Mechanistic role of antioxidants in rescuing delayed gastric emptying in high fat diet induced diabetic female mice

Abstract

Diabetic gastroparesis (DG) exhibits delayed gastric emptying (GE) due to impaired gastric non-adrenergic, non-cholinergic (NANC) relaxation. These defects are due to loss or reduction of nuclear factor (erythroid-derived 2)-like 2 (Nrf2) that causes reduced expression and/or dimerization of neuronal nitric oxide synthase alpha (nNOSα) gene expression and function. We investigated the effect of potent Nrf2 activators (cinnamaldehyde [CNM] & curcumin [CUR]) on GE in obesity-induced diabetic female mice. We fed adult female homozygous Nfe2l2^−/− (Nrf2 KO) and wild-type (WT) female mice with either a high-fat diet (HFD) or a normal diet (ND) for a period of 16 weeks. Groups of HFD mice were fed with CUR or CNM either at 6th or 10th week respectively. Our results demonstrate that supplementation of CNM or CUR restored impaired nitrergic relaxation and attenuated delayed GE in HFD fed mice. Supplementation of CNM or CUR normalized altered gastric antrum protein expression of (1) p-ERK/p-JNK/MAPK/p-GSK-3β, (2) BH₄ (Cofactor of nNOS) biosynthesis enzyme GCH-1 and the GSH/GSSG ratio, (3) nNOSα protein & dimerization and soluble guanylate cyclase (sGC), (4) AhR and ER expression, (5) inflammatory cytokines (TNF α, IL-1β(3, IL-6), (6)TLR-4, as well as (7) reduced oxidative stress markers in WT but not in Nrf2 KO obesity-induced chronic diabetic female mice. Immunoprecipitation experiments revealed an interaction between nNOS and Nrf2 proteins. Our results conclude that Nrf2 activation restores nitrergic-mediated gastric motility and GE by normalizing inflammation and oxidative stress induced by obesity-induced chronic diabetes.

1. Introduction

Diabetic Gastroparesis (DG) is characterized by delayed gastric emptying (GE) in the absence of mechanical obstruction. Hyperglycemia, enteric neuromuscular inflammation, and autonomic neuropathy are main causes of obesity-induced/diabetic gastroparesis [1,2]. Obese and diabetic patients often suffer from delayed or accelerated GE. In the US alone, approximately, 46 million people suffer from diabetes with an expected increase of 35% by 2045 [3]. Several studies have demonstrated that females experience a higher incidence of gastrointestinal disturbance than males, including nausea, vomiting, bloating, and constipation [4–6]. About 75% of diabetic patients experience delayed GE; female patients are more affected than male patients [1,7,8]. Previous studies demonstrate that changes in endogenous sex hormone levels, during menstruation, pregnancy, and menopause may lead to altered gastric motility and slower gastric emptying [9–13]. Recent studies from our laboratory demonstrate that in-vivo supplementation of estradiol-17β restored rapid gastric emptying in high fat diet (HFD)-induced type 2 diabetes (T2D) in ovariectomized female mice [14]. Differences in gastric neuronal nitric oxide synthase (nNOS) dimerization have been proposed as a potential cause of gender bias observed in gastric motility [12,15]. Of all five cofactors, tetrahydrobiopterin (BH₄) known to play a key role for nNOS dimerization and enzyme activity [16]. In the absence of BH₄, nNOS oxidizes NADPH and reduces O₂ but arginine oxidation is blocked, resulting uncoupling of nNOS [16]. Under these circumstances, instead of NO, nNOS generates superoxide, leading to the formation of peroxynitrite (ONOO⁻) [16]. Diminished NO levels and enhanced oxidative stress due to the lack of BH₄ contribute to pathophysiologic conditions, including gastric dysmotility in diabetic females [17]. The importance of BH₄ as a regulator of nNOS function renders BH₄ a logical therapeutic target for DG [7,12,18,19]. BH₄ oxidation contributes to oxidative stress leading to nNOS uncoupling [20]; hence, it is a major contributor to diabetes and cardiovascular diseases.

Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) a basic leucine zipper transcriptional factor plays a role on induction of antioxidant enzymes which display beneficial effects in treating various complications [21]. Nrf2 signaling attenuated the elevated levels of pro-inflammatory cytokines and chemokines in-vivo [22]. Nrf2 is strongly associated with immune regulation and oxidative stress, rendering it a reasonable therapeutic target [22]. We have previously reported that loss of Nrf2 in female mice (Nrf2 KO, Nfe2/2^−/−) reduced BH₄ levels inhibits nNOS-mediated gastric nitrergic relaxation, and decreased nitrite levels, subsequently causing delayed GE [18]. Cinnamaldehyde (CNM) supplementation inhibited systemic cytokine expression through Nrf2 activation [23]. Hence, Nrf2 activation mediated by naturally occurring compounds alleviate symptoms of diseases associated with obesity/diabetes.

Curcumin (CUR (1,7-bis-(4-hydroxy-3-methoxyphenyl)—1, 6-hepta-diene-3,5-dione)) is a bioactive molecule present in the rhizome of Curcuma longa, commonly known as turmeric. CUR has been shown to possess many health beneficial properties in various organs [24]. In addition, CUR molecule has shown to prevent diabetes-induced oxidative stress, functions as a gastroprotectant, and prevents vascular dementia via its antioxidant activity [25–28]. Cinnamomum zeylanicum (cinnamon) activates Nrf2 via the AKT/JNK signaling and is a popular traditional medicine for diabetes management [21,29]. Our previous study using obese female mice revealed that CNM, the bioactive molecule in cinnamon, reversed GE in WT but not in Nrf2 KO mice [23]. However, it is unknown whether chronic type 2 diabetes alters delayed gastric emptying by reducing Nrf2 and nitrergic mediated gastric motility in female rodents. We hypothesize that CNM and CUR suppress elevation of gastric inflammatory cytokines and balance antioxidants/oxidative stress ratios, thus restoring nitrergic-mediated gastric emptying through Nrf2-dependent signaling pathways in HFD-induced chronic T2DM female mice.

2. Materials and methods

2.1. Animals

The Institutional Animal Care and Use Committee (IACUC) at Meharry Medical College (MMC) approved all animal experiments in this study (protocol # 17-09-764). Adult (9 weeks old) female homozygous Nfe2l2^−/− mice (B6.129X1 -Nfe212 tm1Ywk/J, Nrf2 KO) and their wild-type (WT) littermates were purchased from Jackson Laboratories (Bar Harbor, ME). The mice were fed either a high-fat diet (HFD, 70% energy as fat, 19% protein, and 11% carbohydrate, HFD; 5SSV, Test Diet, St. Louis, MO) or a normal diet (ND, 6.2% energy as fat, Teklad Global 2018, Teklad Diets, Madison, WI) and were allowed access to food and water ad libitum. All procedures involving animals were performed in accordance with relevant guidelines set by according to IACUC at MMC.

2.2. Experimental design

Adult female mice were randomly assigned to one of seven treatment groups (n = 8 per group): (i) WT-ND, (ii) WT-HFD, (iii) WT-HFD + CNM (50 mg/kg, i.p., 6 weeks, starting at 10th week), (iv) WT-HFD + CUR (200 mg/kg, p.o., 10 weeks, starting at 6th week), (v) KO-ND; (vi) KO-HFD, and (vii) KO-HFD + CUR (200 mg/kg, p.o., 10 weeks, starting at 6th week). CNM and CUR were purchased from Sigma-Aldrich (St. Louis, MO). Previous studies from our laboratory and others have shown that supplementation of CNM (50 mg/kg, i.p) for 12 weeks start from day one of HFD shown to be effective in normalizing GE in diabetic mice [23,30]. CUR at 200 mg/kg b.w., has been shown to be effective in preventing diabetic retinopathy in diabetic rats [31]. CUR was dissolved in 1% carboxymethyl cellulose (CMC, Sigma-Aldrich, St. Louis, MO) and administered every day for 10 weeks. CNM was prepared in mineral oil (carrier solution) and administered three times a week for 6 weeks starting at week 10. At week 16, all mice were sacrificed by CO₂ asphyxiation. Serum and stomach tissues were snap-frozen and stored at − 80 °C.

2.3. Fasting blood glucose determination, glucose tolerance test, and insulin tolerance test

To determine changes in glucose and insulin levels, the following methods were employed in groups of mice with or without HFD and treatment interventions. Fasting blood glucose levels were monitored every alternate week for the duration of the study period using standard protocol [32]. Intraperitoneal glucose tolerance test (IPGTT) was performed at week 14. In addition, insulin tolerance test (ITT) was performed at week 15 [32]. Data were expressed as the absolute values of blood glucose concentrations.

2.4. Estrus cycle assessment

The progression of two consecutive estrus cycles (proestrus, estrus, metestrus, and diestrus) were assessed among all groups of animals (10 days before final experimental period, 16 weeks) by vaginal smears [33]. Vaginal smears were collected with smooth-edged glass pasteur pipettes filled with 10 μl of normal saline (0.9% NaCl). Cells from the smears were placed on untreated glass microscopic slides and viewed at 20x and 40x magnification to determine the cycle stages.

2.5. Assays for serum 17β-estradiol (E2), insulin, and nitrite

Animals were sacrificed by CO₂ asphyxiation and serum was separated and stored at − 80 °C to be used for subsequent biochemical analysis [23]. Serum nitrite and 17β-estradiol (E₂) were analyzed as reported earlier [23]. Serum insulin levels were measured using ELISA kits (Crystal Chem, Elk Grove Village, IL, USA) per manufacturers’ guidelines. Nitrite levels in the serum were analyzed as total nitrite (metabolic by-product of NO) according to manufacturer’s protocol (Bio vision, Milpitas, CA). Serum 17β-estradiol (E₂) levels were assayed using an ELISA kit (Enzo Life Sciences, Farmingdale, NY) following manufacturer’s instructions. Absorbance at 405 nm was obtained using a micro-plate reader (BioTek, Winooski, VT). To assess insulin resistance, the homeostatic model of insulin resistance (HOMA-IR) index was calculated as fasting serum glucose × fasting serum insulin/22.5 [14].

2.6. Solid GE studies

Solid GE studies were performed as described previously [18]. Briefly, the mice were first deprived from solid food overnight (water consumption allowed). The next day, each mouse was caged separately and were fed with a known amount of food for 3 h with water ad libitum. Thereafter, each mouse was placed in a separate clean cage and starved for 2 h without food and water. The remaining unconsumed food was dried at 37 °C for about 20 min and weighed to determine how much food each mouse (food intake, FI) consumed. This procedure is critical to minimize and avoid measuring the content of gastric secretions that may differ between healthy and experimental groups. At the end of 2 h fasting, all mice were euthanized by cervical dislocation. To determine gastric content (GC), the weights of full and empty stomachs were recorded for each animal. The rate of GE was calculated as follows: GE (% in 2 h) = [1 − gastric content (GC)/food intake (FI)] × 100.

2.7. Organ bath studies

Electric field stimulation (EFS)-induced non-adrenergic non-cholinergic relaxation (NANC) was examined in circular gastric antrum neuromuscular strips in WT mice (n = 4/group) [23]. The circular gastric antrum neuromuscular strips were mounted in 10 ml Krebs buffer at 37 °C and NANC-dependent nitrergic relaxation (nNOS function) was determined at 2 Hz [23] (DMT Technologies, Nottingham, UK). The NO dependence of nitrergic relaxation was confirmed with NG-nitro-L-arginine-methyl ester treatment (L-NAME, 100 μM, 30 min). Comparison between groups was performed by measuring the area under the curve (AUC/mg of tissue) of the EFS-induced relaxation (AUCR) curve at 1 min and the baseline (AUCB) curve at 1 min, as follows: (AUCR – AUCB)/weight of tissue (mg) = AUC/mg of tissue.

2.8. Reduced/oxidized glutathione (GSH/GSSG) assay

Reduced/oxidized glutathione (GSH/GSSG) ratio were measured using a HT Glutathione Assay kit (Trevigen, Gaithersburg, MD). The gastric tissues were washed with cold isotonic saline (150 mM Nacl), dried, weighed, homogenized with ice cold 5% (wt/vol) metaphosphoric acid at 20 ml/g tissue, and centrifuged at 14,000g for 15 min at 4 °C. Supernatants were collected and used for the assay according to manufacturer’s protocol.

2.9. nNOSα dimerization in the gastric neuromuscular tissues

Levels of nNOSα monomer and dimer were quantified by western blotting via low temperature (LT)-polyacrylamide gel electrophoresis (PAGE) of gastric antrum homogenates [18,23]. A polyclonal anti-nNOSα antibody (N-terminal) (1:500, Thermo Fisher Scientific, Waltham, MA) and anti-rabbit IgG conjugated with horseradish peroxidase (1: 10,000, Sigma Chemical, St. Louis, MO) were used as the primary and secondary antibodies, respectively.

2.10. Subcellular fractionation, immunoprecipitation and immunoblotting

Subcellular fractionation was carried out according to Ramsey et al. [34]. The tissue lysates were suspended in fractionation buffer (10 mM HEPES [pH 7.9], 10 mM KC1, 1.5 mM MgCl₂, 0.1% NP-40, 0.5 mM NaF, 200 mM Na₃V0₄, and 1x protease inhibitor cocktail). The cells were incubated on ice for 15 min with shaking. Lysates were centrifuged at 2600g at 4 °C, and supernatants representing cytosolic fraction were collected. The precipitates then were resuspended with the modified RIPA buffer containing 1x protease inhibitor cocktail and incubated on ice for 20 min with periodic vortexing. The lysates were then cleared by centrifugation at 10,000g at 4 °C, and supernatants were used as the nuclear fractions.

Immunoprecipitation experiments were performed as previously described [35]. For immunoprecipitation (IP), cytosols and nuclear extracts (50-μg protein) were incubated with a desired IP antibody (2 μg), and the immune complexes were pulled down by incubation with Protein A Dynabeads (Thermo Fisher Scientific, Waltham, MA) and centrifugation. These and other proteins in the gastric antrum neuromuscular tissue lysates were separated by SDS-PAGE. The membrane was immunoblotted with polyclonal anti-Nrf2, ER α, ER β, sGC α, sGC β, IL 6, TNF-α, IL-1β, TLR-4 and GCH-1 from Santa Cruz Biotechnology (Dallas, TX), anti-p-ERK, p-JNK, P38/MAPK from Cell Signaling (Danvers, MA), and anti-nNOSα (N-terminal) from Abeam (Cambridge, MA). Anti-rabbit/anti-mouse IgG conjugated with horseradish peroxidase were used as secondary antibodies (1: 10,000) (Sigma-Aldrich, St. Louis, MO). All primary antibodies were used at 1:500 or 1:1000 dilution following vendor recommendation. Antibody binding was detected using an enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech, Piscataway, NJ) following manufacturer’s instructions. Stripped blots were re-probed with polyclonal anti-β-actin antibody (Sigma-Aldrich, St. Louis, MO). Band intensities were analyzed using ImageQuant LAS 500 (GE Health Sciences, Pittsburgh, PA).

2.11. PCR microarray

Total RNA was isolated from mouse gastric neuromuscular tissues using a single-step guanidine thiocyanate method with Trizol (Invitrogen, Carlsbad, CA). RNA quality was determined by NanoDrop™ (Thermo Fisher Scientific). The iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) was used to synthesize cDNA. RT-qPCR amplification was performed using the SYBR-Green method (Bio-Rad, Hercules, CA). PCR arrays were performed using the PAMM-018ZA mouse oxidative stress Pathway RT² Profiler PCR Array (Qiagen, Germantown, MD). This array profiles the expression of 84 oxidative stress-related genes, including glutathione peroxidases and peroxiredoxins, as well as genes involved in reactive oxygen species (ROS) and superoxide metabolism. Data analysis were performed according to manufactures protocol. All studies were conducted in the MMC Molecular Core Laboratory.

2.12. Statistics

Data were presented as the mean ± standard error (SE). Statistical comparisons between groups were performed using the Student’s t-test or Tukey’s test after one-way or two-way analysis of variance (ANOVA) with GraphPad Prism Version 5.0 (GraphPad Software, San Diego, CA). A p-value of less than 0.05 was considered statistically significant. Heat map was generated using R Studio Version 1.2.5001.

3. Results

3.1. Time-dependent effects of HFD on body weight gain, fasting blood glucose levels, GE, and nitrergic relaxation

Young adult HFD-fed female mice started gaining weight at week three and showed incremental weight gain for the remaining experimental period. Their fasting glucose levels increased to 190–200 mg/dL from week 10 onwards and remained elevated for the 16-week experimental period. We began noticing a delay in solid GE at week six (HFD: 65.9 ± 6.3 vs ND: 77.4 ± 3.4%), and the effect was most significant at week 12 and 16 (29.21 ± 4.9 and 16.5 ± 4.8%, respectively) (Fig. 1A). No change in GE was noticed throughout the study in ND group.

Fig. 1.

Effect of HFD on solid GE and nitrergic relaxation of circular gastric antrum strips following EFS (2 Hz). (A) Solid GE at different time points in WT HFD-fed mice. (B) Influence of HFD on nitrergic relaxation at different time points. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. Data are mean ± SEM (n = 6). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared between HFD-fed mice at 6th week vs 12th week; ^$p < 0.05 compared between HFD-fed mice at 12th week vs 16th week.

We next examined if nitrergic relaxation in the gastric antrum, measured by area under the curve/mg tissue, was impaired, and if so, correlated with delayed GE. HFD-fed mice showed a significant decrease (− 0.48 ± − 0.04) in nitrergic relaxation at week six compared with age-matched ND-fed mice (− 0.57 ± − 0.06) (Fig. 1B). Mice at early stages of diabetes as measured by fasting glucose levels at week 12 showed a severe reduction in nitrergic relaxation (− 0.27 ± − 0.04) that continued to decrease during chronic diabetes at week 16 (− 0.15 ± − 0.04).

3.2. CNM/CUR supplementation normalizes the body weight gain, estrus cycle, and blood glucose levels in female mice with obesity-induced chronic diabetes

Our results show that CUR/CNM significantly (p < 0.05) reduced the body weight gain in both WT and Nrf2 KO mice fed with HFD (Table 1). Our data show that CNM/CUR supplementation significantly (p < 0.05) reduced fasting blood glucose levels, HOMA-IR indices, (Table 1) and glucose tolerance in HFD-fed mice (Fig. 2A). In addition, supplementation of CUR/CNM normalized insulin sensitivity in HFD fed mice (Fig. 2B).

Table 1.

Body weight, blood glucose and insulin levels in WT and Nrf2 KO female mice at 16 weeks.

	WT				KO
	ND	HFD	HFD + CNM	HFD + CUR	ND	HFD	HFD + CUR
Body weights (g)	25.6 ± 1.8	45.2 ± 1.1*	27.1 ± 1.9#	33.5 ± 3.6#	27.9 ± 1.01	48.2 ± 3.3*	29.5 ± 1.2#
Body weight gain (g)	3.6 ± 0.6	23.8 ± 4.3*	9.6 ± 1.9#	14.7 ± 2.1#	7.0 ± 1.4	28.6 ± 2.8*	17.8 ± 1.9#
Fasting blood glucose (mg/dL)	110 ± 5.6	274 ± 8.9*	138 ± 10.1#	125 ± 6.6#	148 ± 11.5	303 ± 10.6*	178 ± 12.2#
Fasting insulin (ng/ml)	0.28 ± 0.04	1.1 ± 0.2*	0.8 ± 0.05#	0.7 ± 0.07#	0.5 ± 0.05	1.3 ± 0.17*	1.0 ± 0.09#
HOMA-IR	1.36 ± 0.1	13.39 ± 0.8*	4.9 ± 0.3#	3.8 ± 0.2#	3.3 ± 0.2	17.5 ± 0.8*	7.9 ± 0.4#
Length of estrus cycle (days)	04	09*	06#	05#	05	09*	08
Serum estradiol (ng/L)	35.8 ± 4.4	66 ± 6.5*	42 ± 4.2#	48 ± 2.9#	27.4 ± 2.9	40.6 ± 5.8*	38 ±3.2
Serum nitrate levels (μM)	12.7 ± 1.3	6.10 ± 0.9*	12.6 ± 1.0#	11.9 ± 0.8#	7.4 ± 1.1	6.2 ± 0.7	6.8 ± 1.4

Results are expressed as mean ± SEM.

^*p < 0.05 compared to normal diet.

^#p < 0.05 compared to high fat diet respectively.

Fig. 2.

Effect of CNM/CUR on glucose tolerance and insulin sensitivity in HFD-fed mice. (A) Effects of CNM/CUR on intraperitoneal glucose tolerance test (IPGTT). Shown here is the profile of blood glucose concentration versus time in WT and Nrf2 KO HFD-fed mice. (B) Effects of CNM/CUR on blood glucose concentration (percentage of initial value) as a function of time upon intraperitoneal injection of insulin in WT and Nrf2 KO HFD-fed mice. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. The values are mean ± SE (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared with HFD-fed mice.

Estrus cycle assessment revealed similar abnormalities in the cycling phases and cycle length, as reported in our earlier study [23]. CNM/CUR supplementation restored cycle predominance as well as decreased cycle length only in WT but not in Nrf2 KO mice (Table 1).

3.3. CNM/CUR reversed obesity/diabetes-induced delay in GE

Next, we investigated whether CNM/CUR could restore GE in HFD-induced chronic diabetes. As shown in Fig. 3A, HFD-fed WT mice showed a significant delay (p < 0.05) in solid GE compared to ND-fed WT mice (17.5 ± 2.3% vs 81.8 ± 2.4%, p < 0.05). CNM/CUR supplementation restored GE (CNM: 54.5 ± 4.9% and CUR: 69.6 ± 8.4% vs 17.5 ± 2.3%, p < 0.05) in WT HFD-fed mice almost to the level in ND-fed mice (Fig. 3A). On the other hand, CUR supplementation only partially restored GE (25.3 ± 4.5% vs 13.5 ± 3.3%) but not significantly in Nrf2 KO HFD-fed mice (Fig. 3B). As reported earlier, WT-ND supplemented with CNM/CUR did not show any significant change in GE when compared with WT-ND [23].

Fig. 3.

Effect of CNM/CUR on solid GE and nitrergic relaxation of circular gastric antrum strips following EFS (2 Hz). (A) Effects of CNM/CUR on solid GE at 16 weeks in WT HFD-fed mice. (B) Effects of CUR on solid GE at 16 weeks in Nrf2 KO HFD-fed mice. (C) Effect of CNM/CUR on nitrergic relaxation at 16 weeks in WT HFD-fed mice. (D) Effect of CUR on nitrergic relaxation in Nrf2 KO HFD-fed mice. WT and Nrf2 KO female mice fed with ND served as controls. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. The values are mean ± SE (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared with HFD-fed mice.

3.4. CNM/CUR attenuated obesity/diabetes-induced impairment of gastric nitrergic relaxation

Impairment of gastric nitrergic relaxation was observed in WT HFD-fed mice compared to WT-ND (− 0.22 ± − 0.03vs − 0.60 ± − 0.04, p < 0.05). CNM and CUR restored nitrergic relaxation in HFD-fed WT mice (− 0.39 ± − 0.02 and − 0.51 ± − 0.03), respectively, compared to HFD (− 0.22 ± − 0.03) (Fig. 3C). However, CUR did not restore nitrergic relaxation in HFD-fed Nrf2 KO mice (− 0.15 ± − 0.03 vs − 0.16 ± − 0.03) (Fig. 3D). Our data shows that non-selective NOS inhibitor L-NAME reduced relaxation more than 80% in gastric tissue used suggesting that NANC mediated stomach motility exist among all experimental groups.

3.5. CNM/CUR restored gastric Nrf2 protein expression in obesity-induced chronic diabetes

We have previously shown that loss of Nrf2 delays GE and disrupts nitrergic relaxation [18,23]. Here, we demonstrated that CNM/CUR supplementation restored total, cytoplasmic, and nuclear Nrf2 expression in the gastric antrum of obese/diabetic WT mice (Fig. 4). Specifically, we detected marked Nrf2 accumulation in the nucleus after treatment with CNM and CUR. The absence of nuclear marker lamin B in the cytoplasmic fraction eliminated the possibility of cross-contamination.

Fig. 4.

Nrf2 protein expression in the nuclear fractions of gastric antrum specimens. (Top panel) Immunoblots for the indicated proteins. (A–C) Densitometry analysis of immunoblots for total Nrf2 (A), cytoplasmic Nrf2 (B), and nuclear Nrf2 (C) in HFD-fed WT female mice. GAPDH and lamin B were used as markers and loading controls for the cytoplasmic and nuclear fractions, respectively. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. Values are mean ± SE (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared with HFD-fed mice.

3.6. CNM/CUR restored AKT, extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38 mitogen-activated protein kinase (MAPK) signaling as well as GSK-3β expression in gastric tissues

ERK, JNK, and p38 have been shown to transduce extracellular signals to the nucleus [36–38]. Nrf2 has been shown to regulate antioxidant and anti-inflammatory pathways by activating ERK/JNK/p38 signaling [36]. Therefore, we next investigated whether these upstream kinases were altered in the obese/diabetic mice; and if so, would supplementation with CNM/CUR restore their levels in gastric antrum specimens. As shown in Fig. 5A–D, gastric antrum neuromuscular tissues from obese WT mice with chronic diabetes exhibited a significant decrease (p < 0.05) in the levels of p-ERK, p-JNK, p-AKT, and p38/MAPK. Supplementation with CNM or CUR significantly attenuated (p < 0.05) diabetes induced changes among these proteins. On the other hand, gastric antrum neuromuscular tissues from Nrf2 KO mice exhibited less phosphorylation in HFD fed mice when compared to Nrf2 KO ND fed mice. CUR supplementation failed to restore these proteins in the Nrf2 KO HFD-fed mice (Fig. 5E–H).

Fig. 5.

Effect of CNM/CUR on the expression of MAPK signaling proteins in gastric antrum specimens of WT and Nrf2 KO female mice. Representative immunoblots and densitometric analysis data for the following in gastric neuromuscular tissues: p-AKT, p38 MAPK, p-JNK, and p-ERK in WT mice (A—D), p-AKT, p38 MAPK, p-JNK, and p-ERK in Nrf2 KO mice (E–H), p-GSK-3β in WT mice (I), and p-GSK3β in Nrf2 KO mice (J). Blots showing same β-actin were stripped and reprobed. Data were normalized with band intensities for β-actin. Bar graphs depict ratios of target proteins to β-actin. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. Values are mean ± SE (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared with HFD-fed mice.

Since p-AKT has been shown to inhibit GSK-3β via phosphorylation at Ser9, it is possible that AKT activation and GSK-3β inactivation up-regulates Nrf2 [39]. Our results show that supplementation of CNM/CUR promoted GSK-3β phosphorylation in gastric neuromuscular tissues from diabetic WT mice but not in Nrf2 KO diabetic mice (Fig. 5I and J).

3.7. Effects of CNM/CUR on obesity/diabetes-induced oxidative stress

The reduced/oxidized glutathione (GSH/GSSG) ratio is a well-known marker for oxidative stress [40]. Here, we observed a significantly lower GSH/GSSG ratio in the gastric antrum neuromuscular specimens (p < 0.05) of HFD-fed WT mice (Fig. 6A). Supplementation with CMN/CUR significantly increased the GSH/GSSG ratio (p < 0.05 respectively) in WT HFD-fed mice (Fig. 6A). However, CUR supplementation had no significant effect in HFD-fed Nrf2 KO mice (Fig. 6B).

Fig. 6.

Effect of CNM/CUR on gastric antrum GSH/GSSG ratio, oxidative stress/anti-oxidative profile and proinflammatory cytokines in HFD-fed mice. (A) GSH/ GSSG ratio in WT mice, (B) GSH/GSSG ratio in Nrf2 KO mice, (C) Heat map of differentially expressed oxidative stress and anti-oxidative genes in WT-HFD-fed mice and HFD-fed mice receiving CNM/CUR supplementation. Data are presented as the average fold change of four mice per group at 16-week time point analyzed. Relative expression values from high to low was shown by gradient of red to green in the heat map. Red indicate higher and green indicating lower expression of mRNAs. (D,E) IL-1β, IL-6, TNF-α, and TLR-4 protein levels in gastric antrum neuromuscular specimens of WT (F,G) IL-1β, IL-6, TNF-α, and TLR-4 protein levels in gastric antrum neuromuscular specimens of Nrf2 KO mice. Blots showing same β-actin were stripped and reprobed. Data were normalized with band intensities for β-actin. Bar graphs depict ratios of target proteins to β-actin. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. Values are mean ± SE (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared with HFD-fed mice. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.8. Identification of oxidative stress and anti-oxidative responsive transcripts

Transcription factor Nrf2 reduces oxidative stress by elevating the expression of antioxidant genes in addition to suppressing the expression of pro-inflammatory cytokines [36]. The down-regulated antioxidative genes and up-regulated oxidative stress genes could lead to the progression of oxidative stress, and thus contribute towards diabetic gastroparesis. In this study, we screened for a list of genes whose expression is responsive to oxidative stress and antioxidant defense mechanism using RT² profiler qPCR array for mouse oxidative stress and antioxidant defense profiler. Our results confirmed that the genes of which the expression was altered are critically involved in oxidative stress and antioxidant defense mechanism in obesity-induced diabetic mice (Fig. 6C). As shown in Table 2, out of 84 genes, 14 genes responsive to oxidative stress were altered by more than 1.5 fold (p < 0.05) in obesity-induced diabetic mice (WT-HFD) compared to WT-ND mice. Supplementation of CNM/CUR suppressed (p < 0.05) some but not all of the oxidative stress genes (thioredoxin reductase 1, proteasome [pro-some, macropain] beta type 5, myeloperoxidase, isocitrate dehydrogenase 1 (NADP⁺), soluble reduced nicotinamide adenine dinucleotide phosphate dehydrogenase, quinone 1). Genes involved in antioxidant defense were downregulated with significant p-values in obesity-induced diabetic mice (WT-HFD) compared to WT-ND mice (Table 3). Supplementation with CNM/CUR restored the expression of most of the antioxidant genes in gastric antrum specimens.

Table 2.

Effects of CNM/CUR on oxidative stress markers in obesity induced chronic diabetic female mice gastric tissues.

Sl. No	Gene symbol	Gene	(HFD vs ND)		(HFD + CNM vs HFD)		(HFD + CUR vs HFD)
			Av. fold change	p-value	Av. fold change	p-value	Av. fold change	p-value
1	NOX 1	Nicotinamide adenine dinucleotide phosphate oxidase 1	1. 9*	0.04	0.9	0.1	1.1	0.4
2	Prdx2	Peroxiredoxin 2	1.5*	0.3	1.1	0.8	1.2	0.3
3	Ercc2	Excision repair cross-complementation group 2	2.1*	0.05	1.2	0.35	1.1	0.6
4	Txnip	Thioredoxin binding protein	2.0*	0.006	2.8#	0.0001	2.9#	0.0006
5	Als2	Amyotrophic lateral sclerosis 2 (juvenile) homolog (human)	2.7*	0.02	3.0#	0.02	2.1#	0.001
6	Ctsb	Cathepsin B	2.4*	0.0005	1.7#	0.002	0.9	0.4
7	Epx	Eosinophil peroxidase	3.5*	0.0001	2.6#	0.006	2.1#	0.001
8	Idh1	Isocitrate dehydrogenase 1 (NADP+), soluble	1.7*	0.03	1.2	0.6	1.2	0.9
9	Krt1	Keratin 1	1.5*	0.05	1.2	0.4	1.3	0.7
10	Mpo	Myeloperoxidase	3.4*	0.009	2.6#	0.008	2.7#	0.007
11	Psmb5	Proteasome (prosome, macropain) subunit, beta type 5	1.8*	0.02	1.3	0.8	1.2	0.4
12	NOX 4	Nicotinamide adenine dinucleotide phosphate oxidase 4	1.5*	0.006	2.9#	0.007	0.8	0.8
13	Txnrd1	Thioredoxin reductase 1	3.9*	0.05	0.9	0.2	1.7#	0.01
14	Nqo1	Nicotinamide adenine dinucleotide phosphate dehydrogenase, quinone 1	1.9*	0.03	1.8#	0.02	1.1	0.2

Results are expressed as average fold change (n = 4).

^*p < 0.05 compared to normal diet.

^#p < 0.05 compared to high fat diet respectively.

Table 3.

Effect of CNM/CUR on anti-oxidative genes in obesity induced chronic diabetic female mice gastric tissues.

Sl. No	Gene symbol	Gene	(HFD vs ND)		(HFD + CNM vs HFD)		(HFD + CUR vs HFD)
			Av. fold change	p-value	Av. fold change	p-value	Av. fold change	p-value
1	Gpx1	Glutathione peroxidase 1	0.7*	0.002	1.7#	0.01	1.5#	0.08
2	Gpx2	Glutathione peroxidase 2	0.6*	0.003	1.7#	0.005	1.7#	0.02
3	Gpx3	Glutathione peroxidase 3	0.5*	0.0002	1.8#	0.01	1.7#	0.07
4	Gpx4	Glutathione peroxidase 4	0.4*	0.001	2.1#	0.0002	2.3#	0.001
5	Gpx5	Glutathione peroxidase 5	0.6*	0.01	2.0#	0.01	2.1#	0.02
6	Gpx6	Glutathione peroxidase 6	0.5*	0.02	2.1#	0.01	2.0#	0.004
7	Gpx7	Glutathione peroxidase 7	0.5*	0.002	2.4#	0.01	2.4#	0.002
8	Gstk1	Glutathione S-transferase kappa 1	0.5*	0.02	1.7#	0.01	1.8#	0.008
9	Ehd2	EH-domain containing 2	0.5*	0.01	1.7#	0.008	1.8#	0.005
10	Gss	Glutathione synthetase	0.6*	0.005	2.0#	0.02	1.8#	0.009
11	Prdx5	Peroxiredoxin 5	0.6*	0.005	1.6	0.1	2.2#	0.0003
12	Prdx6	Peroxiredoxin 6	0.6*	0.01	1.9#	0.005	2.0#	0.01
13	Apc	Adenomatosis polyposis coli*	1.4*	0.0008	1.9#	0.03	0.8	0.1
14	Cat	Catalase	0.6*	0.0006	1.9#	0.003	2.1#	0.0006
15	Lpo	Lacto peroxidase*	1.4*	0.0005	1.7#	0.01	1.7#	0.001
16	Ptgs2 (COX2)	Prostaglandin-endoperoxide synthase 2	1.4*	0.001	1.7#	0.0003	1.8#	0.03
17	Tpo	Thyroid peroxidase	0.6*	0.01	1.9#	0.01	1.7#	0.01
18	Gsr	Glutathione reductase	0.6*	0.004	2.0#	0.002	2.1#	0.007
19	Sod1	Superoxide dismutase 1, soluble	0.5*	0.002	1.9#	0.002	2.0#	0.004
20	Sod2	Superoxide dismutase 2, mitochondrial	0.6*	0.001	1.8#	0.001	1.9#	0.0003
21	Sod3	Superoxide dismutase 3, extracellular	0.5*	0.02	1.9#	0.02	1.8#	0.03
22	Cyba	Cytochrome b-245, alpha polypeptide	0.7*	0.06	1.5#	0.03	1.5#	0.05
23	Nox4	NADPH oxidase 4	1.3*	0.006	2.7#	0.007	0.9	0.9
24	Ucp2	Uncoupling protein 2 (mitochondrial, proton carrier)	0.8*	0.02	1.1	0.1	1.5#	0.04
25	Dnm2	Dynamin 2	0.6*	0.009	1.5#	0.0001	0.9	0.6
26	GcLc	Glutamate-cysteine ligase, catalytic subunit	0.5*	0.002	2.1#	0.0008	1.7#	0.01
27	Gclm	Glutamate-cysteine ligase, modifier subunit	0.5*	0.0003	2.2#	0.0005	1.6#	0.01
28	Hsp90ab 1	Heat shock protein 90 alpha (cytosolic), class B member 1	0.5*	0.0003	1.9#	0.01	1.9#	0.007
29	CCl5	Chemokine (C-C motif) ligand 5	0.6*	0.0008	1.8#	0.02	1.9#	0.0008
30	Prnp	Prion protein	0.7*	0.02	1.9#	0.001	1.8#	0.004
31	Hmox1	Heme oxygenase 1	0.7*	0.07	1.5#	0.01	1.7#	0.007
32	Il22	Interleukin 22*	1.4*	0.0001	1.6#	0.01	1.7#	0.01
33	Fmo2	Flavin containing monooxygenase 2*	1.3*	0.01	1.6#	0.008	1.5#	0.02
34	Il19	Interleukin 19*	1.2*	0.002	1.8#	0.004	1.7#	0.03

Results are expressed as average fold change (n = 4).

^*p < 0.05 compared to normal diet.

^#p < 0.05 compared to high fat diet respectively.

3.9. Anti-inflammatory effect of CNM and CUR in obesity-induced chronic diabetic gastric antrum specimens

Nrf2 is known to play a role in regulating the expression of pro- and anti-inflammatory genes in diabetes [36]. Our data show that enhanced the expression of tumor necrosis factor-alpha (TNF-α), interleukin (IL)—1beta, IL-6, and toll like receptor-4 (TLR-4) in WT HFD fed mice (Fig. 6D and E). Nrf2-KO HFD fed mice showed higher expression of TNF-α, IL-1β, IL-6, and TLR-4 when compared Nrf2-KO mice fed on ND. Supplementation with CNM/CUR significantly reduced the expression of inflammatory cytokines in HFD-fed WT mice, but not in Nrf2 KO HFD-fed mice (Fig. 6F and G).

3.10. CNM and CUR suppressed the expression of aryl hydrocarbon receptor (AhR) and enhanced that of estrogen receptors (ERs) in the gastric

Activation of AhR is linked to major health conditions including obesity and also inhibits the expression of estradiol-17β (E₂)-induced genes [41,42]. Here, we investigated the expression of AhR in gastric antrum neuromuscular specimens of HFD-fed WT and Nrf2 KO female mice. Our results demonstrated that AhR expression was significantly higher in HFD-fed mice. Supplementation with CNM/CUR substantially attenuated AhR expression in the gastric antrum of WT mice, but not significantly in that of Nrf2 KO HFD-fed mice (Fig.7A and B). Previous studies have shown that p38/MAPK and AhR may contribute to E₂-mediated ER binding and function [23,43,44]. As shown in Fig. 7, the expression of both gastric ERα and ERβ was suppressed in HFD-fed mice. CNM/CUR supplementation significantly (p < 0.05) restored both ERβ and ERα expression in WT but not in Nrf2 KO HFD-fed mice (p < 0.05) (Fig. 7 A and B). Although ERα expression showed an increase by CUR in Nrf2-KO, it was not statistically significant.

Fig. 7.

Effects of CNM/CUR on the expression of AhR and ERs in the gastric antrum specimens of WT and Nrf2 KO female mice. Representative immunoblots and densitometric analysis data for AhR, ERα, and ERβ in the gastric neuromuscular tissues of WT (A) and Nrf2 KO (B) female mice. Blots showing same β-actin were stripped and reprobed. Data were normalized with band intensities for β-actin. Bar graphs depict ratios of target proteins to β-actin. Data were analyzed using one-way and two-way ANOVA by using graph pad prism software. The values are mean ± SEM (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < compared with the HFD-fed mice.

3.11. CNM/CUR supplementation normalized nNOSα protein expression and dimerization in the gastric antrum of WT-HFD mice but not in that of Nrf2 KO mice

As depicted in Fig. 8A, nNOSα expression was significantly reduced in chronically obese/diabetic female mice (p < 0.05); whereas CNM/CUR supplementation enhanced nNOSα expression in HFD-fed WT but not in HFD-fed Nrf2 KO mice. To measure whether the decrease in nNOSα expression was due to altered nNOS dimer levels (an indirect measurement for enzyme activity), we investigated dimerization using a LT-PAGE gel. We observed a significant decrease in the dimer/monomer ratio in both WT and Nrf2 KO HFD-fed mice (p < 0.05) (Fig. 8B). CNM/CUR supplementation restored nNOSα dimerization in HFD-fed WT but not in HFD-fed Nrf2 KO mice (Fig. 7B).

Fig. 8.

Effect of CNM and CUR on the expression of NO signaling proteins in gastric antrum specimens. Representative immunoblots and densitometric analysis data for the following in the gastric antrum neuromuscular specimens: nNOSα in WT and Nrf2 KO nice (A), nNOS α dimerization in WT and Nrf2 KO mice (B), GCH-1 in WT and Nrf2 KO mice (C), sGCα and sGCβ in WT (D) and sGCα and sGCβ in Nrf2 KO (E) mice. Blots showing same β-actin were stripped and reprobed. Data were normalized with band intensities for β-actin. Bar graphs depict ratios of target proteins to β-actin. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. The values are mean ± SEM (n = 4). *p < 0.05 compared with ND-fed mice; ^#p < 0.05 compared with the HFD-fed mice.

3.12. CNM/CUR attenuated diabetes-induced decrease in GCH-1 expression

GCH-1 plays a role in BH₄ biosynthesis via a de novo pathway [45]. As depicted in Fig. 8C, obesity-induced diabetes significantly reduced GCH-1 expression in WT but not in Nrf2 KO HFD-fed mice (p < 0.05). Supplementation with CNM/CUR restored GCH-1 expression in WT but not in Nrf2 KO HFD-fed mice.

3.13. CNM/CUR attenuated reduced expression of soluble guanylate cyclases (sGCs), which synthesize second messenger cyclic guanosine monophosphate (cGMP) in NO signaling

The downstream effects of NO on gut smooth muscle cells include the activation of soluble guanylate cyclase (sGC) and production of cGMP, which in turn stimulates protein kinase G (PKG) and results in smooth muscle relaxation due to a decrease in intracellular calcium concentration. Here, we found sGCα and sGCβ levels were significantly (p < 0.05) lower in the gastric neuromuscular tissues from HFD-fed mice than in those from ND-fed mice (Fig. 8D and E). Supplementation with CNM/CUR restored sGCβ protein expressions in WT but not in Nrf2 KO HFD-fed mice (Fig. 8C and D). Interestingly supplementation with CNM but not CUR restored sGCα in WT HFD-fed mice.

3.14. nNOS as a novel Nrf2-interacting protein

Using immunoprecipitation with anti-nNOS antibody followed by immunoblotting with anti-Nrf2 antibody, we found that obesity-induced diabetes significantly (p < 0.05) reduced the level of the nNOS-Nrf2 complex, indicating little or no interaction. We also observed that CNM/CUR supplementation significantly (p < 0.05) increased the levels of nNOS-Nrf2 complex both in total lysates and nuclear fractions, indicating strong interaction between Nrf2 and nNOS (Fig. 9). This complex may be key to restoring nitrergic neuron mediated GE in obesity-induced diabetic female mice.

Fig. 9.

Physical interaction between nNOSα and Nrf2 in gastric antrum specimens. Total tissue lysates (A) and nuclear fractions (B) were subject to immunoprecipitation with anti-nNOSα antibody and then used for immunoblotting with anti-Nrf2 antibody. The experiment was performed in triplicate.

4. Discussion

Obese patients with T2D experience gastric discomfort, gastro-paresis, and intestinal dysmotility due to disruption in the enteric nervous system [46]. Natural dietary bioactive compounds are being explored for management and prevention of diabetes because of their abundant availability, efficacy and possess less or no side effects. Several of these phytonutrients have exhibited anti-obesity and anti-hyperglycemia properties in pre-clinical and clinical screenings [47]. CNM/CUR have demonstrated their ability in preventing inflammation, oxidative stress and attenuating diabetes or diabetes-related complications [48–51]. However, very few studies have examined their ability to protect against DG. Our results suggest that CNM/CUR treatment restored delayed gastric emptying in obese/T2DM.

Here, we demonstrate that female mice developed symptoms of obesity by week four and early diabetes by week 10 fed on HFD. These findings are in agreement with those reported by Stenkamp-Strahm et al. in male mice [46]. In addition, our data show that changes in body weights and fasting blood glucose levels were associated with impaired gastric nitrergic relaxation and delayed GE. Our study demonstrates that HFD caused a delay in GE and an impairment in nitrergic relaxation at the onset of obesity (week six) as well as in early (week 12) and chronic (week 16) diabetes. CNM/CUR supplementation normalized body weight gain, regulated blood glucose levels both in WT and Nrf2 KO and restored gastric nitrergic relaxation, and normalized GE in WT but not in Nrf2 KO mice suggesting the importance of Nrf2 on nitrergic mediated GE. Finally, our findings demonstrate that CNM/CUR supplementation restored GE in obese/diabetic female mice by suppressing gastric inflammatory cytokines and oxidative stress through a Nrf2-mediated pathway.

Gender differences in GE have been reported in both humans and rodents [12]. Endogenous sex steroid hormones may have a significant influence in GE in both humans and rodents [8]. Recent studies from our laboratory have shown that female mice fed with HFD for 12 weeks exhibited abnormalities in estrus cycle, higher levels of serum estrogens and reduced gastric antrum ERs. Supplementation with CNM for an entire 12 weeks effectively regulated the estrus cycle as well as restored levels of serum estrogens and ERs expression [23]. Our current study further shows that serum estrogen levels were elevated and the expression of ERs (α and β) was reduced in the gastric antrum of 16-week HFD-fed female mice. CNM/CUR supplementation attenuated the gastric ERs expression and normalized the estrus cycle as well as serum estrogen and nitrite levels in WT but not in Nrf2 KO mice. As a result, the WT-HFD mice showed improvement in GE. Collectively, our findings suggest that Nrf2 activation restores GE by restoring expressions of gastric ERs in obese/diabetic female mice.

We have previously shown that Nrf2 deletion reduced nitrergic relaxation and delayed GE in mice [18,23]. Recently, we demonstrated that treatment with Nrf2 activator CNM supplemented from day one of HFD significantly improved nNOS function and GE in WT mice fed with HFD for 12 weeks by restoring gastric Nrf2 expression [23]. This finding suggests that Nrf2 plays an important role in nitrergic-mediated gastric motility. In addition, there are other factors involved such as vasoactive intestinal polypeptide (VIP) and purinergic with NANC induced nitrergic relaxation which needs to be explored with Nrf2 activators [52,53]. In this study, we tested the hypothesis that supplementation of Nrf2 activators (CNM and CUR) at later stages of HFD ingestion may restore nitregic neuron mediated gastric motility and GE. Our data demonstrate that CUR/CNM supplementation started at the 6th and 10th week after HFD treatment in WT-HFD, restored Nrf2 expression in the gastric antrum specimens of WT but not Nrf2-KO female mice, thus promoting normal gastric motility and GE. Nrf2 translocation from the cytoplasm to the nucleus triggers a stress-induced defense mechanism. In HFD-fed mice, Nrf2 remains bound to Kelch-like ECH-associated protein 1 (Keap1) in the cytoplasm and undergoes proteosomal degradation. On the other hand, CNM/CUR supplementation leads to Nrf2 dissociation from Keap1 and Nrf2 translocation to the nucleus. Nuclear translocation of Nrf2 results in activation of Nrf2 target genes that are responsible for cytoprotection and maintenance of the cellular homeostasis [54]. Nrf2 activators act on members of MAPK family: ERK, JNK, and p38 [55]. In this study, we observed that activation of ERK/JNK and p38/MAPK signaling only in HFD-fed WT mice supplemented with CNM/CUR but not in Nrf2 KO HFD-fed mice. Therefore, we conclude that CNM and CUR activate ERK/JNK and p38/MAP kinase signaling, which are required for the Nrf2-mediated pathway.

Gene expression varies more considerably across organs within the same species than across species, and tissue-specific genes are often associated with tissue-specific diseases [56]. Animals and humans have adopted highly efficient and adaptive antioxidant defense mechanisms, including the use of anti-oxidative enzymes such as superoxide dismutase (SOD), catalase (CAT) and non-enzyme antioxidant molecules, as well as the GSH oxidation/reduction system [57]. In the present study, we conducted a microarray experiment to investigate the oxidative stress-related transcriptome profile using mRNA samples isolated from the gastric antrum neuromuscular tissues of WT female mice. We observed lower expression of specific anti-oxidative genes in HFD-fed mice with obesity-induced diabetes. The expression of catalase, which catalyzes the removal of hydrogen peroxide from biological tissues, is relatively lower in the gastric tissues [57]. Nrf2 regulates detoxification and antioxidant genes such as Nqo1, HO-1, TixR1, Psmb5, Mpo, and Idh1. Our data shows that disruption of Nrf2 signaling significantly increases HFD-induced oxidative stress. Nqo1 is a major quinone reductase enzyme attenuates oxidative stress under diabetic conditions regulated through Nrf2. Nqo1 acts as a superoxide reductase, modulate the NAD⁺/NADH redox balance, and stabilizes p53. Nqo1 is stimulated by xenobiotic responsive elements (XRE) induced by AhR agonists and Nrf2 activators [58]. Interestingly, XRE and ARE possess promoters of Nqo1 gene hence, there is an overlap between Nrf2 and AhR target genes [58]. Strong evidence indicates that AhR activation leads to oxidative stress [58]. Then AhR translocate into the nucleus forming ARNT complex that binds to XRE in the promoter regions of Nqo1 during oxidative stress [58]. Induction of Nqo1 is shown to be upregulated in WT HFD-fed mice in our current study. The oxidative stress genes were highly upregulated in WT HFD-fed mice with obesity-induced diabetes. Most importantly, CNM/CUR supplementation restored the expression of these anti-oxidative genes. Interestingly, not all of these were suppressed with CNM/CUR supplementation. Furthermore, we speculate that there may be a compensatory mechanism towards gastric oxidative damage. Hence, further studies need to be validated and the potential role of these genes with Nrf2 pathway needs to be explored.

Next, we showed that Nrf2 deficiency in obesity-induced diabetes caused significant delayed GE due to increased levels of inflammatory cytokines and oxidative stress in HFD-fed mice [18,23,59]. Oxidative stress and inflammation are common in obese and diabetic rodents and humans. Obesity and diabetes are known to promote low-grade systemic inflammation, at least in part, by upregulating the production of inflammatory cytokines in the adipose tissue [60]. Diabetes is caused by impaired utilization of insulin resulting from insulin resistance, which is increasingly being characterized as an inflammatory disease [61–63]. Higher levels of circulating cytokines and chemokines have been reported in patients with T2D [64]. Circulating inflammatory cytokines derived from adipose tissue readily reach and damage gastric mucosal lining, causing digestive motor abnormalities [65,66]. In the obesity and prediabetes stage prior to the onset of T2D, high levels of IL-1 receptor antagonist (IL-1RA), inflammatory cytokines (IL-1β, IL-6, TNF-α), and C-reactive protein are predictive of T2D [64]. We previously reported that in mice fed with HFD for 12 weeks, levels of circulatory cytokines were elevated and supplementation with CNM normalized their levels [23]. Here, we show that organ-specific expression of cytokines such as IL-1β, IL-6, and TNF-α was higher in the gastric antrum of HFD-fed mice. The regulation of inflammation by TLR signaling leads to the production of TNF-α, IL-6, chemokines (IL-8 and MIP2), and interferon (type-I) [22].

Studies have shown that TLR-2 and TLR-4 levels are elevated in patients with T2D [67]. In fact, TLR-4 has been reported to play an important role in the pathogenesis of atherosclerosis, diet-induced obesity, and insulin resistance [68]. In the current study, we observed higher gastric TLR-4 expression in HFD-fed WT mice; CNM/CUR supplementation-reduced TLR-4 expression. TLR-mediated innate immune responses and Nrf2-modulated antioxidant system coordinate in various ways to regulate the cross-talk between Nrf2 and TLR [22]. Chao et al., demonstrated that in-vitro supplementation of CNM inhibited TLRs induced activation of cytokine expression in human blood monocytes derived primary macrophages, suggesting that CNM is a potential anti-inflammatory candidate [69]. Our results show that CNM/CUR supplementation elevated ERK/JNK and p38/MAPK signaling. This signaling enhancement correlated with reduced levels of inflammatory cytokines and enhanced levels of antioxidants, resulting in reduction of oxidative stress. CUR has been shown in a randomized, double-blind, crossover trial to suppress IL-4, IL-6, transforming growth factor beta (TGF-β), monocyte chemoattractant protein 1 (MCP-1), and TNF-α [70–72]. CD206-positive M2 macrophages that generate TNF-α have been shown to be important in preventing diabetic gastroparesis through a heme oxygenase-dependent mechanism [73–75]. These changes were not effective in Nrf2 KO HFD-fed mice supplemented with CUR.

GSK-3β phosphorylates Nrf2, which results in the nuclear exclusion and degradation of Nrf2 [76]. This form of Nrf2 regulation switches off the self-protective antioxidant stress response [76] and downregulates the antioxidant defense elicited by Nrf2 [39,77]. We previously detected reduced expression of p-GSK-3β in the gastric tissues of Nrf2 KO mice as well as in the HFD-fed WT mice [23]. It has been well-established that a cross-talk exists between GSK-3β and Nrf2 [76]. Our current study demonstrates that CNM/CUR supplementation activated Nrf2 as well as p-AKT and p-GSK3β expression in WT but not in Nrf2 KO mice. Therefore, we conclude that Nrf2 upregulation by Nrf2 activators leads to activation of AKT, which in turn mediates GSK-3β phosphorylation at Ser9 to inhibit GSK-3β activity.

AhR is a ligand-dependent transcription factor that influences the development of obesity and metabolic disorders. AhR inhibition has been shown to prevent obesity and the incidence of fatty liver in both female and male mice [78]. Activated AhR inhibits ER activity through several different mechanisms [79]. In this study, we demonstrate that CNM/CUR supplementation reduced gastric AhR levels, subsequently enhancing ER expression and protecting WT female mice from obesity-induced diabetes and ensuing metabolic complications. CNM has been shown to inhibit AhR signaling and induce Nrf2-mediated antioxidant activity in human keratinocytes [80]. Estrogen has been shown to cause relaxation of gastric and colon smooth muscles [81,82].

Additionally, estrogen induces smooth muscle relaxation via a process involving the activation of NO/cGMP signaling [82]. As reported earlier, the crosstalk between Nrf2 and AhR exists while Nrf2 controls AhR signaling [83,84]. In addition, AhR also modulates its function on estrogen receptors (ERα and ERβ) [85]. Our data highlight the fact that supplementation of CNM/CUR reduces gastric AhR, thereby enhancing the expression of both ERs.

nNOS activity represents a critical signaling node for regulating gastric motor function. nNOS catalyzes the formation of NO, which initiates smooth muscle relaxation. nNOS activity in turn is regulated by BH₄, a cofactor for nNOS dimerization and enzyme activity that couples electron flow to generate NO in the smooth muscle [86]. Since Nrf2 loss affects BH₄ levels, it is also shown to affect nNOS function [18]. We have previously shown that CNM supplementation from day one of HFD for 12 weeks restored gastric nNOSα levels and nNOSα dimerization in HFD-fed WT mice but not in HFD-fed Nrf2 KO mice [23]. In addition, CNM supplementation restored the expression of GCH-1 (de novo enzyme for BH₄ biosynthesis) in HFD-fed WT mice [23]. We have previously shown the reduction of intracellular BH₄:BH₂ ratio to be the molecular trigger for NO insufficiency in diabetics [87]. In addition, Xue et al. proposed that GCH-1 activation by Nrf2 increases BH₄ levels and inhibits NOS uncoupling, subsequently reducing ROS generation [45]. Relaxation of smooth muscle in the gastrointestinal tract occurs due to the cGMP-mediated inhibitory actions of NO [88]. One factor involved in the inactivation of NO-cGMP-PKG signaling is oxidative stress, which affects NO bioavailability as well as the ability of NO to activate sGC [89]. Our data shows that supplementation of CNM/CUR at later stages of obesity or T2DM restored altered expression of GCH-1, sGCα, sGCβ protein and restored NO production in WT fed with HFD. In contrast, CUR is unable to restore these pathways in Nrf2-KO mice suggesting that Nrf2 may be playing a role in NO mediated gastric emptying in HFD fed female mice. Previously we have reported that reduced nNOS dimerization is caused by heme (cofactor for nNOS) [88]. This heme is oxidized to hemin induce dimerization of DiGeorge Syndrome Critical Region Gene 8 (DGCR8) stimulating miR-28 biogenesis [90]. miRNA 38 thus in turn targets the 3′ untranslated region (3′UTR) of Nrf2 mRNA and decreases Nrf2 expression [90]. Therefore we speculate that CNM/CUR supplementation may inhibit miR 28 biogenesis enhances Nrf2 expression and nNOS dimerization, to thus regulate nNOS-mediated motility of the stomach. This clearly a field of interest that warrants further investigation.

Nrf2 deficiency has been shown to cause reduced nNOS activity and loss of NO generation, which then contribute to the development of gastroparesis [91]. In this study, we investigated the formation of the nNOS-Nrf2 complex. Our data demonstrate that in female mice with obesity-induced diabetes, nNOS-Nrf2 complex formation was restored with supplementation of CNM/CUR. Collectively, the above findings indicate that Nrf2 KO female mice suffered from a defect in nitrergic relaxation and GE, both of which are prevalent in obesity-induced diabetes.

In summary, our current study demonstrates that CNM/CUR supplementation at later stages of obesity and diabetes condition effectively restored gastric nNOSα dimerization by elevating BH₄ synthesis enzyme GCH-1 in WT female mice. Furthermore, CNM/CUR reversed hyperglycemia-induced impairment of gastric nitrergic relaxation by activating p38/MAPK/AKT/ERK/JNK signaling, ER’s and by suppressing GSK-3β, AhR, inflammation and oxidative stress. Collectively, these findings suggest that Nrf2 activation by CNM/CUR enhances the expression of GCH-1 and nNOS dimerization, subsequently normalizing GE in female mice with obesity-induced chronic diabetes.

5. Conclusions

Diabetic gastroparesis is an extremely complex and multifaceted disorder. These results point out potential naturally occurring bioactive compounds (CNM/CUR) alleviated symptoms of obesity-induced chronic T2D in a Nrf2-mediated pathway. One possible mechanism in controlling diabetic gastroparesis is through the modulation of key detoxifying enzymes via Nrf2-mediated gastric ER and nNOS function. Furthermore, Nrf2 activation restored GE by suppressing pro-inflammatory cytokines and oxidative stress markers in the gastric tissue of female mice with obesity-induced diabetes. Taken together, our findings highlight the importance of using CNM and CUR to treat and/or manage diabetic gastroparesis. These compounds are potential candidates in clinical studies that screen for pharmaconutrition agents that can alleviate diabetic gastroparesis.