Tail clamping induces anxiety-like behaviors and visceral hypersensitivity in rat models of non-erosive reflux disease

Abstract

BACKGROUND

Non-erosive reflux disease (NERD), the main gastroesophageal reflux subtype, features reflux symptoms without mucosal damage. Anxiety links to visceral hypersensitivity in NERD, yet mechanisms and animal models are unclear.

AIM

To establish a translational NERD rat model with anxiety comorbidity via tail clamping and study corticotropin-releasing hormone (CRH)-mediated neuroimmune pathways in visceral hypersensitivity and esophageal injury.

METHODS

Sprague-Dawley (SD) and Wistar rats were grouped into sham, model, and modified groups (n = 10 each). The treatments for the modified groups were as follows: SD rats received ovalbumin/aluminum hydroxide suspension + acid perfusion ± tail clamping (40 minutes/day for 7 days), while Wistar rats received fructose water + tail clamping. Esophageal pathology, visceral sensitivity, and behavior were assessed. Serum CRH, calcitonin gene-related peptide (CGRP), 5-hydroxytryptamine (5-HT), and mast cell tryptase (MCT) and central amygdala (CeA) CRH mRNA were measured via ELISA and qRT-PCR.

RESULTS

Tail clamping induced anxiety, worsening visceral hypersensitivity (lower abdominal withdrawal reflex thresholds, P < 0.05) and esophageal injury (dilated intercellular spaces and mitochondrial edema). Both models showed raised serum CRH, CGRP, 5-HT, and MCT (P < 0.01) and CeA CRH mRNA expression (P < 0.01). Behavioral tests confirmed anxiety-like phenotypes. NERD-anxiety rats showed clinical-like symptom severity without erosion.

CONCLUSION

Tail clamping induces anxiety in NERD models, worsening visceral hypersensitivity via CRH neuroimmune dysregulation, offering a translational model and highlighting CRH as a treatment target.

Key Words: Non-erosive reflux disease; Anxiety and depression; Animal model; Tail-clamping; Corticotropin hormones

Core Tip: This study systematically compares two non-surgical modeling approaches, ovalbumin (OVA)/aluminum hydroxide suspension [Al(OH)₃] + acid perfusion vs fructose water + tail clamping, for non-erosive reflux disease (NERD). It demonstrates that a 7-day tail clamping protocol effectively induces anxiety-like behaviors and exacerbates visceral hypersensitivity through corticotropin-releasing hormone-mediated neuroimmune pathways. The modified OVA/Al(OH)₃ model combined with tail clamping can serve as an efficient platform for investigating brain-gut interactions in NERD with psychiatric comorbidity. This novel approach offers significant advantages over traditional chronic stress methods in terms of both induction efficiency (requiring only 40 minutes/day for 7 days) and pathological reproducibility.

INTRODUCTION

Non-erosive reflux disease (NERD), which is characterized by the presence of reflux in patients with heartburn symptoms and no endoscopically evident mucosal breaks of the esophageal mucosa, can account for up to 79.4% of cases of gastroesophageal reflux disease (GERD), and is the most common subtype of GERD[1,2]. Currently, the pathomechanism of NERD has not been fully elucidated; acid exposure, visceral hypersensitivity, and delayed gastric emptying are all thought to be associated with the development of NERD, and visceral hypersensitivity, in particular, is thought to be the most critical factor contributing to the symptoms of patients with NERD[3]. Anxiety and depression are often present in patients with NERD, and preclinical explorations have observed that patients with NERD and comorbid anxiety and depression have elevated serum mast cell tryptase (MCT) and calcitonin gene-related peptide (CGRP) levels (P < 0.05), suggesting the presence of neuroimmune dysregulation. In previous studies, anxiety and depression have been reported to cause sensitization of sensory nerve afferent fibers, exacerbate visceral sensitivity, and lead to more severe esophageal symptom reporting with decreased quality of life scores[2,4-6], and further investigation is needed to investigate if neuroimmune dysregulation is the focal point of mediating the exacerbation of esophageal sensitization by adverse emotions.

Constructing an ideal animal model is an important step in exploring the pathological mechanisms of the disease. Currently, the methods of constructing animal models of NERD can be broadly divided into three: The first one involved intraperitoneal injection of an ovalbumin (OVA) and aluminum hydroxide suspension [Al(OH)₃] adjuvant suspension combined with esophageal acid perfusion. This protocol was designed to induce visceral hypersensitivity and simulate an esophageal acid exposure environment in rats[7,8]. This model is predominantly established in Sprague-Dawley (SD) rats[7]. The second model combined fructose water administration with constrained water immersion. In this model, fructose intake created a gastric hyperacidic environment to simulate the risk factors of reflux, while constrained water immersion induced psychological stress, leading to inflammatory changes in the esophageal mucosa[9]. This model is primarily characterized in Wistar rats[9]. The third type employs incomplete pyloric ligation combined with antral ligation to synergistically induce reflux of gastric contents[10,11]. Compared with the first two methods, the third method is more demanding on the operator, such as controlling the consistency of the caliber of pyloric ligation and the degree of antral ligation in different rats, in addition to the fact that the surgical modeling will also disrupt the original gastrointestinal structure of rats, which is not consistent with the pathological features of NERD. Therefore, non-surgical methods are more frequently used[7]. However, whether there are differences between these two modeling approaches in terms of visceral hypersensitivity, emotional disturbances, or neuroimmune dysregulation have not been systematically compared.

Seven consecutive days of tail-clamp stimulation was found to provoke irritable behavior in rats with reflux esophagitis[12] and lead to neuroimmune dysfunction[13], demonstrating the efficacy of this method in inducing negative emotional states. Compared to prolonged 28-day restrained water immersion, this method is more straightforward and time-efficient. However, the precise central mechanisms by which such stress contributes to visceral hypersensitivity in NERD remain poorly understood. The central amygdala (CeA) is a key brain region regulating emotion, and corticotropin-releasing hormone (CRH) and its receptor CRH-R1 expressed within the CeA play a pivotal role in anxiety and visceral pain modulation[14]. Despite this implication, the specific role of a CRH-mediated neuroimmune pathway in NERD-related visceral hypersensitivity is not yet fully understood.

This study aimed to systematically compare the characteristics of the two aforementioned non-surgical modeling methods and innovatively introduce tail-clamp stimulation as a psychological stressor for methodological refinement. A comprehensive evaluation will be conducted on the behavioral aspects (visceral hypersensitivity, anxiety, and depression), serological profiles (MCT and CGRP levels), CRH pathway components, and histopathological changes in the esophageal tissue of model rats. The study not only provides direct experimental evidence for model selection and optimization in NERD research but also ultimately aims to establish a reliable animal platform that accurately mimics NERD and its emotional comorbidities. Utilizing this model, we seek to preliminarily validate the critical role of neuroimmune dysregulation, particularly involving the CRH pathway, in mediating the exacerbation of esophageal sensitivity by psychological stress, thereby laying a foundation for further mechanistic exploration and the discovery of therapeutic targets.

MATERIALS AND METHODS

Animals

Thirty male SD rats and thirty healthy male Wistar rats of specific pathogen free grade, weighing 230 ± 10 g, were purchased from Spearfish (Beijing) Biotechnology Co., Ltd, Laboratory Animal Production License No. SCXK (Beijing) 2024-0001, and housed in cages at the Animal Experimentation Center of Xiyuan Hospital, China Academy of Traditional Chinese Medicine, under a 12-hour/12-hour light/dark cycle, with the room temperature controlled at 22 ± 2 °C and relative humidity at 35% ± 5%. The human subject study was approved by the institutional review board of Xiyuan Hospital, China Academy of Chinese Medical Sciences (No. 2024XLA247-2). All animal procedures were approved by the same institution's Institutional Animal Care and Use Committee (No. 2024XLC101-2). All rats were acclimatized for 1 week before the start of the experiments.

Experimental design

At the end of acclimatization, both SD and Wistar rats were randomly and equally divided into control (SD-sham, Wistar-sham group, n = 10), model (SD-model, Wistar-model group, n = 10), and modified groups (SD-modified and Wistar-modified, n = 10). A priori power analysis was conducted using G*Power software (version 3.1.9.7; χ² test, α = 0.05, power = 0.8, effect size w = 0.97), which indicated that a minimum sample size of n = 4 per group would be sufficient. Nevertheless, a sample size of n = 10 was utilized to enhance the robustness of the study and account for potential technical challenges associated with CeA tissue sampling. The SD-modified group received OVA (100 mg) combined with Al(OH)₃ adjuvant (200 mg) suspension intraperitoneally injected 1.5 mL to induce visceral hypersensitivity in rats, and hydrochloric acid (0.1 mol/L, 10 mL/hour) esophageal perfusion for 50 minutes to simulate an acid-exposed environment[8]. The SD-modified group was superimposed with tail clamping (40 minutes/day, with the clamp repositioned every 20 minutes and a 5 minutes interval between sessions, for 7 days) to induce an anxious-depressive state in rats[12]. The SD-sham group received saline intraperitoneal injection and esophageal perfusion. The Wistar-modified group was given fructose in their drinking water (200 g/L for 28 days) to simulate the risk factors for reflux, and the rats were placed in restraint cages so that they could not turn back and forth, with the water level maintained at the xiphoid process (2 hours/day for 28 days) to induce psychological stress in rats[9]. In the Wistar-modified group, restraint water immersion for 28 days was replaced by 7 consecutive days of tail clamping to assess whether 7 days of tail clamping could produce the same psychological stress effect. The Wistar-sham group was given free access to drinking water without any treatment. The specific interventions and experimental procedures for different rat groups are illustrated in Figure 1.

Figure 1 Schematic diagram of the animal intervention. SD: Sprague-Dawley; OVA: Ovalbumin.

Data collection and sample collection

Throughout the modeling period, the rats were monitored daily for their general health and mental status. Assessments included posture, coat condition (e.g., fur color and grooming), spontaneous activity levels, responsiveness to mild stimuli, and signs of discomfort such as eye squinting. The occurrence of any abnormal deaths or injuries was also recorded. Body weights were measured and recorded on a weekly basis. Upon completion of the modeling protocol, dietary intake was assessed, and behavioral evaluations were conducted using the open field test (OFT) for anxiety-like behavior, the sucrose preference test (SPT) for anhedonia (a core depressive-like behavior), and the abdominal withdrawal reflex (AWR) scoring system to assess visceral hypersensitivity.

Key strategies for mitigating harm to animals

Several measures were implemented to minimize animal suffering throughout the study. Paraffin oil was used to lubricate both the esophageal perfusion catheter and the balloon of the Foley catheter to reduce discomfort during the perfusion and AWR scoring procedures. For the tail-clamp stimulation, the metal clip was wrapped with gauze to cushion the pressure and mitigate potential pain. The clamped site was treated promptly with povidone-iodine after each session to prevent infection, and the clamping position was alternated regularly between sessions. During the water restraint stress test, the water temperature was strictly maintained at 22 ± 2 °C to avoid hypothermia. Furthermore, adequate anesthesia was ensured by intraperitoneal administration of 1% sodium pentobarbital at a dose of 40 mg/kg body weight. This anesthetic protocol guaranteed that the rats experienced no pain during invasive procedures, including esophageal perfusion, blood collection, and tissue harvesting.

Behavioral tests

To comprehensively evaluate the emotional and sensory phenotypes of the established NERD models, a battery of behavioral tests was selected based on their well-validated efficacy in assessing specific domains in rodents. The SPT was employed to measure anhedonia, a core symptom of depression, which reflects a diminished responsiveness to reward[15]. The OFT was utilized to assess anxiety-like behaviors (e.g., reduced center area exploration) and general locomotor activity[16,17]. Visceral hypersensitivity, a hallmark feature of NERD, was quantitatively evaluated using the AWR score in response to colorectal distension; this metric directly reflects the nociceptive sensitivity of the visceral organs[18-20]. Together, these tests provide a multidimensional behavioral profile encompassing both the emotional comorbidities (anxiety/depression) and the peripheral pathophysiology (visceral hypersensitivity) central to NERD.

SPT

In the acclimatization phase, rats were simultaneously provided with two identical bottles containing 1% (w/v) sucrose solution per cage for 24 hours. Subsequently, one bottle of sucrose solution was replaced with plain water for another 24 hours. Following the acclimatization phase, the rats were food- and water-deprived for 12 hours. They were then simultaneously presented with one bottle of 1% (w/v) sucrose solution and one bottle of plain water. The volumes of sucrose solution and plain water consumed were recorded after 1 hour. The sucrose preference index was calculated using the following formula: [Sucrose solution consumption/(sucrose solution consumption + plain water consumption)] × 100%[15].

OFT

Rats were first allowed to acclimatize to the test room for at least 1 hour in a quiet environment. Subsequently, each rat was placed individually into the center of an open-field arena and allowed to move freely for 5 minutes while being video-recorded. The following parameters were analyzed: Total distance moved (cm), average speed (cm/second), time spent in the center zone (%), number of entries into the center zone, and immobility time (%). The arena was thoroughly cleaned with 75% ethanol and dried after each trial to eliminate olfactory cues. A decrease in total distance moved and average speed, and an increase in immobility time were interpreted as potential depression-like behaviors. A decrease in the time spent in the center zone and the number of center entries, coupled with an increase in time spent in the peripheral zone, were interpreted as potential anxiety-like behaviors[16,17]. All procedures were conducted within the Labmaze Animal Behavior Analysis System (V3.0; Zhongshidi Chuang, China).

AWR

The AWR score of rats in each group was assessed before and after the 7-day tail-clamping period. Prior to testing, rats were allowed to defecate spontaneously to avoid interference. Briefly, rats were gently restrained in a customized device that permitted limited forward and backward movement. A paraffin oil-lubricated Foley catheter (6-Fr; Huayue, China) was inserted into the rectum until the balloon was positioned 8 cm proximal to the anus[18,19]. The catheter was then secured by taping it to the base of the tail. The balloon was progressively inflated with water in increments of 0.2 mL, starting from 0.4 mL. Each volume was maintained for approximately 1 minute. The AWR was scored on a scale from 0 to 4: (1) 0 means no behavioral response; (2) 1 means brief head movement followed by immobility; (3) 2 means mild abdominal muscle contraction; (4) 3 means strong abdominal muscle contraction and lifting of the abdomen; and (5) 4 means body arching and lifting of the pelvis. The minimal distension volume required to consistently elicit an AWR score of 3 was defined as the pain threshold. This measurement was repeated three times for each rat at 5-minute intervals, and the average value was calculated to ensure reliability[20].

Serum collection and analysis

Rats were anesthetized via intraperitoneal injection of 1% pentobarbital sodium at a dose of 40 mg/kg. Upon achieving a surgical plane of anesthesia, the abdominal cavity was opened, and 4-5 mL of blood was collected from the abdominal aorta. The blood samples were centrifuged at 3000 × g for 15 minutes at 4 °C to separate the serum. The supernatant (serum) was aliquoted into microcentrifuge tubes and stored at -80 °C until further analysis. ELISA was performed to quantify serum levels of MCT[21], CGRP[22-24], 5-hydroxytryptamine (5-HT)[25], substance P (SP)[26], and CRH[27,28]. For the Wistar rats, blood glucose levels were measured using a glucose oxidase assay[29].

Esophagus tissue processing and analysis

Following blood collection from the abdominal aorta, the thorax was rapidly dissected. The right auricle was incised, and the rats were perfused transcardially first with 50 mL of phosphate-buffered saline (G4202-500ML; Servicebio, China) and then with 4% formaldehyde fixative (G1101-500ML; Servicebio, China) at a uniform rate until the liver turned pale, indicating successful perfusion[30,31] (this perfusion fixation was performed on n = 4 rats per group for histological analysis). The esophagus was then isolated. A 5-mm segment above the gastroesophageal junction was excised and fixed in formaldehyde for 48 hours for subsequent paraffin embedding, sectioning, and hematoxylin and eosin (H&E) staining. An adjacent 2-mm wide segment was placed in a pre-cooled 4 °C electron microscope fixative solution (G1102-1.5ML; Servicebio, China) for 24 hours for transmission electron microscopy to observe dilated intercellular spaces (DIS) and tight junctions[32-34]. For the remaining rats (not perfused with fixative), immediately after blood collection, the esophagus above the gastroesophageal junction was quickly dissected. The pH of the inner mucosal surface of the esophagus was immediately measured using a planar pH electrode (LabSen371; SanXin, China).

CeA collection and qRT-PCR

While measuring esophageal pH, the rat head was rapidly severed. The CeA was then collected according to “The Rat Brain in Stereotaxic Coordinates—Compact Third Edition” and immediately frozen at -80 °C[35]. Total RNA was extracted from CeA tissues using RNA extraction reagent and a chloroform substitute, precipitated with isopropanol, washed with 75% ethanol, and finally dissolved and quantified using RNA solubilization solution. Reverse transcription was performed using the SweScript All-in-One RT SuperMix (G3337; Servicebio, China), with 200 ng/μL RNA serving as the template for cDNA synthesis; qRT-PCR amplification was carried out using 2 × Universal Blue SYBR Green qPCR Master Mix (G3326; Servicebio, China), with primers synthesized by Wuhan Servicebio Technology Co., Ltd. The expression levels of CRH mRNA and CRH-R1 mRNA in the CeA were quantified via qRT-PCR[14], analyzed by the ΔΔCt method, and normalized to GAPDH as the internal reference[36].

Statistical analysis

Statistical analyses were performed using SPSS 26.0 software. Continuous data are expressed as the mean ± SD. Intergroup comparisons were conducted using the Student’s t-test for two groups and one-way analysis of variance for multiple groups, with the Least Significant Difference test applied for homogeneous variances. For data with heterogeneous variances (P < 0.05), the Kruskal-Wallis test followed by Tamhane’s T2 test was used. A P value less than 0.05 was considered statistically significant. The statistical methodology for this study was reviewed by a professional biomedical statistician.

RESULTS

No animals died during the modeling period. By the end of the procedure, rats in both SD-modified and Wistar-modified groups displayed withered fur, lethargy, reduced activity, and decreased consumption of food and water relative to their sham controls.

In SD rats, pre-clamping body weight change did not differ significantly between groups (P > 0.05). After tail-clamping, however, the SD-modified group showed a marked weight loss (P < 0.01; Figure 2A). In Wistar rats, the modified group showed a tendency toward higher weight gain before clamping compared to the sham group, although this difference was not statistically significant (P > 0.05). After clamping, the modified group exhibited a significant reduction in weight gain (P < 0.01; Figure 2B). At the end of modeling, both Wistar-model and Wistar-modified groups exhibited significantly elevated blood glucose (P < 0.05; Figure 2C) and lower esophageal pH (P < 0.05; Figure 2D) compared to the Wistar-sham group, indicating successful reflux induction via fructose water.

Figure 2 General condition comparisons of the animals. A: Comparison of body weight gain rate in Sprague-Dawley (SD) rats before and after tail clamping; B: Comparison of body weight gain rate in Wistar rats before and after tail clamping; C: Blood glucose levels in Wistar rats at the end of modeling; D: Esophageal pH values in Wistar rats at the end of modeling. ^aP < 0.01 vs post-SD-modified; ^bP < 0.05 vs pre-Wistar-model; ^cP < 0.01 vs pre-Wistar-modified; ^dP < 0.01 vs post-Wistar-sham; ^eP < 0.01 vs Wistar-sham. SD: Sprague-Dawley.

Tail clamping induces anhedonia-like behavior

As shown in Figure 3A, at the end of modeling, the sugar water preference index was significantly lower in the SD-modified group than in the SD-sham group and SD-model group (P < 0.01), and significantly lower in the Wistar-model and Wistar-modified groups than in the Wistar-sham group (P < 0.01). Sugar water preference index was slightly lower in the Wistar-model group than in the Wistar-modified group, with no statistical difference (P > 0.05).

Figure 3 Behavioral assessment of rats after model establishment. A: Sucrose preference test; B: Animal movement trajectories; C: Total distance traveled; D: Average velocity; E: Immobility time (%); F: Time in center zone; G: Number of center zone entries; H: Abdominal withdrawal reflex score. ^aP < 0.01 vs Sprague-Dawley (SD)-modified, ^bP < 0.01 vs Wistar-sham, ^cP < 0.05 vs SD-modified, ^dP < 0.05 vs Wistar-sham, ^eP < 0.01 vs pre-SD-sham, ^fP < 0.05 vs pre-SD-modified, ^gP < 0.05 vs pre-Wistar-model. SD: Sprague-Dawley; AWR: Abdominal withdrawal reflex.

Tail clamping induces anxiety-like behavior and decreases locomotion

As shown in the representative movement trajectories of Figure 3B, which illustrate locomotor patterns across groups, the SD-modified group exhibited a significant reduction in total distance traveled compared to both the SD-sham and SD-model groups (P < 0.01 and P < 0.05, respectively; Figure 3C), as well as a decrease in average speed (P < 0.01 for both comparisons; Figure 3D). Immobility time was significantly increased in the SD-modified group relative to the SD-sham and SD-model groups (P < 0.05 and P < 0.01, respectively; Figure 3E). Although the time spent in the central region (Figure 3F) and the number of entries into the central region (Figure 3G) were reduced, these differences did not reach statistical significance (P > 0.05).

Tail clamping lowers AWR thresholds, indicating worsened visceral hypersensitivity

Before tail clamping, AWR score in both the SD-model and SD-modified groups was significantly lower than that in the SD-sham group (P < 0.01), though there was no significant difference between the two SD experimental groups (P > 0.05). After tail clamping, the SD-modified group showed a significant reduction in AWR score compared to its pre-clamping baseline (P < 0.05; Figure 3H). In Wistar rats, the modified group exhibited a significantly lower AWR score relative to the model group before clamping (P < 0.05), and though AWR score decreased further after clamping, the change was not significant (P > 0.05; Figure 3H).

Esophageal mucosal barrier integrity is impaired following tail clamping

Upon gross observation, no obvious pathological changes were observed in the esophageal mucosa of rats across all groups. Histological examination by H&E staining revealed no significant inflammatory cell infiltration in the esophageal mucosal layer of any group (Figure 4A), though squamous cell hyperplasia was noted in both the SD-modified and Wistar-modified groups. Transmission electron microscopy (Figure 4B) demonstrated widening of DIS in the SD-model, SD-modified, Wistar-model, and Wistar-modified groups. Notably, disruption of intercellular tight junctions was more pronounced in the SD-modified and Wistar-modified groups compared to their respective model groups. Interestingly, while obvious mitochondrial edema was observed in the esophageal mucosa of the Wistar-modified group, the density of intercellular tight junctions appeared greater than that in the Wistar-model group.

Figure 4 Histopathological and ultrastructural changes in the esophageal mucosa. A: Hematoxylin and eosin staining of esophageal mucosa (scale bar: 50 μm); B: Esophageal mucosa under transmission electron microscopy (scale bar: 2 μm). SD: Sprague-Dawley.

Neuroimmune mediators are elevated in serum following tail clamping

As shown in Figure 5A-E, at the end of the modeling period, serum levels of MCT, SP, CRH, CGRP, and 5-HT were significantly increased in both the SD-modified and Wistar-modified groups compared to their respective sham groups (P < 0.01). Furthermore, the SD-modified group showed significantly increased serum CGRP and 5-HT levels compared to the SD-model group (P < 0.05 and P < 0.01, respectively). Similarly, serum MCT and 5-HT levels were significantly increased in the Wistar-modified group compared to the Wistar-model group (P < 0.05 and P < 0.01, respectively).

Figure 5 Multi-level comparison of neuroimmune metrics. A: Serum mast cell tryptase (MCT) levels in rats; B: Serum substance P (SP) levels in rats; C: Serum corticotropin-releasing hormone (CRH) levels in rats; D: Serum calcitonin gene-related peptide (CGRP) levels in rats; E: Serum 5-hydroxytryptamine (5-HT) levels in rats; F: Central amygdala (CeA) CRH mRNA levels in rats; G: CeA CRH-R1 mRNA levels in rats; H: Z-score normalized serum levels of MCT, SP, CRH, CGRP, and 5-HT in healthy volunteers (HV), non-erosive reflux disease patients, and experimental groups; I: MCT, SP, CRH, CGRP, and 5-HT, Δ% change of serum markers relative to HV. ^aP < 0.01 vs Sprague-Dawley (SD)-sham, ^bP < 0.01 vs Wistar-sham, ^cP < 0.01 vs Wistar-model, ^dP < 0.05 vs SD-sham, ^eP < 0.05 vs SD-model, ^fP < 0.01 vs SD-model, ^gP < 0.05 vs Wistar-sham, ^hP < 0.05 vs Wistar-model. SD: Sprague-Dawley; MCT: Mast cell tryptase; SP: Substance P; CRH: Corticotropin-releasing hormone; CGRP: Calcitonin gene-related peptide; 5-HT: 5-hydroxytryptamine; CeA: Central amygdala; NERD: Non-erosive reflux disease.

CRH signaling is activated in CeA following tail clamping

As shown in Figure 5F and G, CRH mRNA expression was significantly increased in the CeA of the SD-modified group compared to the SD-model group (P < 0.01). The expression of CRH-R1 mRNA was also increased in the SD-modified group, but the difference was not statistically significant compared to the SD-sham group (P > 0.05). In the Wistar-modified group, the expression of both CRH and CRH-R1 in the CeA was significantly increased compared to the Wistar-sham group (P < 0.01). Their expression was also higher than that in the Wistar-model group, but these differences were not statistically significant (P > 0.05).

Divergent neuroimmune profiles between acute rat models and chronic NERD patients

We examined the serum concentrations of MCT, SP, CRH, CGRP, and 5-HT in NERD patients with anxiety and depression. In contrast to the elevated levels of these mediators (particularly SP, CRH, and 5-HT) observed in the SD-modified and Wistar-modified rat models, the expression of serum SP, CRH, and 5-HT was significantly decreased in NERD patients compared to healthy volunteers (P < 0.05), as detailed in Figure 5H and I.

DISCUSSION

In this study, we first systematically compared the characteristics of two commonly used non-surgical NERD modeling methods [OVA/Al(OH)₃ intraperitoneal injection + esophageal acid infusion vs fructose water + constrained water immersion], and also evaluated the feasibility of combined tail-clamping stimulation for the induction of NERD with anxiety and depression in an animal model by combining the behavioral, serological, pathological, and neurobiological indices. A key novel finding is that a brief, 7-day tail-clamp stress protocol effectively induced anxiety-like behaviors and exacerbated core NERD pathophysiology, primarily through CRH-mediated neuroimmune mechanisms. This establishes a rapid and efficient modeling strategy that contrasts with traditional chronic stress protocols which require longer durations (e.g., 28 days). The results showed that this combined method significantly exacerbated visceral sensitization (reduced AWR score) and esophageal mucosal damage (widened DIS or mitochondrial edema) in the model rats, and this effect may be closely related to the CRH-mediated neuroimmune pathway.

Specifically, both the SD-modified group [OVA/Al(OH)₃ + clipped tail] and the Wistar-modified group (fructose water + clipped tail) exhibited significant elevation of serum CRH, CGRP, 5-HT, MCT, and SP, and the expression of CRH mRNA was significantly up-regulated in the CeA of the SD-modified group compared with that of the SD-model group, suggesting a central regulatory role of the CRH signaling pathway in the emotion-visceral axis. This coordinated upregulation suggests a plausible mechanistic cascade: Psychological stress (tail-clamp) activates CeA CRH neurons, leading to peripheral CRH release and subsequent mast cell activation (elevated MCT), which in turn sensitizes visceral afferent nerves, promoting the release of neuropeptides (CGRP and SP) and neurotransmitters (5-HT) that collectively drive visceral hypersensitivity and mucosal barrier disruption[3,14,37]. Our study provides a more integrated view of this brain-gut crosstalk in NERD. These findings are consistent with previous studies reporting that NERD is often accompanied by neuroimmune dysregulation[9,37]. However, serum CRH, SP, and 5-HT levels showed opposite expression trends in model rats compared to clinical samples, which is a crucial and novel observation. This divergence likely reflects the critical difference between the acute stress response captured in our rodent model (7-day tail clamp) and the chronic adaptive state in patients with long-standing NERD (usually ≥ 3 months)[38,39]. In acute stress, a rapid surge in these mediators is a typical neuroendocrine response, whereas chronic stress can lead to receptor downregulation, negative feedback inhibition, or metabolic exhaustion, resulting in a return to or even a drop below baseline levels[40]. This insight underscores the importance of considering disease chronicity when interpreting animal models and highlights that our acute-on-chronic model is particularly adept at capturing the initial, dynamic phase of neuroimmune activation, which might be most relevant for studying exacerbations or for targeted therapeutic interventions. For example, under acute stress, human cortisol levels rise along with a significant increase in CRH levels, whereas during chronic inflammatory stress, CRH levels decline back to near baseline[40]. CRH mRNA expression in the CeA was upregulated in the SD-modified group compared to the SD-model group, but did not differ significantly from the SD-sham group. The CeA is a key region in the central regulation of emotion, and CRH and its receptor CRH-R1 are involved in emotion regulation and pain perception in the CeA[41,42]. In previous studies, up-regulation of CRH expression in the CeA was reported to mediate anxiety-depression-like behaviors in rats with visceral hypersensitivity[14], and esophageal sensitization and central sensitization were also reported to be associated with negative emotions in GERD[43,44]. The significant upregulation in the SD-modified group, but not the SD-model group, strongly implies that the combination of peripheral sensitization (OVA/acid) and psychological stress (tail-clamping) is necessary to robustly engage this central emotional hub, leading to the amplified negative affective and sensory outcomes. The results of the present study suggest that visceral hypersensitivity may enhance the psychological stress response in rats, leading to increased dysregulation of central emotion regulation.

Another important finding of the present study is that the 7-day tail-clamping paradigm combined with intraperitoneal injection of OVA/Al(OH)₃ (SD-modified group) was superior in inducing anxiety-depression-like behaviors (decreased sugar-water preference and increased immobility time) and neuroimmune dysregulation (elevated serum CRH and 5-HT, and elevated CRH mRNA in the CeA). This establishes the OVA/Al(OH)₃ + tail-clamp protocol as a preferred method for efficiently generating a comprehensive NERD-anxiety phenotype. The Wistar-modified group (fructose water + tail-clamping) did not show significant superiority in behavioral, expression of CRH mRNA and CRH-R1 mRNA in the CeA, and esophageal pathological changes, although the changes in serological indexes were more obvious than those in the Wistar-model group. These results suggest that the intraperitoneal injection of OVA/Al(OH)₃ in concert with tail-clamping stimulation enhances the neuroimmune response, and is a more appropriate method to construct a rat model of NERD with anxiety and depression. In addition, the changes in CRH, SP, and 5-HT in the Wistar-model group (fructose water + restraint water immersion) were milder than those in the Wistar-modified group (7-day tail-clamping), suggesting that long-term chronic stress (e.g., clinical course) may be closer to the neuroimmune features of human NERD. However, the 28-day constrained water immersion model has a long cycle and high operational complexity, whereas our data demonstrate that tail-clamping stimulation can achieve similar or even stronger anxiety-depression and neuroimmune induction in only 7 days, which is a significant practical advantage for high-throughput drug screening and mechanistic studies where time efficiency is paramount[45].

The present study also has limitations, such as not fully mimicking the clinical chronic disease course. The difference between the acute tail-clamping stimulus (7 days) and the long-term stress of human NERD does not diminish the applicability of the model, as the acute stress still mimics the core pathology of NERD (visceral hypersensitivity and esophageal mucosal injury). Rather, it defines the model's specific utility: It is excellent for studying the exacerbation of NERD by acute psychological stress and for probing the initial, dynamic mechanisms of CRH-mediated neuroimmune crosstalk. The short-term paradigm is more suitable for pharmacological interventions, such as rapid screening of CRH receptor antagonists. The tail-clamping cycle can be further optimized if it is needed to be closer to the clinical chronic course of the disease (e.g., extended to 14-21 days). In addition, the lack of insight into downstream inflammatory mechanisms is also a limitation. Although tail-clamping induced elevated CRH mRNA in serum and the CeA in the model rats, how it regulates pro-inflammatory factors, such as IL-6 and TNF-α, remains unclear and needs to be further analyzed by combining with transcriptomics or proteomics to fully map the inflammatory cascade downstream of CRH activation.

CONCLUSION

Our study confirms that tail-clamping stimulation combined with intraperitoneal injection of OVA/Al(OH)₃ is a preferred regimen for generating a rat model of NERD with anxiety and depression because of its high efficiency in inducing neuroimmune dysregulation and esophageal damage, and its simplicity and short cycle time. This model successfully captures the interplay between psychological stress and visceral hypersensitivity via CRH pathway activation. Despite the differences between the acute model and the clinical chronic course, they are highly aligned in their core phenotypes, such as CRH pathway activation, visceral hypersensitivity, and mucosal barrier disruption. These findings provide an ideal platform for intervention studies targeting CRH (e.g., antagonist or neuromodulatory therapies). Future studies should explore the mechanism at multiple levels by extending the stress cycle or integrating multi-omics analysis to reveal the association between acute stress and chronic disease course, and refine the specific role of inflammatory pathways to further optimize the translational value of the model.