| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institut de Recherche Contre les Canceurs de l’Appareil Digestif-European Institute of Telesurgery (F.R., A.F., M.V., J.M.), Université Louis Pasteur, 67091 Strasbourg, France; Institut National de la Santé et de la Recherche Médicale Unité 549 (P.Z., M.-T.B.-P., D.G.), 75014 Paris, France; Département de Pathologie Moléculaire (C.T.), Institut de Génétique et de Biologie Moléculaire et Cellulaire, Unité Propre de Recherche 6520 Centre National de la Recherche Scientifique/Unité 596 Institut National de la Santé et de la Recherche Médicale/Université Louis Pasteur, 67404 Illkirch, France; and Istituto di Clinica Chirurgica (F.R.), Catholic University, 00168 Rome, Italy
Address all correspondence and requests for reprints to: Francesco Rubino, M.D., Institut de Recherche Contre les Canceurs de l’Appareil Digestif-European Institute of Telesurgery, 1 Place de l’Hopital, 67091 Strasbourg, France. E-mail: f.rubino{at}lycos.com.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Roux-en-Y gastric bypass (RYGB), which is one of the most effective and commonly performed procedures (4), includes reduction of gastric capacity to about 5% of its normal volume and a bypass of 95% of the stomach, the entire duodenum, and part of the jejunum (5, 6). In contrast to pure restrictive operations such as the adjustable gastric banding (AGB) and vertical banded gastroplasty, which simply induce early satiety during a meal (7), the RYGB reduces the number of meals and snacks per day and restricts consumption of calorie-dense foods (7, 8, 9, 10). Hence, satiety after RYGB extends well beyond the immediate postprandial period and therefore cannot be explained by gastric restriction alone. Understanding how RYGB prevents compensatory hunger after surgical weight loss may facilitate the development of new antiobesity drugs and perhaps shed light on the pathophysiology of obesity, which still remains elusive.
In 2002, Cummings et al. (11) found remarkably low plasma levels of ghrelin, a circulating orexigen (appetite stimulant), in patients who had undergone weight loss by RYGB, a finding that is in striking contrast with the fact that weight loss induced by dieting, anorexia, or regular physical exercise characteristically results in high levels of circulating ghrelin (11, 12). These authors suggested that ghrelin suppression might explain the loss of hunger after RYGB and possibly contributes to surgically induced weight loss (11). However, the issue is still controversial, as other series reported in the literature showed unchanged or even increased levels of circulating ghrelin in patients undergoing RYGB (13).
Ghrelin is a peptide hormone whose plasma levels increase before meals and decrease after food intake in humans and rodents (14, 15, 16). Because ingested nutrients are dominant regulators of ghrelin production (7, 12, 17) and the majority of ghrelin-producing tissue is located in the proximal stomach (18, 19), it has been hypothesized that RYGB disrupts ghrelin regulation by permanently excluding the fundus from contact with nutrients (7, 11, 20). Accordingly, differences in surgical technique, such as leaving behind a larger gastric pouch and an incomplete exclusion of the fundic part of the stomach from contact with the nutrient load, might explain why RYGB fails to suppress circulating ghrelin levels in some instances (12).
Although this is a plausible hypothesis, it is also entirely possible that changes in circulating ghrelin are the result of signals originating further downstream the gastrointestinal tract, away from the stomach, possibly as the result of the rerouting of food through the bowel according to the RYGB.
To test the hypothesis that intestinal signals are involved in the regulation of ghrelin production, we used a modified RYGB, which leaves the stomach intact and diverts food from the pyloric area directly to the distal jejunum (Fig. 1A
). With this duodenal-jejunal bypass (DJB), the entire gastric ghrelin-producing tissue is left in digestive continuity, that is, in contact with nutrients. The operation does not involve any type of vagotomy, which could also influence ghrelin secretion (21, 22). Consequently, any change in circulating ghrelin levels occurring after DJB would be secondary to postgastric signals. Because this model does not impose any mechanical restriction on the flow of food, we also aimed to investigate whether the proximal bowel bypass may per se contribute to the control of appetite and weight loss.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Interventions
Zucker ZDF rats underwent one of DJB, GB, and sham operations. Normal controls underwent either DJB or the sham operation. An additional group of age-matched Zucker rats underwent food restriction to provide a model of diet-induced weight loss.
For all surgical procedures, 12-wk-old rats were fasted overnight and anesthetized with 2% isoflurane and air/oxygen.
DJB.
The volume of the stomach was left unperturbed, whereas the entire duodenum and the proximal jejunum were bypassed. This operation is a stomach-sparing bypass that excludes approximately the same amount of foregut excluded in standard RYGB in humans. The length of the bypass was made proportional to human RYGB by measuring the average length of rat intestine, which is approximately 95–100 cm long. The duodenum was separated from the stomach, and the bowel continuity was interrupted at the level of the distal jejunum (10 cm from the ligament of Treitz); the distal of the two limbs was directly connected to the stomach (gastrojejunal anastomosis); the proximal limb carrying the biliopancreatic juices was reconnected downward to the ileum at a distance of 15 cm from the gastrojejunal anastomosis (Roux-en-Y reconstruction). The procedure is illustrated in Fig. 1A
.
GB.
The proximal stomach was exposed, and a nonresorbable 4-mm-wide band made of polytetrafluoroethylene (BardPTFE Braided Tape; Bard, Tempe, AZ) was placed just below the gastroesophageal junction, in a similar fashion to that usually performed during AGB in humans (23). The band was calibrated to a standard length to provide a similar degree of gastric restriction in all animals undergoing this operation. In addition, the band was fixed to the gastric wall by several nonresorbable stitches (3/0 nylon) to avoid slippage or band migration. The appropriateness of band position and calibration was checked at the time of death to exclude displacement of the band, which would influence results. Although this annular banding is not adjustable as is the AGB, it provides significant mechanical narrowing of the rat’s stomach, thus imposing restriction on the flow of food. The operation is illustrated in Fig. 1B.
Sham operation.
A gastrotomy was performed at the level of the gastric antrum and closed with an absorbable suture. Furthermore, transections and reanastomosis of the gastrointestinal tract were performed at the level of the duodenum and distal jejunum. The physiological flow of food through the bowel was left intact.
Food intake restriction.
After measuring the mean daily food intake for 2 wk in an age-matched and ad libitum-fed group of rats, access to food was then restricted to only one third of it for 4 consecutive weeks.
Weight change and food intake monitoring
Weight and food intake were measured daily for the duration of the study.
Sampling for plasma hormone measurements
From the tail of conscious rats, blood was collected in EDTA tubes containing the gastrointestinal preservative Aprotinin. After centrifugation at 3000 rpm and 4 C for 12 min, plasma was immediately separated and stored at –80 C until analysis.
Plasma glucose and hormone measurements (Fig. 2)
All blood sampling tests were performed between 0900 and 1100 h.
|
Postoperative 48-h fasting and RF tests were performed 3 wk after the operations.
Ghrelin.
Plasma levels of total and acylated ghrelin were determined by two laboratory immunoenzymatic assays using polyclonal rabbit antibodies against the C terminus of rat ghrelin (developed at the Institut de Génétique et de Biologie Moléculaire et Cellulaire, Illkirch, France) and N-terminal rat ghrelin (kindly provided by Dr. Hosoda, Osaka, Japan), respectively, and human ghrelin was coupled to acetylcholinesterase (Spibio, Saclay, France) as tracer. The sensitivity was 100 and 20 pg/ml and the intra- and interassay coefficients of variation were 4 and 15%, respectively.
Leptin and insulin.
Plasma leptin and insulin concentrations were determined with RIA kits (Linco Research, St. Charles, MO). The sensitivity was 1 and 0.2 ng/ml, respectively, and the intra- and interassay coefficients of variation were less than 10% (intraassay, 2–4.6 and 1.4–4.6%; interassay, 3–5.7 and 8.5–9.4%, respectively).
Glycemia.
Plasma glucose levels were measured by enzymatic test (glucose oxydase; Beckman analyzer II; Beckman Coulter, Fullerton, CA).
Isolation of RNA and Northern hybridization
A portion of mucosa from the gastric body from each operated rat was taken at the time of and 4 wk after surgery. Surgical biopsies were rapidly frozen in liquid nitrogen. Total RNAs were extracted by the rapid guanidinium isothiocyanate procedure (43). The total RNA was fractionated by electrophoresis on 1% agarose, 2.2 M formaldehyde gels, transferred to nylon membrane (Hybond N; Amersham Corp. Arlington Heights, IL), and immobilized by baking for 2 h at 80 C. RNA loading was visualized by methylene blue staining. Membranes were hybridized at 42 C in 50% formamide, 5x standard saline citrate, 0.4% ficoll, 0.4% polyvinylpyrrolidone, 20 mM sodium phosphate (pH 6.5), 0.5% sodium dodecyl sulfate, 10% dextran sulfate, and 100 µg/ml denatured salmon sperm DNA, for 36–48 h, with the 32P-labeled specific rat ghrelin probe (GenBank accession no. AB029433) diluted to 0.5–1.106 cpm/ml. Stringent washings were performed at 60 C in 0.1x standard saline citrate and 0.1% sodium dodecyl sulfate. Filters were autoradiographed at –80 C for 12–72 h. The effect of different gastric surgeries on the expression level of ghrelin transcripts in the stomach was measured by Northern blot analysis.
Statistical analysis
Results are expressed as mean ± SEM. The changes in food intake, body weight, and plasma ghrelin levels were compared by ANOVA followed by Student’s t test. Comparisons between groups were performed with the unpaired t test. Simple linear regression analysis was used to evaluate correlation between body weight and plasma ghrelin levels. P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
)
|
). However, the most dramatic effect on food intake was seen in DJB-treated obese Zucker rats, which, despite the intact gastric reservoir, showed a lower daily rate of food intake (19.6 + 1.4 g/rat) than both GB and sham animals (P < 0.001) (Fig. 3A). In sharp contrast, DJB did not influence food intake in age-matched Wistar rats as demonstrated by identical average daily rates of food intake between DJB and sham-operated animals during the 4 wk of follow-up (Fig. 3A).
|
). In striking contrast, DJB had no effect on control Wistar rats because they showed the same weight gain after DJB or sham surgery (Fig. 3B). Zucker rats undergoing marked food restriction achieved a decrease in weight gain that was comparable with that of DJB [6.6 ± 0.6 vs. 5.2 ± 5.3%, respectively; P = not significant (NS)]. Hence, this group of rats was an appropriate control for comparing the hormonal effect of diet-induced weight loss vs. surgically induced weight loss.
Ghrelin mRNA expression
To assess for the integrity of ghrelin-secreting tissue in all rats undergoing surgery, we measured ghrelin mRNA in stomach biopsies obtained at the time of the operation and then again 4 wk after surgery. To ensure that biopsies of gastric mucosa were obtained from the same site at both times, the place of the first biopsy was marked by a stitch of nonresorbable suture.
No change in ghrelin mRNA expression could be detected after any type of surgery in ZDF rats (Fig. 4), demonstrating that all the different gastric surgeries performed in this study preserved the integrity of ghrelin-secreting gastric tissue, without apparent modification of ghrelin gene transcription.
|
Plasma total ghrelin levels were negatively correlated with body weight (r = –0.69, P < 0.001) and were significantly decreased by RF in each type of animals (lean and obese) (Table 2). However, mean ghrelin concentrations per gram of body weight were higher in obese Zucker rats than in normal Wistar rats (10.4 vs. 7.9 pg/ml·g body weight; P < 0.05), consistent with previous reports indicating that the obese Zucker rat is characterized by increased ghrelin mRNA in the stomach and peptide release in the circulation (24).
|
).
|
The experiments showed that DJB does not suppress circulating ghrelin levels in both normal and obese animals. Indeed, in fasting conditions, the Zucker rats that had undergone the DJB showed higher concentration of total and acylated ghrelin compared with sham-operated controls and GB rats as expected for their lower mean body weight. In normal Wistar animals, DJB did not have any impact on ghrelin response to eating. In contrast, DJB restored feeding-induced suppression of acylated ghrelin in the obese Zucker rats (–30% with respect to fasting; P = 0.01; Fig. 5), whereas neither the GB nor the sham operation caused significant change in the response of acylated ghrelin to food intake in these animals.
To investigate whether the effect of DJB on the ghrelin response to feeding was a specific outcome of the bypass of the proximal bowel and not secondary to the reduction of weight gain induced by the procedure, we tested the effect of diet-induced weight loss on circulating ghrelin in a group of matched Zucker rats undergoing food restriction for 4 wk. Despite a similar degree of weight gain compared with DJB rats, the diet group maintained the paradoxical increase of acylated ghrelin in response to feeding (330 ± 9 vs. 218 ± 23 pg/ml; P < 0.01).
The effect of surgery on leptin and insulin (Table 2 and Fig. 5)
To evaluate the effect of DJB on the secretion of long-term satiety hormones, we measured leptin and insulin levels in fasted and fed condition. Plasma leptin levels were significantly higher in obese Zucker rats compared with normal Wistar animals in both fasted and fed conditions.
The obese Zucker rats who had undergone DJB had significantly lower levels of plasma leptin compared with sham-operated animals and nonoperated controls; however, this difference disappears when considering the ratio of leptin to body weight (0.062 ± 0.003 vs. 0.066 ± 0.003 vs. 0.062 + 0.007 ng/ml·g body weight, respectively; P = NS). Furthermore, there was no difference in leptin levels in these rats after DJB- or diet-induced weight loss, suggesting that there was no direct impact of DJB on leptin secretion.
As expected, ZDF rats had higher plasma insulin levels than normal Wistar animals in both fasting and fed conditions (Fig. 5). Furthermore, in both obese animals and normal controls, RF after 48 h fasting significantly increased plasma insulin levels. DJB-treated ZDF rats had lower levels of insulin compared with sham-operated animals; however, this difference does not reach statistical significance when considering the insulin to body weight ratio.
The effect of DJB on type 2 diabetes
To investigate the effect of DJB on type 2 diabetes in the Zucker ZDF obese diabetic model, we tested plasma glucose in the fasted and fed states. The diabetic rats that had undergone DJB showed plasma glucose levels comparable with nondiabetic controls in both fasted and fed conditions. Sham-operated ZDF animals instead showed significantly higher glycemia, particularly in the fed state (Fig. 5).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Consistently with previous reports (30), this study found that, in obese Zucker rats, RF is associated with a paradoxical 30% increase of acylated ghrelin levels over fasting concentrations, suggesting that these obese animals are resistant to the meal-induced decrease of circulating acylated ghrelin. English et al. (31) reported that food intake fails to suppress plasma ghrelin levels also in obese humans. This altered endocrine response to eating in obese subjects may contribute to overeating and have implications for antiobesity therapies. Indeed, interventions in the dynamic response of ghrelin to eating, which seems to be altered in rodent and human obese subjects, may be a more effective way to fight obesity.
It has been suggested that RYGB may induce satiety and weight loss through suppression of circulating ghrelin levels (11) and that the mechanism responsible for the effect on ghrelin is the exclusion of the fundus and the isolation of ghrelin-producing cells from direct contact with ingested nutrients (11, 12, 20). However, this explanation is counterintuitive because an empty stomach is usually associated with increased ghrelin levels. We speculated that the rerouting of food through the bowel characteristic of the RYGB may influence ghrelin and appetite independently on fundus exclusion. The rationale for this hypothesis is that ghrelin-secreting cells in the stomach exist in both open and closed types (33), suggesting that they can receive both luminal and neuroendocrine information. Furthermore, recently published experiments put forward the possibility of a postgastric regulation of ghrelin secretion (34).
Unlike the standard RYGB, the DJB that we used in this study is an ideal model to test the hypothesis of an independent intestinal contribution to the regulation of circulating ghrelin levels because it allows one to rule out that any change in ghrelin and/or appetite is related to gastric exclusion/restriction.
Consistent with earlier reports showing no suppression in basal ghrelin levels in humans after RYGB (13), we did not observe suppression of mean total ghrelin levels after DJB. However, in obese Zucker rats, the operation reversed the altered response of circulating acylated ghrelin to feeding. The specificity of this result is demonstrated by the lack of such an effect in GB-treated and sham-operated Zucker rats as well as in the group undergoing diet. These findings support the hypothesis that intestinal signals contribute to the regulation (or dysregulation) of ghrelin secretion. It is possible that DJB influences ghrelin regulation by causing variations in gut hormone peptide YY (PYY) secretion, but other factors may also play a role (i.e. unknown peptides or neural signals from the bowel). Further exploration is necessary to understand how exactly the bowel influences ghrelin regulation.
The effect of DJB on appetite was also noteworthy. In this study, DJB decreased food intake in obese Zucker rats even more than GB, which is a pure restrictive procedure. The theoretical possibility that this effect could be determined by stenosis at the level of the gastrointestinal anastomosis is discounted by the fact that DJB-treated Wistar rats showed the same food intake rates compared with sham-operated controls. Furthermore, obese Zucker rats undergoing DJB still showed the ability to slightly increase their mean daily food intake with growth, a possibility that would be negated by the presence of significant stenosis.
The decreased food intake after the DJB may depend on the increase or enhancement of a factor inducing satiety and/or the reduction or inhibition of an antisatiety signal.
The lack of effect of DJB on food intake in normal Wistar rats, in sharp contrast with what is seen in obese animals, suggests that the second hypothesis is more likely. Previous work from our group showed that DJB does not reduce food intake in Goto-Kakizaki diabetic rats (35), which, like the Wistar rats used in the present study, are nonobese. Altogether, these observations suggest that the DJB may offset an abnormal signal produced by the bowel of hyperphagic/obese animals but not by the intestine of normal and lean rats. This implies the possibility that the gut plays an important role in the pathophysiology of hyperphagia and obesity. We speculate that failure to suppress an orexigenic signal originating in the bowel could result in persistence of the drive to eat even in the setting of positive energy balance, leading to excess energy storage and obesity. Our experiments and the clinical evidence that the most effective surgical operations in maintaining long-term weight loss are those that include a bypass of the foregut (biliopancreatic diversion and RYGB) (36) support the view of obesity as a disease of the bowel.
In this study, the DJB greatly reduced plasma glucose levels in obese diabetic Zucker rats, consistent with the knowledge that RYGB induces remission of type 2 diabetes in the majority of morbidly obese humans (37, 38). A recent study by two of us (F.R. and J.M.) demonstrated that this surgical control of diabetes depends on the bypass of the duodenal-jejunal tract, as shown by the fact that DJB (the same model used in the present study) controls type 2 diabetes in nonobese diabetic rats (Goto-Kakizaki rats), independently of caloric intake and weight loss (35).
Finally, our findings challenge the opinion that RYGB is primarily a restrictive operation (39, 40) and support an endocrine mechanism of action of the procedure (7, 41). Although it is generally believed that the small size of the gastric pouch is essential to reduce caloric intake and guarantee the weight loss effect of RYGB (39, 40, 42), our study suggests that performing an extremely small gastric pouch might be unnecessary. Certainly, a better understanding of the mechanism of action of RYGB is a priority because it may improve the design of bariatric operations.
In summary, our experiments demonstrate that the DJB, an experimental model of proximal bowel bypass with full preservation of the stomach, restores meal-induced suppression of acylated ghrelin and reduces food intake and weight gain in obese Zucker rats. Intriguingly, the operation does not alter food intake in normal Wistar animals. Altogether, these results indicate an intestinal contribution to the regulation of the dynamic ghrelin response to eating and suggest that an abnormal signaling from the proximal bowel may be involved in the pathogenesis of hyperphagia and obesity. Our experimental model, the DJB, may be useful to investigate intestinal factors regulating ghrelin secretion and appetite in the attempt to identify new targets for the development of antiobesity drugs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Results from this work were presented at the Annual Meeting of the American Gastroenterological Association, New Orleans, Louisiana, May 15–20, 2004.
First Published Online December 29, 2004
Abbreviations: AGB, Adjustable gastric banding; DJB, duodenal-jejunal bypass; GP, gastric banding; RF, refeeding; RYGB, Roux-en-Y gastric bypass.
Received September 3, 2004.
Accepted for publication December 20, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. P. Thaler and D. E. Cummings Hormonal and Metabolic Mechanisms of Diabetes Remission after Gastrointestinal Surgery Endocrinology, June 1, 2009; 150(6): 2518 - 2525. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Rubino Is Type 2 Diabetes an Operable Intestinal Disease?: A provocative yet reasonable hypothesis Diabetes Care, February 1, 2008; 31(Supplement_2): S290 - S296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Inoue, F. Rubino, Y. Shimada, V. Lindner, M. Inoue, P. Riegel, and J. Marescaux Risk of Gastric Cancer After Roux-en-Y Gastric Bypass Arch Surg, October 1, 2007; 142(10): 947 - 953. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
