Diabetes mellitus (DM) is one of the most important metabolic diseases of people (1, 2). Type 1 or insulin-dependent diabetes mellitus (IDDM) typically in childhood is the result of a frank deficiency of insulin (3). It is due to destruction of pancreatic ß cells, most likely the result of autoimmunity to one or more components of those cells (4). In clinical, many of the acute effects of this disease can be controlled by insulin replacement therapy (5, 6). However, the diabetic patient who accepted with regular insulin treatment was attached to an often overlooked clinical problem of disordered gastrointestinal (GI) motility (7-9).

Insulin, a well-known hormone, is a popular remedy for DM patients in clinical (10, 11). Some GI problems of DM are mitigated due to the insulin treatment (12, 13). However, no studies have examined the effects of peripherally administered insulin on expression of insulin receptors (IRs) and cyclooxygenase in GI motility in DM rats.

Cyclooxygenase (COX) is an enzyme to convert arachidonic acid (AA, an -6 polyunsaturated fatty acid) physiologically to prostaglandin H2 (PGH2) (14, 15). COX-2, an isoenzyme of COX, is abundant in activated macrophages and other cells at sites of inflammation (16, 17). Although prostaglandin E (PGE) levels are increased by COX-2 during inflammation (18), COX-2 is undetectable in most normal tissues (19, 20). Both up-regulation and vascular smooth muscle contractile hyperreactivity of COX-2 has been demonstrated in spontaneous diabetic db/db mice (21). But the relationship between COX-2 and GI tract is unknown. Non-steroidal anti-inflammatory drugs (NSAIDs) treatment was usually used in clinical (22). There were many GI risks in patients who used NSAIDs constantly (23). NSAIDs can inhibit COX enzyme selectively or non-selectively. It is interesting to see the effects of insulin on GI emptying under NSAID supplement in DM rats.

On the other hand, in neurological mechanism, the parasympathetic nervous system (PSNS) is a division of the autonomic nervous system (ANS) (24). The PSNS uses acetylcholine (ACh) as its neurotransmitter, but other peptides (such as cholecystokinin) may act on the PSNS as a neurotransmitter (25, 26). The ACh acts on two types of receptors, the muscarinic and nicotinic cholinergic receptors (27). Cholinergic-neural system has been shown to be involved in the regulation of blood glucose (28). Many anti-cholinergic agents used in clinical have been found to change GI functions (29). However, the roles and the action mechanisms of muscarinic system on the GI motility are still unclear. Muscarinic acetylcholine receptors, M1 receptor and M3 receptor, are predominantly found to bind to G proteins of class Gq (30). M1 receptor is Gq (Gi /Gs) -coupled and found to affect the secretion from salivary glands and stomach (31, 32). M3 receptor is Gq-coupled and mediates an increase in intracellular calcium, it typically causes constriction of smooth muscle (33).

Disordered GI motility is an often overlooked clinical problem (8). Delayed gastric emptying of solid and/or liquid meal in patients with both type 1 and type 2 DM occurs in approximately 50% of these patients (34). Delayed gastric emptying is very common in patients with DM (8) and it has no direct correlation to blood sugar control, duration of the disease, and upper gastrointestinal symptoms (35). It has been well-known that the hyperglycemia induced by streptozotocin (STZ) inhibits both gastric emptying (36) and GI transit in rats, but the effect was reversed by supplement of insulin (12, 13). Cholecystokinin (CCK), a GI related peptides, is released mainly from duodenum. It has been shown that CCK inhibits the gastric emptying (37) and plays a key role in the GI tract. Therefore , it is interesting to find out the relationship between DM and GI and to see if it is via the mechanism of CCK secretion.

On the other hand, either gastric emptying or cholinergic receptors were related to the smooth muscle contraction (38, 39). PGE2 secretion affects the smooth muscle contraction which was mediated by the COX-1 and COX-2 activation (40). Moreover, COX-2 stimulates smooth muscle contraction in GI tract especially (41). So the western blot analysis was performed to observe the COX-2 expression and direct connection to the smooth muscle contraction in GI tract.

In the present study, we first aimed to examine the effects of insulin on gastric emptying and the nervous action on cholinergic M1 and M3 receptors along with plasma CCK secretion in normal and DM rats. Second, we investigated the changes in contraction of intestinal smooth muscle induced by insulin in DM rats. Finally, we examined the mechanism about expression of IRs and COX-2, and PGE2 production in GI tract under the insulin treatment and the close association with gastric emptying and smooth muscle contraction in DM status.


MATERIALS AND METHODS
Animals Male Sprague-Dawley rats weighing 250-350 g were housed in a temperature (22 ± 1°C) and light (6 a.m. -8 p.m.) controlled environment. Tap water and rat chow were given ad libitum. Animal protocols were approved by the Institutional Animal Care and Use Committee of National Yang-Ming University. All animals received humane care in compliance with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals, published by the National Science Council, Taiwan, R.O.C.

Diabetes induction

Diabetic hyperglycemia was induced by the intravenous injection of the tail vein with freshly prepared STZ (32 mg/kg, Sigma, St. Louis, MO, U.S.A) solution in saline/ 0.01 M citrate buffer (pH 4.5). The onset of DM was confirmed by the rapid appearance of polyuria, weight loss, and glycosuria (Combur-Test U, Boehringer Mannheim, Mannheim, Germany).

Experimental designs

In the Experiments 1-4, rats were divided into 2-6 groups for the experiment of gastric emptying.

Experiment 1. Dose effects of insulin on gastric emptying in normal male rats. Rats were divided into six groups and fasted for 20 h before use. Rats in the first group were injected i.p. with saline. Rats in other groups were injected i.p. with insulin (0-10 IU/kg).

Experiment 2. Interaction between insulin and atropine on gastric emptying in normal and diabetic male rats. Normal and diabetic rats were divided into four groups each and fasted for 20 h before use. On the experiment day, rats were injected i.p. with saline, insulin (0.25 IU/kg), atropine (5 mg/kg) or insulin plus atropine, respectively. Diabetes was induced by the intravenous injection of the tail vein with freshly prepared STZ (32 mg/kg). Some STZ-induced diabetic rats were divided into four groups and fasted for 20 h before use. Finally, the plasma glucose and (CCK) was measured by RIA after decapitation.

Experiment 3. Role of muscarinic receptors in the effects of insulin on GI motility in normal and diabetic male rats. The normal or DM rats were divided into six groups and fasted for 20 h before use. Diabetes was induced by the intravenous injection of the tail vein with freshly prepared STZ (32 mg/kg). Either normal or STZ-induced diabetic rats were injected i.p. with saline, insulin (0.25 IU/kg), pirenzepine (a M1 receptor antagonist, 15 mg/kg), 4-DAMP (a M3 receptor antagonist, 3 mg/kg), insulin plus pirenzepine, or insulin plus 4-DAMP, respectively.

Experiment 4. Role of NSAIDs in the effects of insulin on GI motility in normal and diabetic male rats. The normal or DM rats were divided into eight groups and fasted for 20 h before use. Diabetes was induced by the intravenous injection of the tail vein with freshly prepared STZ (32 mg/kg). Either normal or STZ-induced diabetic rats were injected i.p. with saline, insulin (0.25 IU/kg), indomethacin (the COX general inhibitor, 40 mg/kg), insulin plus indomethacin, respectively.

Experiment 5. Effects of insulin on expression of IRs on stomach in normal and diabetic male rats. The normal or DM rats were injected i.p. with saline or insulin (0.25 IU/kg) once per day for 3 days. Rats were fasted for 20 h before use. On the experimental day, rats were decapitated. Proteins from stomach were extracted by lysis buffer. Protein of IRs was analyzed by the analysis of Western blot.

Experiment 6. Effects of insulin on expression of IRs in stomach of normal, diabetic and hyperglycemic male rats. The normal rats were divided into two groups. One group was injected i.p. with 20% dextrose (2g/kg) once per day for 3 days and another was i.p. with saline for vehicle. The group 3 was DM rats those were injected i.p. with saline for three days. All rats were fasted for 20 h before use. On the experimental day, rats were decapitated. Proteins from stomach and colon were extracted by lysis buffer. Protein of IRs was analyzed by the analysis of Western blot.

Experiment 7. Effects of insulin on expression of COX-2 enzyme in stomach of normal and diabetic male rats. The normal or DM rats were injected i.p. with saline or insulin (0.25 IU/kg) once per day for 3 days. Rats were fasted for 20 h before use. On the experimental day, rats were decapitated. Proteins from stomach were extracted by lysis buffer. Protein of COX-2 enzyme was analyzed by the analysis of Western blot.

Experiment 8. Effects of insulin on the stomach PGE2 concentration in normal and diabetic male rats. The normal or DM rats were injected i.p. with saline or insulin (0.25 IU/kg) once per day for 3 days. Rats were fasted for 20 h before use. On the experimental day, rats were decapitated. The plasma and tissue samples of stomach were collected and acidified by addition of 2 M HCl to pH of 3.5 for PGE2 EIA.

Experiment 9. Effects of PGE2 on stomach smooth muscle contraction in normal and diabetic male rats. The normal or DM rats were injected i.p. with saline or insulin (0.25 IU/kg) once per day for 3 days. Rats were fasted for 20 h before use. On the experimental day, rats were decapitated. The segments of the stomach were quickly removed for experiment of spontaneous contractile activity of smooth muscle strips.

Measurement of gastric emptying

All animals were used after a 20 h fast. On the day of experiment, animals were received i.p. injection of drugs. Fifteen min later, all rats were orally ingested with radioactive Na251CrO4 containing 10 % charcoal via a PE-205 tubing directly into the stomach. Fifteen min after administration of the liquid meal, rats were decapitated. The small intestine was divided equally into ten segments. The radioactivities in the stomach and 10 segments of small intestine were counted by an automatic gamma counter (1470 Wizard, Pharmacia, Turku, Finland). Gastric emptying was determined by measuring the amount of radiolabeled chromium contained in the small intestine as a percentage of the initial amount received.

Measurement of spontaneous contractile activity of smooth muscle strips

Rats were fasted for 20 h and divided into 4 groups. Two groups of them were DM groups (induced by STZ, 32 mg/kg). They were individually received i.p. injection of normal saline (1 ml/kg) or/and insulin (0.25 IU/kg) 30 min before decapitation. Stomach tissues were quickly removed and cultured in Kreb’s solution. The stomach was collected along the mesentery after decapitation. Muscle strips which parallel to the longitudinal fibers were cut into small pieces (3×7 mm) and the mucosa on each strip was removed gently. The muscle strips were suspended in a thermostatically controlled (37°C) tissue chamber containing 5 ml Kreb’s solution and bubbled continuously with 95% O2 and 5% CO2. The composition (in mM) of the Kreb’s solution included NaCl 119, KCl 4.75, KH2PO4 1.2, NaHCO3 25, MgSO4 1.5, CaCl2 2.5 and glucose 11. One end of the strip was fixed to a hook at the bottom of the chamber and another end was connected to an external isometric force transducer. After being stabilized for 30 min, PGE2 dose-response curves were constructed by applying different concentrations (10-6~10-5M) at 5-min intervals. Spontaneous contractile activity of muscle strips (under a initial tension of 1 g) was simultaneously recorded by the PowerLab data acquisition system with Chart software (ADInstruments).

Processing of plasma for measurements of blood glucose, plasma CCK and PGE2 concentrations

The concentration of blood glucose was an indicator of the stable experimental design. On experimental days, rat blood samples were measured for blood glucose (Accu-Chek Advantage II, Mannheim, Germany) immediately and then collected and mixed with EDTA (1 mg/ml of blood) plus aprotinin (500 kiu/ml of blood) after decapitation. Plasma was immediately prepared by centrifugation at 1000 x g for 30 min at 4°C and used for measurement of plasma CCK and PGE2 concentrations.

CCK radioimmunoassay (RIA)

The plasma samples were acidified with an equal volumn of 1% trifluoroacetic acid (TFA) and then centrifuged at 2600 x g for 20 min at 4°C. The SEP-PAK C18 cartridge (Waters Associates, Milford, MA, U.S.A.) was equilibrated with 60% acetonitrile in 1% TFA (1 ml), followed by 1% TFA (3 ml, three times), and then the supernatant from the treated plasma sample was applied. After being washed with 1% TFA (3 ml, twice), the peptide (bound material) was slowly eluted with 3 ml of 60% acetonitrile in 1% TFA. The eluant was collected, lyophilized in a Speed Vac concentrator (Salvant Instruments, Farmingdale, NY, U.S.A.), then stored at -80°C and reconstituted with the appropriate assay buffer before measurement by radioimmunoassay (RIA). The CCK concentration in the extracted sample was measured by RIA using a rabbit anti-CCK antiserum supplied by Dr. K. Y. Francis Pau (Irvine, CA, U.S.A.) and 3H-CCK purchased from Amersham International Plc. In this RIA system, a known amount of unlabeled CCK in a total volume of 0.3 ml of 0.1% gelatin-PBS was incubated at 4°C for 24 h with 100 µl of anti-CCK antiserum, 1: 2,000 dilution in normal rabbit serum and 100 µl of [3H]CCK (~8,000 cpm). Two hundred µl of anti-rabbit gamma-globulin (ARGG) was then added, and incubation continued at 4°C for 24 h. The assay tubes were then centrifuged at 1,000 x g for 20 min. The pellet was dissolved in 400 µl of 1N NaOH, and 80 µl of 5 N HCl was added. The sample was mixed with 3 ml of liquid scintillation fluid, and the radioactivity counted in an automatic gamma counter (Wallac 1409, Pharmacia, Turku, Finland).

Western blot (immunoblotting)

All animals were used after a 20 h fast. On the experimental day, rats were decapitated. Proteins from stomach were extracted by lysis buffer and analyzed by the analysis of Western blot.

The proteins (20 µg each) were separated by 12% SDS-PAGE and then transferred onto polyvinylidenedifluoride (PVDF) membranes. Membranes were blocked with 5% nonfat milk then incubated with primary antibodies (of IR: Santacluz sc-57342 ; COX-2: Santacluz sc-1745; GAPDH: Santacluz sc-32233). Membranes were washed four times with TBS-T and then incubated with secondary antibody (Santacluz). Finally, immunoreactive bands were detected by chemiluminescence.

PGE2 enzyme immunoassay (EIA)

The tissue samples were lysised by lysis buffer in advance. The lysised tissue and plasma samples were acidified by addition of 2 M HCl to pH of 3.5 and at 4°C for 15 min. Samples were centrifuged at 10000 xg for 2 min to remove precipitates. The C18 reverse phase column was prepared by washing with 10 ml of ethanol followed by 10 ml of deionized water. The sample was applied under a slight positive pressure to obtain a flow rate of about 0.5 ml/ min. The colomn was washed with 10 ml of water, followed by 10 ml of 15% ethanol, and finally 10 ml hexane. The sample was eluted from the column by addition of 10 ml ethyl acetate and analyzed immediately by PGE2 EIA kit.

Statistical analysis

The data were expressed as mean ± S.E.M. The treatment means were tested for homogeneity using one-way analysis of variance (ANOVA), and the significance of any difference between means tested using Duncan’s multiple range test. A difference between two means was considered to be statistically significant when P was less than 0.05.


RESULTS
Effects of insulin on gastric emptying and its correlation with blood glucose and muscarinic system in normal and DM rats

The fasted normal range of blood sugar was between 80-100 mg/dl. The level of blood glucose was decreased (about 50 mg/dl) by the i.p. injection of insulin (0.25 IU/kg) and rose markedly (almost 400 mg/dl) after intravenous injection of the tail vein with freshly prepared STZ (32 mg/kg). Insulin suppressed the high level blood glucose in DM rats and returned the value to the levels with no difference from normal range (data not shown). Fig. 1 shows the dose effects of acute administration of insulin on gastric emptying in male rats. Insulin at doses of 0.25, 0.5, 1, 5 and 10 mg/kg significantly (P<0.05) increased gastric emptying comparing with controls (Fig. 1). Administration of atropine (5 mg/kg) in vivo maintained a normal range of blood glucose level (Fig. 2A, the middle panel), but restored the higher level of gastric emptying induced by insulin back to the control level (Fig. 2A, the upper panel). However, insulin did not change the level of plasma CCK in response to atropine, although restored the lower gastric emptying to control levels (Fig. 2A, the lower panel). Administration of STZ (32 mg/kg) in rats inhibited gastric emptying, and insulin significantly restored (P<0.05) the inhibition of gastric emptying. Atropine significantly inhibited (P<0.05) the gastric emptying in DM rats. However, treatment of insulin combined with atropine significantly restored the inhibition of gastric emptying in DM model (Fig. 2B, upper panels). On the other hand, insulin treatment in STZ rats significantly reduced (P<0.05) the higher plasma CCK concentration in DM rats (Fig. 2B, lower panel). Treatment with M1 and M3 receptor antagonists, pirenzepine (15 mg/ml/kg) and 4-DAMP (3 mg/ml/kg), respectively, decreased the insulin-induced up-regulation of gastric emptying (P<0.01) (Fig. 3A). Either M1 or M3 receptor antagonist reduced the effects of insulin in normal and DM rats (Fig. 3A and B).

Fig. 1. Effects of acute administration of insulin on rat gastric emptying: Insulin at doses of 0.25, 0.5, 1, 5 and 10 mg/kg significantly increased gastric emptying comparing with controls (P<0.05 as compared to control level).

Fig. 2. Effects of atropine on the insulin-enhanced gastric emptying in normal and DM rats: Administration of atropine (5 mg/kg) in vivo maintained a normal range of blood glucose level (Fig. 2A, the middle panel), but inhibited the gastric emptying (P<0.01) and reduced the higher level of gastric emptying induced by insulin back to the control level (Fig. 2A, the upper panel, P<0.05). Administration of STZ (32 mg/kg) in rats inhibited gastric emptying, and insulin significantly restored (P<0.05) the inhibition of gastric emptying. Atropine also significantly inhibited (P<0.05) the gastric emptying in DM rats. Treatment of insulin combined with atropine significantly restored the inhibition of gastric emptying in DM model (Fig. 2B, upper panels). Atropine did not affect the blood glucose in DM rats (Fig. 2B, the middle panel). Atropine increased plasma CCK in normal rats but insulin significantly reduced (P<0.05) the higher plasma CCK concentration in DM rats (Fig. 2A and B, lower panel). *, ** P<0.05 and P<0.01 as compared to the control group respectively. +, P<0.05 as compared with insulin-treated group. #, P<0.05 as compared with atropine-treated group.

Fig. 3. Effects of muscarnic M1/M3 receptor antagonists on the insulin-stimulated gastric emptying in normal and DM rats: Treatment with M1 and M3 receptor antagonists, pirenzepine (15 mg/kg) and 4-DAMP (3 mg/kg), respectively decreased the insulin-induced up-regulation of gastric emptying in normal (panel A) and DM rats (panel B). *, ** P<0.05 and P<0.01 as compared to the control levels. +, P<0.05 as compared to insulin-treated levels.

Effects of DM and insulin on the expression of IRs and COX-2 enzymes in rat stomach

Expression of IRS was decreased significantly in the stomach of DM rats (Fig. 4A). Administration of insulin increased the expression of IRs in stomach in normal rats (Fig. 4A). However, insulin reversed the decreased expression of IRs in DM rats. COX-2 enzyme expression was decreased significantly (P<0.05) in the stomach of DM rats (Fig. 5A). Administration of insulin increased the expression of stomach COX-2 in both normal and DM rats (Fig. 5A and B, P<0.05).

Fig. 4. Effects of DM and insulin on the expression of IRs in rat stomach: Expression of IRs was decreased significantly in the stomach of DM rats (panel A, P<0.05). Administration of insulin increased the expression of IRs in stomach in normal rats (panel A, P<0.05). However, insulin reversed the decreased expression of IRs in DM rats. Effects of DM and hyperglycemia on the expression of IRs enzymes in rat stomach: The expression of IR in antrum was decreased in DM rats (P<0.05) but maintain the original level at 20% dextrose treatment (panel B). Protein of IRs was analyzed by the analysis of western blot. * P<0.05 as compared to the normal group.

Fig. 5. Effects of DM and insulin on the expression of COX-2 enzymes in rat stomach: COX-2 enzyme expression was decreased significantly in the stomach of DM rats (panel A, P<0.05). Administration of insulin increased the expression of stomach COX-2 in both normal and DM rats (panel A, P<0.05). Administration of insulin increased the expression of COX-2 in both normal and DM rats (panel A, P<0.05). Protein of COX-2 was analyzed by the analysis of western blot. * P<0.05 as compared to the normal group. Effects of COX inhibitor NSAID, indomethacin, on gastric emptying treated with insulin in normal and DM rats: The gastric emptying decreased in DM rats and reversed by insulin. Administration of indomethacin, the stimulation and reversal of insulin on gastric emptying in DM rats were attenuated, respectively. (panel B, P<0.05).

Effects of DM and hyperglycemia on the expression of IRs enzymes in rat stomach

The IRs expression in antrum was decreased in DM rats (P<0.05) but unchanged at 20% dextrose treatment as compared to normal rats (Fig. 4B).

Effects of COX inhibitor NSAID, indomethacin, on gastric emptying treated with insulin in normal and DM rats

In Fig. 5B, insulin at doses of 0.25 mg/kg significantly (P<0.05) increased gastric emptying comparing with controls. Administration of STZ (32 mg/kg) in rats inhibited gastric emptying, and insulin significantly restored (P<0.05) the inhibition of gastric emptying. When supplement with indomethacin, the gastric emptying was not significant changed. However, indomethacin strongly decreased the effect of insulin on gastric emptying in either normal (P<0.05) or DM (P<0.05) rats.

Effects of DM and insulin on stomach and plasma PGE2 concentrations

In rat stomach, the concentration of PGE2 was increased by the administration of insulin. In DM rats, the concentrations of stomach PGE2 was significantly reduced (P<0.05), but restored by insulin replacement (Fig. 6). The concentration of plasma PGE2 was not altered in the DM rats as compared to normal rats (data not shown). Insulin did not affect the level of plasma PGE2 in normal rats, but enhanced (P<0.05) that in DM rats (data not shown).

Fig. 6. Effects of DM and insulin on stomach PGE2 concentration: The concentration of PGE2 was increased by the administration of insulin in stomach. In DM rats, the concentrations of stomach PGE2 was significantly reduced (*P<0.05), but restored by insulin replacement.* P<0.05 as compared to the normal group).

Effects of PGE2 on the contractile activity of rat stomach smooth muscle

The contraction of stomach smooth muscle stimulated by PGE2 was increased in normal rats. However, the action was exhibited strongly when pretreated with insulin, especially in 10-5M (P<0.01). On the other hand, the effect of PGE2 on stomach smooth muscle contraction in DM rats was not a significant stimulation. But the stimulatory effect of PGE2 was restored when pretreated with insulin plus DM (Fig. 7).

Fig. 7. Effects of PGE2 on the contractile activity of rat stomach smooth muscle: The contractions of stomach smooth muscles stimulated by PGE2 in normal and DM rats. However, the action was exhibited strongly when pretreated with insulin. *P<0.05 as compared to the control group at corresponding level of PGE2.

DISCUSSION
Diabetes is a popular disease in which the body does not produce or properly use insulin. In clinical, diabetic patients usually suffer from GI disorders including nausea, vomiting, bloating, and fullness about dysfunction of gastrointestinal tract (42, 43). Many pregnancy women also have the type 1 diabetes related to the disorder of GI hormone secretion and gene expression (44, 45). However, when blood glucose returns to normal with insulin supplements, the patients have a brief respite in most symptoms above. The relationship between insulin and GI tract in diabetes is an interesting topic. The gastric emptying might be decreased by hyperglycemia and increased by replacement of insulin. The present results showed that administration of insulin (0.25~10 IU/kg) dose-dependently increased gastric emptying from 56% to 98%.

In autonomic nervous system, the PSNS secrets ACh which acts on the cholinergic receptors (46). The important site of muscarinic receptors has been found in endothelial cells of blood vessels (47). It has been known the cholinergic receptors mediated smooth muscle contraction (48). The muscarinic receptors can be blocked by atropine. We also showed that administration of atropine (a peripheral muscarinic receptor antagonist) decreased the gastric emptying strongly. Our study partly showed that administration of atropine in normal male rats increased the levels of plasma CCK but decreased the gastric emptying. It proved that the effects of cholinergic receptors were related to the GI tract mostly. When treatment of insulin plus atropine, their restored effects on gastric emptying mean that insulin affect GI motility via an action on the muscarinic receptors. However, insulin did not change concentration of plasma CCK. We suggested that insulin did not affect CCK secretion in itself. In STZ-DM rats, the gastric emptying was decreased. Insulin reversed the effect of STZ on gastric emptying and atropine decreased the restored action of insulin in DM rats. On the other hand, insulin decreased the stimulation of STZ on concentration of plasma CCK to normal range. Therefore, we suggested that insulin restored gastric emptying at least partly via some association with cholinergic receptors. Insulin attenuated the plasma CCK levels to normal range in DM rats but atropine maintained the higher secretion of CCK in either normal or DM rats. The difference between the phenomena meant that the reverse of insulin on gastric emptying in DM rats partly depended on the plasma CCK secretion, but atropine did not alter the CCK production by itself. It meant that the block of cholinergic receptor did not affect the CCK secretion levels under the insulin treatment in DM rats.

Five subtypes of muscarinic receptors have been determined including M1, M2, M3, M4, and M5 receptors. M1 receptor was found to affect the secretion from salivary glands and stomach (31, 32). M3 receptor was mainly responsible for smooth muscle contraction and increased endocrine-exocrine gland secretions in stomach (33). M2 receptor was abundant in heart (46). M4 and M5 receptors were unclear yet mainly. Some studies have shown that effects on inflammation and proliferation may be associated with M1 and M3 receptors (49). Muscarinic receptor activation by the release of ACh from vagal nerves thus mainly leads to release of gastrin and inhibition of somatostatin release, which together with the direct muscarinic effects on the parietal and chief cells to increase gastric acid production (49). In the present study, pirenzepine (a M1 receptor antagonist, 15 mg/kg) and 4-DAMP (a M3 receptor antagonist, 3 mg/kg) were used to repeat the measurement of gastric emptying. Our data showed that both pirenzepine and 4-DAMP were similar to the atropine to reduce gastric emptying. However, some studies have shown that when pirenzepine at 15 mg/kg inhibited gastric acid secretion in M1-receptor knockout mice (50). However, activation of M1 receptors evoked a gastric smooth muscle relaxation via a NO-mediated mechanism (49). Gastric acid secretion, one mechanism of gastrointestinal motility, was not correlated completely with gastric emptying (51, 52). It may be concerned with pharmacology (53), endocrinology (54) or neurology (55). Administration of insulin plus pirenzepine or 4-DAMP, the returned effects of gastric emptying were reconstructed. It means that the insulin affects GI gastric emptying via an action depending mainly on the muscarinic M1 and M3 receptors, but it still exists other pathways between insulin and gastric emptying. In DM rats, the stimulatory effects of insulin on gastric emptying were also reduced by the administration of pirenzepine and 4-DAMP. Therefore, we suggested that insulin restored gastric emptying at least in part by a mechanism associated with the muscarinic M1 and M3 receptors in diabetes.

Although insulin is thought of as a popular hormone, it is difficult to keep balance between blood glucose control and side effects. However, no studies have examined the effects of insulin on distribution and expression of IRs in GI tract of DM rats. In the present study, we further observed the strong expression of IRs under insulin treatment in stomach. Insulin affected the expression of IRs in stomach of DM rats. We believed the down-regulation of IRs expression in diabetic rats was caused by STZ treatment. In Fig. 2B the middle panel, the group 2 column showed that when treatment with insulin at dose 0.25 IU/kg in DM rats, the blood glucose level was low down but not to the normal range. Moreover, the blood glucose levels stimulated by dextrose were called hyperglycemia. The blood glucose value was very high (>300 mg/dl more) in either DM or hyperglycemic rats. In DM rats, the bcells were damaged by STZ treatment to lack production of insulin. Relatively, insulin secretion was stimulated in hyperglycemia. However, the expression of IRs in antrum was decreased in DM rats but not in the hyperglycemic rats. It suggested that the decreased IR expression in DM rats was mainly due to the deficiency of insulin caused by STZ administration.

COX is a kind of enzyme that can convert AA to PGH2 and control the symptoms of inflammation and pain (18). PGH2 converts to PGE2 via the PGE synthase (56). PGE2 can stimulate the smooth muscle contraction (57). COX-1 is considered as a constitutive enzyme, being found in most cells and tissues (58). COX-2 is undetectable in most normal tissues (20). As an inducible enzyme, COX-2 is almost activated under stimulation (59). It has been known that COX-2 mediated the production of inflammatory eicosanoids in the joints but sparing the endogenous protective eicosanoids in the stomach (60). Many studies have pointed out that COX-2 up-regulates the vascular smooth muscle contractile hyperreactivity (21). The inducible enzyme COX-2 exerts its action at inflammatory site of the joints and muscles chiefly (60). In our study, we found that the expression of COX-2 in stomach was decreased in DM rats. Insulin stimulated the expression of COX-2. Pretreatment of insulin and STZ, COX-2 expression was reversed to the normal range.

Indomethacin, the COX general inhibitor, was an irritative activity of antiinflammatory agents (61). It has been known that indomethacin pretreatment existed different reactive oxygen-scavenging system (ROSS) activity (62). Cytotoxicity of reactive nitrogen oxide species (RNOS) causes cellular damage and leads to the disorders with smooth muscle proliferation (62). It has been well known that administration of indomethacin once daily for 2 wk enhances development of gastric mucosal damage (63). It also increases the gastric emptying and induces gastric lesions (64) as well as gastric ulcers (65) with time-course. It was established that NSAIDs increased the vulnerability of the GI mucosa for the development of peptic lesion and many serious ulcer complications, including erosions, inflammation, ulceration, bleeding and perforation (60). But it has no effect at dose 40 mg/kg within 30 min short-time after administration of indomethacin. In the present study, treatment of indomethacin decreased the stimulation of insulin in gastric emptying in both normal and DM rats. We found that indomethacin attenuated the reversal by insulin of the delayed gastric emptying in diabetic rats. Authentically, insulin affected gastric emptying in DM rats through the COX-2 pathway indeed.

On the other hand, PGE2 production is a key indication of COX-2 activity in GI tract (66, 67). Our data showed that the concentration of stomach PGE2 was decreased in DM rats and the effect was restored by insulin replacement. We suggested that the lost expression of COX-2 in stomach caused the decrease of PGE2 production in DM. It meant that insulin reversed the expression of COX-2 and caused the stomach PGE2 production relatively. In DM model, the PGE2 concentration in stomach was decreased. It resulted in the decreased PGE2 synthesis or production. Therefore, the increased PGE2 caused by insulin is the reason of the contraction of smooth muscle in rat stomach in rats. However, the plasma PGE2 levels were not changed in DM or insulin-pretreated rats (data not shown). It might be defined as the PGE2 is a sort of paracrine and/or autocrine hormones so the plasma PGE2 levels could not be used as an index for the contractility of GI smooth muscles. Finally, we found that PGE2 stimulated the smooth muscle contraction, especially in insulin-pretreated group. Instead, the stimulatory effect was weak in DM rats. When combining with insulin and DM treatment, the restoration was almost to the normal range. Although it has been known that blood glucose may affect the gastric emptying (36), our other data strongly suggested the insulin can affect the gastric emptying and smooth muscle contraction in GI tract of DM rats.

Above all, our study has shown the down regulation of COX-2 / PGE2 production and the insulin receptor expression, in addition to an increase of plasma CCK levels in diabetic rats. These factors were all connected with the delay gastric emptying in diabetic rats. Of course, it might still exist other mechanism between diabetes and GI motility (68, 69).We believed that, in our studies, the more interesting point was focused on the COX-2 / PGE2 pathway because many mechanism up- regulated this COX pathway were unknown, including the newer subject cannabinoid (CB) receptors. On the other hand, NSAIDs were used extensively in clinical and had a variety of side effects.

We conclude that insulin changed the expression of IRs in stomach in DM rats. The delayed gastric emptying in diabetes was at least in part due to the COX-2 and PGE2 pathway (decreased COX-2 and diminished PGE2 production in stomach). The change of COX-2 expression was employed for the index of the smooth muscle contraction in stomach in diabetes. Insulin not only stimulated the smooth muscle contraction through the COX-2 expression plus PGE2 production in stomach but also reversed the delayed gastric emptying via the nervous actions of muscarinic M1 and M3 receptors in DM rats.

Acknowledgement: This study was supported by the grants (95002-62-089 and 96002-62-106) from Taipei City Hospital, Taipei, Taiwan, Republic of China

Conflict of interests: None declared.


REFERENCES

1. Hubinger A. Diabetes mellitus-most frequently occurring metabolic diseases. Krankenpflege (Frankf) 1985; 39: 243-245.
2. Shestakova MV, Butrova SA, Sukhareva O. Metabolic syndrome as a precursor of diabetes mellitus type 2 and cardiovascular diseases. Ter Arkh 2007; 79: 5-8.
3. Kristiansen OP, Pociot F, Bennett EP, et al. IDDM7 links to insulin-dependent diabetes mellitus in Danish multiplex families but linkage is not explained by novel polymorphisms in the candidate gene GALNT3. The Danish Study Group of Diabetes in Childhood and The Danish IDDM Epidemiology and Genetics Group. Hum Mutat 2000; 15: 295-296.
4. Ihm SH, Lee KU, Yoon JW. Studies on autoimmunity for initiation of beta-cell destruction. VII. Evidence for antigenic changes on beta-cells leading to autoimmune destruction of beta-cells in BB rats. Diabetes 1991; 40: 269-274.
5. Langeveld M, de Fost M, Aerts JM, Sauerwein HP, Hollak CE. Overweight, insulin resistance and type II diabetes in type I Gaucher disease patients in relation to enzyme replacement therapy. Blood Cells Mol Dis 2008; 40: 428-432.
6. Os I, Os A, Abdelnoor M, Larsen A, Birkeland K, Westheim A. Insulin sensitivity in women with coronary heart disease during hormone replacement therapy. J Womens Health (Larchmt) 2005; 14: 137-145.
7. Fellows IW, Evans DF, Bennett T, Macdonald IA, Clark AG, Bloom SR. The effect of insulin-induced hypoglycaemia on gastrointestinal motility in man. Clin Sci (Lond) 1987; 72: 743-748.
8. Horowitz M, Edelbroek M, Fraser R, Maddox A, Wishart J. Disordered gastric motor function in diabetes mellitus. Recent insights into prevalence, pathophysiology, clinical relevance, and treatment. Scand J Gastroenterol 1991; 26: 673-684.
9. Yamamoto T, Watabe K, Nakahara M, et al. Disturbed gastrointestinal motility and decreased interstitial cells of Cajal in diabetic db/db mice. J Gastroenterol Hepatol 2008; 23: 660-667.
10. Gao Y, Guo XH, Vaz JA. Biphasic insulin aspart 30 treatment improves glycaemic control in patients with type 2 diabetes in a clinical practice setting: Chinese Present study. Diabetes Obes Metab 2009; 11: 33-40.
11. Jeandidier N, Riveline JP, Tubiana-Rufi N, et al. Treatment of diabetes mellitus using an external insulin pump in clinical practice. Diabetes Metab 2008; 34: 425-438.
12. Keshavarzian A, Iber FL. Gastrointestinal involvement in insulin-requiring diabetes mellitus. J Clin Gastroenterol 1987; 9: 685-692.
13. Keshavarzian A, Iber FL, Vaeth J. Gastric emptying in patients with insulin-requiring diabetes mellitus. Am J Gastroenterol 1987; 82: 29-35.
14. Kulkarni PS. Synthesis of cyclooxygenase products by human anterior uvea from cyclic prostaglandin endoperoxide (PGH2). Exp Eye Res 1981; 32: 197-204.
15. Messina EJ, Rodenburg J, Slomiany BL, Roberts AM, Hintze TH, Kaley G. Microcirculatory effects of arachidonic acid and a prostaglandin endoperoxide (PGH2). Microvasc Res 1980; 19: 288-296.
16. Narita M, Shimamura M, Imai S, et al. Role of interleukin-1beta and tumor necrosis factor-alpha-dependent expression of cyclooxygenase-2 mRNA in thermal hyperalgesia induced by chronic inflammation in mice. Neuroscience 2008; 152: 477-486.
17. Tsatsanis C, Androulidaki A, Dermitzaki E, Gravanis A, Margioris AN. Corticotropin releasing factor receptor 1 (CRF1) and CRF2 agonists exert an anti-inflammatory effect during the early phase of inflammation suppressing LPS-induced TNF-alpha release from macrophages via induction of COX-2 and PGE2. J Cell Physiol 2007; 210: 774-783.
18. Faour WH, Gomi K, Kennedy CR. PGE2 induces COX-2 expression in podocytes via the EP4 receptor through a PKA-independent mechanism. Cell Signal 2008; 20: 2156-2164.
19. Leo C, Faber S, Hentschel B, Hockel M, Horn LC. The status of cyclooxygenase-2 expression in ductal carcinoma in situ lesions and invasive breast cancer correlates to cyclooxygenase-2 expression in normal breast tissue. Ann Diagn Pathol 2006; 10: 327-332.
20. Wu CY, Lee WH, Wang JY, et al. Tissue microarray-determined expression profiles of cyclooxygenase-2 in colorectal adenocarcinoma: association with clinicopathological parameters. Chin J Physiol 2006; 49: 298-304.
21. Guo Z, Su W, Allen S, et al. COX-2 up-regulation and vascular smooth muscle contractile hyperreactivity in spontaneous diabetic db/db mice. Cardiovasc Res 2005; 67: 723-735.
22. Taniguchi M. Physiopathology and treatment of aspirin (NSAID) intolerance. Nippon Naika Gakkai Zasshi 2006; 95: 148-157.
23. Aabakken L. NSAID-associated gastrointestinal damage: methodological considerations and a review of the experience with enteric coated naproxen. Eur J Rheumatol Inflamm 1992; 12: 9-20.
24. Dobrek L, Nowakowski M, Mazur M, Herman RM, Thor PJ. Disturbances of the parasympathetic branch of the autonomic nervous system in patients with gastroesophageal reflux disease (GERD) estimated by short-term heart rate variability recordings. J Physiol Pharmacol 2004; 55(Suppl 2): 77-90.
25. Searl TJ, Cunnane TC. Neurotransmitter release mechanisms in sympathetic and parasympathetic nerve terminals. Biochem Soc Trans 1993; 21: 416-420.
26. Takai N, Shida T, Uchihashi K, Ueda Y, Yoshida Y. Cholecystokinin as neurotransmitter and neuromodulator in parasympathetic secretion in the rat submandibular gland. Ann N Y Acad Sci 1998; 842: 199-203.
27. Saji M, Miura M. Nicotinic and muscarinic property of ACh receptors on cultured brain stem neurons. Jpn J Physiol 1982; 32: 1015-1021.
28. Stone WS, Cottrill KL, Walker DL, Gold PE. Blood glucose and brain function: interactions with CNS cholinergic systems. Behav Neural Biol 1988; 50: 325-334.
29. Bonaz B. The cholinergic anti-inflammatory pathway and the gastrointestinal tract. Gastroenterology 2007; 133: 1370-1373.
30. Luo W, Latchney LR, Culp DJ. G protein coupling to M1 and M3 muscarinic receptors in sublingual glands. Am J Physiol Cell Physiol 2001; 280: C884-C896.
31. Ghayur MN, Khan AH, Gilani AH. Ginger facilitates cholinergic activity possibly due to blockade of muscarinic autoreceptors in rat stomach fundus. Pak J Pharm Sci 2007; 20: 231-235.
32. Proctor GB. Muscarinic receptors and salivary secretion. J Appl Physiol 2006; 100: 1103-1104.
33. Yamashita T, Kokubun S. Calcium channel modulation by M3 receptor in guinea-pig tracheal smooth muscle cells. Jpn J Pharmacol 1992; 58(Suppl 2): 404P.
34. Heptulla RA, Rodriguez LM, Mason KJ, Haymond MW. Gastric emptying and postprandial glucose excursions in adolescents with type 1 diabetes. Pediatr Diabetes 2008; 9: 561-566.
35. Tack J. Gastric motor disorders. Best Pract Res Clin Gastroenterol 2007; 21: 633-644.
36. Woerle HJ, Albrecht M, Linke R, et al. Impaired hyperglycemia-induced delay in gastric emptying in patients with type 1 diabetes deficient for islet amyloid polypeptide. Diabetes Care 2008; 31: 2325-2331.
37. Schwizer W, Borovicka J, Kunz P, et al. Role of cholecystokinin in the regulation of liquid gastric emptying and gastric motility in humans: studies with the CCK antagonist loxiglumide. Gut 1997; 41: 500-504.
38. Lin S, Kajimura M, Takeuchi K, Kodaira M, Hanai H, Kaneko E. Expression of muscarinic receptor subtypes in rat gastric smooth muscle: effect of M3 selective antagonist on gastric motility and emptying. Dig Dis Sci 1997; 42: 907-914.
39. Ohki Y, Tomomasa T, Tabata M, et al. Delayed gastric emptying in a neonate, associated with a partial defect in the gastric smooth muscle. J Pediatr Surg 1995; 30: 1511-1512.
40. Hsieh HL, Sun CC, Wang TS, Yang CM. PKC-delta/c-Src-mediated EGF receptor transactivation regulates thrombin-induced COX-2 expression and PGE2 production in rat vascular smooth muscle cells. Biochim Biophys Acta 2008; 1783: 1563-1575.
41. Kadowaki H, Yamamoto T, Kageyama-Yahara N, Kurokawa N, Kadowaki M. The pathophysiological roles of COX-1 and COX-2 in the intestinal smooth muscle contractility under the anaphylactic condition. Biomed Res 2008; 29: 113-117.
42. Hecking E, Knick B. Nausea and vomiting in diabetes mellitus. Dtsch Med Wochenschr 1972; 97: 1798-1799.
43. Wilm S, Helmert U. The prevalence of fullness, heartburn and nausea among persons with and without diabetes mellitus in Germany. Z Gastroenterol 2006; 44: 373-377.
44. Iciek R, Wender-Ozegowska E, Seremak-Mrozikiewicz A, et al. Leptin gene, leptin gene polymorphisms and body weight in pregnant women with diabetes mellitus type I. J Physiol Pharmacol 2008; 59(Suppl 4): 19-31.
45. Zawiejska A, Wender-Ozegowska E, Brazert J, Sodowski K. Components of metabolic syndrome and their impact on fetal growth in women with gestational diabetes mellitus. J Physiol Pharmacol 2008; 59(Suppl 4): 5-18.
46. Princi T, Delbello G, Grill V, Battigelli D. Role of the parasympathetic autonomic nervous system in digitalis intoxication: electrocardiographic and histological study. Acta Med Croatica 1996; 50: 81-85.
47. Ryberg AT, Selberg H, Soukup O, Gradin K, Tobin G.. Cholinergic submandibular effects and muscarinic receptor expression in blood vessels of the rat. Arch Oral Biol 2008; 53: 605-616.
48. Kitazawa T, Hirama R, Masunaga K, et al. Muscarinic receptor subtypes involved in carbachol-induced contraction of mouse uterine smooth muscle. Naunyn Schmiedebergs Arch Pharmacol 2008; 377: 503-513.
49. Tobin G, Giglio D, Lundgren O. Muscarinic receptor subtypes in the alimentary tract. J Physiol Pharmacol 2009; 60: 3-21.
50. Aihara T, Fujishita T, Kanatani K, et al. Impaired gastric secretion and lack of trophic responses to hypergastrinemia in M3 muscarinic receptor knockout mice. Gastroenterology 2003; 125: 1774-1784.
51. Barboriak JJ, Meade RC. Effect of alcohol on gastric emptying in man. Am J Clin Nutr 1970; 23: 1151-1153.
52. Singer MV, Leffmann C. Alcohol and gastric acid secretion in humans: a short review. Scand J Gastroenterol Suppl 1988; 146: 11-21.
53. Yeomans ND, Dent J. Personal review: alarmism or legitimate concerns about long-term suppression of gastric acid secretion? Aliment Pharmacol Ther 2000; 14: 267-271.
54. Bank S, Marks IN, Louw JH, Tigler-Wybrandi N. Stimulation of gastric-acid secretion by histamine, pentagastrin, and pentagastrin-propantheline after vagotomy in man. Lancet 1967; 2: 67-69.
55. Aihara T, Nakamura Y, Taketo MM, Matsui M, Okabe S. Cholinergically stimulated gastric acid secretion is mediated by M(3) and M(5) but not M(1) muscarinic acetylcholine receptors in mice. Am J Physiol Gastrointest Liver Physiol 2005; 288: G1199-G1207.
56. Kelner MJ, Uglik SF. Mechanism of prostaglandin E2 release and increase in PGH2/PGE2 isomerase activity by PDGF: involvement of nitric oxide. Arch Biochem Biophys 1994; 312: 240-243.
57. Ruan YC, Wang Z, Du JY, et al. Regulation of smooth muscle contractility by the epithelium in rat vas deferens: role of ATP-induced release of PGE2. J Physiol 2008; 586: 4843-4857.
58. Crofford LJ. COX-1 and COX-2 tissue expression: implications and predictions. J Rheumatol Suppl 1997; 49: 15-19.
59. Creminon C, Habib A, Maclouf J, Pradelles P, Grassi J, Frobert Y. Differential measurement of constitutive (COX-1) and inducible (COX-2) cyclooxygenase expression in human umbilical vein endothelial cells using specific immunometric enzyme immunoassays. Biochim Biophys Acta 1995; 1254: 341-348.
60. Dajani EZ, Islam K. Cardiovascular and gastrointestinal toxicity of selective cyclo-oxygenase-2 inhibitors in man. J Physiol Pharmacol 2008; 59(Suppl 2): 117-133.
61. Okabe S, Tabata K, Ishihara Y, Kunimi H, Izumi K, Uchida N. Irritative activity of antiinflammatory agents, betamethasone 17-valerate, beclomethasone 17, 21-dipropionate, betamethasone 17, 21-dipropionate, or indomethacin on the gastrointestinal tract in rats and dogs (author’s transl). Nippon Yakurigaku Zasshi 1978; 74: 773-781.
62. Zayachkivska O, Gzregotsky M, Ferentc M, Yaschenko A, Urbanovych A. Effects of nitrosative stress and reactive oxygen-scavenging systems in esophageal physiopathy under streptozotocin-induced experimental hyperglycemia. J Physiol Pharmacol 2008; 59(Suppl 2): 77-87.
63. Kanatani K, Ebata M, Murakami M, Okabe S. Effects of indomethacin and rofecoxib on gastric mucosal damage in normal and Helicobacter pylori-infected mongolian gerbils. J Physiol Pharmacol 2004; 55: 207-222.
64. Takeuchi K, Ueki S, Okabe S. Importance of gastric motility in the pathogenesis of indomethacin-induced gastric lesions in rats. Dig Dis Sci 1986; 31: 1114-1122.
65. Okabe S, Tabata K. Does indomethacin activate healed gastric ulcers in the dog? Digestion 1981; 21: 179-183.
66. Brzozowski T, Konturek PC, Konturek SJ, et al. Involvement of cyclooxygenase (COX)-2 products in acceleration of ulcer healing by gastrin and hepatocyte growth factor. J Physiol Pharmacol 2000; 51: 751-773.
67. Konturek PC, Kania J, Burnat G, Hahn EG, Konturek SJ. Prostaglandins as mediators of COX-2 derived carcinogenesis in gastrointestinal tract. J Physiol Pharmacol 2005; 56(Suppl 5): 57-73.
68. Fischer H, Heidemann T, Hengst K, Domschke W, Konturek JW. Disturbed gastric motility and pancreatic hormone release in diabetes mellitus. J Physiol Pharmacol 1998; 49: 529-541.
69. Liu TT, Shih KC, Kao CC, Cheng WT, Hsieh PS. Importance of cyclooxygenase 2-mediated low-grade inflammation in the development of fructose-induced insulin resistance in rats. Chin J Physiol 2009; 52: 65-71.