Nrf2 attenuates hyperglycemia-induced nNOS impairment in adult mouse primary enteric neuronal crest cells and normalizes stomach function

Abstract

Enteric neuronal cells play a vital role in gut motility in humans and experimental rodent models. Patients with diabetes are more vulnerable to gastrointestinal dysfunction due to enteric neuronal degeneration. In this study, we examined the mechanistic role and regulation of nuclear factor-erythroid 2-related factor 2 (Nrf2) in hyperglycemia-induced enteric neuronal cell apoptosis in vitro by using adult mouse primary enteric neuronal crest cells (pENCs). Our data show that hyperglycemia (HG) or inhibition of Nrf2 induces apoptosis by elevating proinflammatory cytokines, reactive oxygen species (ROS) and suppresses neuronal nitric oxide synthase (nNOS-α) via PI3K/Nrf2-mediated signaling. Conversely, treating pENCs with cinnamaldehyde (CNM), a naturally occurring Nrf2 activator, prevented HG-induced apoptosis. These novel data reveal a negative feedback mechanism for GSK-3 activation. To further demonstrate that loss of Nrf2 leads to inflammation, oxidative stress, and reduces nNOS-mediated gastric function, we have used streptozotocin (STZ)-induced diabetic and Nrf2 null female mice. In vivo activation of Nrf2 with CNM (50 mg/kg, 3 days a week, ip) attenuated impaired nitrergic relaxation and delayed gastric emptying (GE) in conventional type 1 diabetic but not in Nrf2 null female mice. Supplementation of CNM normalized diabetes-induced altered gastric antrum protein expression of 1) p-AKT/p-p38MAPK/p-GSK-3β, 2) BH4 (cofactor of nNOS) biosynthesis enzyme GCH-1, 3) nNOSα, 4) TLR4, NF-κB, and 5) inflammatory cytokines (TNF-α, IL-1β, IL-6). We conclude that activation of Nrf2 prevents hyperglycemia-induced apoptosis in pENCs and restores nitrergic-mediated gastric motility and GE in STZ-induced diabetes female mice.

NEW & NOTEWORTHY Primary neuronal cell crust (pENCs) in the intestine habitats nNOS and Nrf2, which was suppressed in diabetic gastroparesis. Activation of Nrf2 restored nNOS by suppressing inflammatory markers in pENCs cells. Inhibition of Nrf2 reveals a negative feedback mechanism for the activation of GSK-3. Activation of Nrf2 alleviates STZ-induced delayed gastric emptying and nitrergic relaxation in female mice. Activation of Nrf2 restored impaired gastric BH₄ biosynthesis enzyme GCH-1, nNOSα expression thus regulating nitric oxide levels.

INTRODUCTION

The enteric nervous system (ENS) is a network of neurons and glia that controls various processes in the gastrointestinal (GI) tract, including motility, secretion, local immunity, and inflammation (1). Damage to the ENS can lead to functional GI disorders. Neuronal (sensory, motors, and autonomic) damage is multifactorial, (e.g., chronic diabetes) (2). Diabetic neuropathy directly affects enteric neuronal crest (ENC; composed of enteric glia and neuronal) cells and impairs the entire GI function (3). Symptoms, such as postprandial fullness, nausea, vomiting, bloating, early satiety, and abdominal pain, are typically observed in patients with diabetes (4). Reports from clinical studies and animal models of type 1 diabetes (T1D) and high-fat diet (HFD)-induced type 2 diabetes (T2D) have shown that enteric (myenteric) nerve disturbances involve nitrergic nerves (releases nitric oxide) (3, 5, 6). In addition, impairment in nitrergic relaxation may contribute to gastric neuropathy in diabetic rodents (7–10).

The enteric reflex circuits regulate movement by controlling the activity of both excitatory and inhibitory neurons that innervate the muscle. These neurons have cotransmitters for the excitatory neurons acetylcholine and tachykinins and for the inhibitory neurons nitric oxide (NO), vasoactive intestinal peptide (VIP), and ATP (11). Nitric Oxide is a major nonadrenergic noncholinergic (NANC) inhibitory signal in the peripheral nervous system and in gastrointestinal (GI) tract (12). Neuronal nitric oxide synthase (nNOS) is the primary source of NO in the gut and altered NO production can disrupt normal GI motility. In the myenteric plexus of different mammalian species, nitric oxide synthase (NOS) constitutes ∼25%–40% of the total myenteric neurons (5). Enteric neurons express high levels of nNOS, indicating that nNOS regulates GI neurotransmission (12). It has been reported that nNOS inhibits superoxide production and mediates cardio protection through suppression of reactive oxygen species (ROS) (13). However, the interaction between nNOS and ROS has not been shown in pENCs.

Oxidative stress has been etiologically implicated in the progression of diabetes and contributes to various GI diseases including gastroduodenal ulcers, inflammatory bowel disease (IBD), gastric and colorectal cancer (14). Although several studies have reported a role for nNOS in inhibiting ROS in cardiomyocytes, the effect of oxidative stress on nNOS expression and phosphorylation is not known. Nuclear factor-erythroid 2-related factor 2 (Nrf2) is part of a group of transcription factors that are important for cellular defense against oxidative stress and inflammation (15). Our earlier studies have shown that suppression of Nrf2 in HFD-induced diabetic female mice promoted delayed gastric emptying (16). Cinnamic aldehyde, bioactive compound derived from cinnamon, activates the Nrf2-dependent antioxidant response and is a popular traditional medicine for diabetes management (17). Our previous study demonstrated that natural bioactive compounds such as cinnamaldehyde (CNM) or curcumin normalized gastric emptying by inducing antioxidant markers, restoring nNOS function, while suppressing inflammation in T2D female mice (18). However, little is known concerning the role and regulation of Nrf2 in T1D-associated enteric neuropathy. Therefore, we hypothesized that increased glucose concentration in diabetes and subsequent inhibition of Nrf2 may repress the expression of phosphoinositide 3-kinases/protein kinase B (PI3K/Akt) pathway and reduces pENCs survival, and this may lead to altered gastric motility and delayed gastric emptying.

MATERIALS AND METHODS

Culturing Adult Mouse Primary Enteric Neuronal Crest Cells

Primary cultures for enteric neuronal crest (enteric glia + enteric neuronal: composed of all kinds of neuronal and glial cells) cells were performed according to Ye et al. (19). In brief, pENCs were isolated from adult female mouse [9- to 10-wk-old C57BL/6J wild-type (WT)] colon and ileum myenteric plexus. Primary cells were seeded on Matrigel-coated plates and cultured at 37°C with 5% CO₂ in complete neurobasal-A medium containing B-27 serum-free supplement, 1 mmol/l-glutamine, 1% fetal bovine serum, and 10 ng/mL GDNF. Half of the medium was replaced every day for 5–7 days. After 5–7 days of plating, the pENCs were treated with 25 mM glucose and/or cinnamaldehyde (CNM; 100 µM) for 48 h.

Viability Assay

HG-induced cell toxicity was assessed using MTT (Sigma-Aldrich Inc, St. Louis, MO) colorimetric assay. pENCs were seeded in 96-well plates (5 × 10⁴ cells/well). Cells were cultured as described in Culturing Adult Mouse Primary Enteric Neuronal Crest Cells. After exposure to HG with or without CNM for 24 and 48 h, medium was completely taken out and the cells were incubated with MTT at 5 mg/mL for 3 h. The absorbance was read at 570 nm using a microplate reader (Synergy HT Multi-Mode Reader, BioTek, Winooski, VT). The results were expressed as the percentage of control treatment representing 100% viability.

Cytokine Enzyme-Linked Immunosorbent Assay

Cytokine concentrations in cell culture supernatants were quantified by sandwich enzyme-linked immunosorbent assay (ELISA) using specific pairs of monoclonal antibodies against tumor necrosis factor-α (TNF-α), IL-1α, and IL-1β. The assays were performed according to the manufacturer’s protocols (Signosis, Santa Clara, CA). pENCs were cultured as described above. At the end of 48 h, an aliquot (100 µL) of the supernatant was used for the assay. The concentrations of the cytokines were directly proportional to the color intensity of the samples, respectively. Absorbance was measured spectrophotometrically at 450 nm using a microplate reader (BioTek Synergy HT Multi-Mode Reader, BioTek, Winooski, VT).

Flow Cytometry for Intracellular ROS Production

pENCs were cultured as described in Culturing Adult Mouse Primary Enteric Neuronal Crest Cells. pENCs incubated in Hank’s balanced salt solution (HBSS) with 10 μM 5-(and -6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate, acetyl ester (CM-H2DCFDA, Invitrogen, Cat. No. C6827) for 1 h. Cells were washed with PBS and then suspended in 500 μL phosphate-buffered saline (PBS) and fluorescence detected by flow cytometry using a BD FACSAria II (BD Biosciences). 2′,7′-Dichlorodihydrofluorescein (DCF) fluorescence intensity was measured in the Fluorescein (FITC) channel. The geometric mean of FITC intensity was used to indicate the amount of ROS production. All the data were analyzed with FlowJo 10.1 software (FlowJo, LLC).

Immunofluorescence Staining

pENCs were cultured as described in Culturing Adult Mouse Primary Enteric Neuronal Crest Cells and then harvested and fixed in 4% paraformaldehyde as previously published (20). The cells were blocked for 1 h in blocking and permeabilization solution containing 0.3% Triton X-100 (Bio-Rad) and 5% BSA (Millipore Sigma) in PBS followed by overnight incubation with primary antibodies at 4°C. The primary antibodies used include anti-Nrf2, anti-Caspase (1:500, Cell Signaling, Danvers, MA), and anti-nNOS (1:500, Abcam, Cambridge, MA). Β-tubulin (Tuj-1, 1:1,000, Cell Signaling, Danvers, MA) was used as neuronal marker. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Secondary antibody staining was performed with anti-mouse IgG (Alexa Fluor 488, Abcam, Cambridge, MA; 1:200) and anti-rabbit IgG (Alexa Fluor 546, Abcam, Cambridge, MA; 1:200) diluted in PBS and incubated for 1 h in the dark. Slides were washed and mounted using FluorSave reagent (Millipore, Burlington, MA). Mounted slides were viewed on a Leica DM LB2 microscope with Nikon Digital Sight DS-U2 camera, using ×40 objectives. Images were taken using the software NIS-Elements version 3.0 (Nikon, Japan).

DNA Fragmentation Assay

Gel electrophoresis was performed to determine DNA fragmentation after treatment of cells with 25 mM glucose and/or cinnamaldehyde (100 µM) for 48 h. After the respective incubation period, DNA was extracted using the DNeasy Blood and Tissue Kit following the manufacturer’s instructions (Qiagen, Valencia, CA). After quantification, DNA extracts (50 ng) were mixed with Peq Green and loaded onto 1.5% agarose gel, and then electrophoresis was performed at 50 V for 2 h. The gel also contained a DNA ladder for comparing the sizes of the DNA fragments.

RNA Interference

Small interfering RNAs (siRNAs) specific for Nrf2 (Santa Cruz Biotechnology, Dallas, TX), and transfection of siRNAs was performed according to siRNA transfection protocol (Santa Cruz Biotechnology, Dallas, TX). pENCs were cultured as described above for 5 days. In one tube, 2 μL of Nrf2 siRNA duplex was diluted into 100 μL of siRNA transfection medium (Santa Cruz Biotechnology, Dallas, TX). In a second tube, 2 μL of transfection reagent (Santa Cruz Biotechnology, Dallas, TX) was diluted into 100 μL of siRNA transfection medium. The contents of both tubes were gently mixed to and incubated for 45 min at room temperature. Next, 200 μL of the Nrf2 siRNA transfection cocktail was added to pENCs and further incubated for 6 h at 37°C. After transfection, fresh neurobasal media was changed and incubated for a further 18 h at 37°C.

In Vivo Experimental Animals

The Institutional Animal Care and Use Committee (IACUC) at Meharry Medical College (MMC) approved all experiments, in accordance with recommendations of the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (e-protocol no. 17-09-764). Adult (8–9 wk old) female homozygous Nfe2l2^−/− mice [B6.129X1-Nfe2l2 tm1Ywk/J, Nrf2 knockout (KO)] and their wild-type (WT, Nrf2^+/+) littermates were purchased from Jackson Laboratories (The Jackson Laboratory, Bar Harbor, ME). All animals were housed in the institutional animal care vivarium under standard conditions (4 mice/cage, 12 h light cycle) and allowed access to food and water ad libitum.

Experimental Design

Mice were randomly allocated into three groups (n = 8 per group): 1) control, 2) STZ, and 3) STZ + CNM (50 mg/kg) either in homozygous Nrf2 KO mice or in WT, respectively. The mice received either sodium citrate buffer (control) or streptozotocin (STZ, 50 mg/kg, pH 4.5, dissolved in sodium citrate buffer) through intraperitoneal injection for five consecutive days. Blood was obtained by lancet prick in the tail. Two weeks following STZ injection, fasting blood glucose levels (4-h fast) were measured, and mice with a fasting glucose level above 250 mg/dL were considered diabetic and used for this study. CNM (50 kg/body wt) was administered (intraperitoneal, diluted in mineral oil) three times per week for 9 wk. CNM dose was guided by published literature (16, 21).

Assays for Serum Insulin and Nitrite Levels

Animals were euthanized by carbon dioxide (CO₂) asphyxiation and blood was drawn immediately by cardiac puncture. Serum was separated from the blood by centrifugation (2,000 g at 4°C, 15 min). Aliquots of serum were separated and stored at −80°C to be used for subsequent biochemical analysis. Serum insulin levels were measured using enzyme-linked immunosorbent assay (ELISA kits, Crystal Chem, Elk Grove Village, IL) as per the manufacturer’s 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).

Measurement of Solid Gastric Emptying

Gastric emptying experiments for all groups of mice were performed as published earlier (22). At the end of the 9-wk treatment period, mice were fasted overnight. Each mouse was separately housed and provided a premeasured bolus of food with water ad libitum for 3 h. Afterward, the mice were moved to clean cages and fasted for an additional 2 h; the remaining food was dried and weighed to determine the amount of food ingested (FI). To measure the rate of gastric emptying, mice were euthanized by cervical dislocation and stomachs were carefully dissected. After euthanasia, the abdominal cavity was opened, the pylorus and cardia were clamped, and the stomach was removed. Full and empty stomach weights were recorded; the difference estimates the remaining gastric content (GC) after 2 h of fasting. The rate of GE was measured with the equation of 1 − [(GC/FI) × 100].

Isometric Muscle Recording with Electric Field Stimulation

Circular gastric antrum neuromuscular strips from WT and Nrf2 KO mice (n = 4/group) were used for electric field stimulation (EFS)-induced nonadrenergic noncholinergic relaxation (NANC) as previously described (23). The neuromuscular strips were mounted in 10 mL Krebs buffer at 37°C, and NANC-dependent nitrergic relaxation (nNOS function) was determined at 2 Hz for a duration of 30 s (DMT Technologies, Nottingham, UK). The NO dependence of nitrergic relaxation was confirmed with N^G-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.

Western Blot Analysis

Snap-frozen gastric antrum neuromuscular specimens [full thickness; muscularis layers + serosa (i.e., without mucosal and submucosal layers] were homogenized with sonication using radioimmunoprecipitation assay (RIPA) buffer containing a protease inhibitor (Thermo Fisher Scientific, Rockford, IL). Protein concentrations were estimated in each lysate via bicinchoninic acid assay (BCA) method. Equal concentrations of lysates (40 µg) were separated using either 6% or 12% sodium dodecyl sulfate (SDS) polyacrylamide gels before wet transfer to a nitrocellulose membrane. Each membrane was subsequently blocked with 5% dried nonfat milk for 1 h, then incubated with primary polyclonal antibody [GCH-1, (1:500), DHFR (1:500), interleukin-6 (IL-6, 1:500), interleukin-1β (IL-1β, 1:1,000), Toll-like receptor 4 (TLR4, 1:500), tumor necrosis factor-α (TNF-α, 1:500), Nrf2 (1:1,000), nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB, 1:1,000), p-AKT (1:500) glutamate-cysteine ligase modifier subunit (GCLM, 1:1,000), glutamate-cysteine ligase catalytic subunit (GCLC, 1:1,000) purchased from (Santa Cruz Biotechnology, Dallas, TX)]. p-PI3K (1:1,000), GSK-3β (1:1,000), caspase 3 (1:1,000), cleaved caspase 3 (1:1,000), total poly(ADP-ribose) polymerase (PARP, 1:1,000), cleaved PARP (1:1,000), nNOSα (1:1000, NH₂-terminal), and p38 mitogen-activated protein kinases (p38MAPK, 1:1,000) were purchased from (Abcam, Cambridge, MA) and Cell Signaling, Inc (Danvers, MA). Primary polyclonal antibodies were incubated overnight (∼18 h), respectively. The membranes were washed three times for 10 min each in 0.01% tris-buffered saline (TBS)-Tween, then incubated in horseradish peroxidase-conjugated secondary antibody (1:1,000) for 1 h at room temperature. The blots were visualized with enhanced chemiluminescence (ECL) Western Blotting Detection Reagent (GE Healthcare Bio-Sciences Corp., Piscataway, NJ) and the reactive bands were analyzed quantitatively by optical densitometry. The blots were stripped and reprobed to measure protein expression. Blots were reprobed with β-actin polyclonal antibodies (1:5,000) (Abcam, Cambridge, MA) to enable normalization of signals between samples.

Quantitation and Statistical Analysis

Data were analyzed for normality by two-tailed unpaired or paired Student’s t test or Mann-Whitney test for comparisons of two groups with Prism 5 for Windows (GraphPad Software). One-way analysis of variance (ANOVA) of the repeated experiments followed by Tukey’s post hoc pairwise multiple-comparisons test was also used when appropriate.

RESULTS

Effects of Cinnamaldehyde on the Viability, Oxidative Stress/Neuronal Apoptosis In Vitro of pENCs

To determine the protective effects of cinnamaldehyde (CNM) from hyperglycemia (HG)-induced cell death, pENCs were pretreated with CNM at different concentrations for 3 h before cotreatment with HG for a 24- to 48-h period, respectively. As shown in Fig. 1A, exposure to HG alone significantly (P < 0.05) reduced viable cells in a time-dependent manner (24 h: 68 ± 6% vs. 99 ± 2%; 48 h: 44 ± 7% vs. 98 ± 3%) when compared with untreated (vehicle) cells. Pretreatment of pENCs cultures with various doses of CNM (25, 50, and 100 µM) increased cell viability when compared with HG alone at both 24 and 48 h time points. The maximum beneficial effect of CNM in restoring cell viability was noticed at 100 µM at 48 h. Hence, we have used 100 µM of CNM at 48-h time for all subsequent experiments conducted in this study.

Figure 1.

Cinnamaldehyde (CNM) prevents hyperglycemia-induced oxidative stress, and apoptosis. Enteric neuronal crest isolated from adult mouse intestines were cultured in the presence of hyperglycemia (HG, 30 mM), CNM (100 µg/mL), or both for 24 h. A: screening of cytotoxic effects of HG and CNM on pENCs at 24 h and 48 h incubation. B: reactive oxygen species (ROS) production was analyzed by flow cytometry. C: triple immunofluorescent staining. Representative images of primary enteric neuronal cells fluorescent staining (from the top) for DAPI (nuclear staining), caspase and α-tubulin (Tuj-1) as a neuronal marker. Cells were visualized with 40 magnifications. Scale bars: 50 µm. The arrows indicate the positive staining for caspase 3 in neuronal cells. Western blot was performed to assess expression of total and cleaved-caspase-3 (D) and total and cleaved PARP 1 (E). F: demonstration of apoptosis by DNA fragmentation. DNA extracted from pENCs: control, HG and HG treated with CNM treatment for 48 h. The DNA was separated by electrophoresis using 1.5% agarose gel and visualized using chemdoc image analyzer. Data is for four independent experiments. Bars represent mean values, with error bars representing SD. ^aP < 0.05 compared with control; ^bP < 0.01 compared with HG.

To investigate the molecular mechanisms of hyperglycemia-induced apoptosis and the effects of CNM, we measured oxidative stress using flow cytometry and protein expression of apoptosis related markers such as caspase-3 and poly(ADP-ribosyl) polymerase (PARP) immunoreactivity levels in pENCs (Fig. 1). pENCs were cultured in normal glycemia (NG, 5.5 mmol/L), hyperglycemia (HG, 25 mmol/L) media to mimic HG conditions and in the presence or absence of CNM (100 µM, Nrf2 activator). Our data show that exposure of pENCs to HG increased ROS after 48-h incubation (P < 0.05), whereas CNM treatment reduces ROS generation (Fig. 1B).

pENCs stained with neuronal marker and cleaved caspase-3 showed a significant increased number of positive cleaved caspase-3 neuronal cells in HG group (P < 0.05, Fig. 1C). pENCs exposure to HG increases the expression levels of active caspase-3 (cleaved caspase 3) suggesting that HG causes apoptosis in enteric neurons (Fig. 1D). pENCs treated with CNM showed a decreased number of positive-cleaved caspase-3 neuronal cells. To investigate the enzymatic activation of caspase-3, the cleavage of PARP, which is a caspase-3 substrate, was also measured. When cells were treated with CNM in the presence of HG, the formation of the cleaved PARP was decreased when compared with the cells exposed to HG. This data clearly demonstrated the inhibitory activity of CNM on caspase-3 and its substrate PARP (Fig. 1E). Collectively, the aforementioned results demonstrate that CNM attenuated oxidative stress and apoptosis in pENCs in vitro.

Furthermore, we have measured DNA fragmentation, a hallmark of apoptosis, by incubating pENCs in HG conditions with or without CNM for 48 h. DNA gel electrophoresis showed a higher level of internucleosomal DNA fragments or chromatin cleavage mimicking the typical DNA fragmentation pattern. Treatment with CNM attenuated HG-induced DNA fragmentation in pENCs (Fig. 1F). These results indicate that CNM protected pENCs against HG-induced apoptosis.

Cinnamaldehyde Attenuates HG-Induced Neuronal Inflammation in pENCs In Vitro and Enhances the PI3K Pathway

In this study, we have attempted to elucidate the interplay between oxidative stress parameters and proinflammatory cytokines, which are simultaneously responsible for the neuronal cell loss. We measured various cytokines released in the media in response to HG and found an increase of IL-1α, IL-1β, and TNF-α (Table 1). CNM treatment reduced these cytokine secretions in the media. Furthermore, our data show that exposure to HG increased the expression of TLR4, nuclear NF-κB, and downstream factors such as IL-1β and TNF-α significantly (Fig. 2, A–E) in pENCs, whereas cells treated with CNM significantly (P < 0.05) inhibited this overexpression.

Table 1.

Cinnamaldehyde alleviates cytokine release in pENCs

Cytokine, pg/mL	Control	HG	HG + CNM
IL 1α	25 ± 5	58 ± 6*	49 ± 8#
IL 1β	42 ± 7	78 ± 8*	71 ± 5#
TNF-α	10 ± 3	43 ± 7*	21 ± 6#

*P < 0.05 compared with control. #P < 0.05 compared with HG. Three separate experiments (n = 3), each performed in duplicate. CNM, cinnamaldehyde; HG, high glucose.

Figure 2.

Cinnamaldehyde (CNM) attenuates hyperglycemia-induced TLR4 activation, proinflammatory cytokine secretion and enhances PI3K/AKT/GSK-3β expression in pENCs. Representative immunoblots and densitometric analysis data for TLR-4 (A), cytosol NF-κB (B), nuclear NF-κB (C), IL-1β (D), TNF-α (E), GSK-3β (F), p-PI3K (G), and total AKT and p-AKT (H) in pENCs. 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 are for four independent experiments. ^aP < 0.05 compared with control. ^bP < 0.01 compared with HG. Bars represent mean values, with error bars representing SD. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. HG, hyperglycemia.

PI3K/AKT has been shown to be an important signal transduction factor regulating insulin signal transduction and protecting against apoptosis. Hence, we further explored the molecular mechanism of CNM on upstream markers such as GSK-3β/PI3K/AKT signaling pathway in vitro. As shown in Fig. 2, F–H, HG exposure increased GSK-3β expression compared with the control (P < 0.05) and decreased the phosphorylation of p-PI3K and p-AKT. CNM treatment significantly reversed all of these proteins in pENCs exposed to in vitro HG (P < 0.05), indicating that CNM inhibited apoptosis by regulating PI3K/AKT/GSK-3β signaling pathway.

Cinnamaldehyde Activates Nrf2 in HG-Treated pENCs In Vitro

CNM has been shown to activate Nrf2 signaling pathway in various cell types, but whether CNM has a similar effect on pENCs under HG conditions is unknown. Immunofluorescent staining revealed that pENCs exposed to HG suppressed Nrf2 expression and loss of neuronal cells. CNM treatment showed predominant Nrf2-positive stained neuronal cells (Fig. 3A). Immunoblot data complimented immunofluorescent staining results showing a reduced Nrf2 expression in HG exposed pENCs (Fig. 3B). CNM treatment attenuated the Nrf2 protein expression in pENCs exposed to HG (Fig. 3B). Next, we investigated whether the induction of the Nrf2 pathway by CNM would also affect the Nrf2 downstream genes, such as GCLM and GCLC. As shown in Fig. 3C, the expression of GCLM and GCLC was increased with CNM treatment in HG-exposed pENCs.

Figure 3.

Cinnamaldehyde (CNM) increases Nrf2 [nuclear factor (erythroid-derived 2)-like 2], Phase II enzymes, and nNOS expression in pENCs. A: triple immunofluorescent staining images of primary enteric neuronal cells fluorescent staining for DAPI (nuclear staining), Nrf2 [nuclear factor (erythroid-derived 2)-like 2] and β-tubulin (Tuj-1, neuronal staining) as a neuronal marker. Cells were visualized with ×40 magnifications. Scale bars: 50 µm. The arrows indicate the positive staining for Nrf2 in neuronal cells. Representatives immunoblot and densitometry analysis data for Nrf2 (B), GCLC:GCLM (C), and triple immunofluorescent staining (D). Representative images of primary enteric neuronal cells fluorescent staining for DAPI (nuclear staining), Nrf2 [nuclear factor (erythroid-derived 2)-like 2], nNOS (neuronal nitric oxide synthase), and β-tubulin (Tuj-1, neuronal staining) as a neuronal marker. Cells were visualized with ×40 magnifications. Scale bars: 50 µm. The arrows indicate the positive staining for Nrf2 or nNOS in neuronal cells. Data were normalized for housekeeping gene or protein (β-actin). E and F: GCH-1 and nNOSα protein expression. G: colocalization staining images for Nrf2 and nNOS. Cells were visualized with ×40 magnifications. Scale bars: 50 µm. Stripped blots were reprobed with β-actin. Bar graphs showed a ratio of target gene or protein with β-actin. Data were analyzed using one-way ANOVA by using GraphPad Prism software. Bars represent mean values, with error bars representing SD. Data are for four independent experiments. ^aP < 0.05 compared with control. ^bP < 0.05 compared with HG. HG, hyperglycemia.

Immunofluorescent images show that nNOS is localized in neuronal cells marked with enteric neuronal marker and significantly reduced nNOS-positive cells when pENCs exposed to HG (Fig. 3D). Our data show that CNM significantly increased the nNOS-stained neurons in HG-exposed pENCs (Fig. 3D). Guanylate cyclohydrolase-1 (GCH-1) is responsible for BH₄ biosynthesis via the de novo pathway. Treatment with CNM attenuated HG-induced impairment of GCH-1 protein expression in pENCs (Fig. 3E). As depicted in Fig. 3F, nNOSα expression was significantly reduced in HG-exposed pENCs (P < 0.05), whereas CNM treatment enhanced nNOSα expression in pENCs exposed to in vitro HG.

Colocalization staining revealed that nNOS and Nrf2 were expressed in majority of enteric neuronal cell bodies (Fig. 3G). pENCs exposed to HG suppressed nNOS- and Nrf2-positive enteric neuronal cells, whereas CNM treatment significantly expressed both Nrf2- and nNOS-positive cells (Fig. 3G). These data suggest that nNOS-Nrf2 interaction may be key to restoring nitrergic neuron-mediated enteric neuronal damage.

Inhibition of Nrf2-Induced Apoptosis, Inflammation, and Elevation of GSK-3β

To further demonstrate whether HG-induced alterations are mediated through Nrf2, pENCs were exposed to siRNA for 18 h in vitro. As shown in Fig. 4, suppression of Nrf2 (Fig. 4A) resulted in reduction of GCLC expression (Fig. 4A), elevation of GSK-3β (Fig. 4B), and diminished p-PI3K/p-AKT expression (Fig. 4C) in pENCs. Furthermore, suppression of Nrf2 resulted in cleavage of caspase 3, induction of NF-κB and reduced nNOS-α protein expression in pENCs (Fig. 4, D–F).

Figure 4.

Effects of Nrf2 inhibition by siRNA on the expression of downstream Nrf2 signaling in pENCs in normal glycemic conditions. Representative immunoblots and densitometric analysis data for Nrf2: GCLC (A), GSK-3β (B), PI3k: Akt (C) cleaved caspase 3 (D), NF-κB (E), and nNOS (F). Blots for 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. Bars represent mean values, with error bars representing SD. Data were analyzed with one-way ANOVA using the GraphPad Prism software. The values are from four independent experiments. *P < 0.05 compared with control.

Supplementation of Cinnamaldehyde Improves Metabolic Disorder in an STZ-Induced Chronic Diabetic Model

Based on the in vitro results, we explored the efficacy of CNM supplementation in STZ-induced WT and Nrf2^−/− female mice. The effectiveness of dietary Nrf2 activators in alleviating metabolic disorder was assessed in the STZ-induced diabetic model in both WT and Nrf2^−/− female mice. Diabetes induction showed increased blood glucose levels and reduced body weight gain (Table 2). Supplementation with CNM significantly alleviated all indices of metabolic dysfunction in WT but not in Nrf2^−/− mice, suggesting that the effects of CNM is Nrf2 specific. In addition, STZ injection decreased insulin levels (P < 0.05), which was unaltered by supplementation with CNM regardless of genotype, indicating that the protective effects of CNM was not insulin dependent (Table 2). As shown in Table 2, serum nitrite levels were significantly (P < 0.05) reduced in STZ-induced diabetic female mice. CNM supplementation normalized nitrite levels only in WT but not in Nrf2^−/− diabetic mice.

Table 2.

Changes in body weight, blood glucose, and insulin levels in WT and Nrf2^−/− female mice at 9 wk

	WT			Nrf2−/−
	ND	STZ	STZ+CNM	ND	STZ	STZ+CNM
Body weights, g	25.4 ± 0.4	20.0 ± 0.5a	21.28 ± 0.6b	26.4 ± 0.5	17.9 ± 0.7d	18.5 ± 0.4
Fasting blood glucose, mg/dL	106 ± 10.0	550 ± 12.0a	209 ± 28.0b	130 ± 14.0	585 ± 13.0d	545 ± 25.0
Fasting insulin, ng/mL	0.48 ± 0.08	0.22 ± 0.05a	0.25 ± 0.04b	0.5 ± 0.07	0.23 ± 0.06d	0.24 ± 0.05
Serum nitrate levels, µM	35.8 ± 4.4	21 ± 3.5	28 ± 2.9b	27.4 ± 2.5c	20.6 ± 5.8d	18 ± 3.0

^a,cP < 0.05 compared with WT-ND. ^bP < 0.05 compared with WT-STZ. ^dP < 0.05 compared with Nrf2^−/−-ND. CNM, cinnamaldehyde; ND, normal diet; STZ, streptozotocin; WT, wild type.

Cinnamaldehyde Improves Chronic Diabetes-Induced Delayed Gastric Emptying

To investigate the therapeutic effects of Nrf2 activation on improvement of gastric motility in the STZ-induced diabetic model, functional and pathological changes in the gastric neuromuscular tissue were measured. The results on food intake (FI) in all the groups did not have any significant difference when compared with control group. The ratio of stomach weight to body weight was higher in STZ-induced diabetic groups compared with the control group. Whereas treatment with CNM significantly attenuated the ratio of stomach to body weight in WT but not in Nrf2 KO female mice (Fig. 5A). As shown in Fig. 5B, STZ-induced diabetic mice displayed a significant delay in solid GE compared with control group (76.3 ± 1. 5% vs. 27.7 ± 5.5%). CNM normalized delayed gastric emptying significantly (50.1 ± 6.6% vs. 27.7 ± 5.5%, P < 0.05) in WT-STZ but not in Nrf2 KO STZ mice (Fig. 5B). The above data suggest that FI, weights of stomach, and body weight influence gastric emptying rate in all experimental groups.

Figure 5.

Cinnamaldehyde (CNM) enhances solid GE and nitrergic relaxation of neuromuscular gastric antrum strips. A: effects of CNM on stomach weight and body weight ratio. B: solid GE in WT and Nrf2 KO STZ-induced diabetic mice. C: effect of CNM on nitrergic relaxation in WT and Nrf2 KO STZ-induced diabetic mice. Data were analyzed with one-way and two-way ANOVA using the GraphPad Prism software. The values are means ± SE (n = 4). ^a,cP < 0.05 compared with WT-Vehicle. ^bP < 0.05 compared with WT-STZ. ^dP < 0.05 compared with Nrf2 KO-vehicle. GE, gastric emptying; KO, knockout; L-NAME, N^G-nitro-L-arginine methyl ester; STZ, streptozotocin; WT, wild type.

NO-mediated nitrergic relaxation comprises the primary component of NANC relaxation in gastric smooth muscle. NANC relaxation was severely impaired in the STZ-induced diabetic mice (Fig. 5C). CNM supplementation restored NANC relaxation in gastric antrum of STZ-induced diabetic mice (P < 0.05). Blockade of nNOS activity with l-NAME significantly abolished relaxation to confirm this relaxation was mediated by NO (P < 0.05). Taken together, CNM restored nitrergic function and normalizes GE rates in WT-STZ but not in Nrf2 KO STZ-induced diabetic mice demonstrating Nrf2-dependent response to CNM.

Cinnamaldehyde Attenuates STZ-Induced Diabetic Inflammation in the Gastric Neuromuscular Tissue

Reports suggest that there is a strong link between hyperglycemia, hyperglycemic-induced oxidative stress, inflammation, and the development and progression of both type 1 and type 2 diabetes (T1D and T2D). Inflammatory factors were known to be involved in STZ-induced tissue damage in diabetic mice (24). TLR4 signaling leads to the production of proinflammatory cytokines both in T1D and T2D pathogenesis. In this study, we have investigated the expression levels of TLR4, IL-1β, IL-6, TNF-α, and proinflammatory factors in STZ-induced diabetic gastric tissues (Fig. 6, A–D). Our results showed that STZ-induced diabetic mice had a significant increased levels of TLR4, IL-1β, IL-6, and TNF-α (P < 0.05, Fig. 6), whereas supplementation with CNM significantly inhibited the overexpression of all the above cytokines (P < 0.05).

Figure 6.

Effect of cinnamaldehyde (CNM) on gastric neuromuscular antrum proinflammatory cytokines in STZ-induced diabetic mice. TLR-4 (A), IL-1β (B), IL-6 (C), and TNF-α (D) protein levels in gastric antrum neuromuscular specimens of WT and 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 means ± SE (n = 4). ^a,cP < 0.05 compared with WT-vehicle. ^bP < 0.05 compared with WT-STZ. ^dP < 0.05 compared with Nrf2 KO vehicle. STZ, streptozotocin; WT, wild type; KO, knockout.

Next, we investigated changes in AKT/p38MAPK/GSK-3β signaling pathway in gastric neuromuscular tissues. As shown in Fig. 7, A and B, Western blot results showed that the levels of p-AKT and p38MAPK decreased compared with the control group (P < 0.05). Our data further show that these changes were significantly reversed with CNM supplementation (P < 0.05), indicating posttranslational phosphorylation of Nrf2.

Figure 7.

Effect of CNM on the expression of AKT/MAPK signaling proteins in gastric neuromuscular antrum specimens of WT and Nrf2 KO STZ-induced diabetic mice. Representative immunoblots and densitometric analysis data for the following in gastric neuromuscular tissues: p-AKT (A), p38 MAPK (B), GSK-3β (C), GCLC (D), GCLM (E), GCH-1 (F), and nNOSα (G) in WT and 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 means ± SE (n = 4). ^a,cP < 0.05 compared with WT-vehicle. ^bP < 0.05 compared with WT-STZ. ^dP < 0.05 compared with Nrf2 KO-vehicle. CNM, cinnamaldehyde; STZ, streptozotocin; WT, wild type; KO, knockout.

Glycogen synthase kinase-3β (GSK-3β) participates in the cellular response to oxidative stress, a hallmark of several nervous system disorders. Phosphorylation of GSK-3β at serine 9 residue inhibited GSK-3β activity (25). Our results showed that CNM supplementation enhanced phosphorylation of GSK-3β thus inhibiting its interaction with Nrf2 (Fig. 7C).

Cinnamaldehyde Enhanced Nrf2 and Phase II Antioxidant Enzymes Expression in STZ-Induced Diabetic Mice

Oxidative stress plays an important role in the progression of diabetes. Nrf2 and its downstream Phase II enzyme have emerged as key players in combating cellular oxidative stress insult. Induction of Nrf2 expression and Nrf2-dependent Phase II enzyme expression coupled with the AKT signaling pathway were examined by Western blot analysis. Our data show that gastric Nrf2 protein expression was significantly reduced in STZ-induced diabetic mice. Supplementation of CNM significantly restored (P < 0.05) Nrf2 induction in diabetic gastric tissue. Similar observations were seen with Phase II enzyme induction (Fig. 7, D and E). Catalytic subunit of glutamyl cysteine ligase (GCLC) and glutamate-cysteine ligase modifier (GCLM) were suppressed in STZ-induced diabetic female mice compared with control groups (Fig. 7, D and E). CNM supplementation regulated Phase II antioxidant enzyme expression of GCLC and GCLM in STZ-induced diabetic female mice (Fig. 7, D and E).

Cinnamaldehyde Normalized Gastric GCH-1 and nNOSα Protein Expression in Wild-Type but Not in Nrf2 KO Mice

GCH-1 play a role in BH₄ biosynthesis via de novo pathway. As depicted in Fig. 7F, chronic diabetes significantly (P < 0.05) reduced GCH expression in WT but not in Nrf2^−/−. In addition, supplementation of CNM restored GCH-1 expression. nNOSα expression was significantly (P < 0.05) reduced in STZ-induced diabetic mice, whereas CNM supplementation enhanced the expression of nNOS-α expression in WT mice but not in Nrf2 KO mice (Fig. 7G).

DISCUSSION

Diabetic enteropathy causes GI autonomic nerve dysfunction, which significantly disturbs ENS activity and function. Hyperglycemia mediates ENS damage through various mechanisms, such as polyadenosine triphosphate ribose, advanced glycosylation end (AGE) products, endoplasmic reticulum stress, oxidative stress, inflammation, and ischemia, which result in demyelination and axonal degeneration (26). ENS is composed primarily of neurons and glial cells that control the GI motility functions. Glial cells in the ENS form a cellular and molecular bridge among enteric nerves, enteroendocrine cells, immune cells, and epithelial cells (27). Enteric glial cells have been shown to play an active role in maintaining ENS function and regulating enteric neuronal survival (28). Our current studies demonstrated that Nrf2-regulated signaling cascade is active in adult mouse pENCs and this is associated with changes in gastric emptying (GE) and nitrergic relaxation in chronic hyperglycemic conditions. Whether and how Nrf2 in enteric neurons interact and maintain nNOS signaling remains unknown.

In patients with diabetes, uncontrolled glucose levels known to cause neuronal damage due to glucose neurotoxicity (29). The induction of neuronal cell death by hyperglycemia has been shown to be associated with oxidative stress and apoptosis (30, 31). A growing body of evidence demonstrates that diabetes induces apoptotic cell death in multiple target organs and thus leads to organ dysfunction (32). Several mechanisms have been proposed for increased cell death in hyperglycemic conditions. Studies by Moriyama et al. suggest that exposure of serotonergic neurons to in vitro hyperglycemia activates ATP-sensitive K⁺-channels and leads to hypercalcemia and apoptosis (33). Hyperglycemia decreases neuronal growth factors, which in turn decreases enteric neuronal survival, and this is considered to be another possible mechanism (4). Other potential mechanisms of enteric neuron degeneration in diabetes are excessive ROS generation due to oxidative stress. The nuclear factor-erythroid 2-related factor 2 (Nrf2) is a redox-regulated transcription factor that plays a crucial role in protection against the oxidative damage, hyperoxia, nitrosative stress, and ER stress by inducing the expression of cytoprotective and antioxidant genes and elevating the intracellular glutathione levels. Treatment with antioxidants minimizes or prevents development of these complications in patients with diabetes (34). In the present study, we attempted to investigate the underlying mechanisms associated with apoptosis of pENCs in a hyperglycemic environment in vitro. The results demonstrate that HG-induced apoptosis in pENCs was associated with ROS accumulation and reduction of Nrf2 expression. Nrf2 activation by CNM treatment significantly reduced apoptotic markers and ROS levels. We also showed that HG-induced apoptosis involved DNA fragmentation and caspase-3/PARP-dependent pathway. Earlier, Anitha et al. showed that rat fetal embryonic enteric neuronal cells exposed to in vitro glucose for 24 h caused apoptosis (35). Recently Ye et al. (19) demonstrated that enteric neurons that are exposed to intestinal pathogens and bacterial endotoxins such as lipopolysaccharide (LPS), activated caspase-11-mediated pyroptotic pathway. Our studies for the first time using pENCs demonstrate that activation of Nrf2 protect against HG-induced oxidative stress, DNA fragmentation, and the caspase-3/PARP-dependent apoptosis pathway.

Inflammation and oxidative stress are two synergistic conditions that have a significant negative impact on the function of the ENS. In diabetes, oxidative stress causes excess ROS production, which is considered to be the first step in the intestinal inflammatory cascade and production of cytokines (36, 37). Kobayashi et al. reported that loss of Nrf2 plays a detrimental role in causing exacerbation of inflammation in in vivo experimental mouse models (38). In diabetes, activation of TLR4 triggers inflammatory signaling pathway damaging small blood vessels (39, 40). One possible mechanism in TLR4 activation is due to higher circulating levels of recombinant high-mobility group box protein 1 (HMGB1) commonly observed in diabetes (41). On binding to its receptors, HMGB1 induces nuclear translocation of nuclear factor-κB (NF-κB), leading to secretion of proinflammatory cytokines including tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β) (42). Another mechanism is that increased permeability of the digestive tract to microbial components in patients with diabetes increases circulating LPS levels (19). On the other hand, inhibition of the TLR4 signaling pathway attenuates the damage of neuronal function and the apoptosis of neurons, via TLR4 related downstream proteins (39). Conversely, TLR4 mediates caspase-3 and is known to be involved in nerve injury and neuronal apoptosis (43). High expression of TNF-α in intestinal epithelial cells suggests that TNF-α levels may be indicative of intestinal damage (44). Moreover, our current study also demonstrated that activation of Nrf2 by CNM blocks TLR4 signaling, which leads to a significant decline in the expression levels of NF-κB, which in turn attenuated increased expression of downstream inflammatory factors, such as TNF-α, and IL-1β in adult mouse pENCs exposed to HG.

The phosphatidylinositol 3-kinase (PI3K)/protein kinase B (PKB; also known as Akt) signaling pathway plays a central role in the survival of diverse cell types (45). It is well established that Nrf2 is known to be regulated by protein kinase C and mitogen-activated protein kinases (46). GSK-3β is generally considered a potential downstream gene product of AKT, which is a significant component of the PI3K/AKT pathway. Notably, AKT phosphorylates GSK-3β to an inactive form (47). Here, we showed that HG inhibited the expression of PI3K/Akt, thus enhancing GSK-3β, which led to decreased Nrf2 transcriptional activity expression in pENCs. Furthermore, CNM treatment activated PI3K/Akt pathway, and this mechanism is essential in suppressing GSK-3β activity and restore Nrf2-ARE-dependent antioxidant function in pENCs.

Hyperglycemia can also foster the nonenzymatic formation of advanced glycosylation end products (AGEs), protein cross linking, and ROS formation. ROS can be generated as a result of autooxidation of glucose and formation of AGEs. It has been proposed that AGE products play a role in reduced expression of nNOS, since inhibiting these products can prevent the depletion of NO from nerve terminals. A two-phase model of NO depletion has suggested in which during the first phase, expression of nNOS is reduced in a reversible fashion that, if not treated, will eventually lead to irreversible loss of nitrergic neurons or end stage organ function. It has been reported that nNOS-positive neurons that are colocalized with heme oxygenase-2 (HO-2) were shown to be resistant to high glucose flux, whereas nNOS-positive with depleted HO-2 neurons underwent a change in neuronal shape and size (48). This indicated that HO-2 plays an important cellular defense mechanism against oxidative stress. Here, we showed that pENCs exposed to HG in vitro led to decreased Nrf2 transcriptional activity, GCH-1, and nNOS expression. Activation of Nrf2 restored Nrf2 detoxification mechanisms and BH₄/nNOS function in both in vitro and in vivo diabetic conditions. Furthermore, our current studies demonstrated that Nrf2-regulated signaling cascade is active in adult mouse pENCs and this is associated with changes in GE and nitrergic relaxation in chronic diabetic conditions in vivo. Whether and how Nrf2 in glial cells and/or enteric neurons interact to maintain nNOS signaling warrants an in-depth investigation.

It is known that toll-like receptor signaling and Nrf2 modulate inflammatory responses (49). Our siRNA data demonstrate that loss of Nrf2 induces TLR4 expression and activation of downstream inflammatory pathways. Interestingly, siRNA Nrf2 knockdown data showed a decrease in PI3K/AKT and an increase in GSK-3β, suggesting Nrf2 plays a major role in regulating cell survival and functions. One possible mechanism is that GSK-3 acts as a potential feedback control within the PI3K/AKT/TLR4/NF-κB network. As reviewed by Herminda et al., GSK-3 phosphorylates several upstream and downstream components of the PI3K/AKT/mTOR signaling network, including AKT itself, RICTOR, TSC 1 and 2, phosphatase and tensin homolog deleted on chromosome 10 (PTEN), and IRS 1 and 2 (50). Taguchi et al. reported that GSK-3 phosphorylates PTEN, which then counteracts PI3K activity, participates in repression of Nrf2 activity (51). The bidirectionality relation has been studied most with the AKT-GSK-3 interaction, in which AKT not only inhibits GSK-3 but GSK-3 can also regulate AKT. It has been proposed that GSK3 promotes the activity of protein phosphatase-1 (PP1) by phosphorylating, thereby diminishing their induction of phosphoserine-GSK-3 (52). This activation of PP1 not only increases GSK3 activity, but also may dephosphorylate upstream kinases (52). Moreover, GSK-3β is also known to participate in NF-κB regulation (53). Finally, these cell-signaling complexes regulate Nrf2 activation. Our studies further demonstrate that inhibition of Nrf2 reduced nNOSα expression demonstrating a strong correlation with Nrf2-nNOS complex in pENCs. These data together with our current findings strongly suggest that PI3K/AKT signaling mechanisms promote cell survival and facilitate Nrf2-dependent antioxidant defensive system and nNOS activity in adult mouse pENCs. Future studies need to address the role of the PI3K pathway in maintaining Nrf2 stability, nuclear import and export, and interaction with Keap1 in pENCs.

Gastroparesis gradually develops over a period of time in patients with T1DM and T2DM (54). The most common contributing factors for diabetic gastroparesis (DG) are 1) autonomic neuropathy, 2) chronic hyperglycemia, and 3) abnormalities of gut hormones and neurotransmitters, such as motilin, gastrin, changes in interstitial cells of cajal (ICC), and nitric oxide (55–62). It is evident that hyperglycemia marks an adverse effect on gastric emptying (63). Although both delayed and rapid gastric emptying can be observed in diabetes mellitus, delayed gastric emptying was reported to occur in the majority of patients (64, 65). The stimulation of pyloric contractions and inhibition of antral contractions contribute to the hyperglycemia-induced delayed gastric emptying (58). Taken together, diabetes induces marked structural remodeling of the wall of the GI tract and its neuronal support leading to altered function of the GI tract. As reviewed by Bagyánszki and Nikolett Bódi, the two main observations in diabetic gastroparesis occur via 1) the loss of ICCs and 2) reduced expression of nNOS. Nitrergic neurons are susceptible to injury reacting with ROS, leading to protein dysfunction (66). Other studies support the idea that nitrergic neurons are injured due to inflammation (19, 67). Iwasaki et al. showed depletion of nNOS and substance P staining in the antrum of patients with type-2 diabetes (68). nNOS expression was decreased in the gastric body of patients with gastroparesis (69). We reported earlier that high-fat diet-induced diabetes causes reduced gastric motility via abnormal nitrergic relaxation and delayed gastric emptying in female mice (16). It has been reported that nNOS positive neurons are more vulnerable than any other neurons to changes in the intestinal environment due to high-fat diet (7, 70, 71). In addition, we also showed a possible mechanism involved in controlling diabetic gastroparesis is through modulation of key regulating detoxifying enzymes via Nrf2-mediated gastric nNOS function (16, 18). Reduced nitrergic relaxation due to decreased Nrf2-antioxidants and increased inflammatory cytokines led to delayed gastric emptying in HFD-induced T2 DM female mice (16, 72). In addition, we have reported that obesity-induced diabetes reduced the expression levels of nNOS-Nrf2 complex, indicating little or no interaction using immunoprecipitation methods (18). Whereas CNM or curcumin supplementation increased the expression of nNOS-Nrf2 complex both in total lysates and nuclear fractions, indicating strong interaction between Nrf2 and nNOS (18). However, no such studies have been reported in STZ-induced T1 DM female mice. In the current study, we have demonstrated that activation of Nrf2 by CNM attenuated delayed gastric emptying by normalizing gastric 1) AKT/p38/MAPK, 2) GSK-3β, 3) BH4 (cofactor of nNOS) biosynthesis enzyme GCH-1 (de novo), 4) nNOSα protein, and 5) Nrf2 and Phase II antioxidant enzymes in STZ-induced T1 DM female mice (Fig. 8). Most importantly, our studies for the first time demonstrate that treatment with CNM was able to protect pENCs from high glucose-induced alterations such as ROS generation, apoptosis, and elevated inflammatory markers via activation of Nrf2 and nNOS. Finally, CNM treatment failed to show the protective effects in Nrf2^−/− (null) mice. Collectively, the above studies suggest that activation of Nrf2 plays a vital role in regulating nNOS-mediated gastric function by suppressing inflammation and oxidative stress in pENCs. Whether and how Nrf2 in glial cells and enteric neurons interact to maintain nNOS signaling warrants an in-depth investigation.

Figure 8.

Schematic illustration depicting the effects of hyperglycemia (HG) on mechanistic signaling of Nrf2 in nNOS-mediated gastric motility in STZ-induced diabetic female mice. Hyperglycemia displays gradual changes in PI3K/AKT/Nrf2 signaling in the enteric neuronal crest cells causing reduced nitric oxide levels and increased inflammation/cell apoptosis. Supplementation of CNM restores delayed gastric emptying via 1) PI3K/Akt, 2) GSK-3β, 3) TLR4/NF-κB, 4) BH4 (cofactor of nNOS) biosynthesis enzyme GCH-1 (de novo), 5) nNOSα protein and dimerization, and 6) Nrf2 and Phase II antioxidant enzymes prevent 7) caspase 3 and 8) PARP cleavage and neuronal apoptosis in hyperglycemia. Nrf2 siRNA activates GSK-3β, a potential feedback control inhibiting 1) PI3K/Akt and activating 2) NF-κB enhancing inflammation and apoptosis. Arrow indicates activation, whereas bar indicates inhibition. ↑, upregulation; ⊥, inhibition, respectively. CNM, cinnamaldehyde; STZ, streptozotocin.

In conclusion, our results suggest that diabetes-induced gastrointestinal dysfunction may occur due to the increased hyperglycemia-induced apoptosis in pENCs. Hyperglycemia triggers the apoptosis of pENCs through suppression of Nrf2 and PI3K. CNM modulates pENCs survival by targeting PI3K/Akt pathway. Our results suggest a novel insight that siRNA knockdown of Nrf2 plays a major role regulating the upstream and downstream molecular cell survival targets and nNOS function. Further studies are warranted for better understanding of the cross talk between enteric glia versus enteric neuronal cells in the cell survival pathway and the antioxidant system. Finally, our studies may facilitate the screening and design of novel therapeutic compounds to treat gastrointestinal dysfunction in patients with diabetes.

GRANTS

This study was supported by National Institute of General Medical Sciences (NIGMS) of the National Institutes of Health under award number SC1GM121282 (to P. R. Gangula). The study were also supported by RCMI Infrastructure Core (CRISALIS) Grant U54MD007586.

DISCLOSURE

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.R.G. conceived and designed research; C.S. performed experiments; C.S. and A.V.R. analyzed data; C.S. and P.R.G. interpreted results of experiments; C.S. and A.V.R. prepared figures; C.S. and P.R.G. drafted manuscript; A.V.R., M.L.F., S.S., and P.R.G. edited and revised manuscript; C.S., A.V.R., M.L.F., S.S., and P.R.G. approved final version of the manuscript.