Noninflammatory upregulation of nerve growth factor underlies gastric hypersensitivity induced by neonatal colon inflammation

Abstract

Gastric hypersensitivity is one of the key contributors to the postprandial symptoms of epigastric pain/discomfort, satiety, and fullness in functional dyspepsia patients. Epidemiological studies found that adverse early-life experiences are risk factors for the development of gastric hypersensitivity. Preclinical studies found that neonatal colon inflammation elevates plasma norepinephrine (NE), which upregulates expression of nerve growth factor (NGF) in the muscularis externa of the gastric fundus. Our goal was to investigate the cellular mechanisms by which NE upregulates the expression of NGF in gastric hypersensitive (GHS) rats, which were subjected previously to neonatal colon inflammation. Neonatal colon inflammation upregulated NGF protein, but not mRNA, in the gastric fundus of GHS rats. Western blotting showed upregulation of p110γ of phosphatidylinositol 4,5-bisphosphate 3-kinase (PI3K), phosphoinositide-dependent kinase-1 (PDK1), pAKT(Ser473), and phosphorylated 4E-binding protein (p4E-BP1)(Thr70), suggesting AKT activation and enhanced NGF protein translation. AKT inhibitor MK-2206 blocked the upregulation of NGF in the fundus of GHS rats. Matrix metalloproteinase 9 (MMP-9), the major NGF-degrading protease, was suppressed, indicating that NGF degradation was impeded. Incubation of fundus muscularis externa with NE upregulated NGF by modulating the protein translation and degradation pathways. Yohimbine, an α₂-adrenergic receptor antagonist, upregulated plasma NE and NGF expression by activating the protein translation and degradation pathways in naive rats. In contrast, a cocktail of adrenergic receptor antagonists suppressed the upregulation of NGF by blocking the activation of the protein translation and degradation pathways. Our findings provide evidence that the elevation of plasma NE induces NGF expression in the gastric fundus.

functional dyspepsia (FD) is a complex, multifactorial functional bowel disorder whose symptoms include postprandial epigastric pain/discomfort, early satiation, abdominal bloating, nausea, and vomiting (33). These symptoms result from sensory-motor dysfunctions in the gastric wall and impaired signal processing in the spinal cord and the central nervous system. The underlying cellular and molecular mechanisms of these dysfunctions in FD patients remain unknown, primarily because of the lack of availability of live tissues from human subjects and limitations of interventional approaches for ethical and safety considerations. However, epidemiological studies have identified that adverse early-life experiences (AELE), including abuse (12), gastrointestinal infections/allergy (29), and neonatal gastric suction (2), are risk factors for the development of FD, including the symptom of epigastric pain in children as well as in adults (4, 30, 31). Preclinical studies in rodent models also found that neonatal colon inflammation (NCI) (35) or irritation to the stomach (20) induces hypersensitivity to stomach distension in adult life.

Mechanistic studies in a preclinical model demonstrated that NCI induces gastric hypersensitivity (GHS) in adult life by upregulating expression of nerve growth factor (NGF) in the muscularis externa of the gastric fundus (35). The upregulation of NGF was critical in inducing GHS, because neutralization of NGF by its antibody significantly suppressed GHS. Other studies have reported that inflammation in peripheral tissues upregulates NGF expression, which sensitizes the afferent neurons to induce inflammatory hyperalgesia (3, 15, 21). The increase of peripheral NGF was secondary to the release of proinflammatory cytokines from multiple immune cell types. However, in the preclinical model of NCI, the upregulation of NGF in the fundus muscularis externa was not associated with an inflammatory response, defined by myeloperoxidase activity and upregulation of proinflammatory cytokines (35). Instead, the basal plasma level of norepinephrine (NE) was upregulated and blockade of α₁-, β₁-, and β₂-adrenergic receptors (ARs) suppressed NGF expression as well as GHS (35).

The cellular mechanisms of upregulation of NGF by NE remain unknown. Our initial findings showed that NGF mRNA was not upregulated in the fundus muscularis externa of GHS rats subjected previously to NCI. Therefore, we tested the hypothesis that NE upregulates NGF expression in GHS rats by posttranscriptional mechanisms. We found that NE enhances protein translation of NGF from the existing Ngf mRNA through the phosphatidylinositol 4,5-bisphosphate 3-kinase (PI3K) (p110γ)→phosphoinositide-dependent kinase-1 (PDK1)→phospho-AKT (pAKT)→phosphorylated 4E-binding protein (p4E-BP1) signaling pathway. Concurrently, NE inhibits the expression of matrix metalloproteinase 9 (MMP-9), which is a major NGF-degrading protease (14), and hence attenuates NGF degradation.

METHODS

Reagents.

Phentolamine, propranolol, CL316243, and yohimbine hydrochloride (YOH) were purchased from Bachem Americas (Torrance, CA). NE and 2,4,6-trinitrobenzenesulfonic acid (TNBS) were purchased from Sigma (St. Louis, MO), and MK-2206 was from ChemieTek (Indianapolis, IN).

Animals.

Male Sprague-Dawley rats were used in all experiments. The institutional animal care and use committee at the University of Texas Medical Branch at Galveston approved all procedures performed on animals.

Five-day-old and six-week-old male Sprague-Dawley rats were purchased from Harlan Laboratories (Houston, TX). For neonatal inflammatory insult, TNBS (130 mg/kg, dissolved in 200 μl saline containing 10% ethanol) was injected intrarectally 2 cm into the colon of male pups on postnatal day 10 (PND 10). The animals were kept in a head-down position while the anus was held closed for 1 min to prevent leakage of the TNBS solution. Rats in the control group received 200 μl of saline. Six to nine weeks later, the animals were euthanized by CO₂ inhalation to obtain tissues.

Six- to-nine-week-old naive adult male rats were used for in vivo treatment with 0.1 mg/kg YOH dissolved in 0.5 ml of saline and administered daily to each rat by intraperitoneal (ip) route for 3 days. The control rats received daily injection of 0.5 ml of saline. The animals were euthanized 1 h after the last injection. The mucosa/submucosa was scraped off, and fundus muscularis externa was collected, snap-frozen in liquid nitrogen, and stored at −80°C.

In vitro treatment of rat fundus muscularis externa.

Fundus muscularis externa from adult rats was dissected and immersed in carbogenated Krebs solution with a 5% O₂-95% CO₂ mix (17). After the mucosa/submucosa was gently scraped away, the remaining muscularis externa was cut into small pieces, placed in high-glucose DMEM (HyClone, South Logan, UT) containing 10% FBS, antibiotics, and NE, and incubated for 24 h in a 37°C CO₂ incubator. All treated tissues were snap-frozen in liquid nitrogen and stored at −80°C.

Western blot.

Western blotting was performed as described previously (17). Eighty micrograms of proteins from each sample was loaded into each lane. The antibodies were as follows: NGF rabbit polyclonal (sc-549, 1:200), tissue inhibitor of metalloproteinase 1 (TIMP-1) rabbit polyclonal (sc-5538, 1:200) (Santa Cruz Biotechnology, Dallas, TX); β-actin mouse monoclonal (A5441, 1:5,000), AKT rabbit polyclonal (4685S, 1:1,000), pAKT(Ser473) (9271S, 1:1,000), PI3K p110γ rabbit monoclonal (5405S, 1:1,000), PDK1 rabbit polyclonal (3062S, 1:1,000), p4E-BP1(Thr70) rabbit polyclonal (9455S, 1:1,000), mammalian target of rapamycin (mTOR) rabbit polyclonal (2972S, 1:500), phospho (p)-mTOR(Ser2448) rabbit polyclonal (5536S, 1:500), MMP-9 rabbit polyclonal (3852S, 1:1,000) (Cell Signaling). Secondary antibodies were Alexa Fluor 680 donkey anti-rabbit IgG (1:10,000) (Invitrogen) and anti-mouse IgG IRDye800 Conjugated Antibody Pre-adsorbed (1;10,000) (Rockland).

Real-time RT-PCR.

Total RNA was extracted with the RNeasy Mini Kit (Qiagen, Valencia, CA). One microgram of total RNA was reverse-transcribed with the SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen) (18). Quantification of Ngf mRNA levels by real-time PCR was performed with a StepOnePlus Thermal Cycler and TaqMan probe and primers (Applied Biosystems, Foster City, CA). 18S rRNA was quantified as internal control for the amount and quality of cDNA. All samples were assayed in triplicate in an Optical 96-well reaction plate with Optical Adhesive Covers in a 20-μl volume containing 5 μl (2 μl for 18S rRNA) of diluted cDNA (1:5 dilution in water).

NGF immunohistochemistry.

Tissue was fixed in 10% formalin and was embedded in paraffin. Ten-micrometer sections were treated with NGF primary antibody (sc-532, Santa Cruz Biotechnology; 1:200 dilution) overnight. NGF staining was visualized with the Vectastain ABC kit and 3,3′-diaminobenzidine (DAB) substrates from Vector Labs per the manufacturer's instructions. Sections were counterstained with hematoxylin. No DAB staining was observed in sections without primary antibody.

Statistics.

All data are expressed as means ± SE and were analyzed by two-tailed Student's t-test or by one-way analysis of variance (ANOVA) followed by Fisher post hoc analysis, considering P < 0.05 as significant.

RESULTS

Upregulation of NGF in fundus muscularis externa of adult rats subjected to neonatal colon inflammation.

Of 30 adult rats (7–9 wk old) subjected to NCI, 14 showed a significant increase in NGF protein expression in the muscularis externa of the gastric fundus (1.59 ± 0.18 vs. 1.00 ± 0.13 control, P < 0.05) (Fig. 1, A and B). We reported previously that increase in expression of NGF in the fundus muscularis externa induces GHS (35). We called the responder rats “gastric hypersensitive (GHS) rats,” and the remainder were nonresponder rats. The upregulation of NGF in GHS rats occurred only in the fundus, not in the corpus or the antrum (data not shown). GHS rats and nonresponder rats did not show any significant increase of Ngf mRNA (Fig. 1C) compared with control rats, suggesting a posttranscriptional upregulation of NGF in GHS rats. NGF immunoreactivity was observed in the myenteric plexus, smooth muscle cells, and a subset of cells in the submucosa (Fig. 2). NGF staining appeared more intense in the muscle layers and submucosa from GHS vs. control rats.

Fig. 1.

Neonatal colon inflammation upregulates expression of nerve growth factor (NGF) protein (A and B), but not mRNA (C), in the gastric fundus muscularis externa of gastric hypersensitive (GHS) rats vs. nonresponder (NR) rats and control (Ctr) rats [those treated with saline instead of 2,4,6-trinitrobenzenesulfonic acid (TNBS) as neonates; n = 15 Ctr rats, n = 16 NR rats, and n = 14 GHS rats. *P < 0.05 vs. NR and Ctr rats. NR, nonresponder rats; Ctr, control rats; GHS, gastric hypersensitive rats.

Fig. 2.

NGF immunoreactivity (IR) in brown in sections from Ctr (A and B) and GHS (C and D) rats counterstained with hematoxylin. NGF IR was observed in myenteric plexus, smooth muscle and a subset of cells in the submucosa (lower magnification; A and C). Higher magnification shows close-up of staining in the muscle layers (B and D). Staining of keratinized epithelium was nonspecific. NCI, neonatal colon inflammation; M/SM, mucosa/submucosa; CM, circular muscle; LM, longitudinal muscle.

Neonatal inflammation activates PI3K/PDK1/AKT signaling pathway in GHS rats.

Protein expression is regulated at multiple steps, including during the translation of existing mRNA and protein degradation (13). The AKT signaling pathway has pronounced effects on translational regulation (26). We investigated whether neonatal inflammation constitutively modulates PI3K/PDK1/AKT signaling in the rat fundus to upregulate NGF expression. Western blotting showed that total AKT protein expression remained unchanged in the fundus of the GHS rats (Fig. 3A); however, the phosphorylation of AKT at Ser473 was significantly upregulated (1.00 ± 0.06 vs. 1.47 ± 0.09, P < 0.05; n = 6) (Fig. 3B). The catalytic subunit p110γ of PI3K was concurrently upregulated (1.00 ± 0.10 vs. 1.48 ± 0.14, P < 0.05; n = 6) (Fig. 3C). There was also a significant increase of PDK1 in the fundus of GHS vs. control rats (1.00 ± 0.14 vs. 1.41 ± 0.11, P < 0.05; n = 6) (Fig. 3D). However, both pAKT(Ser473) and PDK1 were not significantly altered in the fundus muscularis externa of nonresponder rats (P = 0.204 and 0.261, respectively).To demonstrate that AKT activation plays a critical role in upregulating NGF, we treated the GHS rats with 10 mg/kg MK-2206, a potent and selective AKT inhibitor, for 4 consecutive days (daily ip). The augmentation of NGF protein expression in GHS rat fundus was blocked by MK-2206 treatment (Fig. 3, E and F).

Fig. 3.

Effects of neonatal colon inflammation on protein translation signaling pathway in the gastric fundus of GHS rats. A: total AKT protein level was not affected. B: phospho (p)AKT(Ser473) expression was significantly upregulated in GHS vs. Ctr rats. C: phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) p110γ was elevated in GHS vs. Ctr rats. D: phosphoinositide-dependent kinase-1 (PDK1) protein was elevated in GHS vs. Ctr rats; n = 6. *P < 0.05 vs. control rats. E and F: AKT inhibitor MK-2206 blocked upregulation of NGF in the gastric fundus of GHS rats; n = 5 (Ctr), 10 (NR), 6 (GHS), and 4 (GHS + MK-2206). *P < 0.05 vs. Ctr and NR rats, #P < 0.05 vs. GHS rats.

Increased phosphorylation of 4E-BP1 contributes to translation of NGF protein.

Eukaryotic translation initiation factor 4E (eIF4E)-binding protein 1 (4E-BP1), a translational inhibitor, is the most abundant form of the three 4E-BPs in a majority of tissues, except the brain (34, 36). It regulates the activity of eIF4E by preventing its assembly into the eIF4F complex and thus inhibits the association of the small ribosomal subunit with the mRNA, leading to the suppression of translation (11, 27). Hyperphosphorylation at Thr37, Thr46, Ser65, and Thr70, regulated by AKT/mTORC1 signaling, abolishes binding to eIF4E and activates mRNA translation. We detected p4E-BP1(Thr70) protein levels in the fundus muscularis externa of GHS rats. The protein level of p4E-BP1(Thr70) was significantly upregulated in the fundus of GHS rats (1.00 ± 0.09 vs. 1.59 ± 0.17 control, P < 0.05; n = 6) (Fig. 4A). The expression of mTOR (Fig. 4B) and p-mTOR(Ser2448) (Fig. 4C), however, was not altered by NCI.

Fig. 4.

Expression of phosphorylated 4E-binding protein 1 [p4E-BP1(Thr70); A], but not mammalian target of rapamycin (mTOR; B) or phospho (p)-mTOR(Ser2448) (C), was upregulated in the fundus of GHS rats vs. Ctr rats. Matrix metallopeptidase 9 (MMP-9; D) protein was suppressed, whereas tissue inhibitor of metalloproteinase 1 (TIMP-1; E) was unchanged; n = 6. *P < 0.05 vs. Ctr rats.

MMP-9 suppression contributes to NGF stabilization.

Slower protein decay is an important contributor to protein upregulation. Mature NGF is enzymatically degraded by activated MMP-9 (6). We investigated whether NCI alters MMP-9 expression in the rat fundus. We found that the expression of MMP-9 protein was significantly suppressed in the fundus muscularis externa of the GHS rats (0.44 ± 0.11 vs. 0.12 ± 0.02 control, P < 0.05; n = 6) (Fig. 4D). In contrast, TIMP-1, the inhibitor of MMP-9, was not altered by the neonatal inflammatory insult (Fig. 4E).

Norepinephrine triggers posttranscriptional mechanisms to upregulate NGF expression in fundus muscularis externa.

Previous studies found that NCI upregulates plasma NE, which in turn upregulates the expression of NGF in the gastric fundus of GHS rats (35). We investigated whether NE triggered posttranscriptional mechanisms to upregulate the elevation of NGF in GHS rats. We incubated rat fundus muscularis externa in complete DMEM culture medium containing 0, 0.1, 1, or 10 μM NE for 24 h and examined NGF protein levels. As shown in Fig. 5, A and B, NE upregulated NGF protein expression at the concentrations of 1 and 10 μM; Ngf mRNA was not altered (data not shown), which supports our hypothesis that NGF overexpression is a result of posttranscriptional mechanisms. pAKT(Ser473) level was markedly increased (Fig. 5C) only by 1 μM and 10 μM concentrations of NE.

Fig. 5.

A–C: norepinephrine (NE) upregulated NGF (A and B) and pAKT(Ser473) (C) in the fundus muscularis externa strips from adult naive rats at concentrations of 1 and 10 μM; n = 4. *P < 0.05 vs. vehicle control. D–G: naive rats were treated with yohimbine (YOH; 0.1 mg·kg⁻¹·day⁻¹ for 3 days). Blood was drawn before the first injection and 1 h after the last injection. D: YOH treatment significantly increased plasma NE. E and F: YOH enhanced NGF protein expression in the rat fundus. G: YOH treatment suppressed MMP-9 in fundus tissue; n = 5. *P < 0.05 vs. vehicle control.

To further confirm the regulatory role of NE on NGF protein expression in the gastric fundus muscularis externa, we treated naive adult rats systemically with 0.1 mg/kg YOH (daily ip), an α₂-AR antagonist. YOH elevated plasma NE by more than fivefold (0.63 ± 0.12 vs. 3.51 ± 0.48 control, P < 0.05; n = 5) (Fig. 5D). Importantly, NGF protein expression was significantly upregulated (1.00 ± 0.15 vs. 1.88 ± 0.20 control, P < 0.05) (Fig. 5, E and F), while MMP-9 protein level was markedly suppressed (1.00 ± 0.12 vs. 0.36 ± 0.07 control, P < 0.05) (Fig. 5G), indicating that NE mediates posttranscriptional upregulation of NGF. YOH treatment did not significantly increase pAKT (P = 0.183), suggesting that the increase of NGF in YOH-treated rats may largely be due to the suppression of MMP-9 resulting in decrease in the rate of NGF degradation.

Further support for the posttranscriptional regulation of NGF upregulation by NE in GHS rats came from interventional experiments using a cocktail of AR antagonists, 2 mg/kg phentolamine (α₁- and α₂-AR antagonist), 2 mg/kg propranolol (β₁- and β₂-AR antagonist), and 2 μg/kg CL316243 (β₃-AR antagonist). Systemic administration of AR antagonists reversed the upregulation of NGF (Fig. 6, A and B), pAKT(Ser473) (Fig. 6C), and PDK1 (Fig. 6D) and the suppression of MMP-9 (Fig. 6E) in the fundus muscularis externa of GHS rats.

Fig. 6.

Cocktail (2 mg/kg phentolamine, 2 mg/kg propranolol, and 2 μg/kg CL316243) of adrenergic receptor antagonists (AA) blocked the upregulation of NGF as well as signaling proteins of the translation and decay pathways in the fundus muscularis externa of GHS rats. AA blocked the upregulation of NGF (A and B), pAKT(Ser473) (C), and PDK1 (D) in the fundus muscularis externa of GHS rats. E: at the same time, the AA cocktail reversed the suppression of MMP-9; n = 7. *P < 0.05 vs. vehicle control, #P < 0.05 vs. GHS rats.

DISCUSSION

Several prominent symptoms of FD patients, including postprandial epigastric pain/discomfort and early satiety, relate to GHS, which sends aberrant signals to the higher centers when the proximal stomach distends after ingestion of a meal. Experimental clinical studies show that subsets of FD patients demonstrate GHS to balloon distension of the upper stomach (5, 7, 9, 10, 16, 28). The suppression of GHS in FD patients is an obvious option to mitigate the postprandial symptoms caused by it. Interestingly, FD patients show an increased sympathetic activity at baseline and in response to stress or a meal (8). In addition, a single subcutaneous dose of clonidine, which inhibits the release of NE from the sympathetic neurons, significantly suppresses the meal-related symptoms in FD patients (32). Together, the clinical findings suggest that the elevation of plasma NE may play a critical role in the sensory symptoms of FD patients. It should be noted that complex diseases, such as FD, may have multiple etiologies; the normalization of plasma NE levels may work only in those patients who demonstrate elevated plasma levels of NE.

Our findings in a preclinical model of GHS show that an increase in plasma NE may upregulate the expression of NGF in the gastric fundus muscularis externa by posttranslational mechanisms that include increase in translational efficiency and decrease in protein decay. NE is a stress hormone; employment of posttranscriptional, rather than transcriptional, upregulation may allow faster response times for the stress effects.

One of the goals of preclinical studies is to provide biological evidence for potential therapeutic approaches and suggest improvements in the design of epidemiological studies and clinical trials. Our findings provide biological evidence for the use of clonidine to mitigate the postprandial symptoms of epigastric pain/discomfort (32). Our findings suggest also that future epidemiological studies correlate the severity and timing of gastrointestinal infections in children with the severity of sensory symptoms in FD patients. We induced colon inflammation on PND 10 in rat pups. Early-life programming targets genes vulnerable at the time of the insult during development. The window of vulnerability of genes targeted by early-life gastrointestinal infections would be different between rodents and humans because of the different stages of the development of the neural networks and the endocrine systems in the two species. It is noteworthy that the annual episodes of diarrhea in US children younger than 5 yr of age ranges from 20 to 35 million; 22,000 of these infections are severe enough to result in hospitalization (24, 25).

It remains uninvestigated whether NGF expression is increased in the fundus muscularis externa of all or a subset of FD patients, specifically those who report AELE. However, in light of the efficacy of clonidine in improving postprandial symptoms of epigastric pain and discomfort (32), our findings have identified additional potential targets that may suppress the increase in fundus NGF expression. These targets suppress NGF expression by blocking the increase in translational efficiency and normalizing NGF protein decay. It is noteworthy that these blockers normalized the NGF expression; they did not block it completely, which would impair the normal physiological functions of this NGF.

The role of NGF as the mediator of inflammation-induced pain and neuropathic pain is well established (22, 23). Proinflammatory cytokines trigger transcriptional upregulation of NGF (1, 19). We reported previously that there is no significant increase of proinflammatory mediators in the fundus muscularis externa of GHS rats (35). Our findings show that NE upregulates the expression of NGF in the fundus muscularis externa under noninflammatory conditions.

In conclusion, the constitutive increase of plasma NE induced by NCI upregulates NGF expression in the gastric fundus muscularis externa by posttranscriptional mechanisms including increase in translation of existing Ngf mRNA by activating protein kinase AKT followed by the activation of 4E-BP1 and decrease in degradation of NGF by suppression of MMP-9 (Fig. 7). Blockade of increase in translational efficiency or reversal of decrease in NGF decay may serve as a potential target to normalize the expression of NGF to minimize gastric hyperalgesia induced by NCI.

Fig. 7.

Schematic diagram showing the activation of translation and degradation pathways by NE to enhance NGF expression and hence GHS in rats subjected previously to neonatal colon inflammation (NI).

Perspectives and Significance

Epidemiological studies have identified AELE as a potential etiologic component of several complex diseases, including FD. The precise disease or organ dysfunction depends on the nature and severity of the AELE as well as its timing during early-life development. AELE trigger epigenetic programming that alters transcription of genes vulnerable at the time of the adverse experiences. A previous finding showed that upregulation of NGF in the gastric muscularis externa was critical to the development of GHS in adult rats subjected to colon inflammation as neonates. Our present findings show that NGF upregulation was not due to transcriptional mechanisms; it was due to posttranscriptional mechanisms triggered by the elevation of plasma NE.

GRANTS

This work was supported in part by National Institute of Diabetes and Digestive and Kidney Diseases Grants 5R01 DK-088796 (S. K. Sarna) and DK-32346 (S. K. Sarna).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: Q.L. performed experiments; Q.L. analyzed data; Q.L., J.H.W., and S.K.S. interpreted results of experiments; Q.L. prepared figures; Q.L. drafted manuscript; J.H.W. and S.K.S. conception and design of research; J.H.W. and S.K.S. edited and revised manuscript; S.K.S. approved final version of manuscript.