Angiotensin-(1–7) [Ang-(1–7)] is a major vasoactive metabolite of angiotensin I (Ang I), both being important components of the renin-angiotensin system (RAS). Ang-(1–7) acting via Mas receptor was documented in kidneys, heart, brain, and gastrointestinal (GI)-tract. We studied the gastroprotective activity of exogenous Ang-(1–7) in rats exposed to water immersion and restraint stress (WRS) without or with A-779 [d-Ala7-Ang-(1–7), an antagonist of Ang-(1–7) Mas receptors], AVE 0991 (5-formyl-4-methoxy-2-phenyl-1[[4-[2-(ethylaminocarbonylsulfonamido)-5-isobutyl-3-thienyl]-phenyl]-methyl]-imidazole), the agonist of Ang-(1–7) receptor, as well as the inhibition of nitric-oxide (NO) synthase, the suppression of cyclo-oxygenase (COX)-1 (indomethacin, SC-560 [5-(4-chloro-phenyl)-1-(4-methoxyphenyl)-3-trifluoromethyl-pyrazole]), the activity COX-2 (rofecoxib), and denervation with capsaicin. The mRNA expression of constitutively expressed nitric-oxide synthase (cNOS), inducible nitric-oxide synthase (iNOS), interleukin (IL)-1β, and tumor necrosis factor (TNF)-α was analyzed by reverse transcription polymerase chain reaction. The WRS lesions were dose-dependently reduced by pretreatment with Ang-(1–7), which also caused an increase in gastric blood flow (GBF) and luminal content of NO. COX-1 and COX-2 inhibitors or L-NNA (N5-[imino(nitroamino)methyl]-L-ornithine) reversed the reduction in lesion number and the rise in GBF evoked by Ang-(1–7). Ang II augmented the WRS lesions, decreased GBF and increased the plasma IL-1β and TNF-α levels. Capsaicin denervation attenuated the reduction of Ang-(1–7)-induced gastric lesions and the rise in GBF; these effects were restored by supplementation with calcitonin gene–related peptide (CGRP). The cNOS mRNA was upregulated while iNOS, IL-1β and TNF-α mRNAs were downregulated in Ang-(1–7)-pretreated rats. We conclude that Ang-(1–7), in contrast to Ang II, which worsened WRS ulcerogenesis, affords potent gastroprotection against WRS ulcerogenesis via an increase in GBF mediated by NO, endogenous prostaglandins, sensory neuropeptides, and anti-inflammatory action involving the inhibition of proinflammatory markers iNOS, IL-1β, and TNF-α.
Renin-angiotensin system (RAS) is a classic endocrine system involved in physiologic regulation of blood pressure and water-mineral balance (Paul et al., 2006). The components of RAS appear to be functionally active in numerous organs including kidneys, heart, brain, reproductive organs, and skin. Angiotensin I (Ang I) and angiotensin II (Ang II) play an important role in control of gastrointestinal (GI)-functions such as the fluid and electrolyte homeostasis, maintenance of regional blood flow, mucosal absorption of glucose, gastrointestinal motility, mucosal secretion, gastric inflammation, and carcinogenesis (Fandriks 2011; Garg et al., 2012). Recently, the essential Ang I and Ang II metabolites have been identified throughout the GI tract, including stomach, colon, pancreatic islets, and liver (Carl-McGrath et al., 2009; Olszanecki et al., 2009; Hasegawa et al., 2009).
Ang II is the central product of RAS and potent constrictor of vascular smooth muscles (Heinemann et al., 1999). Ang II acts via angiotensin receptor type 1 (AT1) and contributes to vasoconstriction, inflammation, vascular and cardiac hypertrophy, and extracellular tissue remodeling by inhibition of cell growth and stimulation of apoptosis (Lemarie et al., 2009). Stimulation of the AT1 receptors activates membrane NADPH oxidase in vascular smooth muscle cells (VSMCs), enhances the production of reactive oxygen species such as superoxide and hydrogen peroxide (H2O2), and inactivates NO pathway (Mehta and Griendling, 2007). Ang II-activating phospholipase C (PLC) and protein kinase C (PKC) or phospholipase A2 enhanced synthesis of vasoconstrictive leukotrienes and smooth muscle cell contraction (Mehta and Griendling, 2007; Lemarie et al., 2009). Increased reactive oxygen species (ROS) and decreased blood flow play fundamental roles in the pathogenesis of GI mucosal injury (Bregonzio et al., 2003; Nakagiri et al., 2010).
Exposure to stress is commonly recognized as a risk factor of microbleeding and gastric mucosal injury. Reaction to stress is mediated via two distinct but unrelated systems: the hypothalamic-pituitary-adrenocortical (HPA) system and the sympathoadrenal system (Goldstein and McEwen, 2002; Saavedra et al., 2006). Ang II receptor subtypes AT1 and AT2 were detected in the human esophageal, gastric, small intestinal, and colonic mucosa (Hirasawa et al., 2002; Casselbrant et al., 2009; Hallersund et al., 2011). The antagonists of Ang II AT1 receptors attenuated gastric injury induced by ischemia-reperfusion, cold stress, and indomethacin-induced damage in rodents due to an inhibition of sympathoadrenal axis and the attenuation of vasoconstrictor and proinflammatory activity of Ang II (Pavel et al., 2008; Morsy et al., 2009; Gemici et al., 2010; Saavedra et al., 2011; Shiotani et al., 2011).
Angiotensin-(1–7) [Ang-(1–7)] is a downstream peptide generated from angiotensin I through angiotensin-converting enzyme (ACE) homolog ACE2 or neutral endopeptidase (NEP, also known as neprilysin). Since the discovery of Ang-(1–7) in 1976, the presence of this hectapeptide has been detected in brain, blood vessels, heart, kidney, liver, and stomach (Santos et al., 2005; Xu et al., 2011). Ang-(1–7) acting via its own G protein-coupled receptor called Mas (Santos et al., 2003; Stegbauer et al., 2004) exhibit the vasodilatory, antihypertensive, cardioprotective, and antifibrotic effects. Ang I is quickly degraded to Ang-(1–7) in the rat stomach and the formation of Ang-(1–7) may even precede Ang I conversion to Ang II (Olszanecki et al., 2009). Mas-receptor knockout mice developed hypertension due to dysfunction of vascular endothelium, decrease in NO synthesis, and downregulation of eNOS expression, suggesting a link between Ang-(1–7) and Mas receptor (Xu et al., 2008). The vasoconstrictive action of Ang II in hypertension is limited by vasoactive Ang-(1–7) and bradykinin (Oliveira et al., 1999; Greco et al., 2006; Sampaio et al., 2007). Ang-(1–7) exhibited esophagoprotection against reflux esophagitis (Pawlik et al., 2012), but whether Ang-(1–7) protects the gastric mucosa against stress lesions due to an increase of NO and the activity of prostaglandin (PG)/COX-1 and PG/COX-2 pathways and/or sensory nerves has not been extensively studied so far.
We compared the effects of exogenous Ang-(1–7) and Ang II on stress-induced gastric lesions and accompanying changes in the gastric blood flow (GBF). The involvement of endogenous PG and NO as well as the activity of afferent sensory nerves in the mechanism of gastroprotection induced by Ang-(1–7) was investigated by testing the effect of exogenous Ang-(1–7) against stress ulcerogenesis in the presence of NO-synthase inhibitor L-NNA, nonselective and selective COX-1 and COX-2 inhibitors, as well as in rats with capsaicin denervation. We also assessed the effect of Ang-(1–7) on the expression of mRNA for constitutively expressed nitric-oxide synthase (cNOS), inducible nitric-oxide synthase (iNOS), proinflammatory cytokines interleukin (IL)-1β and tumor necrosis factor (TNF)-α, and plasma levels of these cytokines during stress ulcerogenesis.
Materials and Methods
Male Wistar rats (total 254) with weight averaging about 250 g were used in this study. Rats were fasted for 24 hours with free access to drinking water before the exposure to WRS. The study was approved by the Institutional Animal Care and Use Committee of Jagiellonian University Medical College in Cracow and run in accordance with the statements of the Helsinki Declaration regarding handling of experimental animals.
Stress-Induced Gastric Lesions, Chemicals, and Drugs Application.
To induce gastric lesions, rats were immobilized in individual Bolman cages and immersed in the cold water (23°C) for 3.5 hours to the rat xyphoid level as reported by our group previously (Brzozowski et al., 2000, Konturek et al., 2001). Seven major experimental groups of rats (A–G) were selected. Thirty minutes before exposure to water immersion and restraint stress (WRS), rats in series A–C received pretreatment with either: A) exogenous Ang-(1–7) (6.25–50 μg/kg i.p.); B) Ang II (50 μg/kg i.p.); and C) perindopril (5 mg/kg i.p.), the long-lasting ACE inhibitor (Jawien et al., 2012). The angiotensin Mas receptor antagonistic and agonistic activities were determined in a separate group of rats (series D) treated with A-779 (5 mg/kg i.p.), the selective Ang-(1–7) Mas receptor antagonist (Bayorh et al., 1999; Santos et al., 2003; Castro et al., 2005) with or without the combination with Ang-(1–7) in rats exposed 30 minutes later to 3.5 hours of WRS or AVE 0991 (50 μg/kg i.p.), the nonpeptide Ang-(1–7) receptor agonist (Pinheiro SV et al., 2004; Santos and Fereira, 2006), respectively.
In series E, the effects of cotreatment with Ang-(1–7) or perindopril, with or without the combination with L-NNA (20 mg/kg i.p.), a competitive inhibitor of NO-synthase activity, on WRS lesions and alterations in the GBF were determined.
The involvement of endogenous PG in the gastroprotective effects of exogenous Ang-(1–7) or vehicle (control) was investigated in rats (series F) treated with indomethacin (5 mg/kg i.p.), the nonselective COX-1 and COX-2 inhibitor, or SC-560 (5 mg/kg i.p.), the selective inhibitor of COX-1, and rofecoxib (10 mg/kg i.g.), the selective inhibitor of COX-2 activity as reported in our previous studies (Brzozowski et al., 2000, 2006; Satoh et al., 2013). In another subgroup with COX-1 and COX-2 inhibitors, rats of series F were coadministered with exogenous prostaglandin E2 (PGE2; 5 μg/kg i.g.) in the presence of Ang-(1–7).
In series G, the effect of blockade of sensory nerves induced by large dose of capsaicin (total 125 mg/kg s.c.) on the protective and hyperemic activity of Ang-(1–7) was examined. Capsaicin was injected for 3 consecutive days at a respective dose of 25, 50, and 50 mg/kg s.c. approximately 2 weeks before the experiment to induce the functional ablation of sensory nerves as described previously (Konturek et al., 2009; Kwiecien et al., 2012a). In separate subgroup of series G with capsaicin denervation, the involvement of calcitonin gene–related peptide (CGRP), the major rat neuropeptide released from sensitive afferent nerve endings in protective action of exogenously administered Ang-(1–7) against WRS lesions, was determined. In one of the subgroups of series G, the capsaicin-denervated rats received supplementation with exogenous CGRP (10 μg/kg s.c.) combined with Ang-(1–7) and 30 minutes later were exposed to onset of WRS as in other groups described above.
All tested drugs and compounds were of analytical grade and were purchased from Sigma-Aldrich Laborchemikalien (Schelldorf, Germany) except of SC-560 and rofecoxib purchased from Cayman Chemical (Ann Arbor, MI) and Pfizer (Illertissen, Germany), respectively.
Measurement of GBF and Determination of Gastric Lesion Number.
At the termination of 3.5 hours WRS, rats were anesthetized with pentobarbital (60 mg/kg i.p.), the abdomen was opened, and GBF measured by means of H2-gas clearance technique as reported before (Brzozowski et al., 2004, 2006; Kwiecien et al., 2007). The GBF was measured in the fundic part of the gastric mucosa not involving mucosal lesions. Average values of three measurements were determined and expressed as a percentage of change of the value determined in intact rat stomach. Gastric lesions number was determined on photographed stomachs with computerized planimetry (Morphomat, Carl Zeiss, Berlin, Germany) (Kwiecien et al., 2012a) by a blinded investigation.
Determination of Luminal NO Content and Plasma Level of IL-1β and TNF-α.
The luminal concentration of NO was quantified indirectly as nitrate (NO3–) and nitrite (NO2–) levels in the gastric contents using the nitrate/nitrite kit purchased from Cayman Chemical as described in detail in our previous studies (Brzozowski et al., 2008; Pawlik et al., 2011; Kwiecien et al., 2012b).
The blood samples (~3 ml) were taken from the vena cava for the measurement of plasma proinflammatory cytokines IL-1β and TNF-α as described previously (Kwiecien et al., 2012b). In brief, the plasma TNF-α and IL-1β was determined by a solid-phase sandwich enzyme-linked immunosorbent assay (ELISA; BioSource International Inc., Camarillo, CA) according to the manufacturer’s instructions. Each sample (50 μl) was incubated with biotinylated antibodies specific for rat TNF-α and IL-1β, washed three times with assay buffer, and finally conjugated with streptavidin peroxidase to form a complex with a stabilized chromogen as described before (Kwiecien et al., 2012b).
The expression mRNA of cNOS, iNOS, IL-1β and TNF-α in the rat gastric mucosa determined by reverse transcriptase-polymerase chain reaction.
The stomachs were removed from rats exposed to WRS without or with the pretreatment with Ang-(1–7) alone or combined with A-779 to determine mRNA expression of cNOS, iNOS, IL-1β, and TNF-α by reverse transcriptase-polymerase chain reaction (RT-PCR) with specific primers. Mucosal specimens were scraped off using a glass slide and immediately snap-frozen in liquid nitrogen and stored at –80°C until analysis. Total RNA was extracted from mucosal samples by a guanidium isothiocyanate/phenol chloroform method using a kit from Stratagene (Heidelberg, Germany). The total RNA concentration in each sample was determined by 1% agarose-formaldehyde gel electrophoresis and ethidium bromide staining. Aliquoted RNA samples were stored at –80°C until analysis.
Single-stranded cDNA was generated from 5 μg of total cellular RNA using StrataScript reverse transcriptase and oligo(dT) primers (Stratagene). In brief, 5 μg of total RNA was uncoiled by heating (65°C for 5 minutes) and then reversed by transcribing into complementary DNA (cDNA) in a 50-μl reaction mixture that contained 50 IU of Moloney murine leukemia virus reverse transcriptase (MMLV-RT), 0.3 mg of oligo(dT) primer, 1 ml of RNase block ribonuclease inhibitor (40 IU/μl), 2 ml of a 100 mM mixture of deoxyadenosine triphosphate (dATP), deoxyribothymidine triphosphate (dTTP), deoxyguanosine triphosphate (dGTP), and deoxycytidine triphosphate (dCTP), 5 ml of 10× RT buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl, 5 mM MgCl2). The resultant cDNA (2 μl) was amplified in a 50-μl reaction volume containing 0.3 ml (2.5 IU) Taq polymerase, 200 mM (each) dNTP (Pharmacia, Germany), 1.5 mM/l MgCl2, 5 ml 10× polymerase chain reaction buffer (50 mM KCl, 10 mM Tris-HCl, pH 8.3), and primers used at final concentration of 0.5 mM. The mixture was overlaid with 25 μl of mineral oil to prevent evaporation. The polymerase chain reaction mixture was amplified in a DNA thermal cycler (Perkin-Elmer/Cetus, Norwalk, CT) in the area dedicated for performing PCR reaction. The nucleotide sequences of the primers for cNOS, iNOS, IL-1β, TNF-α, and β-actin presented in Table 1 were constructed based on the published cDNA for these factors. The primers were synthesized by Gibco/Life Technologies (Eggenstein, Germany).
Polymerase chain reaction products were detected by electrophoresis on a 1.5% agarose gel containing ethidium bromide. Location of predicted products was confirmed by using DNA 100-bp ladder (Gibco) as a standard size marker. The intensity of bands was quantified using densitometry (LKB Ultrascan, Pharmacia, Uppsala, Sweden) as described in detail in our previous studies (Brzozowski et al., 2008; Konturek et al., 2009). The signals for cNOS, iNOS, IL-1β, and TNF-α mRNAs were standardized against the β-actin signal for each sample, and results were expressed as cNOS, iNOS, IL-1β, and TNF-α mRNA/β-actin mRNA ratios.
Results of the experiment were expressed as mean ± S.E.M. and the statistical analysis was performed with two-way analysis of variance (ANOVA) test and Tukey post hoc test where appropriate. Differences between estimates of effects were considered significant at P < 0.05. All results in the treated animals were compared with the appropriate control group, which had been established for each set of experiments. Dependent variables were expressed both in percentage of control for GBF and in absolute values for lesion number. The control rats did not differ from experimental groups in terms of relevant characteristics, such as source of purchase, gender, age, weight, diet, and housing conditions. As there was no individual pairing of animals, the paired statistical tests were not used.
Mean Lesion Number and GBF in Rats Pretreated with Ang II or Ang-(1–7).
Exposure of vehicle-pretreated control rats to 3.5 hours of WRS caused gastric mucosal lesions (hemorrhagic erosions) accompanied by a significant fall in GBF (Fig. 1.). The pretreatment with Ang II applied in a dose of 5 μg/kg failed to significantly affect the mean lesion number and GBF compared with vehicle-control. The administration of Ang II in higher doses ranging from 10 to 40 μg/kg dose-dependently increased the mean lesion number and produced a significant and a dose-dependent decrease in GBF (Fig. 1). The pretreatment with Ang-(1–7) administered i.p. in graded doses ranging from 6.25 to 50 μg/kg, dose-dependently attenuated WRS-induced gastric lesions, while producing a significant and a dose-dependent increase in GBF and luminal NO concentration (Fig. 2). The dose of Ang-(1–7) inhibiting WRS lesions by 50% (ID50) was 27 μg/kg. Since the dose of 50 μg/kg afforded the maximal protective response (P < 0.05), this dose of Ang-(1–7) was used in all our determinations. The absolute values for GBF expressed in ml/min per 100 g are presented in Table 2. Exposure to WRS in rats pretreated with vehicle-control significantly decreased the GBF (P < 0.05) compared with the values in the intact gastric mucosa. This fall in GBF under WRS conditions was significantly worsened by the pretreatment with Ang II. In contrast, the pretreatment with Ang-(1–7) resulted in a significant increase in the GBF (P < 0.05) compared with those pretreated with vehicle. The Ang-(1–7)-induced protection and the accompanying rise in the GBF and luminal NO content, observed at the 50 μg/kg dose of this peptide, were completely reversed by the pretreatment with A-779 (50 μg/kg i.p.) combined with intraperitoneal treatment with Ang-(1–7) (Fig. 2; Table 2).
Effect of AVE 0991, the Agonist of Ang-(1–7) Mas Receptor, on WRS-Induced Gastric Lesions and Alterations in the GBF.
As shown in Fig. 3, the pretreatment with AVE 0991 (50 μg/kg i.p.) significantly reduced the mean lesion number (P < 0.05) and caused a significant increase in the GBF (P < 0.05) compared with the respective values in vehicle-control pretreated rats. This decrease in lesion number and an increase in the GBF induced by AVE 0991 were completely abolished in rats treated with the combination of A-779 and AVE 0991 (P < 0.05).
Effect of Suppression of NO-Synthase on Ang-(1–7)- and Perindopril-Induced Gastroprotection and Alterations in GBF in Rats Exposed to WRS.
Figure 4 shows that pretreatment with Ang-(1–7) (50 μg/kg i.p.) significantly reduced the WRS-induced gastric lesions and increased GBF, with the extent similar to the respective values presented in Fig. 2. The pretreatment with perindopril (5 mg/kg i.p.) also significantly decreased the number of WRS-induced gastric lesions (P < 0.05) and significantly increased GBF compared to vehicle-controls. Administration of L-NNA (20 mg/kg i.p.), which by itself failed to significantly affect the lesion number and GBF compared to vehicle-treated control, reversed the reduction in lesion number and the rise in GBF evoked by Ang-(1–7) or perindopril (Fig. 4).
Effect of COX-1/PG and COX-2/PG Suppression on Ang-(1–7)-Induced Gastroprotection against WRS-Induced Gastric Damage and Alteration in GBF.
As shown in Fig. 5, the pretreatment with Ang-(1–7) (50 μg/kg i.p.) caused a similar decrease in the mean number of WRS-induced gastric lesions accompanied by a significant rise in the GBF as presented in Fig. 2. The pretreatment with COX-1 and COX-2 inhibitors alone significantly increased the mean lesion number and produced a significant fall in GBF compared with vehicle-treated animals exposed to WRS (data not shown). The reduction of lesion number by Ang-(1–7) (50 μg/kg i.p.) was significantly attenuated by pretreatment with indomethacin (5 mg/kg i.p.), rofecoxib (10 mg/kg i.g.), and SC-560 (5 mg/kg i.g.) (P < 0.05), and these effects were accompanied by a significant fall in GBF (Fig. 5). The addition of PGE2 (5 μg/kg i.g.) to Ang-(1–7) restored the gastroprotective effect of this peptide in the presence of COX-1 and COX-2 inhibitors (P < 0.05), and these effect were accompanied by an increase in GBF similar to that recorded in Ang-(1–7)-treated animals without concomitant treatment with COX-1 and COX-2 inhibitors (P < 0.05) (Fig. 5).
Effect of Capsaicin Denervation with or without Exogenous CGRP on Ang-(1–7)-Afforded Gastroprotection and Hyperemia against WRS-Induced Gastric Damage.
As shown in Fig. 6, the pretreatment with Ang-(1–7) (50 μg/kg i.p.) caused a similar decrease in the number of WRS-induced gastric lesions accompanied by a significant rise in the GBF as presented in Figs. 4 and 5. The exogenous administration of CGRP (10 μg/kg s.c.) in rats with intact sensory nerves resulted in a significant decrease of WRS-induced gastric damage (P < 0.05) and significant increase in the GBF (P < 0.05) compared with respective values achieved with Ang-(1–7) (Fig. 6). The capsaicin denervation tended to increase the mean lesion number and to decrease GBF compared to rats with intact sensory nerves. The reduction in lesion number and an increase in the GBF caused by Ang-(1–7) in rats with intact sensory innervation were almost completely lost in those with capsaicin denervation. The concurrent administration of CGRP combined with Ang-(1–7) significantly reduced the mean lesion number (P < 0.05) and significantly increased GBF in capsaicin-denervated rats (P < 0.05); however, these values were still significantly different from those attained with Ang-(1–7) in rats with intact sensory nerves (Fig. 6).
Effect of Pretreatment with Ang-(1–7) or Ang II on Plasma Levels of Proinflammatory Cytokines IL-1β and TNF-α in Rats Exposed to WRS.
As shown in Fig. 7, the plasma levels of IL-1β and TNF-α were negligible in intact rats not exposed to WRS. In contrast, the plasma TNF-α and IL-1β levels were significantly increased in vehicle-pretreated rats exposed to WRS (P < 0.02). The further significant rise in plasma levels of IL-1β and TNF-α was observed in the group administered with Ang II (50 μg/kg i.p.) compared with those pretreated with vehicle and exposed to WRS (Fig. 7). In contrast, Ang-(1–7) (50 μg/kg i.p.) significantly decreased (P < 0.05) the plasma levels of IL-1β and TNF-α compared to vehicle-control group and Ang II-pretreated group (Fig. 7).
Gastric Mucosal Expression of cNOS, iNOS, IL-1β, and TNF-α mRNAs in Rats Treated with Ang-(1–7) without or with the Combination with A-779.
Figure 8 (left panel) shows the expression of cNOS mRNA in gastric mucosa of vehicle-pretreated rats and those administered Ang-(1–7) with or without the combined administration of A-779 and then exposed to WRS. In vehicle-pretreated rats exposed to WRS, the signal for cNOS mRNA was weak compared with that observed in intact gastric mucosa. In contrast, the strong signal for the cNOS mRNA was observed in rats pretreated with Ang-(1–7) (50 μg/kg i.p.) and exposed 30 minutes later to 3.5 hours of WRS compared with those pretreated with vehicle (Fig. 8). Ratio of cNOS mRNA over β-actin confirmed that cNOS mRNA was significantly increased in Ang-(1–7)-pretreated gastric mucosa over that observed in the vehicle-control gastric mucosa exposed to WRS (Fig. 8, right panel). A weak signal of cNOS mRNA was recorded in rats with combined administration of A-779 and Ang-(1–7) compared with that in animals treated with Ang-(1–7) alone. Ratio of cNOS mRNA over β-actin confirmed that cNOS mRNA was significantly decreased (P < 0.05) in rats treated with the combination of A-779 and Ang-(1–7) compared with those administered with Ang-(1–7) alone (Fig. 8, right panel).
As shown in Fig. 8 (left panel) both IL-1β and TNF-α mRNAs were strongly detected in vehicle-treated gastric mucosa, and the ratio of IL-1β or TNF-α mRNA over β-actin mRNA (Fig. 8, right panel) confirmed that IL-1β and TNF-α mRNAs were significantly upregulated in WRS-induced gastric mucosa. These effects were significantly attenuated in those pretreated with Ang-(1–7) (Fig. 8, right panel). In contrast, strong signals for IL-1β and TNF-α mRNAs were observed when rats received the combination of A-779 and Ang-(1–7) compared with those treated with Ang-(1–7) alone (Fig. 7, left panel). The ratio of IL-1β and TNF-α over β-actin confirmed that Ang-(1–7) significantly decreased expression of mRNAs for IL-1β and TNF-α and this effect was reversed in animals administered with the combination of A-779 and Ang-(1–7) (Fig. 8, right panel).
Figure 9 (upper panel) demonstrates that the signal for iNOS mRNA was negligible in the intact gastric mucosa, but mRNA for iNOS was detected as strong signal in gastric mucosa exposed to WRS, and this effect was significantly decreased in those pretreated with Ang-(1–7). The ratio of iNOS mRNA over β-actin confirmed that mRNA for iNOS was significantly increased in rats exposed to WRS when compared with that in the intact gastric mucosa and this effect was significantly attenuated in those pretreated with Ang-(1–7) (Fig. 9, lower panel). The decrease in iNOS mRNA expression observed in Ang-(1–7)-pretreated animals was reversed in those concomitantly treated with A-779. The ratio of iNOS mRNA over β-actin confirmed that mRNA for iNOS was significantly increased when A-779 was combined with Ang-(1–7) (Fig. 9, lower panel).
Our study indicates for the first time that Ang-(1–7), one of the major metabolites of Ang II, contributes to the mechanism of gastroprotection against gastric lesions induced by stress, which is one of the important risk factors for peptic ulcer, hemorrhagic erosions, and microbleedings in animals and humans (Pavel et al., 2008; Konturek et al., 2011). We have shown that parenteral administration of Ang-(1–7) ameliorated in a dose-dependent manner the severity of WRS-induced gastric lesions and this effect was accompanied by the increase in GBF and rise in luminal NO content. Blockade of Mas receptor by A-779 inhibited the Ang-(1–7)-induced protection and hyperemia, while AVE 0991, the agonist of Ang-(1–7) receptors, mimicked the gastroprotective and hyperemic actions of Ang-(1–7). Our results provide the evidence that NO-NOS system and PG-COX pathways could be involved in the protective and hyperemic activities of this RAS metabolite because this protection and an increase in GBF were reversed by the NOS activity inhibitor L-NNA, and by either nonselective or selective COX-1 and COX-2 inhibitors. We have demonstrated that these protective and hyperemic effects of Ang-(1–7), which disappeared in COX-1- and COX-2-treated animals, have been restored by PGE2 coadministered with this peptide in the presence of COX-1 and COX-2 inhibitors. The involvement of NO in gastroprotection and the hyperemic actions of Ang-(1–7) is further supported by the fact that expression of cNOS was upregulated while expression of iNOS, considered as proinflammatory marker, was downregulated in the gastric mucosa of Ang-(1–7)-pretreated rats. This gastroprotective and hyperemic effect of Ang-(1–7) was similar to those exhibited by perindopril, a long lasting ACE inhibitor. The protective and hyperemic effects of Ang-(1–7) were lost in rats with capsaicin denervation consistent with the notion that this peptide may trigger the sensory afferent endings to release vasodilatory and protective CGRP. Indeed, the pretreatment with CGRP coadministered with Ang-(1–7) enhanced the protective activity of this Ang I metabolite, resulting in gastric hyperemia but also counteracted the capsaicin-induced gastric impairment and the accompanying fall in the gastric GBF observed in Ang-(1–7)-treated rats with deactivated sensory nerves. These findings indicate that sensory neuropeptide CGRP can cooperate with PG and NO in the mechanism of Ang-(1–7)-induced gastroprotection and gastric hyperemia against WRS-induced gastric lesions (Fig. 10).
Since stress causes gastric damage of poorly recognized mechanism and etiology, and RAS has been implicated in the pathogenesis of gastric mucosal integrity (Brzozowski et al., 2012) and stress ulcerogenesis (Ender et al., 1993; Kwiecien et al., 2007; Konturek et al., 2011), we determined the effect of vasoactive Ang-(1–7) against stress-induced gastric lesions and compared it with that of Ang II. In clear contrast to Ang-(1–7), the pretreatment with Ang II failed to exert gastroprotection and exacerbated the WRS-induced gastric lesions accompanied by the fall in the GBF. Moreover, Ang-(1–7) markedly decreased the expression and release of proinflammatory cytokines IL-1β and TNF-α (Szlachcic et al., 2013) suggesting that the anti-inflammatory properties of Ang-(1–7) contribute to protective activity of this Ang I metabolite in the rat stomach (see Fig. 10).
Previous studies documented that AT1-receptor antagonists help to maintain the proper gastric blood perfusion via the reduction of sympathetic neural activity and by attenuation of inflammatory mediators (Ender et al., 1993; Chung et al., 2010; Garg et al., 2012). Bregonzio et al. (2004) observed that AT1 blockade led to increase in adrenal corticosterone, reduction in TNF-α and intercellular adhesion molecule 1 (ICAM-1) expression, and neutrophil infiltration in stressed animals. However, the blockade of AT1 receptors does not influence gastroprotective action of glucocorticoids released during stress (Filaretova et al., 1998; Pavel et al., 2008). Similarly, AT1-receptor antagonists dose-dependently attenuated gastric ulcers in rodents (Merai et al., 2009; Morsy et al., 2009) and counteracted the effects of ischemia and inflammation by the reduction in mucosal neutrophil infiltration and expression of gastric intercellular adhesion molecule 1 and TNF-α (Saavedra et al., 2005, 2006). It is not excluded that the beneficial effect of AT1-receptor antagonists could depend on enhancement of the concentration of angiotensin metabolites Ang-(1–7) and Ang-(1−9) (Neves et al., 2000; Olszanecki et al., 2009), but this hypothesis requires further studies.
Our results show that WRS increased the expression and plasma levels of TNF-α and IL-1β and that the plasma level of these proinflammatory cytokines was augmented by Ang II, suggesting that, in contrast to Ang-(1–7), Ang II, known as a potent vasoconstrictor, aggravated WRS-induced gastric damage due to its proinflammatory action. This is corroborative with the observation that high levels of circulating Ang-(1–7) ameliorated the metabolic stress induced by a high-fat diet via decrease in the proinflammatory profile of adipose tissue cytokines (Santos et al., 2012). Ang-(1–7) decreased body weight, increased HDL cholesterol, and decreased expression of COX-2 and IL-1β in abdominal fat of overweight rats (Santos et al., 2012). Moreover, Clark et al. (2001) reported a direct binding of Ang-(1–7) to the Ang II receptor type 1 resulting in downregulation of these receptors. In keeping with these findings, we observed decreased expression and plasma levels of IL-1β and TNF-α in rats pretreated with Ang-(1–7), indicating the apparent difference between Ang-(1–7) and Ang II with respect to proinflammatory cytokines. Moreover, the endogenous Ang II could contribute to pathogenesis of cold-restraint stress ulcer in obstructive jaundice rats (Mou et al., 1998). Enalapril, an inhibitor of ACE, reduced both the plasma and gastric mucosal Ang II level, decreased gastric blood flow, and increased the extent of mucosal damage (Mou et al., 1998). Furthermore, Ang-(1–7) acting as an endogenous inhibitor of ACE, enhanced the vasodilator effects of bradykinin (Tom et al., 2003). In our study, perindopril significantly decreased WRS-induced gastric lesions and raised GBF with an extent similar to that observed with Ang-(1–7). L-NNA reduced the gastroprotective and hyperemic activity of perindopril, suggesting that this protection and rise in the GBF caused by ACE inhibitor might be also mediated by NO. Finally, the luminal content of NO and gastric mucosal expression of mRNA for cNOS were both increased by Ang-(1–7), suggesting that NO derived from cNOS pathway contributes to the beneficial effect of Ang-(1–7) against stress ulcerogenesis. In contrast, the mRNA expression of iNOS was downregulated in these rats, which is consistent with the notion that Ang-(1–7) inhibits WRS lesions due to its potent anti-inflammatory activity.
We clearly demonstrated that Ang-(1–7) significantly and dose-dependently attenuated WRS-induced gastric damage while increasing GBF, and these effects were abolished by d-Ala7-Ang-(1–7) (A-779), the selective antagonist of Mas receptors. Interestingly, the antagonist A-779 has been shown to inhibit most of the physiologic effects of Ang-(1–7) (Santos et al., 2003). Liao et al. (2011) revealed that cardioprotective effect of Ang-(1–7) against ischemia-reperfusion damage is mediated by COX/PG system responsible for the attenuation of malonyldialdehyde content and rise in superoxide dismutase activity. The intestinal mucosal COX-2 expression is regulated by both AT1 and AT2 receptors (Tani et al., 2008). Ang-(1–7) stimulated PGE2 release from spontaneously hypertensive rat vascular smooth muscle cells (Jaiswal et al., 1993). In our study, the gastroprotection and increase of GBF evoked by Ang-(1–7) were counteracted by pretreatment with COX-1 and COX-2 inhibitors. For many years, PGs have been considered major cytoprotective mediators that play an important role in various aspects of gastroduodenal protection and ulcer healing (Robert, 1979; Tarnawski et al., 1988; Brzozowski et al., 2006; Takeuchi, 2010). Yousif et al. (2012) revealed that PGs are important intermediaries of the beneficial effects of Ang-(1–7) in cardiac recovery and vascular reactivity in diabetes. Herein, exogenous PGE2 added to Ang-(1–7) in the presence of COX-1 and COX-2 inhibitors restored the gastroprotective and hyperemic activities of this metabolite. Thus, the mechanism through which the Ang-(1–7)/Mas receptor axis induced gastroprotection depends on the activation COX/PG system and endogenous PG.
Sensory nerves were implicated in the mechanism of gastroprotection against various gastric damaging factors, including stress and Helicobacter pylori lipopolysaccharide (LPS) (Brzozowski et al., 2004; Kwiecien et al., 2007). The gastroprotective and hyperemic activities of Ang-(1–7) were markedly impaired in rats with capsaicin-induced functional ablation of sensory fibers. This indicates that besides NO and PG afferent sensory fibers and the major sensory neuropeptide CGRP released from rat sensory nerve endings might mediate Ang-(1–7)-induced protection and hyperemia. Exogenous CGRP in the presence of Ang-(1–7) restored this protection in part, and gastric hyperemia in rats with capsaicin denervation; however, this increase in GBF was significantly less pronounced in capsaicin-denervated rats compared with those with intact sensory nerves. Thus, it is reasonable to conclude that CGRP, which is a potent vasodilator and protective factor in the stomach, can cooperate with Ang-(1–7) in this protection.
In summary, Ang II and Ang-(1–7) show opposite action against stress ulcerogenesis, because Ang II enhanced stress ulcerogenesis but Ang-(1–7) afforded protection against stress lesions. The mechanism of Ang-(1–7)-induced protection against stress may involve activation of NO/cNOS and PG/COX systems and vasodilatory and gastroprotective sensory neuropeptides such as CGRP. In contrast to Ang II, Ang-(1–7) exhibits vasodilatory and anti-inflammatory properties that inhibit expression and release of proinflammatory cytokines IL-1β and TNF-α. Further studies in experimental animals and humans are warranted to further determine the therapeutic efficacy of Ang-(1–7) in various gastrointestinal disorders.
Participated in research design: Ptak-Belowska, Kwiecien, Brzozowski.
Conducted experiments: Magierowski, Jasnos, Pawlik, Kwiecien.
Contributed new reagents or analytic tools: Krzysiek-Maczka, Olszanecki, Korbut.
Performed data analysis: Magierowski, Jasnos, Brzozowski.
Wrote or contributed to the writing of the manuscript: Magierowski, Kwiecien, Brzozowski.
- Received June 17, 2013.
- Accepted September 18, 2013.
This work was supported by National Science Centre in Poland [Grant N N402 479937; 4799/B/P01/2009/37 (to T. B.)].
- angiotensin-converting enzyme
- Ang II
- angiotensin II
- angiotensin receptor type 1
- AVE 0991
- calcitonin gene–related peptide
- constitutively expressed nitric-oxide synthase
- gastric blood flow
- inducible nitric-oxide synthase
- nitric-oxide synthase
- renin-angiotensin system, SC-560, [5-(4-chloro-phenyl)-1-(4-methoxyphenyl)-3-trifluoromethyl-pyrazole]
- tumor necrosis factor
- water immersion and restraint stress
- Copyright © 2013 by The American Society for Pharmacology and Experimental Therapeutics