Chronic atrophic gastritis and intestinal metaplasia induced by high-salt and N-methyl-N'-nitro-N-nitrosoguanidine intake in rats.

The aim of the present study was to induce chronic atrophic gastritis (CAG) with intestinal metaplasia (IM) in rats by administering saturated salt and methyl-N'-nitro-N-nitrosoguanidine (MNNG) via oral gavage. Changes in gastric mucosal blood microcirculation and activation of the cyclo-oxygenase-2 (COX-2)/hypoxia inducible factor-1α (HIF-1α)/vascular endothelial growth factor (VEGF) signaling pathway during CAG and IM development were investigated. After administering saturated salt and MNNG for 25 weeks, mild atrophy was detected in the stomach of model rats using hematoxylin and eosin staining. CAG with IM was successfully induced in the gastric mucosa of the model rats after 35 weeks. Gastric mucosal blood flow was decreased in comparison with controls as early as 15 weeks after treatment to induce CAG and the mRNA expression levels of COX-2, HIF-1α, vascular endothelial growth factor receptor (VEGFR)1 and VEGFR2 were increased in comparison with untreated rats as early as 25 weeks after treatment. HIF-1α, COX-2 and VEGFR2 expression levels were increased as early as 25 weeks after CAG induction treatment when compared to controls and HIF-1α, COX-2, VEGFR1 and VEGFR2 expression levels were significantly increased after 35 weeks. These findings indicated that administering saturated salt and MNNG by gavage for 35 weeks successfully induced CAG and IM in rats. Furthermore, the microcirculation was disturbed before activation of the COX-2/HIF-1α/VEGF signaling pathway. Copyright: © Yin et al.

Introduction

Gastric cancer (GC) is one of the most common malignant gastrointestinal tumors and the third leading cause of cancer deaths worldwide (1). The majority of reported cases of GC are in East Asia, Eastern Europe and South America, and more than half of GC cases occur in developing countries, with the majority in China (1-3). The global incidence rates of chronic atrophic gastritis (CAG) range from 0-10.9% yearly (4), and it takes several years for chronic gastritis to develop into gastric cancer. Dysplasia and intestinal metaplasia (IM) occur after CAG and are considered premalignant lesions of GC (5). CAG and IM greatly increase the risk of GC, as they promote the development of dysplasia (3). Chronic Helicobacter pylori (Hp) infection is one of the most important risk factors for GC development. According to the anatomic site and criteria, GC is divided into cardia GC, which arises 2-5 cm from the gastric mucosa distal to the esophagogastric junction, and non-cardia GC which originates from the gastric mucosa distal to the cardia (6). Hp is thought to cause 65-80% of all gastric cancer cases (3), however, Hp is not a risk factor for cardia GC (1,3,7-9). Environmental factors, such as high dietary salt intake and oncogenic agents such as methyl-N'-nitro-N-nitrosoguanidine (MNNG), are also important risk factors, particularly under Hp infection-free conditions (1,3,5,10). In China, the mean sodium intake is 5,400 mg/day, which is much higher than the World Health Organization's recommended daily intake of 2,000 mg (11). Several meta-analyses have shown that excess dietary salt intake is a health hazard worldwide and is associated with CAG and IM (12-14). Studies conducted in Japan and Korea, where residents tend to have high-salt intakes owing to their dietary habits of eating salt-rich traditional foods such as miso soup, found that high-salt intake was related to an increased risk of CAG and IM (12-17).

Animal models are important for drug screening and for studying GC, CAG and IM. Several animal models of CAG are used, such as rats, mouse or Mongolian gerbil (18,19). Gastric mucosal injuries, due to Hp infection, surgery, ethanol or indomethacin, and oncogenic agents (primarily MNNG) are the main causes of CAG with precancerous lesions (20,21). Several studies have shown that a synergistic interaction between Hp and a high-salt diet accelerates chronic inflammation and GC development in Mongolian gerbils (22,23) and a high-salt intake is also associated with CAG and IM (14). However, animal models of CAG induced by MNNG and saturated NaCl (to simulate high-salt intake) are rare (22,23); thus, developing a model of CAG with IM in rats treated with oncogenic agents and saturated NaCl may be beneficial for future research and development of drugs to treat CAG and IM.

Chronic inflammation plays an important role in GC development and progression. Moreover, high-salt intake and interleukin-17 (IL-17) synergistically induce vascular endothelial growth factor (VEGF)-A expression through nuclear factor of activated T cells 5 (NFAT5)/signal transducers and activators of transcription 3 (STAT3) interaction in breast cancer cells (24). High-salt intake also promotes inflammation in the tumor microenvironment and enhances angiogenesis and VEGF expression (25). A previous study showed that microcirculatory disorders existed in CAG (26). However, whether high-salt intake with MNNG can induce microcirculatory disorders and whether the hypoxia-inducible factor (HIF)-1α pathway is activated during the pathological process of CAG with IM remains uncertain.

In the present study, a new rat model of CAG with IM was developed by administration of saturated salt and MNNG by gavage. Dynamic changes in the gastric mucosal blood microcirculation and activation of the cyclo-oxygenase-2 (COX-2)/HIF-1α/VEGF signaling pathway during the development of CAG with IM were investigated.

Materials and methods

Materials

MNNG (95%) and sodium chloride (99.8%) were purchased from Sinopharm Chemical Reagent Co., Ltd. The primary antibody against VEGF receptor (VEGFR)1 (cat. no. ab184784) was purchased from Abcam. The primary antibodies against COX-2 (cat. no. 12282S), VEGFR2 (cat. no. 9698S) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; cat. no. 5174S) were provided by Cell Signaling Technology, Inc. The primary antibody against HIF-1α (cat. no. 610958) was supplied by Becton-Dickinson and Company.

Animals

A total of 30 male Wistar rats (160-180 g; 5 weeks old) were obtained from Shanghai Laboratory Animal Center of the Chinese Academy of Science (Shanghai, China), and housed in the Laboratory Animal Center of Shanghai University of Traditional Chinese Medicine (Shanghai, China). All rats were housed under a 12-h light/dark cycle at room temperature (25±2˚C) and humidity (60±2%) with free access to food and water. All animal experiments were conducted according to protocols approved by the Animal Ethics Committee of Shanghai University of Traditional Chinese Medicine (approval no. SZY201703012). For euthanasia, rats were anesthetized by an intraperitoneal injection of 3% pentobarbital sodium (30 mg/kg body weight), once fully anesthetized, rats were sacrificed by cervical dislocation and verification of death was defined by cessation of breathing and faded eye color.

CAG model induction

Rats were randomly divided into the control group (n=10) and 4 model groups (n=5/group). Model rats were treated with MNNG at 200 mg/kg body weight by oral gavage on days 0 and 14. Saturated NaCl (1 ml per rat) was administered 3 times per week by oral gavage for the first 3 weeks. MNNG (600 µg/kg) and saturated NaCl (1 ml per rat) were then administered by gavage on alternate days (Fig. 1). From week 5, 1 group of 5 model rats was killed every 10 weeks (at weeks 5, 15, 25 and 35).

Figure 1

CAG induction protocol in rats. CAG, chronic atrophic gastritis; MNNG, methyl-N'-nitro-N-nitrosoguanidine.

Measurement of gastric mucosal blood flow (GMBF)

Rats were fasted for 24 h, then anesthetized with 3% pentobarbital sodium (30 mg/kg body weight). After exposing the stomach, a fiber-optic probe for laser Doppler flowmetry (Moor Instruments Ltd.) was placed on the stomach wall of the fundus, gastric body and antrum to measure the blood flow in the stomach. The voltage number was recorded to represent the relative GMBF.

Specimen collection

After measuring the GMBF, anesthetized rats were sacrificed by cervical dislocation. The stomach was then quickly removed, cut along the greater curvature and washed with 0.9% sodium chloride. The antral tissues were then separated into two or three parts. A part was fixed for 24-48 h in 10% formalin at room temperature for hematoxylin and eosin (H&E) and alcian blue-periodic acid-Schiff (AB-PAS) staining. The other two parts were used for western blotting and quantitative PCR analysis and stored at -80˚C.

Morphological assay

Tissue specimens were fixed for 24-48 h with 10% formalin at room temperature, processed, embedded in paraffin and cut into 5-µm sections. All sections were stained with H&E and AB-PAS (12), observed under a light microscope and scored according to the histopathological grading standard. The degrees of atrophy and metaplasia were assessed according to the updated Sydney classification as: 1, absent; 2, mild; 3, moderate; or 4, severe (12).

Western blot analysis

Stomach tissue was homogenized on ice with CellLytic™ MT mammalian tissue lysis reagent (Sigma-Aldrich) containing protease and phosphatase inhibitor cocktails and centrifuged at 13,523 x g for 15 min at 4˚C. The supernatant was then collected, and the protein concentration was determined via BCA assay. Next, 30 µg of protein per sample was loaded and separated by SDS-PAGE (8 for HIF-1α, VEGFR1 and VEGFR2 or 15% for COX-2, GAPDH), then transferred onto polyvinylidene fluoride (PVDF) membranes by a wet-transfer system (Bio-Rad Laboratories, Inc.). The PVDF membranes were then blocked with 5% bovine serum albumin (BSA; cat. no. G611BA0007; Sangon Biotech Co., Ltd.) in 1X phosphate buffered saline with 0.1% Tween-20 (PBST) for 1 h. PVDF membranes were incubated with primary antibodies (each, 1:1,000) against COX-2, HIF-1α, VEGFR1, VEGFR2 and GAPDH at 4˚C overnight. The membranes were further incubated with goat anti-rabbit horseradish peroxidase-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, Inc.; cat. no. 111-035-003; 1:1,000) for 1 h at room temperature. Thereafter, the protein bands were visualized using an ECL-prime kit (EMD Millipore; cat. no. WBKLS0500) in a Tanon-5200 (Tanon Science & Technology Co., Ltd.). Quantification of the target proteins was normalized to that of GAPDH within the same sample.

Reverse transcription quantitative PCR (qPCR) analysis

Total RNA was extracted from the stomach tissues using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) as per the manufacturer's instructions. cDNA was reverse transcribed from RNA (1 µg) using the RevertAid First Strand cDNA Synthesis kit (cat. no. K1622; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The cDNA (1 µl) was diluted with nuclease-free water 5 times for the qPCR reaction. RT-qPCR was performed using SYBR Premix EX Taq (Roche Diagnostics) on the Quant Studio 6 Flex System (Applied Biosystems; Thermo Fisher Scientific, Inc.) under the following conditions: 95˚C for 30 sec, followed by 40 cycles of 95˚C for 5 sec, 60˚C for 34 sec, 95˚C for 15 sec, 60˚C for 1 min and 95˚C for 15 sec. The target genes were quantified via the 2-ΔΔCq method (27). Table I lists the sequences of primers obtained from Shanghai GeneRay Biotech Co., Ltd. The relative expression of the individual target genes were normalized to that of GAPDH in the same sample.

Table I

Primer sequences used in the qPCR analysis.

Genes	Forward primer (5'-3')	Reverse primer (5'-3')
VEGF	CCTCTCCCTACCCCACTTCCT	CACTTTCTCTTTTCTCTGCCTCCAT
VEGFR1	TTGATGGTAGGCTGAGGGATG	AGATGTAACTGCCGAGGATGC
VEGFR2	GAGTTGGTGGAGCATTGGGAA	ATACAGGAAACAGGTGAGGTAGGCA
HIF-1α	CCCATTCCTCATCCATCAAACATT	CTTCTGGCTCATAACCCATCAACTC
COX-2	TGAAATATCAGGTCATCGGTGGAG	CATACATCATCAGACCCGGCAC
Ang-1	TTGGTTACTCGTCAGACATTCATC	TCTTCTTCTCTTTTTCCTCCCTTTA
Ang-2	AAGTCAACGCTGCCATCTTCC	GACCTTCCCCAACTCCACAGA
GAPDH	TCTCTGCTCCTCCCTGTTC	ACACCGACCTTCACCATCT

COX-2, cyclooxygenase-2; HIF, hypoxia inducible factor; VEGFR, vascular endothelial growth factor receptor; Ang, angiotensin.

Statistical analysis

All data are expressed as the mean ± SEM. Differences between two groups were analyzed using an unpaired Student's t-test. Differences among more than two groups were analyzed using one-way analysis of variance with Dunnett's multiple comparison test. Analyses were performed using GraphPad Prism 6 (GraphPad Software, Inc.). Differences were considered significant at P<0.05.

Results

Atrophic changes in the gastric mucosa

In control rats, stomach glands were arranged neatly and were the same size. After treatment for 5 and 15 weeks, the glands were arranged in order, and the number of glands remained similar (Fig. 2). CAG was detected by HE staining developed in the model rats after 25 weeks. Changes in the glands were much more evident after 35 weeks of treatment than at earlier weeks and compared with the control rats, the glands were reduced in number and visibly disordered. IM, detected by AB-PAS staining, was induced after 35 weeks of induction (P<0.01 vs. control).

Figure 2

CAG is induced by intragastric administration of saturated NaCl and MNNG. Hematoxylin and eosin staining (magnification, x100) in (A) control group, (B) week 5, (C) week 15, (D) week 25, (E) week 35. Values are expressed as the mean ± SEM (n=5/group). Data were analyzed using one-way analysis of variance (ANOVA). **P<0.01 vs. model group. CAG, chronic atrophic gastritis; MNNG, methyl-N'-nitro-N-nitrosoguanidine.

IM changes in the gastric mucosa

The AB-PAS staining results showed that, compared with the control rats, glands were decreased in number, but mild IM was not obvious at 25 weeks of treatment (Fig. 3). After 35 weeks of induction, moderate and severe IM appeared in the antrums of the model rats (P<0.001 vs. control).

Figure 3

Gastric intestinal metaplasia is induced by intragastric administration of saturated NaCl and MNNG (Alcian blue-periodic acid-Schiff staining, x100 magnification). Gastric intestinal metaplasia changes were stained purple and blue. (A) Control group, (B) week 25, (C) week 35. Values are expressed as the mean ± SEM (n=5/group). ***P<0.01 vs. model group.

GMBF changes at different time points

After 5 weeks of treatment, the GMBF of the fundus, gastric body and antrum did not differ between the control and model rats (Fig. 4). After treatment for 15-35 weeks, the GMBF of the fundus, gastric body and antrum in the model rats was decreased time-dependently, compared with that of the control rats (P<0.01; P<0.001 vs. control).

Figure 4

Gastric mucosal microcirculation disorder was induced by intragastrically administering saturated NaCl and MNNG. (A) GMBF of the fundus, (B) GMBF of the gastric body, (C) GMBF of the antrum, (D) GMBF of the whole stomach. Values are expressed as the mean ± SEM (n=5/group). **P<0.01, ***P<0.001 vs. model group. GMBF, gastric mucosal blood flow.

Activation of the COX-2/HIF-1α/VEGF pathways at different time points

As early as 25 weeks after treatment, the mRNA expression levels of HIF-1α (P<0.05 vs. control), COX-2 (P<0.01 vs. control), VEGF (P<0.001 vs. control), VEGFR1 (P<0.05 vs. control) and VEGFR2 (P<0.05 vs. control) were increased compared with those of the control rats. However, the mRNA expression levels of Ang-1 and Ang-2 were not obviously affected (Fig. 5).

Figure 5

Gene expression levels of COX-2, HIF-1α, VEGFR1, VEGFR2, Ang-1 and Ang-2 in the gastric tissue. (A) Week 15, (B) week 25 and (C) week 35. Values are expressed as the mean ± SEM (n=5/group). *P<0.05. **P<0.01, ***P<0.001 vs. model group. COX-2, cyclooxygenase-2; HIF, hypoxia inducible factor; VEGFR, vascular endothelial growth factor receptor; Ang, angiotensin.

Further study demonstrated that the protein expression levels of HIF-1α (P<0.001 vs. control), COX-2 (P<0.05 vs. control) and VEGFR2 (P<0.05 vs. control), but not VEGFR1 were significantly enhanced in the model rats after 25 weeks compared with those of the control rats (Fig. 6). The protein levels of HIF-1α (P<0.01 vs. control), COX-2 (P<0.05 vs. control), VEGFR1 (P<0.05 vs. control) and VEGFR2 (P<0.05 vs. control) were obviously upregulated in the model rats after 35 weeks compared with those of the control rats.

Figure 6

Protein expression levelsof COX-2, HIF-1α, VEGF and VEGFR1 in the gastric tissue. (A) Week 15, (B) week 25 and (C) week 35. Values were expressed as the mean ± SEM (n=5/group). *P<0.05. **P<0.01, ***P<0.001 vs. model group. COX-2, cyclooxygenase-2; HIF, hypoxia inducible factor; VEGFR, vascular endothelial growth factor receptor.

Discussion

CAG with IM and dysplasia are the most significant risk factors for GC and considered the two key types of precancerous lesions in GC (1-3). However, the molecular mechanisms of CAG with precancerous lesions are unclear. Suitable animal models that are similar to clinical patients are important for determining the underlying molecular mechanisms of CAG and drug screening and development to treat CAG with precancerous lesions. At present, several methods exist to induce CAG with precancerous lesions in animals. The main methods include Hp infection with or without MNNG (28,29) in Mongolian gerbils, surgery with or without 150 g/l NaCl paste at 50-70˚C (30,31), and MNNG with or without the other two or more factors (32-37) (Table II). However, multiple factors, complex operations, long procedure times, high costs and high death rates have resulted in no standard model for CAG with precancerous lesions. Oncogenic agents, such as MNNG, can easily penetrate the pylorus and gastric mucosa of the stomach to cause DNA damage; thus, MNNG in the drinking water is used as a specific carcinogen to induce GC (38). However, it takes a long time for MNNG in the drinking water to induce precancerous lesions in animals, and the success rate is low (35). Thus, other stimulating factors, such as ammonia, sodium deoxycholate, salicylic acid and ethanol, are also used to promote the development of CAG with precancerous lesions (Table II).

Table II

Methods of establishing chronic atrophic gastritis with precancerous lesions in rats.

Method
Gastrogavage	Water	Food	Operation	Others	Time (week)	Mortality rate (%)	Success rate (%)	(Refs.)
H. pylori	-	-	-	-	72	0	100	(29)
H. pylori	MNNG 20 m.m.p	-	-	-	50	0	60	(28)
-	-	-	Gastrojejunostomy	-	36	0	71.4	(31)
50-60 Paste containing 150 g/l NaCl, 2 ml	-	-	Spring pyloric implantation	-	16	0	100	(30)
-	MNNG 100 µg/ml	-	-	-	50	0	64.3	(35)
Sodium deoxycholate 20 mmol/l
Ethanol 60%, 8 ml/kg
Indomethacin 0.05%, 8 ml/kg	Ammonia 0.05-0.1%	-	-	-	42	0	100	(33)
MNNG 2 mM, 10 ml/kg
Ethanol 5 ml/kg	-	-	-	Fasting food Tail clamp	32	0	100	(36)
MNNG 120 µg/ml, 10 ml/kg	Ammonia 0.1%	Ranitidine 0.03%	-	-	28	30	70	(32)
MNNG 120 µg/ml, 10 ml/kg
Ethanol 35%, 3 ml/rat	Salicylic acid 2%	Ranitidine 0.03%	-	-	28	15.4	25	(37)
-	MNNG 100 µg/ml	Nacl 8%	-	-	42	45	31.3	(34)

MNNG, methyl-N'-nitro-N-nitrosoguanidine.

High-salt diets can directly damage the gastric mucosa and induce hypergastrinemia, leading to parietal cell loss and GC progression (39). High-salt diets also promote inflammatory cell infiltration and increased COX-2 expression in Hp-infected Mongolian gerbils (40). Transgenic COX-2 expression and high-salt intake enhance susceptibility to methylnitrosourea (MNU)-induced GC development in mice (39,41). A 10% NaCl diet significantly increased the incidence of GC induced by oncogenic agents, such as MNNG, in drinking water (42). These findings indicate that high-salt intake plays an important role in GC progression. Reports on the induction of CAG with precancerous lesions in animals using MNNG and NaCl are rare. A previous study reported that combining 100 µg/ml MNNG in the drinking water and chow pellets with 8% NaCl induced CAG with precancerous lesions in rats after modeling for 42 weeks. However, the mortality rate was 45% with only a 31.3% success rate (34).

Therefore, in the present study, only MNNG and saturated NaCl were intragastrically administered to induce CAG with IM in rats. At week 25, the success rate reached 60% with no deaths. At week 35, the success rate reached 100% with no deaths. Moreover, long-term saturated NaCl is similar to high-salt intake in clinical patients with CAG with IM (43,44). Therefore, the protocol described in the present study of inducing CAG with IM is simple, easy, controllable, less costly and repeatable.

Angiogenesis and microcirculatory disorders have been found in both CAG patients and animals (45,46). As a transcription factor, HIF-1α directly regulates VEGF gene expression in cancer angiogenesis, especially under hypoxia (47). The HIF-1α signaling pathway plays a vital role in blood microcirculatory disorders, including ischemia, hypoxia, inflammation and tumor angiogenesis (48-51). Gastric mucosal injury can induce upregulation of HIF-1α, VEGF and COX-2(52). COX-2 can enhance the expression of HIF-1α and VEGF (53). HIF-1α can also upregulate COX-2 expression in human endothelial cells (54). However, whether high-salt and MNNG intake can induce gastric microcirculatory disturbance and activation of the HIF-1α pathway remains uncertain. In the present study, GMBF disturbance first appeared without significant activation of the HIF-1α pathway at week 15. This demonstrated that blood microcirculatory problems began to appear before the development of HIF-1α pathway dysfunction and CAG with IM, indicating that high-salt-induced gastric injury may be involved in blood microcirculatory disorders of the stomach. At week 25, CAG was successfully established, and the blood microcirculatory disorder was much more severe. The gastric mucosal blood flow was reduceded, indicating that the decreased speed of the GMBF induced hypoxia in the model rats. HIF-1α and COX-2 mRNA expression levels were significantly increased, indicating that blood microcirculatory disorder-induced hypoxia activated the HIF-1α pathway and COX-2. After 35 weeks, moderate or severe atrophic gastritis with IM occurred in the model rats, indicating that high-salt-induced gastric mucosal injury may induce microcirculatory disorders, then microcirculatory disorder-induced hypoxia would induce abnormal HIF-1α pathway expression and gastric inflammation evidenced by COX-2 upregulation. This would further enhance angiogenesis and consequently enhance the GC process (48,50,55,56). VEGFR1 and VEGFR2 are VEGF receptors (57). Upregulated VEGFR1 and VEGFR2 expressions in the gastric tissues of rats with CAG and IM are consistent with overactivation of the COX-2/HIF-1α/VEGF pathway. In this study, blood microcirculatory disorders occurred during CAG with IM in rats. Thus, this animal model can help clarify the molecular mechanisms of CAG with IM and help develop drugs to treat CAG with IM.

In conclusion, a new rat model of CAG with IM was stably and effectively established by intragastrically administering saturated NaCl and MNNG. Activation of the HIF-1α pathways and gastric inflammation, resulted from high-salt-induced stomach microcirculatory disorders, might be involved in the pathological process of CAG with IM induced by high-salt and MNNG intake. However, the inconsistencies between replicate experiments is a limitation of the present study, which should be improved in future work.