Sepiapterin alleviates impaired gastric nNOS function in spontaneous diabetic female rodents through NRF2 mRNA turnover and miRNA biogenesis pathway

Abstract

An impaired nitrergic system and altered redox signaling contribute to gastric dysmotility in diabetics. Our earlier studies show that NF-E2-related factor 2 (NRF2) and phase II antioxidant enzymes play a vital role in gastric neuronal nitric oxide synthase (nNOS) function. This study aims to investigate whether supplementation of sepiapterin (SEP), a precursor for tetrahydrobiopterin (BH₄) (a cofactor of NOS) via the salvage pathway, restores altered nitrergic systems and redox balance in spontaneous diabetic (DB) female rats. Twelve-week spontaneous DB and age-matched, non-DB rats, with and without dietary SEP (daily 20 mg/kg body wt for 10 days) treatment, were used in this study. Gastric antrum muscular tissues were excised to investigate the effects of SEP in nitrergic relaxation and the nNOS-nitric oxide (NO)-NRF2 pathway(s). Dietary SEP supplementation significantly (P < 0.05) reverted diabetes-induced changes in nNOS dimerization and function; nitric oxide (NO) downstream signaling molecules; HSP-90, a key regulator of nNOSα activity and dimerization; miRNA-28 that targets NRF2 messenger RNA (mRNA), and levels of microRNA (miRNA) biogenesis pathway components, such as DGCR8 (DiGeorge Syndrome Critical Region Gene 8) and TRBP (HIV1-1 transactivating response RNA-binding protein). These findings emphasize the importance of the BH₄ pathway in regulating gastric motility functions in DB animals by modulating nNOSα dimerization in association with changes in enteric NRF2 and NO downstream signaling. Our results also identify a new pathway, wherein SEP regulates NRF2 mRNA turnover by suppressing elevated miRNA-28, which could be related to alterations in miRNA biogenesis pathway components.

NEW & NOTEWORTHY This study is the first to show a causal link between NF-E2-related factor 2 (NRF2) and neuronal nitric oxide synthase (nNOS) in gastric motility function. Our data demonstrate that critical regulators of the miRNA biosynthetic pathway are upregulated in the diabetic (DB) setting; these regulators were rescued by sepiapterin (SEP) treatment. Finally, we show that low dihydrofolate reductase expression may lead to impaired nNOS dimerization/function-reduced nitric oxide downstream signaling and elevate oxidative stress by suppressing the NRF2/phase II pathway through miRNA; SEP treatment restored all of the above in DB gastric muscular tissue. We suggest that tetrahydrobiopterin supplementation may be a useful therapy for patients with diabetes, as well as women with idiopathic gastroparesis.

INTRODUCTION

Abnormalities in gastrointestinal (GI) motility occur in patients with both type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetes. The majority of patients (80%) with diabetic (DB) gastroparesis are women (25). The pathogenesis of gastric dysfunction and the mechanism for its sex bias are not well characterized. Earlier, we provided evidence that streptozotocin (STZ)-induced diabetes affects the ratio of gastric tetrahydrobiopterin (BH₄) to total biopterins associated with delayed gastric emptying, nitrergic relaxation [neuronal nitric oxide synthase (nNOS) function], and nNOS-α (nNOSα) dimerization in female rats (15, 16). It has been shown that supplementation with BH₄, a cofactor for nNOS, or sepiapterin (SEP), a precursor for BH₄ via salvage pathway, reversed delayed gastric emptying in DB female rats (15, 16). BH₄ may be synthesized mainly by two pathways: de novo and salvage. GTP cyclohydrolase I (GCH) is the rate-limiting enzyme in the de novo synthetic pathway of BH₄. Pyruvoyl-tetrahydropterin synthase and SEP reductase are two additional enzymes required for the final production of BH₄ (8). In the salvage pathway, BH₄ that is oxidized to 7,8-dihydrobiopterin (BH₂) may be recycled back to BH₄ by the activity of dihydrofolate reductase (DHFR). This conversion is particularly crucial, as elevated levels of BH₂ may compete with BH₄ to bind to nNOS and, thus, lead to uncoupling of the enzyme and superoxide production. We have reported earlier that inhibition of the de novo or salvage pathway impaired nNOS dimerization and nitrergic relaxation (15, 16).

Oxidative stress, through the production of reactive oxygen species (ROS), has been proposed as the cause underlying the development of diabetes mellitus (11). Although a relationship between ROS and diabetes is well established, the sources of ROS production remain controversial. Diminished nitric oxide (NO) and enhanced oxidative stress, due to lack of BH₄, play a central role in several pathophysiological pathways. NRF2 (NF-E2-related factor 2) is a redox-sensitive, basic leucine zipper transcriptional factor, which binds to the antioxidant response element (ARE) in the promoter region of phase II antioxidant enzymes, leading to an upregulation of antioxidant gene expression. Heme oxygenase-1 (HO-1) and the enzymes involved in glutathione (GSH) biosynthesis are two examples of phase II genes regulated by NRF2 (30). Induction of the phase II enzyme HO-1 lowered ROS and elevated nNOS in gastric antrum interstitial cells of Cajal, thereby mitigating diabetes-induced gastroparesis (7). However, no systemic studies regarding nNOS/NRF2 signaling in the stomach, as well as concerning gastroparesis, has been performed to date.

Relaxation of smooth muscle in the GI tract occurs due to the inhibitory actions of NO, mediated by the cyclic guanosine monophosphate (cGMP) system (12, 49). One of the mechanisms involved in the inactivation of the NO-cGMP-PKG pathway is oxidative stress, which affects NO bioavailability, as well as NO’s ability to activate soluble guanylate cyclase (sGC) (35). Low BH₄ levels impair the production of NO and lead to increased superoxide radical production, due to nNOS uncoupling (38). Our earlier studies demonstrate that the diminished intracellular BH₄:BH₂ ratio is the molecular trigger for NO insufficiency in diabetes (15).

MicroRNAs (miRNAs) are small noncoding RNAs with 22 nucleotides in length that control diverse biological functions by promoting degradation or inhibition of translation of target mRNAs (22). miRNAs are dysregulated in almost all human diseases, including diabetes (22). First, the primary transcript of miRNA (pri-miRNA), which is 5′ capped and 3′ polyadenylated in structure, mediated by RNA polymerase II, will be capped (17). The pri-miRNAs are cleaved in the nucleus into precursor miRNA (pre-miRNA) by a microprocessor complex consisting of DiGeorge Syndrome Critical Region Gene 8 (DGCR8), an RNA-binding protein, and Drosha, a type III RNase. Pre-miRNAs are processed by Dicer, an RNase III enzyme complex, in the cytoplasm into ~20–22-nucleotide-length mature miRNA. The mature miRNAs then promote the association of a large protein complex—termed the RNA-induced silencing complex (RISC)—with specific regions in the 3′-untranslated region (3′UTR) of protein-coding genes, and repress their expression by translational inhibition and/or promoting mRNA degradation (14). miRNA biogenesis is regulated at different stages, including, but not limited to, miRNA transcription and Drosha-/Dicer-mediated processing (18). Alterations in miRNA expression often lead to severe pathological consequences and are frequently observed in human diseases, including diabetes (13). However, the connection between miRNAs and NRF2 in the DB gastroparesis setting has yet to be investigated. Therefore, in the recent study, we examined the link between miRNA and NRF2 regulation in diabetes and the effect of SEP in these signaling pathways in DB female rats.

In the current study, we hypothesized that if SEP were used to increase the DHFR expression (salvage pathway enzyme) by exogenous supply, it would restore the gastric nNOS function via NRF2 and miRNA biogenesis in DB female rats. Therefore, we have investigated whether SEP supplementation attenuates an impaired gastric nNOS-NO-NRF2 signaling pathway. In addition, we also have investigated whether restoration of this pathway-normalized nitrergically mediated gastric motility in DB female rats.

MATERIALS AND METHODS

Experimental Rats and Induction of Diabetes

Control (BBn) and DB (BBd) rats (12–14 wk), derived initially from the Wistar-Furth strain, were obtained from the Sir Frederick Banting Research Center (Ottowa, ON, Canada). Diabetes incidence in these rats occurred spontaneously between 60 and 120 days of age. From the onset of diabetes, rats were given small daily doses of ultralente insulin (0.3–3.0 units) to prevent ketoacidosis and to maintain hyperglycemic levels between 300 and 400 mg/dl. The animals were maintained in the Meharry Medical College (MMC) Animal Care Facility (Nashville, TN) under controlled temperature, humidity, and 12-h:12-h light-dark cycle, with free access to a purified diet (Purina LabDiet, St. Louis, MO) and water. All experiments in this study were approved by the Institutional Animal Care and Use Committee at MMC, in accordance with recommendations of the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. Blood glucose levels were checked on alternate days, and diabetes was diagnosed by a plasma glucose level greater than 250 mg/dl. All experiments were performed 12 wk after the onset of diabetes.

Experimental Design

Animals.

Animals were divided into three groups: control female rats, DB female rats, and sepiapterin (SEP)-supplemented DB female (DB+SEP) rats. DB+SEP rats were provided with SEP tablets (20 mg·kg body wt^–1·day^–1) for 10 days. On the last day of SEP supplementation, animals were used for various experiments or euthanized to collect gastric antrum muscular tissue for future analysis. The tissue samples collected from animals were snap frozen in liquid nitrogen and stored at –80°C until analyzed.

Organ Bath Studies

Electric field stimulation (EFS)-induced nonadrenergic noncholinergic relaxation (NANC) was studied in circular gastric muscle strips (15, 16). Gastric muscle strips obtained from the antrum region of the stomach were tied with silk thread at both ends and were mounted in 10-ml water-jacketed organ baths containing Krebs buffer at 37°C and continuously bubbled with 95% O₂-5% CO₂ (DMT Technologies, Nottingham, UK). Tension for each muscle strip was monitored with an isometric force transducer and analyzed by a digital recording system (DMT Technologies). A passive tension equal to 2 g was applied to each strip in a 1-h equilibration period through an incremental increase (0.5 g, four times, at 15-min intervals). Strips were exposed to atropine, phentolamine, and propranolol (10 µmol each; Sigma Chemicals, St. Louis, MO) in bath solution for 1 h to block cholinergic and adrenergic responses. 5-Hydroxytryptamine (100 µmol) precontracted strips were exposed to EFS (90 V, 2 Hz, 1-ms pulse for the duration of 1 min) to elicit NANC relaxation. Relaxation response, elicited by low-frequency (2 Hz) stimulus under NANC conditions, as used in this study, was demonstrated as predominantly nitrergic in origin (15). The NO dependence of nitrergic relaxations was confirmed by preincubation with N^G-nitro-l-arginine methyl ester (l-NAME, 100 µM; 30 min) (33). At the end of the experiment, each muscle strip was blotted dry with filter paper and weighed. Comparisons between groups were performed by measuring the area under the curve (AUC/mg of tissue) of the EFS-induced relaxation (AUC_R) for 1 min and the baseline for 1 min (AUC_B), according to the formula (AUC_R – AUC_B)/weight of tissue (mg) = AUC/mg of tissue.

nNOSα Dimerization in Rat Gastric Muscular Tissue

Levels of the neuronal nitric oxide synthase-α (nNOSα) monomer and dimer were quantified by Western blot analysis via low-temperature (LT)-PAGE in gastric antrum homogenates, as described previously (15, 16). LT-SDS-PAGE was performed on ice. The LT process was used to identify nNOS dimers and monomers in the native state, as LT is known to prevent monomerization of nNOS dimers. For the LT process, 30 µg of protein in the standard Laemmli buffer at 4°C was used for SDS-PAGE. The mixture was incubated at 0°C for 30 min before LT-SDS-PAGE using a 6% separating gel. All gels and buffers were preequilibrated to 4°C before electrophoresis, and the buffer tank was placed in an ice bath during electrophoresis to maintain the gel temperature below 15°C. A polyclonal antibody specific to nNOSα (Thermo Fisher Scientific, Waltham, MA) and anti-rabbit IgG conjugated with horseradish peroxidase (Sigma Chemicals) were used as the primary and secondary antibodies, respectively.

Western Blot Analysis

Proteins were measured by Bio-Rad protein assay (Bio-Rad, Hercules, CA), and 30 μg of protein was separated by SDS-PAGE. The membrane was immunoblotted with polyclonal heat shock protein-90 (HSP-90), NRF2, DHFR, glutamate-cysteine ligase catalytic subunit (GCLC) and a modifier subunit (GCLM), PKG (Cell Signaling, Danvers, MA), HO-1 (Assay Designs, Ann Arbor, MI), soluble guanylate cyclase-α (sGC-α), and sGC-β (sGC-β) (Sigma Chemicals) primary antibodies, respectively. All of the primary antibodies were probed at 1:1,000 dilutions. Anti-rabbit IgG was conjugated with horseradish peroxidase (1:10,000; Sigma Chemicals) as a secondary antibody. Binding of antibodies to the blots was detected using an enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech, Piscataway, NJ) following the manufacturer’s instructions. Stripped blots were reprobed with β-actin-specific polyclonal antibodies (Sigma Chemicals) to enable normalization of signals between samples. Band intensities were analyzed using Bio-Rad ImageLab (Bio-Rad).

RT-Quantitative PCR

Total tissue RNA was isolated from the rat gastric muscular tissues by a single-step guanidine thiocyanate method using the reagent TRIzol (Invitrogen, Carlsbad, CA). The quality of RNA was determined by NanoDrop (Thermo Fisher Scientific), and the quantity was estimated by an Agilent 2100 bioanalyzer (Agilent Technologies, Houston, TX). To eliminate any contaminating DNA, RNA was treated with RNase-free DNase (Invitrogen). One microgram of DNase-treated RNA was used for cDNA synthesis. The iScript cDNA synthesis kit (Bio-Rad) was used to synthesize cDNA. One microliter of cDNA was used for each reaction, and the following primers were used: 1) NRF2 (F: 5′-CATTTGTAGATGACCATGAGTCGC-3′ and R 5′-TCCTGCCAAACTTGCTCCAT-3′); 2) DHFR (F: 5′-ACCCTGGCCTCTGCT CAGGA-3′ and R: 5′-AGCTGCCTCCGACTACCCAGA-3′); 3) GCH1 (GTP cyclohydrolase I) (F: 5′-GCCGAC CTGCCGAGAAGTCC-3′ and R: 5′-GACGAGTAAGCGGCCGCCAG-3′); 4) DGCR8 (DiGeorge Syndrome Critical Region Gene 8) (F: 5′-CAAGCAGGAGACATCGGA-3′ and R: 5′-CACAATGGACATCTTGGGC-3′); and 5) TRBP (HIV1–1 transactivating response RNA-binding protein) (F:5′-GGGCTGCCTAGTATAGAG-3′ and R:5′-GACCCGGAAGGTGAA ATTAG-3′).

RT-quantitative PCR (RT-qPCR) amplification was performed using the SYBR Green (Bio-Rad) method. Cycling conditions were 95°C for 3 min, followed by 45 cycles of 95°C for 30 s and 55°C for 1 min. Relative amounts of mRNA were normalized to GAPDH (F: 5′-GGGGCTCTCTGCTCCTCCCTG-3′ and R: 5′-AGGCGTCCGATACGGCCAAA-3′), and threshold cycle (CT) numbers were calculated (i.e., 2^–ΔΔCT, the C_t method), according to the manufacturer’s instructions (Bio-Rad). All studies were performed in the MMC Molecular Core Laboratory.

RT-qPCR assays were performed for measurement of the expression levels of DGCR8 and TRBP mRNAs and mature miRNA-28. Total RNA was isolated from tissues as described above and subjected to reverse transcription with the iScript cDNA synthesis kit (Bio-Rad), according to the manufacturer’s instructions. Quantitative RT-qPCR was performed using the SYBR Green Master Mix (Thermo Fisher Scientific) for mRNA or TaqMan MicroRNA assay kit (Applied Biosystems, Foster City, CA) for miRNA-28 on an ABI StepOne Real-Time PCR System (Thermo Fisher Scientific), in accordance with the manufacturer’s protocol. Data analysis was performed using the comparative C_t method. All reactions were carried out in triplicate. Results were normalized to GAPDH or U6 small nuclear RNA (snRNA). All studies were performed in the MMC Molecular Core Laboratory.

Statistical Analysis

Data were presented as means ± SE. Statistical comparisons between groups were performed by the Student’s t-test or the Tukey test after one-way ANOVA, using the GraphPad Prism Version 5.0 (GraphPad Software, San Diego, CA). A P value of <0.05 was considered statistically significant.

RESULTS

Effect of SEP Supplementation on Blood Glucose and Body Weight in DB Female Rats

Table 1 shows the role of SEP in the body weights and blood glucose levels in diabetic (DB) female rats. Fasting blood glucose levels were elevated significantly in female rats in both the DB (411 ± 36.0 mg/dl; P < 0.05) and DB+SEP (359.5 ± 10.5 mg/dl; P < 0.05) groups compared with control groups (Table 1). Although blood glucose levels were reduced upon SEP supplementation in DB female rats, this was not statistically significant. No significant weight loss was noted in the DB rats. SEP supplementation significantly elevated the BWs of DB female rats (226 ± 6.0) (Table 1).

Table 1.

Blood glucose level and body weight in the control and DB female rats

	C	C+SEP	DB	DB+SEP
Body weight, g
Before SEP treatment	236 ± 10.4	242 ± 7.7	223 ± 6.3	199 ± 2.0
After SEP treatment				226 ± 6.0*
Blood glucose, mg/dl
Before SEP treatment	109.8 ± 8.1	114.5 ± 2.3	394.3 ± 17.9†	411 ± 36.0†
After SEP treatment				359.5 ± 10.5†

Values are means ± SE. C, control group; DB, diabetic; SEP, sepiapterin.

^†P < 0.05 compared with control;

^*P < 0.01 compared with baseline before SEP treatment.

SEP Attenuated Diabetes-Induced Impairment of Gastric Nitrergic Relaxation

The effects of SEP on decreased nitrergic relaxation of antrum strips from female rats due to induction of diabetes followed by electric field stimulation (2 Hz) are presented in Fig. 1. Induction of diabetes caused a significant decrease (–0.37 ± 0.08) in nitrergic relaxation compared with control rats (–1.0 ± 0.15). As shown in Fig. 1, SEP supplementation did not affect the nitrergic relaxation in control rats (–0.93 ± 0.2), whereas a significant induction (–0.75 ± 0.05) in the nitrergic relaxation was observed when DB animals were supplemented with SEP.

Fig. 1.

Effect of sepiapterin (SEP) on nitrergic relaxation in diabetic female rats in vivo. Nitrergic relaxation in diabetic rats was measured following daily exposure to SEP (20 mg/kg body wt) for 10 days. Values are expressed as means ± SE of 4–6 animals. *P < 0.05 compared with the control group. #P < 0.05 compared with the diabetic group.

SEP Attenuated Diabetes-Induced Impairment of nNOSα Protein Expression and nNOSα Dimerization

According to Fig. 2A, the protein expression of nNOSα, the only functional isoform of nNOS in gastric muscular tissue, was decreased significantly (P < 0.05) following induction of diabetes. SEP supplementation to DB female rats resulted in significant restoration of the nNOSα protein. Further, we performed the dimerization study by LT-PAGE gel to establish whether the decreased nNOSα was the result of altered nNOSα dimer levels in DB rats. As depicted in Fig. 2B, a significant decrease in the dimer/monomer ratio of nNOSα was seen in DB female rats compared with controls (1.4 ± 0.06 vs. 0.97 ± 0.03). SEP supplementation resulted in complete reversal of diabetes-induced alteration of the dimer/monomer level (1.33 ± 0.03) of nNOSα.

Fig. 2.

Effect of sepiapterin (SEP) on neuronal nitric oxide synthase-α (nNOSα) protein expression and nNOSα dimerization of diabetic rat gastric tissues. nNOSα protein expression and nNOSα dimerization were measured following daily exposure to SEP (20 mg/kg body wt) for 10 days. Representative immunoblot and densitometric analysis data for nNOSα protein expression (A) and nNOSα protein dimerization (B) in female rat gastric pylorus. Values are expressed as means ± SE of four samples in each group. *P < 0.05 compared with the control group. #P < 0.05 compared with the diabetic group.

Diabetes Decreased Gastric Dihydrofolate Reductase Protein Expression

In addition to the de novo biosynthesis of BH₄ (tetrahydrobiopterin), mammalian cells may also generate BH₄ using an alternate pathway, whereby oxidized BH₂ (7,8-dihydrobiopterin) is converted to BH₄ by SEP reductase and dihydrofolate reductase (DHFR) (8). Increased BH₂ may compete with BH₄ and, thus, uncouple NOS. The diabetes-induced reduction of DHFR mRNA and protein expression in gastric muscular tissues from female rats is presented in Fig. 3, A and B, respectively. Induction of diabetes caused a significant decrease in DHFR levels compared with control rats. In contrast, neither diabetes nor SEP treatment affected the mRNA and protein expression of GCH1 (the enzyme critical for the de novo synthesis of BH₄) (Fig. 3, C and D).

Fig. 3.

Effect of sepiapterin (SEP) on mRNA and protein expression of dihydrofolate reductase (DHFR; A and B) and GTP cyclohydrolase I (GCH1; C and D) in diabetic rat gastric muscular tissues. Representative immunoblot and densitometric analysis data for DHFR (A and B) and GCH1 (C and D) mRNA and protein expression in female rat gastric muscular tissue. Values are expressed as means ± SE of four samples in each group. *P < 0.05 compared with the control group. #P < 0.05 compared with the diabetic group.

SEP Attenuated Diabetes-Induced Reduction of HSP-90 Protein Expression

HSP-90 was a key player in regulating nNOS and endothelial NOS (eNOS) activity, dimerization, and nitrergic relaxation (2, 32). SEP improved eNOS-HSP-90 association in persistent pulmonary hypertension of newborns (45). Our results show that the protein expression of HSP-90 was reduced significantly (P < 0.05) in DB gastric muscular tissue compared with that of the control group. In vivo SEP supplementation reversed HSP-90 protein expression to control levels (Fig. 4).

Fig. 4.

Effect of sepiapterin (SEP) on gastric muscular heat shock protein-90 (HSP-90) protein expression in diabetic female rats. Representative immunoblot and densitometric analysis data for HSP-90 protein expression in female rat gastric muscular tissue. Values are expressed as means ± SE of four samples in each group. *P < 0.05 compared with the control group. #P < 0.05 compared with the diabetic group.

To demonstrate that the changes noticed in nNOS expression and structure were of functional significance, we performed experiments to test further nitric oxide (NO) downstream signaling pathway.

SEP Reversed Diabetes-Induced Reduction in the Gastric Smooth Muscle Response to the Activation of the Cyclic Guanosine Monophosphate Pathway

To investigate whether downregulation of nNOS/NRF2 pathway impairs gastric smooth muscle relaxation, we investigated NO downstream signaling molecules related to smooth muscle relaxation in DB female rats. Fig. 5 shows the effects of SEP on the diabetes-induced reduction of NO-cGMP-mediated downstream signaling molecules in gastric muscular tissues from female rats. According to Fig. 5, A and B, induction of diabetes caused a significant decrease in the protein expression of soluble guanylate cyclase-α (sGC-α) (0.02 ± 0.01) and sGC-β (sGC-β) (0.01 ± 0.09) compared with control rats (0.31 ± 0.03 and 0.86 ± 0.11, respectively). SEP supplementation resulted in complete restoration of both sGC-α and sGC-β protein expressions (0.37 ± 0.34 and 0.85 ± 0.18, respectively). No significant change in the level of PKG in gastric muscle was observed in DB female rats (Fig. 5D).

Fig. 5.

Effect of sepiapterin (SEP) on soluble guanylate cyclase (sGC) (α and β), and PKG protein expression of diabetic rat gastric muscular tissues. Protein expression was measured following daily exposure to SEP (20 mg/kg body wt) for 10 days. Representative immunoblot and densitometric analysis data for sGC-α (A) sGC-β (B and C) and PKG (B and D); in female rat gastric muscular tissue. *P < 0.05 compared with control group. #P < 0.05 compared with the diabetic group.

Attenuation of a Diabetes-Induced Decrease in the NRF2 Protein and mRNA Expression, as Well as the Phase II Enzyme Level, by SEP

Our studies have shown that suppression of NRF2 and selective phase II enzyme protein expression in the low-density lipoprotein receptor (LDLR)-KO and ApoE-null mouse stomachs were restored by SEP treatment in vivo (39). However, it is unknown whether hyperglycemia may affect this path, and if so, whether it may be attenuated by SEP treatment. Next, we investigated both NRF2 and its downstream molecules in DB female rats. According to Fig. 6, A and B, the mRNA and protein levels of NRF2 in gastric muscular tissue were decreased significantly (P < 0.05) following DB induction. SEP supplementation to DB female rats resulted in significant restoration of NRF2 protein levels. DB induction also caused a significant (P < 0.05) decrease in HO-1 protein levels (1.3 ± 0.45) in female rat gastric muscle tissues (Fig. 6C) compared with control rats (2.04 ± 0.23), and this was restored by SEP supplementation (1.9 ± 0.28). Glutamate-cysteine ligase (GCL) is the rate-limiting enzyme in GSH synthesis and is composed of a catalytic subunit (GCLC) and a modifier subunit (GCLM) (4). This is another group of phase II enzymes regulated by NRF2. No significant change was observed in GCLC expression in any of the conditions, whereas a significant decrease in the level of GCLM protein (Fig. 6D) was observed in DB female rats (0.44 ± 0.13) compared with controls (0.78 ± 0.1). The diabetes-induced decrease in the GCLM level was restored (0.66 ± 0.14) by SEP supplementation.

Fig. 6.

Effect of sepiapterin (SEP) on NF-E2-related factor 2 (NRF2) mRNA, protein, and heme oxygenase-1 (HO-1) protein expression of diabetic rat gastric muscular tissues. Both mRNA and protein expression was measured following daily exposure to SEP (20 mg/kg body wt) for 10 days. Representative immunoblot and densitometric analysis data for NRF2 mRNA expression (A), NRF2 protein expression (B), HO-1 protein expression (C), and GCLM protein expression (D) in female rat gastric muscular tissue. Values are expressed as means ± SE (n = 4). *P < 0.05 compared with control group. #P < 0.05 compared with the diabetic group.

Effect of Diabetes on miRNA-28 and DiGeorge Syndrome Critical Region Gene 8 HIV1-1 Transactivating Response RNA-Binding Protein mRNA Expression Levels

We found decreased levels of NRF2 mRNA and protein in gastric muscular tissue following DB induction (Fig. 6, A and B). miRNA-28 has been reported to target NRF2 mRNA and protein by binding to 3′UTR of NRF2 mRNA in breast cancer cells (50). Ectopic expression of miRNA-28 also reduced NRF2 mRNA and protein levels. An inverse correlation was found between miRNA-28 expression and NRF2 mRNA levels (50). Therefore, we investigated whether a miRNA-28 level has a role in targeting NRF2 in DB condition and whether SEP supplementation attenuated the expression of miRNA-28. DB condition upregulated the expression level of miRNA-28 in gastric muscular tissue samples (Fig. 7A), and we noticed significant restoration (reduction) of miRNA-28 after SEP supplementation to DB female rats (Fig. 7A).

Fig. 7.

Effect of sepiapterin (SEP) on miRNA-28, DiGeorge Syndrome Critical Region Gene 8 (DGCR8) and HIV1-1 transactivating response RNA-binding protein (TRBP) mRNA expression of diabetic rat gastric muscular tissue. Expression levels of miRNA-28 (A), DGCR8 (B), and TRBP (C) mRNAs were measured following daily exposure to SEP (20 mg/kg body wt) for 10 days. Values are expressed as means ± SE; n = 4. *P < 0.05 compared with control group. #P < 0.05 compared with the diabetic group.

miRNA biogenesis is regulated at different levels, including at the level of processing by Drosha and DGCR8 proteins in the nucleus, Dicer and TRBP proteins in the cytoplasm (51). Double-stranded RNA-binding proteins, such as DGCR8 and TRBP, have key roles in the processing of miRNAs (41). DGCR8 processes pri-miRNA to pre-miRNA in the nucleus (41), whereas TRBP plays an essential role in the maturation of pre-miRNA to mature miRNAs in the cytoplasm (21).

To check the possibility of involving miRNA biogenesis components such as Drosha, DGCR8, Dicer, TRBP2, and Ago2 in miRNA-28 regulation, we sought to determine these mRNA levels in gastric muscular tissue samples using RT-qPCR. As shown in Fig. 7, B and C, DGCR8 and TRBP mRNA levels were upregulated in response to diabetes but not other proteins (data not shown). Interestingly, these mRNAs were downregulated significantly by SEP treatment (Fig. 7, B and C).

DISCUSSION

Diabetes in the BBn rat (a model of human type I diabetes) results from an autoimmune disease, whereby insulin-producing β-cells of the pancreas are destroyed (31). We reported earlier that diabetes induced by streptozotocin (STZ) causes the reduction in nitrergic regulation of gastric motility and leads to delayed gastric emptying in female rats (15). In addition, we also showed a decrease in intracellular BH₄ (tetrahydrobiopterin) availability in gastric muscular tissue, accompanied by a reduction in the expression and dimerization of the nNOSα protein. This is an effect that could be reversed by BH₄ or SEP supplementation in STZ DB female rats, as well as in LDLR knockout female mice, a model for hyperlipidemia and moderate oxidative stress (8, 16, 39). Finally, studies from our laboratory demonstrated that both nNOS function and the NRF2-phase II system was reduced in ApoE knockout female mice gastric muscular tissue (39). A schematic summary of the results obtained from this study was shown in Fig. 8. The data from the study show that gastric motility is impaired due to the loss of nNOS dimerization and nitrergic relaxation causing downregulation of NO downstream signaling pathway at the onset of diabetes, whereas SEP supplementation restores gastric motility (Fig. 8).

Fig. 8.

A schematic diagram of the role of neuronal nitric oxide synthase (nNOS), miRNA, and NF-E2-related factor 2 NF-E2-related factor 2 (NRF2) pathway in the induction of gastric motility in diabetes. Loss of nNOS dimerization increases the levels of miRNA processing proteins DiGeorge Syndrome Critical Region Gene 8 (DGCR8) and HIV1-1 transactivating response RNA-binding protein (TRBP), leading to the biogenesis of miRNA-28, which targets NRF2, causing reduced nitrergically mediated gastric motility. Supplementation of sepiapterin normalizes the gastric function by restoring the pathways outlined above. Arrow indicates activation, whereas bar indicates inhibition. ↑ and ↓ marks in the parenthesis indicate increase and decrease in levels, respectively.

We previously demonstrated that diabetes induction by STZ reduced nitrergic relaxation and expression of the nNOSα protein and dimer level of nNOSα, and that the BH₄ stabilizes the functionally active, dimeric form of nNOS in the pylorus (15). The results demonstrated that spontaneous diabetes causes the decrease in nitrergic relaxation (Fig. 1) and the protein expression of nNOSα (Fig. 2). SEP treatment restored both nitrergic relaxation and nNOSα protein and dimerization in gastric antrum (Figs. 1 and 2). Selective proteolytic degradation of NOS is a mechanism for regulation of the enzyme. The rapid proteolytic degradation by calpain has been a reason for the absence of nNOS in skeletal muscle sarcolemma of muscular dystrophy patients (26). Moreover, nNOS has been found in ubiquitin conjugates in lactacystin-treated human embryonic kidney-293 cells or rat brain homogenates (5), strongly suggesting that ubiquitin-proteasome pathway regulates the degradation of nNOS in vivo. Studies with the use of reticulocyte lysates and purified NOS indicate that the monomeric form of nNOS is preferentially ubiquitinated (5), which may complement our observations explaining the reduction of nNOSα monomer in the DB tissue (Fig. 2). It is possible that SEP may protect degradation of the enzyme, as well as improve the stabilization of both the nNOS monomer and dimer levels. Additional studies are warranted to address this in detail.

In addition to the de novo biosynthesis of BH₄, mammalian cells may generate BH₄ via an alternate pathway, whereby SEP is converted to BH₄ by SEP reductase and DHFR (5). Cellular BH₄ levels have increased both in vitro and in vivo by exogenous supply of BH₄ via a salvage pathway (8). This information led us to find the molecular mechanism by which SEP supplementation normalizes the gastric dysmotility in DB rats. BH₂ reductase activity of DHFR is crucial in determining cellular BH₄ homeostasis, NO bioavailability, and ultimately NOS coupling. Therefore, it is likely that DHFR expression is critical to cells that do not contain the apparatus required for efficient synthesis of BH₄ (9). SEP is the precursor for BH₄ biosynthesis in the salvage pathway. To avoid production of superoxide and to promote NO synthesis, BH₂ in the body is rapidly converted back to BH₄ through the action of DHFR (8). To complement this finding, we specifically measured the level of DHFR in gastric muscular homogenates obtained from control and DB rats. Our results demonstrate that diabetes caused a significant reduction of DHFR but not the GCH1 levels (Fig. 3), which may ultimately decrease the BH₄ content. Our results have been consistent with earlier reports in which Ren et al. (40) showed that DHFR protein levels are reduced significantly in STZ-induced DB mice.

Recent studies have shown that the nNOSα activity, dimerization, and nitrergic relaxation are regulated by HSP-90 (2, 32). Others indicate that SEP improved cellular eNOS-HSP-90 association and eNOS function, as well as restored angiogenesis in persistent pulmonary hypertension of newborns (45). The results from this study further demonstrate that SEP treatment restored HSP-90 protein expression in DB gastric muscular tissue (Fig. 4). Collectively, these data suggest that SEP may also improve nitrergic function directly by posttranslationally providing BH₄ and, thus, attenuating the DB-induced impairment of nNOSα dimerization.

It has been well established that nNOS present in nonadrenergic, noncholinergic neurons plays a vital role in the production of NO (44). Thereafter, NO diffuses to the GI smooth muscle cells, and relaxation of smooth muscle in the GI tract occurs due to the inhibitory actions of NO. The downstream effects of NO on GI smooth muscle cells include the activation of sGC and the production of cGMP, which, in turn, stimulates PKG and results in smooth muscle relaxation due to a decrease in intracellular calcium concentration (47). As SEP attenuated the diabetes-induced reduction of nNOS activity (Fig. 1), we determined the effect of SEP on the NO downstream signaling pathway. Diabetes induction significantly reduced sGC (both α and β) levels in DB rats compared with the control. In vivo, SEP supplementation significantly attenuated the impaired gastric sGC (α and β) level (Fig. 5, A and B). We did not find any significant alterations in the levels of PKG (Fig. 5C), suggesting that activity of PKG may be altered due to reduced synthesis of cGMP.

BH₄ undergoes auto-oxidation by reactive oxygen species (ROS) and may be depleted under oxidative stress conditions; subsequently, nNOS reactions may be uncoupled and produce ROS instead of NO, causing a further increase in oxidative stress (34). In our previous study, we have reported that biopterin oxidation (BH₂ and B) would diminish the enzyme activity by uncoupling of electron flow of nNOS, leading to superoxide production and degrading NO, forming peroxynitrite that oxidizes BH₄ to BH₃⁺ and subsequently to BH₂ and B (33). The peroxynitrite formation also nitrates protein tyrosine residues to 3-nitrotyrosine, which disrupts NO signaling and inactivates voltage-gated ion channels, disrupting muscle contraction (1, 37). The bZIP transcription factor NRF2 has emerged as a pivotal regulator of intracellular redox homeostasis by controlling the expression of many endogenous antioxidants and phase II detoxification enzymes (23). We recently demonstrated that nNOS dimerization and function have been reduced, and this was accompanied by elevated oxidized biopterin levels in NRF2-null mice gastric muscular tissue (33). It has been reported that the activation of NRF2 increases the availability of NO in naive and stressed cells (20). This suggests that induced oxidative stress, such as cultivation in high glucose, apparently fails to trigger the NRF2 safeguard mechanism and results in reduced NO production. These studies further report a direct role of BH₄ in maintaining eNOS dimerization by activation of the NRF2 pathway in vitro (6). Various NO donors are also inducing a rapid nuclear translocation of NRF2 (29). Furthermore, NO increases HO-1 expression via the NRF2/ARE (antioxidant response element) pathway and promotes vascular smooth muscle survival. We observed that the NRF2 levels (Fig. 6, A and B), as well as the phase II enzymes HO-1 (Fig. 6C) and GCLM (Fig. 6D), were reduced in DB female rats. The data also show that SEP supplementation restores phase II enzymes in the DB female rats (Fig. 6). Our findings agree with the earlier report by Choi et al., who showed that the expression of HO-1 was downregulated in spontaneous DB gastric muscular tissue (7). Furthermore, they report that induction of HO-1 reduced elevated oxidative stress and restored gastroparesis, suggesting the importance of the NRF2-dependent phase II pathway in gastric motility function (7). Recent studies also suggest that HSP-90 influences NRF2 release from Keap1 under stress conditions (43). GCH1, a BH₄ biosynthetic enzyme, is stabilized by HSP90, a client protein (42). Although the underlying molecular mechanism of SEP on restoring NRF2 and phase II enzymes is currently unknown, these studies may reveal that NO elicits a signal that presumably passes through the Keap1-NRF2 complex, resulting in the dissociation of Keap1-NRF2, followed by nuclear accumulation of NRF2.

miRNAs are a recently discovered class of small noncoding RNA gene products that have been implicated in a variety of physiological processes, including glucose homeostasis. miRNAs are believed to regulate the gene expression by mRNA degradation and the inhibition of translation initiation (28). miRNAs are estimated to regulate the translation of more than 60% of protein-coding genes.

Biogenesis of miRNAs takes place through a multistep process that involves the RNase III enzymes Drosha and Dicer. miRNAs are mostly transcribed from intragenic or intergenic regions by RNA polymerase II into primary transcripts called pri-miRNAs (28). The primary transcripts undergo further processing by the ribonucleases Drosha and DGCR8, a complex in the nucleus, thereby resulting in a hairpin intermediate (~70–100 nt) called pre-miRNA (27). The pre-miRNA is then transported from the nucleus to the cytoplasm by exportin 5 (24). In the cytoplasm, the pre-miRNA is processed by another ribonuclease, Dicer, which ultimately results in the production of mature miRNAs of ~22 nt (10). These molecules are loaded by the Dicer-TARBP2 (TAR RNA-binding protein 2; also known as TRBP) complex into a member of the Argonaute protein subfamily to form the RISC, and they direct the regulation of mRNA by recognizing a complementary sequence located at the 3′UTR in the targeted mRNA. Loading of miRNAs into RISC and the function of miRNA machinery are tightly regulated (10, 24). In human diseases, particularly cancer, it has been shown that epigenetic and genetic defects in miRNAs and their processing machinery are a common hallmark of illness (36).

Emerging research has suggested that miRNAs play a critical role in the pathogenesis of diabetes and its related cardiovascular complications (46, 48). In addition, several lines of evidence suggest that elevated levels of specific miRNAs downregulated NRF2 and DHFR mRNA expression (4). However, this path has not been shown in the diabetes gastroparesis setting.

In the current study, we provided evidence for the first time that critical regulators of the miRNA biosynthetic pathway are abnormally expressed in the DB setting; these regulators were rescued by SEP treatment. However, the cause of abnormal expression of these proteins in the DB environment remains to be determined. The microprocessor also has an additional role in destabilizing DGCR8 mRNA by cleaving the hairpin structures embedded in DGCR8 mRNA. DGCR8 stabilizes the Drosha protein by protein–protein interaction (19). It has also been suggested that miRNAs themselves may regulate expression of biosynthetic components (3). However, this is clearly an area of research that warrants further investigation (25) to answer the following questions: Why and how these proteins are aberrantly expressed (15)? Why is NRF2 targeted by miRNA-28 (16)? Does NRF2 have a role in processing or induction of miRNA-28? Is there feedback regulation of miRNA-28 and NRF2 in the DB setting and restoration of these parameters by SEP (8)?

In summary, for the first time, we report that low DHFR expression may lead to impaired nNOS dimerization/function-reduced NO downstream signaling and elevate oxidative stress by suppressing the NRF2/phase II pathway through miRNA; SEP treatment restored all of the above in DB gastric muscular tissue. We suggest that BH₄ supplementation may be a useful therapy for DB, as well as idiopathic, gastroparesis in women.

GRANTS

Research reported in this publication was supported by the National Institute of General Medical Sciences of the National Institutes of Health (NIH) Grants SC1GM121282 (to P. Gangula) and R21DKO76704 (to P. Gangula) and Meharry Translational Research Center Grant 5U54MD007593-09. Dr. Challagundla’s laboratory is supported, in whole or part, from the NIH Grant 1K22CA197074. Finally, we thank the Meharry Office for Scientific Editing and Publications for scientific editing support (Grant S21MD000104).

DISCLOSURES

P. R. Gangula has filed a patent application for the use of BH₄ in gastroparesis subjects through the University of Texas Medical Branch, Galveston, TX.

AUTHOR CONTRIBUTIONS

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