A total of 52 adult male Sprague Dawley rats (weighing 250–350 g) were used in the study. They were housed under normal laboratory conditions of 22°C and 12 h dark/light cycle, and were given food and water ad libitum. All animal protocols were approved by the Animals Care and Use Committee of Huazhong University of Science and Technology (Wuhan, China). The rats were obtained from the Experimental Center of Tongji Medical College (Wuhan, China). They were randomly divided into two groups: The control group, which involved incisions of the abdomen; and the subdiaphragmatic vagotomy group. These groups were further divided into subgroups. The control group consisted of early control group (ECN, 7 days, n=6) and terminal control group (TCN, 56 days, n=6). The experimental group contained the early subdiaphragmatic vagotomy group (ESDV, 7 days, n=20) and the terminal subdiaphragmatic vagotomy group (TSDV, 56 days, n=20). Finally, the ESDV and TSDV were divided into three groups, which were either treated with sham gastric electrical stimulation (n=6), short-term SGES (30 min/day, 7 days, n=7) or long-term SGES (30 min/day, 21 days, n=7), respectively.

Following fasting the subdiaphragmatic vagotomized rats for 24 h, the rats were anesthetized with an intraperitoneal injection of 1% pentobarbital sodium (40 mg/kg; Boster Biological Technology, Ltd., Wuhan, China). Under aseptic conditions, the stomach and subdiaphragmatic esophagus were exposed. Ventral and dorsal subdiaphagmatic vagi were cut completely, in addition to surrounding mesenteries. Simultaneously, a pair of temporary cardiac pacing wires (Medtronic, Dublin, Ireland) was placed on the serosal surface of the stomach. The pair was mounted in the middle of the greater curvature for stimulation. The implanted stimulating electrodes travelled separately from the subcutaneous layer of the backside to the top of the head. Penicillin (Boster Biological Technology, Ltd.) was used locally to prevent infection. The muscular layer was stimulated following one week of recovery.

SGES was performed as previously described by Yang et al (14). Briefly, each stimulation consisted of a long pulse (300 msec, 4 mA) followed with five short pulses (0.33 msec, 100 Hz, 4 mA). The stimulation was synchronized with the peak of gastric intrinsic slow waves.

The authors used a modified gastric emptying model, as described previously (15). Phenol red (0.5 mg/ml) and carboxymethylcellulose (15 mg/ml) were thoroughly mixed as test meal. At the scheduled time, the rats were fasted for 24 h before 2 ml phenol red solution was administered to their stomachs. The stomachs were removed with the gastroesophageal junction and the pylorus harvested after 30 min. Gastric content was rinsed in physiological saline up to 20 ml; then 20 ml NaOH (0.5 M) was added. The solution was mixed and allowed to stand for 1 h at room temperature. Following this, 5 ml supernatant was removed and placed in a centrifuge tube with 0.5 ml trichloroacetic acid (20%, w/v). Centrifugation (1,050 × g, 4°C, 10 min) separated the phenol red solution (supernatant) from solution containing 5 ml supernatant and 0.5 ml trichloroacetic acid. The absorption value of phenol red was identified using a spectrophotometer at 560 nm. Meanwhile, 2 ml phenol red solution, 18 ml physiological saline, 20 ml NaOH (0.5 mol/l) and 4 ml trichloroacetic acid (20%, w/v) represented a standard sample. Absorption value of the standard sample was measured as above. The gastric emptying rate was obtained by: 1-(phenol red absorption value from the stomach of animals sacrificed 30 min after the test meal/standard sample absorption value).

RT-qPCR was used to measure the expression levels of the S100B/glial fibrillary acidic protein (GFAP) gene. RNA was isolated from stomach tissue using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Single-stranded cDNA was synthesized with PrimeScript™ RT Master Mix (Takara Biotechnology Co., Ltd., Dalian, China). Primer sequences were as follows: S100B, forward 5′-GAGCAGGAAGTGGTGGACAAA-3′ and reverse 5′-CACTCCCCATCCCCATCTT-3′; GFAP, forward 5′-TGACCGCTTTGCTAGCTACATC-3′ and reverse 5′-GCGCCTTGTTTTGCTGTTC-3′; and GAPDH, forward 5′-GTATGACTCTACCCACGGCAAGT-3′ and reverse 5′-TTCCCGTTGATGACCAGCTT-3′. GAPDH acted as an internal control. A total of 10 µl PCR reaction volume was used, and included: 0.5 µl upstream primer (Invitrogen; Thermo Fisher, Scientific, Inc.), 0.5 µl downstream primer (Invitrogen; Thermo Fisher Scientific, Inc.), 5 µl SYBR-Green (Qiagen GmbH, Hilden, Germany), 1 µl cDNA, 3 µl ddH₂O. All PCR reactions following the established procedure were quantified using the ABI-OneStep real-time system (Applied Biosystems; Thermo Fisher Scientific, Inc.). Gene relative expression levels were calculated with the 2^−∆∆Ct method (16).

Gastric tissues were double-labeled for enteric glial markers, S100B and GFAP protein. The acquired tissues were fixed in 4% paraformaldehyde for 6–24 h and cut into small segments of ~5 mm. The segments were then embedded into paraffin in a vacuum and sliced into at a thickness of 5 µm. Sections were boiled 1–2 min in citrate buffer for antigen retrieval after dewaxed in xylene and hydrated in the graded ethanol solutions. Following this, nonspecific binding sites were blocked by 5% bovine serum albumin (Boster Biological Technology, Ltd.) for 30 min at room temperature. The sections were then incubated with the following primary antibodies: Rabbit anti-S100B (1:300; cat. no. sc-136061; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and goat anti-GFAP (1:300; cat. no. ab53554; Abcam, Cambridge, MA, USA). Each section was incubated at 4°C overnight. Sections were washed 3 times in PBS following re-warming to room temperature for 60 min the following day. Sections were then incubated with secondary antibodies for Alexa Fluor 488-donkey anti-rabbit IgG (1:100 dilution; AntGene Biotech Co., Ltd, Wuhan, China) and AlexaFluor 594-donkey anti-goat IgG (1:200 dilution; AntGene Biotech Co., Ltd.) and kept in the dark at 37°C for 90 min. Following washing 3 times in PBS, sections were treated with Hoechst stain (1:1,500) at 37°C for 10 min. Sections were sealed with fluorescence quenching agent (Boster Biological Technology, Ltd.) and observed using laser scanning confocal microscopy (Nikon Corporation, Tokyo, Japan). Image analysis was performed using Image Pro5.1 (Media Cybernetics, Inc., Rockville, MD, USA). Quantification was reported as mean signal per field, from 6 random fields per sample.

The specimens of the antrum were fixed in 2.5% glutaraldehyde (pH 7.4) for 2 h at 4°C and rinsed with 0.1 M PBS twice and immersed in 1% O_sO₄ (pH 7.4) for 1 h. Then they were dehydrated with graded alcohol, embedded in Epon (Boster Biological Technology, Ltd.), cut into ultra-thin sections with an ultramicrotome (Leica Microsystems GmbH, Wetzlar, Germany), stained with lead citrate for 10 min. Finally, these sections were viewed using a transmission electron microscope (Tecnai G2 12, FEI; Thermo Fisher Scientific, Inc.).

The mean value was used for statistical analysis and the results were presented as mean ± standard error. One-way analysis of variance, followed by Bonferroni post hoc test, was performed to evaluate the difference between normal control groups and the subgroups of subdiaphragamtic vagotomy. P<0.05 was considered to indicate a statistically significant difference. All calculations were performed using SPSS software (version 17.0; SPSS, Inc., Chicago, IL, USA).

Gastric emptying rates in different subdiaphragmatic vagotomy groups are presented in Fig. 1. In the early groups, ESDV demonstrated a faster gastric emptying when compared with the ECN group (P<0.05). In the terminal groups, gastric emptying was decreased in TSDV, when compared with TCN (P<0.05) and long-term SGES significantly improved delayed gastric emptying compared with TSDV (P<0.05). In addition, long-term SGES was more effective in accelerating delayed gastric emptying than short-term SGES (P<0.05).

The mRNA expression of S100B and GFAP was measured by RT-qPCR (Fig. 2). During the course of subdiaphragmatic vagotomy, expression levels of S100B and GFAP were decreased gradually. However, mRNA expression of S100B and GFAP decreased significantly in the TSDV group compared with the TCN group (P<0.05). In the terminal groups, long-term SGES increased mRNA expression of S100B and GFAP and expression levels were higher compared with short-term SGES (P<0.05).

S100B and GFAP proteins were evaluated by double-labeling immunofluorescence (Fig. 3A). There was no significant change between the early groups (P>0.05; Fig. 3B). S100B and GFAP protein expression was decreased gradually in the course of subdiaphragmatic vagotomy (Figs. 3B). The protein expression decreased significantly in the TSDV group, when compared with those of the TCN group (P<0.05; Fig. 3B). Long-term SGES significantly upregulated S100B and GFAP protein expression (P<0.05) and was more effective at S100B and GFAP than short-term SGES. (P<0.05; Figs. 3B).

The ultrastructure of EGC in the myenteric plexus was demonstrated by transmission electron microscopy (Fig. 4). There were abundant cell organelles containing mitochondria, filaments and the smooth and rough endoplasmic reticulum in the cytoplast of EGCs in the ECN and TCN groups (Fig. 4). The damaged ultrastructural features involved swelling of mitochondria, dilation of the endoplasmic reticulum, decreased numbers of filaments and condensed chromatin, and were identified in the ESDV and TSDV groups (Fig. 4) and the latter was marked than the former. In the short-term or long-term SGES, the numbers of mitochondria and filaments increased (Fig. 4). Unexpectedly, an apoptotic enteric neuron was observed in the TSDV group (Fig. 4).