± 0.247 vs. 1.533 ± 0.133, p < 0.0001, Figure 2b) and 5‐HTR1F (2.239 ± 0.217 vs. 1.500 ± 0.346, p < 0.01, Figure 2c) in MKN45 cells was significantly higher, while the mRNA expression of 5‐HTR1E (0.590 ± 0.075 vs. 0.783 ± 0.126, p < 0.05, Figure 2d), 5‐HTR2B (0.567 ± 0.050 vs. 0.700 ± 0.090, p < 0.05, Figure 2f), and 5‐HTR3D (0.575 ± 0.080 vs. 0.950 ± 0.142, p < 0.001, Figure 2g) was significantly lower. Similar for the AGS cells, there were a higher mRNA expression of 5‐HTR1D (2.187 ± 0.217 vs. 1.533 ± 0.133, p < 0.001, Figure 2b) and 5‐HTR1F (2.144 ± 0.343 vs. 1.500 ± 0.346, p < 0.01, Figure 2c), a lower mRNA expression of 5‐HTR1E (0.492 ± 0.090 vs. 0.783 ± 0.126, p < 0.001, Figure 2d) and 5‐HTR2B (0.459 ± 0.082 vs. 0.700 ± 0.090, p < 0.001, Figure 2f) than those of GES‐1 cells. The results showed that the mRNA of 5‐HTR1D was markedly highly expressed in both gastric cancer cell lines.

FIGURE 2.

mRNA expression profiles of 5‐HT receptor subtypes in the human normal gastric epithelial cell line GES‐1 and human gastric cancer cell lines MKN45 and AGS. (a–i) Expression levels of 5‐HTR1A, 5‐HTR1D, 5‐HTR1F, 5‐HTR1E, 5‐HTR2A, 5‐HTR2B, 5‐HTR3A, 5‐HTR4, and 5‐HTR7 in GES‐1, MKN45, and AGS cells. Data are presented as mean ± SD (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001 compared to GES‐1.

3.3. 5‐HT Induced Non‐Selective Cation Currents Involving 5‐HTR1D and TRPV4 in GES‐1, AGS, and MKN45

To investigate the role of 5‐HTR1D in 5‐HT‐mediated ion channel activity in gastric cells, whole‐cell patch‐clamp recordings were conducted on the normal gastric epithelial cell line GES‐1 and the gastric cancer cell lines AGS and MKN45. To exclude potential solvent effects, vehicle control experiments were performed first. Application of the vehicle alone did not induce any detectable inhibition on the membrane currents in any of the three cell lines (Figure 3a–d).

FIGURE 3.

5‐HTR1D and TPRV4 were involved in 5‐HT‐induced currents in gastric cells. Representative vehicle control recordings showing that application of the vehicle (≤ 5% DMSO in extracellular solution) alone did not induce detectable membrane currents in GES‐1 (a), AGS (b), or MKN45 (c) cells. (d) The change rate of currents induced by vehicle in GES‐1, AGS, and MKN45 cells. (e–g) Representative whole‐cell currents recorded in GES‐1 (e), AGS (f), and MKN45 (g) cells in response to 5‐HT (30 μM) and subsequent inhibition by the selective 5‐HTR1D antagonist LY310762 (10 μM). (h) The change rate of currents induced by LY‐310762 in GES‐1, AGS, and MKN45 cells. (i–k) Representative whole‐cell current recordings in GES‐1 (i), AGS (j), and MKN45 (k) cells showing the response to 5‐HT (30 μM) and its suppression by the selective TRPV4 antagonist RN‐1734 (10 μM). (l) The change rate of current induced by RN‐1734 in GES‐1, AGS, and MKN45 cells. (m,q) qRT‐PCR analysis of TRPV4 mRNA expression following transfection with three independent TRPV4 siRNA sequences in AGS and MKN45 cells, respectively; siRNA‐2 exhibited the highest knockdown efficiency and was used for subsequent electrophysiology experiments. (n–p) Representative whole‐cell currents evoked by 5‐HT in AGS cells transfected with negative control siRNA or TRPV4 siRNA‐2, and corresponding quantification of peak current amplitudes (p). (r–t) Representative 5‐HT‐triggered current traces from MKN45 cells transfected with negative control siRNA or TRPV4 siRNA‐2, along with statistical analysis of maximum current magnitudes (t). Data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 versus GES‐1 or scrambled siRNA control.

As illustrated in Figure 3e–g, application of 5‐HT (30 μM) induced transmembrane non‐selective cation currents in all three cell types. These currents were effectively inhibited by co‐application of the selective 5‐HTR1D antagonist LY310762 (10 μM). Quantitative analysis revealed that the peak current densities in both AGS and MKN45 cells were significantly higher than the density in GES‐1 cells (p < 0.05 and p < 0.001, respectively, Figure 3h), suggesting enhanced 5‐HTR1D‐mediated signaling in gastric cancer cells.

Based on the characters of Ca²⁺ conductance [50] and relevance of TRPV4 with gastric cancer [37], it was explored whether the TRPV4 channel was involved in 5‐HT‐induced currents in gastric cancer cells. The selective TRPV4 antagonist RN‐1734 (10 μM) was applied after administration of 5‐HT in GES‐1, AGS, and MKN45 cells. As illustrated in Figure 3i–k, RN‐1734 markedly inhibited the currents elicited by 5‐HT (30 μM). Quantitative analysis indicated that the inhibitory effect of RN‐1734 was significantly greater in AGS and MKN45 cells compared to GES‐1 cells (Figure 3l, both p < 0.0001). These pharmacological data suggest the involvement of TRPV4 in 5‐HT‐induced cation currents in gastric cancer cells.

To verify the role of TRPV4 in 5‐HT‐induced cation currents, TRPV4 was knocked down by siRNA in MKN45 and AGS cells. TRPV4 siRNA transfection significantly reduced TRPV4 mRNA expression (Figure 3 m and q) and markedly attenuated 5‐HT‐induced non‐selective cation currents compared to scrambled siRNA controls (Figure 3n–p,r–t, both p < 0.001). Collectively, these results demonstrate that TRPV4 is required for the full 5‐HT/5‐HTR1D‐evoked currents in gastric cancer cells.

3.4. Inhibition of 5‐HTR1D or TRPV4 Ameliorates Intestinal Metaplasia and Dysplasia With Concomitant Reduction of AKT/mTOR Activation

The GPL mouse model was established by drinking MNU solution freely (Figure 4a). Compared to the control group, the body weight of mice in the GPL group began to decrease significantly from the 10th week (24.15 ± 15.58 vs. 7.093 ± 4.622, p < 0.05, Figure 4b). After the 10th week, administration of LY310762 and RN‐1734 to the GPL mice could alleviate the body weight loss caused by MNU (7.093 ± 4.622 vs. 10.75 ± 8.406, p < 0.05, Figure 4b), but it was still lower than that of the control group (7.093 ± 4.622 vs. 11.60 ± 8.159, p < 0.05, Figure 4b). The findings demonstrated that the inhibition of 5‐HTR1D or TRPV4 led to an amelioration of MNU‐induced weight loss in mice.

FIGURE 4.

5‐HTR1D and TRPV4 are involved in intestinal metaplasia and dysplasia of stomach with concomitant reduction of AKT/mTOR activation. (a) Schematic diagram of the GPL modeling protocol and experimental timeline. (b) Administration of 5‐HTR1D antagonist or TRPV4 antagonist significantly attenuated the GPL‐induced decrease in body weight (n = 12). (c) Representative gastric mucosal morphology of mice in 4 groups. (d) Representative AB‐PAS staining of gastric mucosa of mice in 4 groups. (e) 5‐HTR1D mRNA expression in gastric tissues from control and GPL mice, determined by qRT‐PCR (n = 6). (f) Representative Western blot bands showing protein levels of AKT, p‐AKT, mTOR, and p‐mTOR in gastric mucosal tissues of 4 groups. (g–j) Quantitative analysis of AKT (g), p‐AKT (h), mTOR (i), and p‐mTOR (j) expression levels (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001 versus Control group; ^# p < 0.05, ^## p < 0.01 versus GPL group.

After the experiment, the representative gastric tissues of mice in each group were stained with H&E (Figure 4c). Gastric mucosal glands of control mice were closely packed, well‐organized, and exhibited regular glandular architecture. In contrast, the GPL group displayed disordered glandular arrangement, crowded glands with indistinct luminal boundaries, and gastric epithelial cells that varied in size and shape. Basal cells showed marked atypia, characterized by nuclear enlargement, hyperchromasia, loss of polarity, prominent nucleoli, and an increased nuclear‐to‐cytoplasmic ratio, consistent with precancerous lesion phenotypes. Both LY310762 and RN‐1734 groups exhibited histological alterations relative to the control group, including irregular gland morphology, hyperchromatic epithelium, nuclear enlargement, elevated nuclear‐to‐cytoplasmic ratio, and reduced polarity. However, compared with the GPL group, these aberrant features were significantly attenuated in mice treated with either LY310762 or RN‐1734, suggesting that inhibition of 5‐HTR1D or TRPV4 alleviates gastric precancerous lesion progression.

AB‐PAS stain of representative gastric tissues was used to assess the degree of intestinal metaplasia (Figure 4d). The results revealed that in control mice, the gastric mucosal epithelium stained purplish‐red, indicating secretion of neutral mucins and absence of intestinal‐type metaplasia. In contrast, the GPL model group exhibited prominent blue‐stained lesions in the gastric mucosa, reflecting the secretion of acidic glycoproteins and polysaccharide‐rich mucus, consistent with the presence of intestinal metaplasia. Administration of LY310762 or RN‐1734 to GPL mice resulted in a visible reduction in both the extent and size of blue‐stained lesions compared with the untreated GPL group. These findings indicate that there is a substantial increase in intestinal metaplasia within the GPL group, and pharmacological blockade of 5‐HTR1D or TRPV4 attenuates the development of intestinal metaplasia.

To determine whether 5‐HTR1D expression is upregulated during GPL development, 5‐HTR1D mRNA levels in gastric mucosa from control and GPL mice were examined by qRT‐PCR. 5‐HTR1D mRNA expression was significantly elevated in the GPL group compared to the control group (p < 0.05; Figure 4e), indicating that 5‐HTR1D upregulation occurs as early as the precancerous stage.

Furthermore, alterations in the AKT/mTOR signaling pathway were investigated by western blot analysis. Compared with the control group, the GPL model group exhibited significantly increased protein expression of AKT (p < 0.001, Figure 4f,g), p‐AKT/AKT ratio (p < 0.01, Figure 4f,h), mTOR (p < 0.01, Figure 4f,i), and p‐mTOR/mTOR ratio (p < 0.01, Figure 4f,j). Treatment with LY310762 significantly reduced the expression of AKT (p < 0.05, Figure 4f,g), p‐AKT/AKT (p < 0.05, Figure 4f,h), and mTOR (p < 0.05, Figure 4f,i) compared with the GPL group, while the p‐mTOR/mTOR ratio remained unchanged (p > 0.05, Figure 4f,j). Similarly, administration of RN‐1734 also led to a significant decrease in AKT (p < 0.05, Figure 4f,g), p‐AKT/AKT (p < 0.01, Figure 4f,h), and mTOR (p < 0.05, Figure 4f,i) expression, with no significant alteration in the p‐mTOR/mTOR ratio (p > 0.05, Figure 4f,j). The results suggest that alterations in AKT/mTOR signaling occur concomitantly with the progression of gastric precancerous lesion associated with both 5‐HTR1D and TRPV4.

3.5. Increased M3R/5‐HT Co‐Localization in GPL Mice Is Related to Elevated Expression of Ach in Stomach and Over‐Activated Cholinergic DMV Neurons

ELISA results showed that the concentration of 5‐HT in gastric mucosa was significantly higher in GPL group compared to control group (Figure 5a, p < 0.05). Based on the effect of ACh on 5‐HT secretion from chromaffin cells via M3R [51], the co‐localization of M3R/5‐HT and the concentration of ChAT in gastric mucosa were further examined. As shown in Figure 5b,c, the rate of M3R/5‐HT co‐labeled cells in total 5‐HT positive cells was markedly higher in GPL group than in control group (p < 0.01). ChAT level was significantly elevated in GPL group compared to control group (Figure 5d, p < 0.05). The findings suggest that the increased 5‐HT expression in GPL mice may be promoted by the surrounding cholinergic system.

FIGURE 5.

Expression of 5‐HT and ACh and M3R/5‐HT co‐localization in gastric antral tissues of GPL model mice. (a) 5‐HT levels were measured by ELISA. (b) Representative images showing M3R/5‐HT co‐localization in gastric antral mucosa, as detected by immunofluorescence. (c) The increased percentage of M3R/5‐HT co‐localization neurons in GPL mice. (d) ACh expression was determined using ELISA. (e) The schematic diagram of FG retrograding test. (f) The FG labeling ChAT‐positive neurons in the DMV 7 days after FG injection (n = 6). (g) Representative images showing c‐Fos expression of ChAT‐positive neurons in control, GPL, LY310762 and RN1734 groups. (h) The percentage of c‐Fos⁺ DMV cholinergic neurons in all groups. Data are presented as mean ± SD (n = 6). Scale bar: 100 μm. *p < 0.05, **p < 0.01, ***p < 0.001 versus Control; #p < 0.05, ##p < 0.01 versus GPL group.

It is well known that the stomach directly receives acetylcholinergic inputs from the DMV [52]. Here, FG retrograde tracing revealed that part of DMV ChAT‐immunopositive neurons were co‐labeled with FG (Figure 5e,f). Further, expression of c‐Fos was used to show the activation of DMV cholinergic neurons. The proportion of c‐Fos‐positive neurons among ChAT‐expressing DMV cells was significantly higher in the GPL group compared to the control group (p < 0.001, Figure 5g). Treatment with either LY310762 or RN‐1734 significantly reduced this activation (Figure 5g,h, p < 0.01 and p < 0.05, respectively). The results indicate that DMV cholinergic neurons are over‐activated in GPL mice and might directly modulate the expression of neurotransmitters in the stomach, such as ACh, 5‐HT.

3.6. Inhibition of DMV Cholinergic Neurons Reduces Expression of ChAT and 5‐HT in Gastric Mucosa and Alleviates Gastric Pathology in GPL Mice

To further examine the role of central cholinergic input on the stomach of GPL mice, chemogenetic inhibition was employed to inhibit the activation of cholinergic neurons in the DMV (Figure 6a). Immunofluorescence confirmed efficient AAV infection, with more than 90% ChAT‐positive neurons expressing the mCherry reporter (Figure 6b).

FIGURE 6.

Chemogenetic suppression of DMV cholinergic neurons alleviates lesion of gastric mucosa in GPL mice. (a) Schematic illustration of injecting rAAV‐ChAT‐hM4D(Gi)‐mCherry into the DMV. (b) Co‐labeling of rAAV‐ChAT‐hM4D(Gi)‐mCherry and ChAT in DMV. Scale bar: 100 μm. (c–e) Immunofluorescence analysis showing that chemogenetic inhibition of DMV ChAT‐positive neurons significantly decreased both 5‐HT immunoreactivity and ChAT‐positive neuron density in the gastric mucosa of GPL mice. (f) Representative images of H&E staining of gastric mucosa in each group mice. (g) Representative images of AB‐PAS staining of gastric mucosa in each group mice. Data are presented as mean ± SD, n = 6, **p < 0.01 versus hM4D + NS group. Scale bar: 50 μm and 25 μm.

Immunofluorescence analysis revealed that the density of ChATimmunopositive neurons per mm²‐ was significantly lower in the hM4D + CNO group compared with the hM4D + NS group, whereas no difference was observed between the mCherry + NS and mCherry + CNO groups (Figure 6c,d). Correspondingly, the concentration of 5‐HT in the gastric mucosa was also markedly decreased in the hM4D + CNO group relative to the hM4D + NS group (Figure 6e). The findings indicate that suppression of aberrantly activated DMV cholinergic neurons attenuates gastric 5‐HT expression in GPL mice.

Histological analysis of gastric mucosa revealed that mice in the mCherry+NS, mCherry+CNO, and hM4D + NS groups exhibited dense, disorganized mucosal glands with indistinct glandular boundaries, pleomorphic epithelial cells, atypical basal cells featuring enlarged and hyperchromatic nuclei, loss of polarity, and an increased nuclear‐to‐cytoplasmic ratio (Figure 6f), but mice in the hM4D + CNO group (where DMV cholinergic neurons were chemogenetically inhibited) showed markedly milder mucosal lesions (Figure 6f). AB‐PAS staining demonstrated magenta‐colored (neutral mucin‐dominant) epithelium in the hM4D + CNO group, whereas the other three groups displayed predominantly blue‐stained (acidic mucin‐rich) epithelium, an indication of intestinal metaplasia (Figure 6g). Thus, the results indicate that chemogenetic inhibition of DMV cholinergic neurons in GPL mice reduces gastric mucosal injury with intestinal metaplasia.

4. Discussion

GPL, which includes intestinal metaplasia and dysplasia arising from chronic atrophic gastritis, constitutes a crucial transitional phase in gastric carcinogenesis [53, 54]. Preventing or delaying the progression of GPL is considered a strategic measure for GC prevention [55]. The role of 5‐HTR1D in cancer progression has been documented in multiple tumor types. A recent study by Wu et al. further confirmed that 5‐HTR1D is overexpressed in gastric cancer and that its high expression is significantly associated with poor prognosis across multiple independent cohorts [56]. In GC, 5‐HTR1D activation exacerbates malignant phenotypes, and suppression curbs tumor development [15]. In colorectal cancer, 5‐HTR1D blockade downregulates Axin1 and inhibits metastasis driven by the Wnt/β‐catenin pathway [57]. 5‐HTR1D also facilitates tumor growth and metastasis in hepatocellular carcinoma [58]. Consistent with the bioinformatic data in this study, 5‐HTR1D mRNA was also upregulated in the gastric mucosa of GPL mice, suggesting that 5‐HTR1D overexpression occurs at the precancerous stage and may represent an early molecular event in gastric cancer.

TRPV4 is frequently overexpressed in GC, where it promotes proliferation, colony formation, and invasion [37]. Our study found that the patch‐clamp experiments showed activation of 5‐HTR1D was associated with TRPV4 channel opening and the induction of distinctive currents in GES‐1, MKN45, and AGS cells. The results uncover a possible 5‐HTR1D–TRPV4 pathway through which 5‐HT signaling may influence gastric epithelial and cancer cells.

We also detected higher levels of ChAT and 5‐HT in the gastric tissue of GPL mice induced by MNU, accompanied by an increase in cells co‐expressing 5‐HT and M3R. In GPL mice, treatment with either the 5‐HTR1D antagonist LY310762 or the TRPV4 antagonist RN‐1734 alleviated weight loss, gastric mucosal damage, and intestinal metaplasia, indicating that pharmacological inhibition of 5‐HTR1D or TRPV4 may attenuate GPL progression, further underscoring the potential of the 5‐HTR1D antagonist and TRPV4 antagonist as therapeutic targets in the precancerous stage. Furthermore, siRNA‐mediated knockdown of TRPV4 significantly attenuated 5‐HT‐induced currents in gastric cancer cells, providing genetic evidence that TRPV4 contributes to the full 5‐HT/5‐HTR1D‐evoked response. Together with the pharmacological data, these results support a functional role for TRPV4 downstream of 5‐HTR1D activation.

Activation of TRPV4 channels increases Ca²⁺ influx, initiating downstream signaling mediated by Ca²⁺/calmodulin‐dependent protein kinase kinase (CaMKK). CaMKK in turn activates downstream effectors, such as CaM‐kinase I, CaM‐kinase IV, AKT, 5′‐AMP‐activated protein kinase, and others. AKT serves as a central player in regulating cell growth, proliferation, and angiogenesis. Its activation drives tumor progression and confers resistance to chemotherapeutic agents [59]. In our study, GPL mice exhibited marked upregulation of AKT and mTOR protein levels, along with elevated phosphorylation. Treatment with LY310762 or RN‐1734 reduced the expression and activation of both AKT and mTOR. High levels of phosphorylated AKT (p‐AKT) are observed in numerous cancers, promoting cell cycle advancement, angiogenesis, and inhibition of apoptosis [60]. In GC, elevated expression of both AKT and p‐AKT supports continuous proliferation and disease advancement [61]. The mammalian target of rapamycin (mTOR), a downstream effector of AKT, is commonly dysregulated in GC, and its overactivation plays a key role in AKT‐mediated oncogenesis [62]. Activation of the PI3K/AKT/mTOR pathway has been observed in GPL models, including 1‐Methyl‐3‐nitro‐1‐nitrosoguanidine‐induced GPL rats and Atp4a‐deficient GPL mice with intestinal metaplasia [63]. In vitro, 5‐HT stimulates the AKT/mTOR pathway in gastric cancer cells and enhances proliferation [64]. Furthermore, VIP receptor‐mediated TRPV4 activation elevates intracellular Ca²⁺, leading to increased phosphorylation of components in the PI3K/AKT pathway [38]. Our findings align with the earlier studies, indicating that the AKT/mTOR signaling pathway might be involved in the contribution of the 5‐HTR1D‐TRPV4 pathway to GPL progression, at least in part.

Although the present study demonstrates that TRPV4 inhibition reduces AKT/mTOR activation in GPL mice, the direct causal link between TRPV4‐mediated Ca²⁺ influx and downstream AKT/mTOR signaling was not directly investigated. A previous study has shown that TRPV4 activation mediates Ca²⁺ influx in gastric cancer cells and that elevated intracellular Ca²⁺ can activate CaMKK, leading to AKT phosphorylation independently of PI3K [39]. Moreover, the PI3K/AKT/mTOR pathway is activated in other GPL models [63]. Based on these findings, it is plausible that TRPV4 contributes to GPL progression through a Ca²⁺/CaMKK/AKT/mTOR cascade. Direct validation of this signaling axis, using Ca²⁺ imaging and genetic disruption of CaMKK or AKT, will be an important direction for future studies.

Elevated proportion/counts of ChAT and c‐Fos double‐labeled neurons in the DMV suggested increased vagal cholinergic signaling. LY310762 and RN‐1734 treatments reduced ChAT neuronal activation, further supporting a modulatory role of vagal cholinergic activity in GPL progression.

Emerging evidence underscores the significance of intratumoral nerves in cancer progression [65, 66, 67]. In GC, cholinergic nerves contribute to tumor development. For example, vagotomy markedly reduces tumor growth, partially attributable to diminished muscarinic signaling [30]. ACh induces gastric mucosal cells to release nerve growth factor, which promotes neural hyperinnervation in the gastric mucosa. This excessive cholinergic stimulation accelerates gastric carcinogenesis through YAP phosphorylation signaling [68]. Additionally, a malignant symbiotic relationship between nociceptive nerves and GC has been described, whereby a feedback loop involving cancer cells and sensory neurons further promotes tumor progression [69].

Apart from acting directly on tumor cells, ACh also indirectly influences gastric pathophysiology by stimulating the secretion of other neurotransmitters. Specifically, it enhances 5‐HT release from enterochromaffin cells, which subsequently affects mucosal cells, smooth muscle, and enteric neurons to modulate gastrointestinal function. We proposed that excessive 5‐HT release resulting from vagal overactivity contributes to the initiation and progression of GC. Hence, inhibiting ACh release from vagal efferent neurons in the DMV should lower 5‐HT levels and suppress GPL development. To evaluate this, chemogenetic techniques were employed to selectively inhibit ChAT‐positive neurons in the DMV in a GPL mouse model. This intervention reduced vagally derived ACh, decreased gastric concentrations of 5‐HT and ChAT, and ameliorated gastric mucosal pathology, as evidenced by improved H&E and AB‐PAS staining. These results suggest that suppressing cholinergic activity in the DMV may delay GPL onset, reinforcing the concept that dysregulated ACh and 5‐HT signaling may be involved in precancerous transformation. Accumulating evidence highlights extensive cross‐talk between tumors and nerves. Tumor cells remodel the tumor microenvironment (TME) to support their growth. As key components of the TME, nerves undergo adaptive changes including axonal outgrowth and altered neurotransmitter secretion [70, 71, 72]. Recent work indicates that neurons and cancer cells communicate via paracrine signaling and electrochemical interactions. Neural activity can also reshape immune cell function and activate stromal cells within the TME, creating a multi‐layered nerve‐tumor network. Through these mechanisms, neurogenic signals contribute to malignancy by driving tumor initiation, promoting proliferation and invasion, facilitating metastasis, enhancing therapy resistance, stimulating pro‐tumor inflammation, and dampening anti‐tumor immunity [73]. In gastric cancer models, vagotomy or pharmacological vagal inhibition significantly reduces tumor incidence and progression [30]. In the present study, GPL mice showed elevated gastric levels of ChAT and 5‐HT, together with an increased proportion of cells co‐expressing 5‐HT and M3R. These results suggest that excess ACh acting on M3R may contribute to abnormal 5‐HT secretion, which could be one molecular mechanism underlying GPL development.

In summary, our data suggest that 5‐HT/5‐HTR1D signaling may promote TRPV4‐involved cation currents, contributing to the electrophysiological changes in gastric epithelial cells and the mucosal pathology and intestinal metaplasia observed in GPL models. Cholinergic neurons in the DMV may facilitate gastric 5‐HT secretion, potentially contributing to GPL development (Figure 7). Collectively, the findings uncover a neural–molecular pathway linking vagal cholinergic output to serotonergic signaling that may be involved in GPL progression and highlight 5‐HTR1D and TRPV4 as candidate targets for early intervention. However, several limitations should be noted. The mechanistic link between 5‐HTR1D and TRPV4 is supported by pharmacological inhibition and siRNA knockdown experiments, yet the direct gating mechanism coupling receptor activation to channel opening remains to be elucidated. The relationship between TRPV4 and AKT/mTOR signaling is currently correlative, and direct validation of the proposed Ca²⁺/CaMKK/AKT cascade awaits further investigation. Moreover, this study is based on the MNU‐induced mouse model, which does not fully reflect the multifactorial etiology of human GPL, including H. pylori infection. Whether the neuro‐epithelial interactions described here operate in human GPL requires validation using clinical specimens. Future studies addressing the molecular coupling between 5‐HTR1D and TRPV4, the Ca²⁺‐dependent signaling pathway linking TRPV4 to AKT/mTOR, the upstream circuits driving DMV cholinergic hyperactivity, and the clinical relevance of ACh/5‐HT signaling will be necessary to evaluate the translational potential of these findings.

FIGURE 7.

The graphical abstract of the proposed mechanism through which the central neural system may modulate the GPL process. Vagal cholinergic neurons may promote gastric 5‐HT secretion, which may contribute to gastric precancerous lesions through 5‐HTR1D and TRPV4, highlighting these as candidate targets for early intervention.

Author Contributions

Shengli Gao: conceptualization, methodology, investigation, formal analysis, supervision, funding acquisition, visualization, writing – original draft, writing – review and editing. Xufei Feng: methodology, visualization, writing – original draft. Nana Zhang: investigation, validation, formal analysis. Jia Su: methodology. Yuyan Wang: methodology. Linghan Kong: methodology. Jianing Zhou: methodology. Hui Dong: writing – review and editing. Feifei Guo: methodology, funding acquisition.

Funding

This work was supported by the Natural Science Foundation of Shandong Province, ZR2025MS458. Qingdao University Undergraduate Training Programs for Innovation and Entrepreneurship, X2025110650696.

Disclosure

Animal studies: All animal procedures, such as housing, experimentation, and disposal, were conducted in accordance with Qingdao University's Guidelines for the Care and Use of Laboratory Animals. The procedures were approved by the Animal Ethics Committee of Qingdao University's Medical College (Approval Document Number: NO. 2‐23‐5‐1Balb/c9620250125148).

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Sequences of primers for quantitative real‐time PCR.

Table S2: Sequences of TRPV4 small interfering RNA.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.