SEARCH

SEARCH BY CITATION

Keywords:

  • colitis;
  • dextran sodium sulphate;
  • gastric emptying;
  • ghrelin knockout mice

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Abstract  Ghrelin is an important orexigenic peptide that not only exerts gastroprokinetic but also immunoregulatory effects. This study aimed to assess the role of endogenous and exogenous ghrelin in the pathogenesis of colitis and in the disturbances of gastric emptying and colonic contractility during this process. Dextran sodium sulphate colitis was induced for 5 days in (i) ghrelin+/+ and ghrelin−/− mice and clinical and histological parameters were monitored at days 5, 10 and 26 and (ii) in Naval Medical Research Institute non-inbred Swiss (NMRI) mice treated with ghrelin (100 nmol kg−1) twice daily for 5 or 10 days. Neural contractility changes were measured in colonic smooth muscle strips, whereas gastric emptying was measured with the 14C octanoic acid breath test. Inflammation increased ghrelin plasma levels. Body weight loss, histological damage, myeloperoxidase activity and IL-1β levels were attenuated in ghrelin−/− mice. Whereas absence of ghrelin did not affect changes in colonic contractility, gastric emptying in the acute phase was accelerated in ghrelin+/+ but not in ghrelin−/− mice. In agreement with the studies in ghrelin knockout mice, 10 days treatment of NMRI mice with exogenous ghrelin enhanced the clinical disease activity and promoted infiltration of neutrophils and colonic IL-1β levels. Unexpectedly, ghrelin treatment decreased excitatory and inhibitory neural responses in the colon of healthy but not of inflamed NMRI mice. Endogenous ghrelin enhances the course of the inflammatory process and is involved in the disturbances of gastric emptying associated with colitis. Treatment with exogenous ghrelin aggravates colitis, thereby limiting the potential therapeutic properties of ghrelin during intestinal inflammation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Ghrelin, the natural ligand of the growth hormone (GH) secretagogue receptor, was isolated from the rat stomach in 1999.1 Ghrelin is not only a positive regulator of the somatotropic axis but also has profound effects on energy homeostasis in rodents and humans. It increases body weight2 by stimulating food intake and decreasing fat utilization.3 In addition, ghrelin is a potent gastroprokinetic agent. It induces phase III of the migrating motor complex,4,5 accelerates gastric emptying6,7 and overcomes postoperative or septic ileus.8,9

Given the widespread distribution of functional ghrelin receptors on various immune cells (B cells, T cells, neutrophils, monocytes), it was hypothesized that this peptide also exerts immunoregulatory effects.10,11 Administration of ghrelin to stimulated human monocytes or T cells decreases the expression of pro-inflammatory cytokines.11 These anti-inflammatory effects of ghrelin were recently disputed by Zhao et al.12 who showed that ghrelin stimulates interleukin-8 gene expression through activation of a protein kinase C-mediated NF-κB-dependent pathway in human colonic epithelial cells.

Besides these in vitro studies, several groups have examined the effects of ghrelin on inflammation in vivo. Treatment with ghrelin has therapeutic effects in rats with endotoxic shock,13,14 sepsis15 or gastric damage.16–18 However, poor attention has been given to the role of ghrelin in inflammatory bowel disease (IBD). One study demonstrated anti-inflammatory effects of exogenous ghrelin in mice with trinitrobenzene sulfonic acid (TNBS)-induced colitis.19 Recent studies suggest that ghrelin may participate in the early pathophysiology of colonic inflammation. Ghrelin levels are increased in patients with IBD with active disease and are normal in patients with IBD in remission.20,21 In addition, an upregulation of ghrelin and its receptor has been observed in the colon of mice with TNBS-induced colitis12 and in mucosal biopsy specimens from patients with IBD.22

To get more insight into the pathophysiological role of ghrelin during intestinal inflammation, we addressed for the first time the role of endogenous ghrelin during the different phases of the inflammatory process by inducing dextran sodium sulphate (DSS)-colitis in ghrelin knockout mice. The results from the knockout study were validated by testing the effect of treatment with exogenous ghrelin in Naval Medical Research Institute non-inbred Swiss (NMRI) mice. For both experiments a multiparametric analysis was performed in which changes in disease activity, histological damage, myeloperoxidase (MPO) activity and colonic IL-1β levels were monitored. As ghrelin is an important endocrine regulator of gastrointestinal motility,23 we investigated by means of in vitro contractility studies whether ghrelin is involved in the changes of colonic contractility known to occur in smooth muscle preparations from patients with IBD24,25 and in animal models of colitis.26,27 We mainly focused on the loss of inhibitory innervation associated with colitis.28,29 It is also known that, during IBD, alterations in gut motility occur at sites remote from the inflamed region.30–33 We therefore investigated whether endogenous ghrelin is involved in the disturbances of gastric emptying occurring during DSS treatment.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Animals

Male (25–30 weeks old) wild-type (ghrelin+/+) and ghrelin knockout (ghrelin−/−) mice (C57Bl/6 genetic background) were developed by Lexicon Genetics Incorporated (The Woodlands, TX, USA).34 Male (15 weeks old) NMRI mice were purchased from Janvier (Le Genest Saint Isle, France). All mice were housed in a temperature controlled (20–22 °C) environment, standard commercial mouse chow and tap water were available ad libitum. The Ethical Committee for Animal Experiments of the Catholic University of Leuven approved all experiments.

Induction of colitis

Experiment 1: role of endogenous ghrelin during DSS-induced colitis  Colitis was induced in ghrelin+/+ (= 22) and ghrelin−/− (= 23) mice by adding 3% DSS (MP Biomedicals Inc., Strassbourg, France) to the drinking water for 5 days. Mice were then further divided in three subgroups: an acute group – studied at the end of the 5 days DSS treatment period (ghrelin+/+: = 8, ghrelin−/−: = 8), a recovery group – studied 5 days after DSS treatment was stopped (ghrelin+/+: = 9, ghrelin−/−: = 8) and a chronic group – studied 21 days after DSS treatment was stopped (ghrelin+/+: = 5, ghrelin−/−: = 7). A healthy control group of six ghrelin+/+ and six ghrelin−/− mice received normal drinking water during the entire observation period. There was no difference in water/DSS intake between the experimental groups (inflamed ghrelin+/+ mice: 9.5 ± 1.0 mL vs inflamed ghrelin−/− mice: 8.9 ± 0.7 mL). At the appropriate time points mice were killed by cervical dislocation.

Experiment 2: effect of treatment with exogenous ghrelin during DSS-induced colitis  Colitis was induced in NMRI mice (= 31) by adding 3% DSS to the drinking water for 5 days, after which mice received normal drinking water for 5 days. A healthy control group (= 31) that received normal tap water for 10 days was included as well. Eight hours after DSS/H2O administration, both groups were divided in two subgroups: the first subgroup (DSS-treated: = 19, healthy control: = 17) received an intraperitoneal injection with 0.9% saline twice daily, while the second subgroup (DSS-treated: = 12, healthy control: = 14) received (10:00 am and 6:00 pm) an intraperitoneal injection with ghrelin (100 nmol kg−1) (Tocris Bioscience, Bristol, UK) twice daily. Preliminary experiments showed that this dose of ghrelin stimulated food intake. Mice were sacrificed either at day 5 or 10 by cervical dislocation. There was no difference in water/DSS intake between the experimental groups (inflamed saline-treated mice: 11.0 ± 2.0 mL vs inflamed ghrelin-treated mice: 10.2 ± 0.9 mL).

Ghrelin plasma levels

Blood samples (cardiac puncture) were mixed with 500 KIU mL−1 trasylol (Bayer, Haywards Heath, UK). Plasma was separated by centrifugation, acidified (0.1 mol L−1 HCl), frozen and stored at −20 °C until radioimmunoassay (RIA). The RIA was performed with [125I]rabbit ghrelin as tracer (Bachem, Torrance, CA, USA) and with a rabbit antibody raised against human ghrelin(14–28) (final dilution 1 : 8000; Eurogentec, Seraing, Belgium), which recognizes both octanoylated and desoctanoylated ghrelin. The inter-assay coefficient of variation of the assay was 6.4% and the intra-assay coefficient of variation was 7.7%. The detection limit of the assay was 15.6 pg mL−1. The plasma ghrelin levels between control ghrelin+/+ mice and control NMRI mice treated during 10 days with saline were not significantly different [ghrelin+/+: 177 ± 39 pgmL−1 (= 7) vs NMRI mice: 291 ± 58 pg mL−1 (= 7)].

Clinical assessment of colitis

Body weight, food intake, stool consistency and presence of rectal blood in fresh faeces were determined daily at 10:00 am between days 0 and 10 and every other day between days 11 and 26 after the start of the experiment. Stool consistency and rectal bleeding (Hemoccult test; Beckman Coulter Inc., Fullerton, CA, USA) of fresh faeces were evaluated with a score between 0 and 4. At different time points, fat pads around the kidneys, the genitals and the limbs were dissected and weighed. The total fat mass was expressed as a percentage of the body weight.

Histological evaluation of colitis

Histological damage was determined on haematoxylin and eosin (H&E) stained colonic sections taken 5.5 cm from the rectum. Histological scoring was assessed by two investigators blinded to treatment codes, by means of a modification of a validated scoring scheme35 (Table S1). Each of the three parameters (severity of inflammation, extent of inflammation and crypt damage) was multiplied by the extent of intestinal involvement and a subsequent addition was performed to determine the total histological score. The thickness of the mucosa and the muscularis propria were measured separately.

Measurement of MPO activity

To determine the infiltration of neutrophils, MPO activity was measured in colonic tissues (taken 4 cm from the rectum) according to the method of Bradley et al.36

Interleukin-1β level in the colon

Interleukin-1β levels were determined in homogenates from colonic tissue taken 2.5 cm from the rectum with an enzyme-linked immunosorbent assay (ELISA) kit (Perbio Science, Cramlington, UK) according to the manufacturer’s instructions.

Colonic contractility

Strips from the distal colon (0.5 × 10–15 mm, taken at 8.5 cm from the rectum), freed from mucosa, were cut and suspended along their circular axis in an organ bath containing Krebs buffer (37 °C; 120.9 mmol L−1 NaCl, 2 mmol L−1 NaH2PO4, 15.5 mmol L−1 NaHCO3, 5.9 mmol L−1 KCl, 1.25 mmol L−1 CaCl2, 1.2 mmol L−1 MgCl2, 11.5 mmol L−1 glucose) gassed with 95% O2 and 5% CO2. After equilibration at optimal stretch, the effect of DSS treatment on the inhibitory neural response was studied in precontracted strips (27.5 mmolL−1 KCl) stimulated under non-adrenergic non-cholinergic (NANC) conditions (5 × 10−7 mol L−1 atropine and 3 × 10−6 mol L−1 guanethidine; Sigma-Aldrich, St Louis, MO, USA) by electrical field stimulation (EFS; 1–16 Hz, 0.5 ms, train 10 s, 6 V) as previously described.29 In addition, the effect of colitis on the excitatory neural off-contractions, which appeared after the cessation of EFS stimulation, was studied. The contractile response to 10−4 mol L−1 acetylcholine (ACh) and/or to 10−6 mol L−1 substance P (SuP) (Sigma-Aldrich) was determined as well. Contractions were normalized for the cross-sectional area of the strip.

Gastric emptying

Gastric emptying was measured with the 14CO2 octanoic acid breath test as described in detail by Kitazawa et al.6 From the 14CO2 excretion curve, Thalf, which is the time at which 50% of the total amount of 14CO2 was excreted, was calculated.

Statistical analysis

Data are presented as mean ± SEM. Survival analysis (according to the Kaplan–Meier method) and area under the curve calculations were performed with the GraphPad Prism 4.0 software (San Diego, CA, USA). Scores, plasma ghrelin and colonic IL-1β levels were analysed by the non-parametric Mann–Whitney test. Changes in contractility were analysed with repeated-measures anova. In case of significant factor effects, tests with contrasts were performed to locate pairs of factor levels with significant differences in the examined variables. All other parameters from the knock-out study were analysed with anova analysis followed by tests with contrasts. Data were analysed with Statistica 6.0 (StatSoft, Tulsa, OK, USA) and significance was accepted at the < 0.05 level.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Plasma ghrelin levels

Plasma ghrelin levels were increased in wild-type mice from 177 ± 39 pg mL−1 at day 0 to 436 ± 62 pg mL−1 at day 5 (< 0.01) and 401 ± 53 pg mL−1 at day 26 (< 0.01). No plasma ghrelin levels were detected in the ghrelin−/− mice. In NMRI mice treated with exogenous ghrelin during 10 days, plasma ghrelin levels were increased 17-fold in the healthy control group and 14-fold in the DSS-treated group.

Clinical assessment of DSS-induced colitis

Survival  Neither the absence of ghrelin nor the treatment with exogenous ghrelin had an effect on the survival rate of mice (Fig. S1A, B). Mortality amounted to 24% and 27% at the end of the observation period in the ghrelin+/+ and ghrelin−/− mice respectively. In the experiment with exogenous ghrelin, 10% of the saline-treated NMRI mice and 25% of the ghrelin-treated NMRI mice died during the experiment.

Body weight  No difference in body weight was observed between healthy control ghrelin+/+ and ghrelin−/− mice (Fig. 1A). Dextran sodium sulphate treatment induced a marked loss of body weight in both genotypes with maximal effects at day 11 in the wild-type mice (27 ± 3%) and at day 9 in the mutant mice (22 ± 2%). Between days 11 and 21, body weight loss was less pronounced in the mutant mice (< 0.001) than in the wild-type littermates.

image

Figure 1.  Time-dependent changes in body weight (A, B) and food intake (C, D) and fat mass (E, F) of healthy control and inflamed ghrelin+/+ and ghrelin−/− mice (= 15–22 mice per group; A, C, E) and of healthy control or inflamed NMRI mice treated with saline or ghrelin during 10 days (= 6–13 mice per group; B, D, F). *< 0.05, **< 0.01, ***< 0.001 vs day 0; †< 0.05, ††< 0.01, †††< 0.001 ghrelin+/+vs ghrelin−/− or saline vs ghrelin. Results are represented as mean ± SEM.

Download figure to PowerPoint

As expected, healthy control NMRI mice treated with ghrelin gained more weight than their saline-treated littermates (< 0.001; Fig. 1B). During DSS-treatment, body weight loss was more pronounced in the ghrelin-treated than in the saline-treated NMRI mice (< 0.01).

Food intake  The absence of ghrelin had no effect on 24 h food intake in healthy control mice (Fig. 1C). Dextran sodium sulphate treatment markedly reduced food intake between days 5 and 11, with maximal effects at day 7 in the ghrelin+/+ mice and at day 8 in the ghrelin−/− mice. However, also under inflammatory conditions, no difference in food intake was observed between both genotypes.

Whereas daily ghrelin treatment increased body weight in healthy control NMRI mice, it had no effect on 24 h food intake (Fig. 1D). Dextran sodium sulphate treatment reduced food intake both in the saline- (< 0.01) and ghrelin-treated NMRI mice (< 0.001) and this decrease was more pronounced (< 0.05) in the ghrelin-treated than in the saline-treated NMRI mice.

Fat mass  Interestingly, fat pad mass of healthy control ghrelin−/− mice was lower than that of the ghrelin+/+ mice (ghrelin+/+: 2.6 ± 0.3% of body weight vs ghrelin−/−: 1.5 ± 0.2% of body weight, < 0.01) (Fig. 1E). In ghrelin+/+ mice, DSS-treatment induced a loss of fat mass during the different phases of the inflammatory process that paralleled the changes in body weight. In ghrelin−/− mice, colitis did not affect fat pad mass, although at day 10, the loss in fat mass reached borderline significance (= 0.058) compared to healthy control mice.

Ghrelin treatment during 10 days in healthy control NMRI mice did not significantly affect fat pad mass. In the inflamed mice, fat pad mass was reduced from 2.0 ± 0.2% to 1.2 ± 0.3% body weight in saline-treated mice (< 0.05) and from 2.3 ± 0.3% to 0.5 ± 0.2% body weight in ghrelin-treated NMRI mice (< 0.001) (Fig. 1F). Although loss of fat pad mass tended to be more pronounced in the ghrelin-treated mice than in the saline-treated NMRI mice, this failed to reach significance (= 0.089).

Stool consistency and rectal bleeding  Healthy control ghrelin+/+ and ghrelin−/− mice had normal stools (Fig. S2A, C). Dextran sodium sulphate treatment induced loose stools in both genotypes with pronounced effects between days 2 and 15. During the first 10 days of the inflammatory process, rectal bleeding was observed, with gross bleedings at day 5. Both parameters did not differ significantly between both genotypes.

Treatment of healthy control NMRI mice with exogenous ghrelin neither affected stool consistency nor induced rectal bleeding (Fig. S2B, D). During DSS-treatment, administration of ghrelin did not affect stool consistency. However, rectal bleeding was more pronounced in ghrelin-treated inflamed than in saline-treated inflamed NMRI mice, as determined from changes in the area under the curve (saline: 11.9 ± 2.3 vs ghrelin: 19.4 ± 1.2, < 0.01).

Histological evaluation of colonic damage

In both ghrelin+/+ and ghrelin−/− mice, the thickness of the mucosa was increased at days 10 and 26 after the induction of colitis, whereas the thickness of the muscularis propria was only increased at day 26 (data not shown). No differences were observed between both genotypes. The total histological score of the H&E-stained colonic sections was increased at all investigated time points in both genotypes (Fig. 2A). However, at day 5, the histological score of colonic sections from ghrelin+/+ mice was 1.6-fold higher (< 0.05) than that from ghrelin−/− mice. Representative H&E stained sections of the colon of inflamed ghrelin+/+ and ghrelin−/− mice at day 5 are shown in Fig. 2B, C respectively.

image

Figure 2.  (A) The effect of DSS-induced colitis on the total histological score of sections of healthy control and inflamed ghrelin+/+ and ghrelin−/− mice (= 5–9 mice per group; *< 0.05, **< 0.01, ***< 0.001 vs day 0, †< 0.05 ghrelin+/+vs ghrelin−/−) and (D) of healthy control and inflamed NMRI mice after treatment with saline or ghrelin for 5 or 10 days (= 6–13 mice per group; *< 0.05, **< 0.01 control vs DSS). Representative H&E stained sections of the colon of ghrelin+/+ (B) and ghrelin−/− mice (C) at day 5 after the induction of colitis. Representative H&E stained sections of the colon of healthy control (E, F) and inflamed NMRI mice (G, H) treated with saline (E, G) or ghrelin (F, H). Results are represented as mean ± SEM.

Download figure to PowerPoint

Treatment of healthy control NMRI mice with exogenous saline or ghrelin during 5 or 10 days neither affected the thickness of the mucosa nor that of the muscularis propria. In the inflamed NMRI mice, the thickness of the mucosa was increased to a similar extent in both saline- and ghrelin-treated mice at day 10, whereas the thickness of the muscularis propria remained unchanged (data not shown). The histological score was not affected by treatment with exogenous saline or ghrelin during 5 or 10 days in healthy control NMRI mice. Inflamed NMRI mice treated with ghrelin had an increased histological score at days 5 (< 0.05) and 10 (< 0.01), whereas the saline-treated NMRI mice only showed an increased histological score at day 10 (< 0.05) (Fig. 2D). However, at both investigated time points, the histological score between saline- and ghrelin-treated inflamed NMRI mice did not differ. Fig. 2E–H shows representative H&E-stained sections of the colon of healthy control (Fig. 2E, F) and inflamed (Fig. 2G, H) NMRI mice treated with saline (Fig. 2E, G) or ghrelin for 10 days (Fig. 2F, H).

Myeloperoxidase activity

Myeloperoxidase activity, a marker for neutrophils, was increased in ghrelin+/+ mice at days 5 and 10 and was normalized at day 26 (Fig. 3A). In contrast, in the ghrelin−/− mice, MPO activity was only significantly increased at day 10, indicating a delayed infiltration of neutrophils in the absence of ghrelin.

image

Figure 3.  Myeloperoxidase activity (A, B) and levels of IL-1β (C, D) in colonic homogenates of healthy control and inflamed ghrelin+/+ and ghrelin-/- mice (A, C) (= 5–9 mice per group) and of healthy control or inflamed NMRI mice after treatment with saline or ghrelin for 5 or 10 days (B, D) (= 6–13 mice per group) *< 0.05, **< 0.01, ***< 0.001 vs day 0 or control vs DSS; †< 0.05, ghrelin+/+vs ghrelin-/- or saline vs ghrelin. Results are represented as mean ± SEM.

Download figure to PowerPoint

Treatment with exogenous saline or ghrelin during 5 days did not affect MPO activity neither in the healthy control mice nor in the inflamed NMRI mice (Fig. 3B). However, after 10 days treatment, MPO activity was 1.7-fold higher in the ghrelin-treated than in the saline-treated inflamed NMRI mice (Fig. 3B).

Colonic IL-1β levels

In both ghrelin+/+ and ghrelin−/− mice DSS-treatment increased IL-1β levels in the colon at all investigated time points (Fig. 3C). At day 5, the increase of IL-1β in the colon of the mutant mice was 2.4-fold lower than in the wild-type mice.

Treatment with exogenous saline or ghrelin during 5 days did not affect colonic IL-1β levels of healthy control or inflamed NMRI mice. However, treatment with ghrelin during 10 days increased IL-1β levels both in healthy control (< 0.05) and inflamed NMRI mice (< 0.05) (Fig. 3D).

In vitro colonic contractility

The role of endogenous ghrelin on the inhibitory neural responses during DSS-treatment was investigated in precontracted strips under NANC conditions. Under these conditions, EFS of colonic strips evoked a large on-relaxation followed by an off-contraction (Fig. 4A). In healthy control ghrelin+/+ and ghrelin−/− mice, the magnitude of the inhibitory response did not differ (Fig. 4B) and was of purinergic and nitrergic origin (data not shown). Dextran sodium sulphate treatment reduced the on-relaxations to a similar extent in both genotypes at day 10 (Fig. 4B).

image

Figure 4.  (A) Representative tracing of neural responses elicited under NANC conditions (5 × 10−7 mol L−1 atropine and 3 × 10−6 mol L−1 guanethidine) by EFS in precontracted (27.5 mmol L−1M KCl) colonic muscle strips from healthy control ghrelin+/+ (upper tracing) and ghrelin−/− (lower tracing) mice. (B, C) Effect of DSS-induced colitis on the EFS-induced on-relaxations in colonic smooth muscle strips from healthy control and inflamed (day 10) ghrelin+/+ and ghrelin−/− mice (B) (= 14–25 strips from 5–9 mice; *< 0.05, **< 0.01, ***< 0.001) and from NMRI mice after treatment with saline or ghrelin for 10 days (C) (= 9–14 strips from 4–7 mice; *< 0.05, †< 0.05, saline vs ghrelin). Results are represented as mean ± SEM.

Download figure to PowerPoint

Unexpectedly, treatment with exogenous ghrelin decreased the inhibitory responses in healthy control NMRI mice but was without effect on the loss of inhibitory neural responses observed in NMRI mice treated with DSS (Fig. 4C). In addition, the excitatory off-responses in strips from healthy control NMRI mice were decreased by treatment with exogenous ghrelin but not in strips from DSS-treated NMRI mice (Fig. 5A). In healthy control NMRI mice, the decrease in neural excitatory responses was accompanied by a decreased smooth muscle response to SuP but not to ACh (Fig. 5B, C).

image

Figure 5.  Effect of DSS-induced colitis on EFS-induced neural excitatory off-responses (A) and on the maximal response to acetylcholine (10−4 mol L−1) (B) and substance P (10−6 mol L−1) (C) in colonic smooth muscle strips from healthy control and inflamed NMRI mice treated with saline or ghrelin for 10 days (= 4–17 strips from 2–6 mice; *< 0.05, **< 0.01, ***< 0.001 vs day 0,†< 0.05, †††< 0.001, saline vs ghrelin). Results are represented as mean ± SEM.

Download figure to PowerPoint

In vivo gastric emptying

Gastric emptying did not differ between ghrelin+/+ and ghrelin−/− mice before the induction of colitis. Five days after DSS-treatment, gastric emptying was accelerated in the ghrelin+/+ mice as reflected by a decrease in Thalf from 79 ± 5 to 49 ± 6 min (< 0.05), whereas gastric emptying in ghrelin−/− mice was not affected (Fig. 6). At the other investigated time points, gastric emptying was not affected, neither in the wild-type nor in the mutant mice.

image

Figure 6.  Effect of DSS-induced colitis on gastric emptying (Thalf) in ghrelin+/+ and ghrelin−/− mice (= 12 mice per group) (**< 0.05 vs day 0, †††< 0.001, ghrelin+/+vs ghrelin−/−). Results are represented as mean ± SEM.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Besides its profound role in the regulation of energy homeostasis, ghrelin also plays an important role in the immune system. In this study, the role of both endogenous and exogenous ghrelin during DSS-induced colitis was thoroughly investigated. Given that previous studies with exogenous ghrelin pointed towards an anti-inflammatory role of ghrelin in various models of inflammation, we expected an aggravation of the disease process in the absence of ghrelin. In contrast, we found that ablation of the ghrelin gene has rather beneficial effects and that long-term treatment with exogenous ghrelin aggravates the disease process.

Our results are at variance with the study of Gonzalez-Rey et al.19 in which mice with TNBS- or DSS-induced colitis were treated with ghrelin. The most important difference between both studies is the dose of ghrelin and the treatment schedule: single dose vs prolonged treatment. Gonzalez-Rey et al.19 administered one dose of ghrelin either 12 h after TNBS-instillation (70 nmol kg−1) or at days 4 and 6 after DSS-administration (30 nmol kg−1). We administered 200 nmol kg−1 day−1 ghrelin during 10 days. Gonzalez-Rey et al.19 showed that the anti-inflammatory effect of ghrelin in mice with colitis was mediated by the inhibitory effect of ghrelin on the production of multiple chemokines and cytokines, studied 60 h after the administration of a single dose of ghrelin. It is known that a single administration of ghrelin increases GH levels whereas during continuous treatment with ghrelin this effect is blunted.37 Thus, it is possible that ghrelin’s acute anti-inflammatory effects may be mediated via its GH-releasing activity as GH is known to reduce the levels of inflammation during colitis.38,39

Our study in ghrelin−/− mice, although performed in a different mouse strain (C57Bl/6 vs NMRI), also points to a pro-inflammatory role of ghrelin during colonic inflammation. Indeed, body weight loss, colonic damage, MPO activity and IL-1β levels in the colon were attenuated in the ghrelin−/− mice. Remarkably, although DSS-treated ghrelin−/− mice lost less body weight than the ghrelin+/+ mice, this was not reflected in a difference in the decrease of food intake or fat mass between both genotypes. Conditions of cachexia are often associated with a loss of lean mass and it cannot be excluded that the distribution of lean and fat mass is different between both genotypes during inflammation. A peripheral dual-energy X-ray absorptiometry (pDEXA) analysis would give more insight in this issue.

In previous studies, a role for endogenous ghrelin in the pathophysiology of colonic inflammation has already been suggested. Plasma ghrelin levels are increased in patients with IBD with active disease and are normal in patients with IBD in remission.20,21 The increased plasma ghrelin levels observed in the present study in inflamed wild-type mice at day 5 mimicked the increased plasma ghrelin levels observed in patients with active disease. Nevertheless, we also found that plasma ghrelin levels were still increased at day 26 and it is difficult to assess whether this can be extrapolated to patients with IBD in remission.

This study showed that ghrelin has important effects on two types of immune cells: neutrophils and macrophages. Myeloperoxidase activity was decreased at day 5 in inflamed ghrelin−/− mice compared with ghrelin+/+ mice, whereas it was increased in ghrelin-treated compared with saline-treated inflamed NMRI mice at day 10. This suggests that both endogenous and exogenous ghrelin stimulate the infiltration of neutrophils. It is unclear whether this effect is due to a direct interaction of ghrelin with its receptor expressed by neutrophils10 or to an indirect effect mediated by the release of chemokines. It was shown in human colonic epithelial cells, over-expressing the ghrelin receptor, that ghrelin stimulates the release of the neutrophil-attracting chemokine IL-8.12 This situation may be mimicked in the present study where the increased ghrelin plasma levels could also lead to an overstimulation of the ghrelin receptor.

It is tempting to speculate that ghrelin is involved in the recruitment of macrophages as IL-1β levels of ghrelin−/− mice were lower than that of ghrelin+/+ mice at day 5 after DSS treatment, whereas ghrelin-treated inflamed NMRI mice had higher IL-1β levels than saline-treated inflamed NMRI mice (day 10). This is in contrast with previous in vitro and in vivo studies, which showed that ghrelin inhibited the release of pro-inflammatory cytokines. Indeed, 24 h incubation of activated human monocytes and T cells with ghrelin inhibited the release of IL-1β, IL-6 and TNF-α.11 In addition, in mice challenged with LPS and treated with a single dose of ghrelin, IL-1β and IL-6 levels were attenuated in both spleen and liver after 4 and 24 h.11 In these studies, cytokine levels were measured after short-term exposure to a single dose of ghrelin, which is in sharp contrast with the long-term treatment schedule used in the present study. It therefore appears that the treatment schedule is a critical factor determining whether ghrelin will work in a pro- or anti-inflammatory way. Moreover, the time point at which ghrelin affects MPO activity or IL-1β levels differed between the knockout study (day 5) and the study with exogenous ghrelin (day 10). This may be due to the differences in mouse strain and age between both groups, which may influence the susceptibility to DSS.

A recent study suggested that peripheral T cells from IBD patients have an altered reactivity to ghrelin when compared to those from healthy control subjects.22 In this study, ghrelin increases the production of IL-4 and IL-13 from peripheral T cells from healthy control subjects, but failed to alter the expression of both cytokines from T cells from patients with Crohn’s disease.22 The dysregulation of the reactivity of T cells induced by ghrelin may result in sustained inflammation due to Th-1 dominant responses in patients with Crohn’s disease. Clearly, more studies are needed to compare the effect of ghrelin on immune cells from healthy control and inflamed mice.

It is well recognized that IBD patients display a disturbed gastrointestinal motility pattern.24,25 One of the key events occurring in animal models of colitis is the loss of inhibitory responses, due to reduced activity and synthesis of nNOS in the myenteric plexus28,29 and a loss of purinergic responses.29 As ghrelin is a potent regulator of gastrointestinal motility23 and its receptor is present in the enteric nervous system of the rat colon,40 we investigated the role of ghrelin in the changes in gastrointestinal contractility during DSS-induced colitis. Our results demonstrate that neither endogenous nor exogenous ghrelin is involved in the reduction of inhibitory responses during DSS-induced colitis.

Unexpectedly, a pronounced loss of the inhibitory NANC responses and of the excitatory neural responses was observed in healthy control NMRI mice treated with ghrelin for 10 days. The latter was due to a decreased response to SuP. As in these mice, colonic IL-1β levels were increased, one could hypothesize that IL-1β either reduced the release of SuP or induced a desensitization of the SuP receptor. The first hypothesis is rather unlikely as previous studies showed that the incubation of colonic muscle strips with IL-1β increases SuP release.41 On the other hand, NK1 and NK2 receptors can be desensitized by protein kinase C (PKC) induced phosphorylation of serine/threonine residues located in the C-terminal tail of the receptor.42In vitro studies have already shown that ghrelin can induce PKC activation,43 which may trigger PKC-dependent desensitization of SuP receptors. Why this desensitization does not occur under inflammatory conditions is not clear yet. However, this indicates that prolonged treatment with ghrelin or its agonists may activate an undesired immune response and may therefore hamper the potential therapeutic properties of ghrelin in general.

Finally, alterations in gut motility at sites remote from the inflamed region are observed in patients with IBD and in animal models.30–33 As ghrelin potently accelerates gastric emptying in mice,6,34 we investigated whether ghrelin is involved in the disturbances of gastric emptying during DSS colitis. In the acute phase, increased plasma ghrelin levels observed in the wild-type mice were associated with accelerated gastric emptying, whereas in the mutant mice, no changes in gastric emptying were observed. In contrast, in the chronic phase, elevated ghrelin levels did not affect gastric emptying. This may be due to the desensitization of the ghrelin receptor, which occurs rapidly in vitro44 or to a change of the constitutive activity of the ghrelin receptor.

In conclusion, our study showed that the increased endogenous plasma ghrelin levels, as observed in patients with active IBD, may enhance the course but not the outcome of the inflammatory process. With a long-term treatment schedule, we could not confirm the anti-inflammatory effects previously reported with single-dose administration of ghrelin, thereby limiting the potential therapeutic applications of ghrelin during intestinal inflammation. Although ghrelin does not seem to be involved in the functional alterations in colonic motility occurring during colitis, ghrelin does play a role in the changes in gastric emptying. In addition, the increased release of IL-1β and the functional changes in colonic contractility observed during treatment with ghrelin in healthy controls may lead to unwarranted side effects and limit the therapeutic use of ghrelin agonists in general.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Supported by grants from the Flemish Foundation for Scientific Research (contracts FWO G.0144.04 and 1.5.125.05) and the Belgian Ministry of Science (contracts GOA 03/11 and IUAP P5/20).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Table S1 Histological grading of colitis (adapted from Dieleman et al.(35))

Figure S1 The effect of DSS-induced colitis on survival in ghrelin+/+ and ghrelin-/- mice (n = 22-23 mice/group) and (B) in NMRI mice treated with saline or ghrelin for 10 days (n = 8-10 mice/group).

Figure S2 Time-dependent changes in stool consistency (A, B) and rectal bleeding (C, D) of healthy control and inflamed ghrelin+/+ and ghrelin-/- mice (A, C) (n = 15-22 mice/group) and of healthy control or inflamed NMRI mice treated with saline or ghrelin during 10 days (B, D) (n = 8-10 mice/group). †† P < 0.01 saline vs ghrelin. Results are represented as mean ± SEM.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
NMO_1184_sm_SUPPLEMENTARY MATERIAL.doc218KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.