Address for Correspondence J. Wang, Department of Physiology and Pathophysiology, Shanghai Medical College of Fudan University, Shanghai, 200032, China. Tel: +86 21 54237609; fax: +86 21 54237405; e-mail: firstname.lastname@example.org
Background Excessive greater splanchnic nerve (GSN) activation contributes to the progression of gastric ischemia-reperfusion (GI-R) injury. This study was designed to investigate the protective mechanism of cerebellar fastigial nucleus (FN) stimulation against GI-R injury.
Methods The GI-R injury model was induced in rats by clamping the celiac artery for 30 min, and then reperfusion for 30 min, 1, 3, 6, or 24 h, respectively.
Key Results Microinjection of l-Glu (3, 6, 12 μg) into the FN dose-dependently attenuated GI-R injury and GSN activity. In addition, there was an enhancement of gastric mucosal blood flow in GI-R rats. Pretreatment with the glutamic acid decarboxylase antagonist into the FN, the GABAA receptor antagonist into the lateral hypothalamic area or lesion of superior cerebellar peduncle all reversed the protective effects of the FN stimulation. Furthermore, the FN stimulation reduced the TUNEL-positive gastric mucosal cell and Bax-positive gastric mucosal cell in GI-R rats.
Conclusions & Inferences These results indicate that the protective effects of the FN stimulation against GI-R injury may be mediated by attenuation of the excessive GSN activation, gastric mucosal cell apoptosis, and Bax expression in GI-R rats.
Many animal models of stress-related mucosal injury have been adopted to elucidate the pathogenesis of gastric mucosal injury.1 Ischemia, major surgery and peripheral tissue trauma-induced gastric mucosal injury, and gastrointestinal (GI) dysmotility are important clinical phenomena.2,3 Reperfusion of ischemic tissues aggravates the injury process resulting in the glial distortion, neuronal changes, and the release of free oxygen radicals, proinflammatory mediators.4–6 Based on these findings, a number of anti-inflammatory and antireactive oxygen species agents have been applied to prevent acute gastric mucosal injury in rats.7,8 However, little is known about the regulative effects of the central nervous system on the acute gastric mucosal injury and gastric motility. The previous studies9,10 showed that the paraventricular nucleus (PVN) is a specific hypothalamic nucleus that attenuate gastric ischemia-reperfusion (GI-R) injury, but little is known about the role of other brain areas on GI-R injury in rats.
Interestingly, the cerebellum, a traditional considered subcortical somatic motor center, also participates in some other non-somatic basic functions such as respiratory, cardiovascular, cognition, learning, immune, and micturition.11–16 The fastigial nucleus (FN) of cerebellum is the phylogenetically oldest nucleus, holds a key position in the ultimate outputs of the spinocerebellum, and has been found to regulate various GI activities.17 Although the underlying neural mechanisms are still understood poorly, a series of neuroanatomical studies has indicated that the bidirectional and direct cerebellar–hypothalamic circuits between cerebellum and hypothalamic nuclei may mediate the cerebellar non-somatic visceral functions.18 Moreover, the neurophysiologic studies demonstrated that the FN functionally impinges on the glycemia-sensitive neurons in the lateral hypothalamic area (LHA),19 which is one of the brain nuclei regulating the viscera activities. γ-aminobutyric acid (GABA), the primary inhibitory neurotransmitter in the mature brain, is not only excitatory, but also plays a more robust role in generating spike activity in hypothalamic neurons.20 Excitatory and inhibitory GABA synapses in the cerebellar are coexistence21 and the activation of glutamate receptors evoke an increase in the release of GABA from cerebellar stellate cells.22 Meanwhile, the previous study demonstrated that the LHA also participated in regulating GI tract activities.23 Based on the previous studies, in this study we focused on the possible neuronal circuit between cerebellar FN and the LHA, an autonomic structure involved in the protective effects on GI-R injury, and the important role of GABA, which is an inhibitory neurotransmitter shown to attenuate cold-restraint stress-induced gastric ulceration.24 Although the related study25 demonstrated the cerebellar fastigial nucleus protected gastric mucosal injury from restraint and water immersion because the pathogenesis between GI-R injury and water immersion-restraint gastric injury is completely different, so we attempt to investigate whether the FN stimulation protect gastric injury from ischemia-reperfusion as similar mechanism as the neuro-protective effects of the FN on water immersion-restraint gastric injury. Furthermore, we also investigate the antiapoptosis effect of the FN stimulation on mucosal cells induced by GI-R injury and the related cell signaling pathway such as mitochondrial apoptosis pathway.
According to the above reports, we surmise that the GABAergic neurons in the FN were activated via the FN stimulation, then released inhibitory neurotransmitter GABA, and the released GABA interact with GABAAR of the LHA through cerebellar–hypothalamic circuit. Finally, the inhibited neurons in the LHA attenuated the greater splanchnic nerve (GSN) activity. Meanwhile, we also surmise that the FN stimulation attenuates the gastric mucosal cell apoptosis, which may be mediated by mitochondrial antiapoptosis pathway by promoting the Bcl-2 overexpression in the gastric mucosa.
Materials and methods
Experiments were carried out with male Sprague–Dawley rats, weighing between 220 and 300 g (Fudan University Experimental Animal Center, Shanghai, China). The procedures were approved by the Experimental Animal Care and Use Committee of Fudan University and complied with the Guide for the Care and Use of Laboratory Animals (NIH Publication no. 85–23). All rats were housed under controlled conditions (12 h of light starting at 20 : 00; 22–24 °C) and given access to water ad libitum for the duration of the study. The animals were fasted for 24 h before the experiment and were allocated to the different groups for different purposes in the experiments. All efforts were made to minimize animal suffering and reduce the number of animals used. After experiments, animals were deeply anesthetized with chloral hydrate (10%, Solvent: 0.9% normal saline) and euthanized by cervical dislocation followed by decapitation.
l-Glu, 3-mercaptopropionic acid (3-MPA), Bicuculline methbromide were obtained from Biomol (Plymouth Meeting, PA, USA), kainic acid (KA) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Anti-Bcl-2 and anti-Bax were purchased from Abcam (San Francisco, CA, USA). Superoxide dismutase (SOD), malondialdehyde (MDA), xanthine oxidase (XOD), and Hydroxyl free radical (-OH) detection kit from Nanjing Jiancheng Bioengineering Institute (China).
Truncal vagotomy and sympathectomy
To investigate whether the protective effect of the FN stimulation on GI-R injury is mediated via a parasympathetic or sympathetic pathway, sympathectomy and truncal vagotomy were performed 1 week before FN stimulation. The rats were anesthetized with an intraperitoneal injection of chloral hydrate (400 mg kg−1). After a midline incision was made in the abdominal wall, the lower part of the esophagus was exposed, and anterior and posterior branches of the vagal nerve were incised above the hepatic and celiac branches. In sham-operated rats, the vagal trunks were similarly exposed. For sympathectomy, the roots of the celiac and superior mesenteric arteries were exposed, then the prevertebral ganglia among these arteries were completely removed. Sham-operated rats were subjected to an identical surgical procedure except for denervation. After surgery, the incisions were closed, and rats were treated with antibiotics (gentamicin, 4 × 104 U kg−1, i.p. once daily for 3 days) and returned to their home cage for a 1-week recovery.
Electrical stimulation of the FN
Prior to experiment, rats were anesthetized with chloral hydrate (400 mg kg−1), then fixed onto a stereotaxic apparatus. The coordinate for stimulating the FN, according to the Rat Brain in Stereotaxic Coordinates,26 was AP 11.6 mm, LR 1.0 mm, and H 5.6 mm.
The concentric bipolar stimulating electrode consisted of an insulated outer barrel from the needle of a 0.4-mm diameter medical syringe and an inner Ni-Cr wire, insulated except the tip, which protruded approximately 0.2 mm from the barrel. The stimulation electrode was inserted into the FN and fixed to the cranium by dental cement.
The FN stimulation was performed 30 min prior to GI-R. The parameters of the stimulation were as follows: uniphasic, constant current square wave pulses (pulse width 0.2 ms, intensity 0.4 mA, frequency 50 Hz, duration 1 min) repeated five times at 5-min intervals. In the sham FN-stimulation group, only the electrode was inserted and no current was passed.
Chemical stimulation of the FN
The coordinate for the FN treatment was the same as the above. l-Glu (3, 6, 12 μg in a volume of 0.3 μL saline for the FN)27 was microinjected 10 min before GI-R via a cannula connected to a microsyringe with polyethylene tube. The microinjection lasted for 2 min, and the microinjection cannula was left for an additional 10 min to prevent backflow.
Microinjection of 3-MPA and bicuculline methbromide
The rats were anesthetized with chloral hydrate (400 mg kg−1, i.p.), then fixed onto a stereotaxic apparatus. The locations of the FN and LHA, which were obtained according to The Rat Brain in Stereotaxic Coordinates,26 are FN: AP 11.6 mm, LR 1.0 mm, H 5.6 mm; LHA: AP 2.8 mm, LR: 1.5 mm, H: 8.3–8.5 mm. The incisor bar was positioned 3.3 mm below the center of the aural bars. The glutamic acid decarboxylase (GAD) inhibitor, 3-MPA (20 μg in a volume of 0.3 μL saline for cerebellar fastigial nucleus)28 and GABAA receptor antagonist, Bicuculline methbromide (10 μg in a volume of 0.3 μL saline for LHA)29 were microinjected via a cannula connected to a microsyringe with polyethylene tube. The injection lasted for 2 min, and the injection cannula was left for an additional 10 min to prevent backflow. l-Glu was microinjected into the FN 10 min before GI-R, and 3-MP was microinjected into the cerebellar fastigial nucleus 15 min before l-Glu microinjection. l-Glu was microinjected into the FN 10 min before GI-R and bicuculline methbromide was microinjected into the LHA 15 min before l-Glu microinjection. In the control groups, the vehicle (saline) was microjected into the FN or the LHA.
Electrolytic destruction of decussation of superior cerebellar peduncle (XSCP) and chemical lesion of the LHA or PVN
Three days before the experiment, the rats were anesthetized with chloral hydrate (400 mg kg−1, i.p.) and mounted onto a stereotaxic apparatus. The coordinate of XSCP for the placement of destructing electrodes was taken from the atlas of Paxinos and Watson as follows: AP: 7.4 mm, LR: 0 mm, H: 7.8–8.0 mm, and the incisor bar was positioned 3.3 mm below the center of the aural bars. The destructing electrode was a single stainless steel wire, 0.3 mm in diameter, with a naked tip about 0.3 mm, and void of insulation. Lesion was then made by passing a positive DC of 1 mA for 10 s. Sham operations were performed with the same surgical procedures, and no current was passed.
Three days before the experiment, ablations of LHA or PVN were produced by microjection of kainic acid (KA, 0.3 μg in a volume of 0.3 μL saline for the LHA or the PVN) (Blake DJ, et al., 1986), the coordinates of LHA and PVN were taken from the atlas of Paxinos and Watson as follows, LHA: AP 2.8 mm, LR 1.5 mm, H 8.3–8.5 mm; PVN: AP 1.5 mm, LR 0.4 mm, 7.8 mm. In the control groups, the vehicle (saline) was microinjected into the LHA or the PVN.
GI-R injury model
Gastric ischemia-reperfusion was performed following l-Glu microinjection according to our previously reported method.30 In brief, the abdominal cavity was cut open, then the celiac artery was carefully isolated from its adjacent tissues. The celiac artery was clamped with a small vascular clip for 30 min, and then reperfusion was established by removal of the clip for 30 min, 1, 3, 6, or 24 h.
Sham-operated animals underwent an identical surgical procedure except for celiac artery clamping. At the end of experiments, the rats were euthanized under deep anesthesia. The stomachs were rapidly removed and they were cut open along the greater curvature, then the gastric mucosa was carefully inspected for ulcers.
Assessment of gastric mucosal injury index
Gastric mucosal injury was measured as described by our previously reported method.29 The stomach was incised along the greater curvature and washed with phosphate-buffered saline. The gastric mucosal injury index (GMII), based on a cumulative-length scale, on which an individual lesion limited within the mucosal epithelium (including the pinpoint erosions, ulcers, and hemorrhagic spots) is scored according to its length. Scores are counted as follows: 1, for a lesion ≤1 mm; 2, for a lesion >1 and ≤2 mm; 3, for a lesion >2 and ≤3 mm. For lesions with a width >1 mm, the lesion score is doubled. The sum of the scores represents the gastric mucosal injury index. To avoid researcher bias, the gastric mucosal injury index was measured by a researcher who was blind to treatments.
Histological verification sites of the FN, LHA, and XSCP
At the end of the experimental period, the rats were sacrificed by cervical dislocation, and then the brain of the rats were removed and fixed in 10% formalin for 5 days. Frozen coronal serial sections (30 μm) were cut across the cerebellar, hypothalamus, mounted and stained with 1% neutral red to verify the microinjection sites and lesion sites under a light microscope. Only the data from rats whose injection or lesion sites were correct were used for analysis.
Measurement of GSN activity
The rat was anesthetized with chloral hydrate (400 mg kg−1 i.p.) and mounted onto a stereotaxic apparatus. The coordinates of FN was the same as the above. l-Glu (6 μg in a volume of 0.3 μL saline for the FN) was microinjected via a cannula connected to a microsyringe with polyethylene tube. The injection lasted for 2 min, and the injection cannula was left for an additional 10 min to prevent backflow. The central end of the GSN was placed on thin bipolar platinum electrodes. The nerve–electrode junction was fixed electrically insulated from surrounding tissues with a Vinyl Polysioxane Impression Material-auto-mixture. The rectified output from the amplifier was displayed using the PowerLab system to record the raw nerve discharge. Basal nerve activity (baseline) was determined by efferent GSN at the beginning of the experiment, and background noise was determined by nerve activity recorded at the end of the experiment. Nerve activity during the experiment was calculated by subtracting the background noise from the recorded value. The GSN activity response to the FN stimulation was expressed as the percentage change from the basal value.
Measurement of gastric mucosal blood flow (GMBF)
GMBF was measured with laser-Doppler flowmeter (LDF-2; Nankai University, Nankai, China). In brief, the rats were anesthetized with chloral hydrate (400 mg kg−1 i.p.), the abdomen was opened, the stomach was exposed and transected, and the gastric contents were slightly evacuated to the exterior through the cut (5 mm) made in the stomach. Next, the laser probe was placed 0.5 mm above and perpendicular to the mucosal surface to monitor GMBF displayed in mV (value of Doppler signal voltage) on the digital panel of the flowmeter. After GMBF was stable, four points were selected for measurement (one point for 1 min) and the average value was calculated and expressed as U/mV.
Bcl-2 and Bax immunohistochemistry staining
The fixed stomach were sliced into 5-μm thick sections and mounted on glass slides. The immunohistochemistry was performed with a PowerVision two-step immunohistochemistry detection kit. The sections were stained with 3,3′-diaminobenzidine (DAB), then counterstained using hematoxylin. The sections were examined with a microscope (Model IX71; Olympus, Tokyo, Japan). Gastric mucosal cells with brown granules visible in the cytoplasm or nucleus were considered positive. The number of positive cells per section was counted in 10 random lower power (×10) fields, and the percentage of positive cells (positive cells/total cells × 100%) was calculated. Three non-consecutive sections were selected from each specimen and those indexes were averaged.
Measurement of ROS generation in gastric mucosa
As an index of lipid peroxidation, the determination of the levels of MDA was carried out. The gastric mucosa was excised and homogenized in saline at 4 °C. The homogenate was centrifuged at 3000 g for 10 min and the supernatant was retained. The protein concentrations were detected by Coomassie brilliant blue protein assay. XOD was determined spectrophotometrically at 530 nm using a commercial XOD kit (Jiancheng Bioengineering Institute, Nanjing, China) and the XOD activity is expressed in Units g−1 of protein. SOD activity and –OH inhibitory ability were determined spectrophotometrically at 550 nm by the xanthine oxidase and the fenton reaction method, respectively. SOD activity and –OH inhibitory ability are expressed in Units mg−1 of protein.
Immunohistochemical assay of gastric mucosal apoptosis and proliferation
Gastric mucosal cell apoptosis was visualized with ApopTag® Peroxidase In Situ Apoptosis Detection Kit (Chemicon International Inc., Temecula, CA, USA). PCNA was used as a marker for gastric mucosal cell proliferation, and the immunohistochemical staining was performed with PowerVisionTM two-step immunohistochemistry detection kit (Zhongshan Golden Bridge Biotech Co., Beijing, China). The sections were stained with DAB, counterstained using hematoxylin, and then analyzed under an inverted research microscope (Model IX71; Olympus, Tokyo, Japan). Cells with brown granules visible in the cytoplasm or nucleus were considered positive. The number of positive cells per section was counted in ten random high-power (×40) or lower power (×10) fields, and the percentage of positive cells (positive cells/total cells × 100%) was calculated. Three inconsecutive sections were selected from each specimen and those indexes were averaged.
Measurement of gastric acid and gastrin secretion in gastric mucosa
Basal gastric acid secretion was collected every 15 min for 60 min in GI-R rats. In FN+GI-R group, gastric acid secretion was collected in plastic tubes by gravity every 15 min, and at the end of each period, the stomach was washed with 2 mL of distilled water and the washing was collected. The concentrations of gastric acid were measured by titrating the collected sample plus the gastric washings with 0.01 N NaOH using an automatic titrator with a PH meter.
In a separate study, the rat stomach was removed immediately, kept on ice, and opened along the greater curvature. The antrum was dissected and frozen on dry ice. Each sample was homogenized, diluted 1 : 5 with distilled water, and boiled in a water bath for 30 min. These samples were centrifuged and the supernatant was saved at −20 °C for measuring concentrations of gastrin. The tissue gastrin was measured by radioimmunoassay.
Data were analyzed using one-way anova and the t-test, and are presented as mean ± SD. Prism software was used (GraphPad Software, San Diego, CA, USA), and statistical significance was defined as P < 0.05.
Histological verification of injection sites
All injection sites of FN, LHA, and XSCP were verified histologically, using the stereotaxic atlas of Paxinos and Watson26 (Fig. 1). The data from animals with a lesion or microinjection in an erroneous location were excluded from the analysis.
Effect of FN stimulation on gastric injury
At first, we sought to determine whether FN stimulation exerts its protective effect on GI-R injury. We found that unilateral microinjection of l-Glu (3, 6, or 12 μg in a volume of 0.3 μL saline) into the FN dose-dependently attenuated GI-R injury (Fig. 2). The GMII was significantly reduced from 120.4 ± 20.34 to 39.4 ± 19.79 (P < 0.05) after 6 μg of l-Glu microinjection, and to 36.8 ± 13.37 (P < 0.05) after 12 μg injection. The results indicated that 6 μg of l-Glu was the optimal gastria protective dose, and this dose was used in the following experiments. Electrical stimulation (ES) of the FN at 0.4 mA resulted in a decrease in GMII from 120.4 ± 20.34 to 39.20 ± 16.51.
Effects of FN stimulation in a rat model of ischemic gastria at different time points after reperfusion
Next, we investigated the protective effects of FN stimulation in a rat model of ischemic gastria at different points after reperfusion and the FN stimulation was performed prior to GI-R. We found that 30 min temporary celiac occlusion results in gastric mucosal damage consisted of both ischemia and reperfusion components. The structure of gastric mucosa was partly damaged and was diminished by GI-R until 6 h after reperfusion, but was almost recovered 24 h after reperfusion. However, a time difference was shown about the recovery of gastric mucosal injury index (Fig. 3).
The LHA, XSCP lesion, and sympathectomy reversed the protective effect of FN stimulation but the PVN lesion and vagotomy did not
The FN+GI-R group was taken as the control, the GMII of three groups (FN+GI-R+LHA lesion, FN+GI-R+XSCP lesion, and FN+GI-R+ sympathectomy) increased (P < 0.05). Compared with FN+GI-R group, the GMII of two groups (FN+GI-R+PVN lesion and FN+GI-R+vagotomy) did not increase (P > 0.05). These results indicate that the LHA and the peripheral sympathetic nerve rather than PVN and vagotomy mediating the protective effects of FN stimulation on GI-R injury (Fig. 4). How did the cerebellar–hypothalamic circuit participate in this regulative progress?
The GAD antagonist and GABAA receptor antagonist reversed the protective effects of FN stimulation on GI-R injury in rats
To determine if GABA-mediated neurotransmission is a major way to regulate the effects of FN stimulation, GABA synthetic enzyme GAD antagonist 3-MPA (20 μg in a volume of 0.3 μL saline) was microinjected into the FN before FN stimulation and GI-R. The GABAAR antagonist, bicuculline methbromide was microinjected into LHA before FN stimulation and GI-R. Compared with FN+GI-R group, the GMII of two groups (FN+GI-R + 3-MPA and FN+GI-R+bicucullin) increased obviously (P < 0.05) (Fig. 5). These results suggested that GABAergic neurons projected from the FN to the LHA mediated the effects of FN stimulation.
Effect of FN stimulation, XSCP lesion, GAD antagonist, and GABAA receptor antagonist on the change in GMBF induced by GI-R
The FN+GI-R group was taken as the control, the GMBF of four groups (Vehicle, FN+GI-R + 3-MPA, FN+GI-R+XSCP lesion, and FN+GI-R+bicucullin) decrease immediately for the initial 6 h. In the FN-stimulation group, gastric mucosal blood flow showed a significant improvement and this improvement was maintained for the initial 6 h (P < 0.05) (Fig. 6).
Effect of FN stimulation on change in GSN activity induced by GI-R
Fig. 7 shows examples of responses of GSN to GI-R injury and FN stimulation. Mean data demonstrating an increase in GSN activity in response to the GI-R injury is shown in Fig. 7A. In addition, the GSN activity obviously decreases in FN+GI-R group compared with GI-R group (P < 0.05, Fig. 7B).
Effect of FN stimulation on the changes in ROS generation in gastric mucosa
Xanthine oxidase and MDA are regarded as indexes to mucosal injury from ROS. The low activity of XOD and the scarcity of MDA were detected in normal mucosa. The MDA content and the XOD activity obviously increased in GI-R group compared with the normal group and FN+GI-R group (P < 0.05). As Table 1 shows, the SOD activity in Sham group remained at a high level whereas it is significantly decreased in the GI-R group. Compared with GI-R group, the FN stimulation enhanced SOD activity. 3-mercaptopropionic acid microinjection into the FN and bicuculline microinjection into the LHA both attenuated SOD activity as compared with the FN+GI-R group (P < 0.05). The pattern of the change in –OH inhibitory ability was similar to that of the SOD activity in the corresponding groups.
Table 1. Effect of fastigial nucleus (FN) stimulation on the changes in ROS generation in gastric mucosa
Immunohistochemical assay of quantitative variations in apoptotic and proliferative cells in the gastric mucosa
As shown in Fig. 8A,B, compared with sham group, the percentage of apoptotic cells of the GI-R group obviously increased (P < 0.05). The FN+GI-R group was taken as the control, the percentage of apoptotic cells of three groups (FN+GI-R + 3-MPA, FN+GI-R+XSCP lesion, and FN+GI-R+bicucullin) increases (P < 0.05). However, as shown in Fig. 8C,D there was no significant difference among four groups (FN+GI-R, FN+GI-R + 3-MPA, FN+GI-R+XSCP, and FN+GI-R+bicucullin) on mucosal cells proliferation percentage. These results indicate that the protective effects of FN stimulation on GI-R injury may be mediated by antiapoptotic pathway but not promoting cell proliferative pathway.
Effect of FN stimulation on gastric secretion in GI-R rats
As Table 2 shows, compared with GI-R group, the gastric acid output and antral gastrin concentration significantly decrease in the FN+GI-R group (P < 0.05).
Table 2. Effect of fastigial nucleus (FN) stimulation on gastric secretion in gastric ischemia-reperfusion (GI-R) rats
*P < 0.05, compared with the GI-R group.
Maximal gastric acid output (μEq 15 min−1)
6.31 ± 0.35
3.93 ± 0.26*
Antral gastrin concentration (μg g−1)
7.03 ± 0.78
4.26 ± 0.23*
Immunohistochemical assay of protein expression of Bcl-2 and Bax in gastric mucosa
Fig. 9A,C shows the Bcl-2 and Bax protein levels in the gastric mucosa in different groups of the study. As shown in Fig. 9B, the expression of Bcl-2 protein among the four groups (GI-R, FN+GI-R + 3-MPA, FN+GI-R+XSCP lesion, and FN+GI-R+bicucullin) was significantly lower than that of the two groups (Sham and FN+GI-R) (P < 0.05). Compared with Sham and FN+GI-R groups, the Bax protein level increased in the four groups (GI-R, FN+GI-R + 3-MPA, FN+GI-R+XSCP lesion, and FN+GI-R+bicucullin) (Fig. 9D, P < 0.05).
The macroscopic photography of gastric tissue in different groups
To clarify the protective effects of FN stimulation on GI-R injury, representative macroscopic photograph was showed in Fig. 10.
The present study demonstrates that microinjection of l-Glu into the FN attenuates GI-R injury in a dose-dependent manner and the electrical stimulation of the FN also attenuates GI-R injury, these results indicated that the neurons in the FN but not the crossing nerve fibers were involved in the regulative effects of FN on GI-R injury. Furthermore, the effects of the FN stimulation may be mediated by hypothalamic LHA but not PVN, which was supported by the evidence that LHA lesion abolished the protective effect of FN stimulation but PVN lesion could not. These results support the notion that the FN, which has been implicated in regulating gastric motility, may be a responsive site of the brain involved in maintaining the integrity of gastric mucosa.31,32
Evidence is accumulating that hypothalamic nucleus may be intrinsically involved in integrating information from the peripheral nerve and cerebellar nucleus.33 Cerebellar–hypothalamic circuit, the direct projections from the cerebellar nuclei to the hypothalamus has been revealed by using anterograde and retrograde tracing techniques.19 The cerebellohypothalamic projections, which arised from all cerebellar nuclei, ascendingly project to the hypothalamus via the superior cerebellar peduncle. Moreover, the neurons in LHA and PVN were inhibited by cerebellar FN stimulation.34 As yet, the detailed mechanism is not clear. GABA, which is the chief inhibitory neurotransmitter, plays an important effect in regulating neuronal excitability throughout the nervous system. Growing evidences35 support the notion that GABA and glycine were considered as major inhibitory neurotransmitter to hyperpolarize target neurons. Immunohistochemical labeling of glycine and GABA in the cerebellum, spinal cord, and brainstem36,37 revealed the coexistence of these two amino acids in neuronal somata and boutons. Our present study demonstrated that the effect of FN stimulation on LHA neurons activities may be mediated by neurotransmitter GABA. The possible detailed mechanism is that chemical stimulation activates neuron of the FN to release GABA, which reacts with GABAA receptor in the LHA. These results were supported by the fact that the GABA synthetic enzyme GAD antagonist 3-MPA microinjected into FN abolished the protective effects of FN stimulation. The evidence that the GABAA receptor antagonist bicuculline methbromide microinjected into LHA abolished the protective effects of FN stimulation also supported the above results.
Importantly, our findings indicate that cerebellar–hypothalamic circuit was involved in the regulative effects of FN on GI-R injury, which were supported by the our results that XSCP lesion reversed the protective effects. Meanwhile, these results further verify that the GABA mediated the effect of FN via cerebellar–hypothalamic circuit. Haines and his colleagues reported that nuclei of the hypothalamus directly projecting to the cerebellum are widespread. Neurons that project directly to the cerebellum are found primarily in the lateral (LHA), posterior (PHA), and dorsal (DHA) hypothalamic areas; the dorsomedial (DMN) and ventromedial (VMN) nuclei; and in the periventricular zone (PVZ).38
How do LHA neurons regulate gastric responses? Neuroanatomy has indicated that LHA links to the parasympathetic and sympathetic nerves.39 The previous study enunciated that the autonomic nervous system participated in regulation of mucosal function against GI-R injury.9 Moreover, GABAAR on the LHA neurons play an important role in autonomic regulation of gastric function in the rat.40 Although related studies demonstrated that the vagal nervous systems mediate electrical stimulation of the PVN in regulating the GI-R injury,9 our present study found that the protective effect of FN stimulation against GI-R injury was mediated by sympathetic nerve but not the vagal pathways. This difference could be due to the complicated regulation of gastric mucosal function by autonomic nervous system and also due to the different central circuits involved in regulating sympathetic-vagal descending pathways.
It is well known that inhibiting sympathetic activity ameliorate organ ischemia-reperfusion injury and is associated with cellular apoptosis.41,42 We consider that the sympathetic inhibition caused by FN stimulation partially contributes to the improvement of gastric ischemia-reperfusion injury and gastric mucosal cell apoptosis. In this study, we found that FN stimulation attenuated the GSN activity and enhanced GMBF. These results suggest that sympathetic outflow might directly mediate the GI-R injury or GI motor activity by contracting or relaxing gastric vasculature or inhibiting the acetylcholine release, which are consistent with the previous study.43,44 In addition, we found that FN stimulation attenuates the gastric mucosal cell apoptosis induced by ischemia-reperfusion and promote Bcl-2 protein expression in gastric mucosa, which is supported by the findings that Bcl-2 overexpression attenuated gastric injury.45 It is worthwhile to note that 3-MPA, XSCP lesion, and bicuculline had no significant effect on FN stimulation of promoting gastric mucosal cell prolieration, indicating that the improvement of GI-R injury was independent of the gastric mucosal cell prolieration. These results also elucidate that the antiapoptosis effect of FN stimulation on GI-R injury may be mediated by mitochondrial antiapoptosis pathway, which promote the Bcl-2 overexpression in gastric mucosa. Previous relevant studies have demonstrated that the SOD, MDA, ROS, and XOD are crucial in the protection on organ injuries,45,46 which was consistent with one of our above findings (as shown in Table 1) that the FN stimulation had antioxidative effects against GI-R injury. Meanwhile, our results indicate that the protective effects of FN stimulation are associated with a decrease in gastric acid output and antral gastrin concentration, which was in accord with previous report.47
In conclusion, the FN stimulation attenuates GI-R injury by neuroendocrine pathway and intracellular signal transduction pathway. Firstly, the effects of FN stimulation may be mediated by central nerve pathway (LHA) and peripheral sympathetic pathway, and the possible regulative mechanism is that the FN stimulation trigger neurons to release inhibitory neurotransmitter GABA, which react with GABAA receptor on the LHA via cerebellar–hypothalamic circuit, then decreased the activity of GSN, then improved the gastric mucosal blood flow to alleviate the gastric ischemia-reperfusion injury in rats (Fig. 11). Secondly, the protective effects of FN stimulation on GI-R injury may be mediated by intracellular signaling pathway such as mitochondrial antiapoptosis pathway.
This work was supported by the National Natural Science Foundation of China (No. 31100838) and the Mingdao Program of Fudan University (NO.MDJH2012001).
The authors have declared that no competing interests exist.
DSD, JW and DNZ contributed to the design of the manuscript and analyzed the data. The first and the last two authors wrote the final versions of the manuscript and decided in consultation with the other authors to submit the manuscript for publication. All authors substantially contributed to the design of the study, interpretation of the data, and the writing of the manuscript. All authors vouch for the completeness and accuracy of the data.