The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Objective: Mechanisms that promote effective and sustained weight loss in persons who have undergone Roux-en-Y gastric bypass surgery are incompletely understood but may be mediated, in part, by changes in appetite. Peptide YY (PYY) is a gut-derived hormone with anorectic properties. We sought to determine whether gastric bypass surgery alters PYY levels or response to glucose.
Research Methods and Procedures: PYY and ghrelin levels after a 75-gram oral glucose tolerance test were measured in 6 morbidly obese patients 1.5 ± 0.7 (SE) years after gastric bypass compared with 5 lean and 12 obese controls.
Results: After substantial body weight loss (36.8 ± 3.6%) induced by gastric bypass, the PYY response to an oral glucose tolerance test was significantly higher than in controls (p = 0.01). PYY increased ∼10-fold after a 75-gram glucose load to a peak of 303.0 ± 37.0 pg/mL at 30 minutes (p = 0.03) and remained significantly higher than fasting levels for all subsequent time-points. In contrast, PYY levels in obese and lean controls increased to a peak of ∼2-fold, which was only borderline significant. Ghrelin levels decreased in a symmetric but opposite fashion to that of PYY.
Discussion: Gastric bypass results in a more robust PYY response to caloric intake, which, in conjunction with decreased ghrelin levels, may contribute to the sustained efficacy of this procedure. The findings provide further evidence for a role of gut-derived hormones in mediating appetite changes after gastric bypass and support further efforts to determine whether PYY3–36 replacement could represent an effective therapy for obesity.
Presently available antiobesity medications typically result in only modest (∼10%) loss of body weight, whereas Roux-en-Y gastric bypass surgery (GBP)1 induces more substantial and sustained weight loss (1). However, the mechanisms by which sustained weight loss occur are incompletely understood. In addition to mechanical factors (1), changes in hormonal milieu may also be important by altering appetite, thus contributing to the sustained efficacy of gastric bypass. This notion is supported by the recent observation that levels of ghrelin (a stomach-derived hormone that increases food intake over the short-term) are decreased in patients after gastric bypass (2, 3).
Peptide YY3–36 (PYY3–36), a gut-derived hormone, reduces food intake over the short term in animals (4, 5) and humans (6) by stimulating hypothalamic neuropeptide Y receptors. However, not all studies have been able to replicate the anorectic effect of PYY3–36 in animals (7). Obese persons have been found to have lower baseline PYY levels than lean persons but are sensitive to the anorectic effects of exogenously administered PYY3–36, suggesting a role for PYY deficiency in the pathogenesis of obesity (6). This remains controversial, however, because others have not shown differences in baseline PYY levels between lean and obese persons, although an early exaggerated rise after a test meal was observed in gastric bypass patients, indicating an important role in regulating appetite in the surgically treated patients (8). Thus, the physiology of PYY regulation and the potential efficacy of PYY3–36 as a therapeutic agent for obesity remain controversial.
We sought to examine the effect of GBP on basal and stimulated PYY levels, as well as on ghrelin levels to study the relationship between PYY and other appetite-regulating hormones and to help elucidate the mechanism(s) by which appetite suppression occurs after gastric bypass. A role for PYY in appetite suppression after gastric bypass would support efforts to develop this hormone or analogs as an antiobesity medication.
Research Methods and Procedures
The study protocol was approved by the local institutional review boards, and subjects gave written informed consent to participate. Six obese subjects (3 men and 3 women) who had undergone GBP for morbid obesity 1.5 ± 0.7 years before enrolling, 12 obese controls (5 men and 7 women), and 5 lean controls (1 man and 4 women) had a 75-gram oral glucose tolerance test (OGTT) at 8:00 am after a 12-hour fast. Blood samples for PYY and ghrelin were collected at baseline and 30, 60, and 120 minutes, and for PYY, at 90 and 180 minutes after the 75-gram oral glucose load (350 kcal). PYY measurements were not performed in two obese controls and one lean control because of insufficient serum.
PYY was measured using an enzyme-linked immunosorbent assay (DSL, Webster, TX) with an intra-assay coefficient of variation of 3%, sensitivity of 15 pg/mL, and 100% cross-reactivity for the full-length peptide (PYY1–36) and the truncated PYY3–36, both of which have biological activity. Total ghrelin was measured by radioimmunoassay (Phoenix Pharmaceuticals, Belmont, CA) with a sensitivity of 20 pg/mL and an intra-assay coefficient of variation of 9%, as previously described (3).
Data are presented as means ± SE. The area under the curve (AUC) for PYY and ghrelin during the OGTT was calculated using the standard trapezoidal method. Non-parametric (Wilcoxon signed rank) tests were used to compare hormone levels between baseline and each time-point after the glucose load. One-way ANOVA was used to compare hormone levels (baseline and at each time-point after the glucose load) and/or BMI between the three groups of subjects, with post hoc tests by least significant differences. Analyses were carried out using SPSS 8.0 (SPSS, Inc., Chicago, IL).
Gastric bypass patients were morbidly obese before surgery, with a preoperative BMI of 55.6 ± 1.0 kg/m2, but had lost 36.8 ± 3.6% of their body weight by 1.5 ± 0.7 years after surgery (p = 0.03). At the time of evaluation, BMI was stable at 35.0 ± 1.8 kg/m2. The BMI of the three groups was significantly different (p = 0.002 by ANOVA), with lean subjects having lower BMI (22.0 ± 0.8 kg/m2) than both gastric bypass (p = 0.003) and obese groups (39.6 ± 2.3 kg/m2, p < 0.001). Obese subjects were recruited to be similar in BMI to the postsurgical group (p = not significant).
Despite the differences in BMI, fasting PYY levels differed only modestly and not significantly among the groups (GBP: 34.7 ± 9.8 pg/mL; obese controls: 55.4 ± 10.4 pg/mL; lean controls: 17.4 ± 1.6 pg/mL; p = 0.07 by ANOVA). In contrast, the AUC of PYY after the OGTT (30, 016 ± 8485 pg/mL · min) in GBP patients was higher compared with obese (11, 755 ± 2462 pg/mL · min) and lean groups (4669 ± 583 pg/mL · min; p = 0.01 by ANOVA; Figure 1 A).
After a 75-gram glucose load, PYY levels were significantly higher in GBP patients compared with lean and obese controls at 30 (p = 0.002), 60 (p = 0.01), and 90 minutes (p = 0.04) but not at 120 (p = 0.13) or 180 (p = 0.21) minutes (Figure 1B). PYY increased ∼9-fold after the glucose load to a peak of 303.0 ± 37.0 pg/mL at 30 minutes (p = 0.03 vs. baseline) and remained significantly higher than fasting levels for the remainder of the OGTT (all time-points: p < 0.05 vs. baseline; Figure 1C). In contrast, the PYY response to the glucose load was much less pronounced in obese and lean groups (peak of ∼2-fold at 30 minutes: 95.4 ± 23.2 and 37.0 ± 11.6 pg/mL, respectively) and was only borderline significant (both p = 0.07 vs. baseline; Figure 1C).
As predicted based on their opposing actions, ghrelin showed an opposite circulating pattern to that of PYY after the glucose load, with higher levels in lean and obese subjects compared with GBP patients (Figure 2). Fasting and mean postprandial ghrelin levels after the OGTT were 231.1 ± 30.1 and 177.6 ± 21.0 pg/mL in GBP patients, 311.2 ± 27.7 and 278.7 ± 24.7 pg/mL in obese controls, and 533.2 ± 44.0 and 418.2 ± 36.9 pg/mL in lean controls, respectively (p = 0.003 by ANOVA for comparison of fasting levels across groups). Interestingly, while PYY levels peaked at 30 minutes during the OGTT in GBP patients, ghrelin levels in these same subjects decreased from a baseline level of 231.1 ± 30.1 pg/mL to a nadir of 152.5 ± 20.0 pg/mL at 60 minutes (p = 0.03), showing that secretory patterns differ for the two hormones.
We showed that patients who have lost a substantial amount of weight after GBP surgery have an increased PYY response to carbohydrate intake compared with both lean and obese persons. Furthermore, ghrelin changes in an opposite fashion to that of PYY levels. These findings further support a role of gut-derived hormones in mediating appetite changes after GBP, which likely contributes to the sustained effect on weight loss in these morbidly obese patients.
PYY is secreted from gastrointestinal cells in levels that correspond to the amount of calories ingested (9). The demonstration that exogenous PYY3–36 administration in rodents and humans inhibits short-term food intake and/or reduces weight gain has generated substantial scientific interest in the potential use of this hormone as a therapeutic agent for obesity (4, 6). In contrast, because most obese humans are resistant to the adipocyte-secreted hormone leptin (10), which also acts on hypothalamic neurons (including neuropeptide Y neurons) to decrease food intake, exogenous leptin administration in obese subjects had only modest effects to induce weight loss (11). Although it has been suggested that PYY deficiency may be a feature of obesity (6), this was not confirmed by an independent study (8) or by our findings. Based on the discrepant findings of these small studies, which may lack power to detect significant differences between different study groups, future larger studies are warranted to clarify the issue of whether obesity represents a PYY-deficient state. In addition, the systems regulating body weight and energy homeostasis are complex and often redundant, and the distinct possibility exists that PYY may still play an important role in appetite regulation even if baseline levels do not differ significantly between lean and obese individuals.
In agreement with a previous study evaluating the effect of GBP on the response of PYY and ghrelin levels to a mixed meal (8), we also found an exaggerated PYY response to carbohydrate intake in patients who had undergone GBP, confirming that the physiology of appetite regulation is altered after this procedure. Our findings showed an ∼9-fold increase in PYY levels from baseline in GBP patients compared with the prior study that found a 3-fold increase in PYY after GBP that also peaked at 30 minutes (8). Importantly, the macronutrient content seems to be of importance for both the quantitative and qualitative response of PYY levels (9), and, thus, further studies using isocaloric nutrient stimuli with different macronutrient content are warranted to clarify the macronutrients that are most effective in stimulating PYY release. The actual peak PYY levels achieved differed between our study (∼300 pg/mL) and the prior study (∼150 pg/mL) (8), but, notably, different assays were used, which may make comparison of actual levels across studies problematic until standardized assays are developed for these newly discovered or described hormones. Other potential factors contributing to differences in PYY levels include differences in the study population [inclusion of men and women in our study vs. all women in the prior study (8)] and postoperative interval, as well as the small cohort size of both studies.
Changes in appetite-regulating hormones resulting from altered gastrointestinal anatomy and/or innervation may contribute to the success of GBP in contrast to the plateau in weight loss and later regain in body weight often associated with food restriction or medical therapy for obesity. This hypothesis is supported by a study in rats showing that transposition of the terminal ileum to a proximal insertion site between the duodenum and jejunum resulted in greater weight loss and increased release of PYY and glucagon-like peptide-1 compared with sham-operated controls (12). This suggests that greater exposure of the distal small intestine to nutrients may increase the release of anorectic ileal hormones, which contributes to the observed weight loss (12). It has also been hypothesized that a decline in the stomach-derived ghrelin after GBP may play a role in the efficacy of this procedure (2, 3). Thus, alterations observed herein may be caused by regulation of ghrelin by PYY, regulation of PYY by ghrelin, and/or regulation of both by a third factor (e.g., central neural mechanism). PYY, which is secreted mainly by cells in the distal small intestine and colon, increases after a meal much faster (within 15 minutes after food intake, with peak at 30 to 60 minutes) than would be expected if secretion were caused solely by direct stimulation by intraluminal gut contents. PYY levels remain elevated for several hours (9), suggesting that neural and/or hormonal mechanisms may be responsible for the immediate postprandial release, with direct gut stimulation contributing to later sustained levels.
While it is possible that bypass of ghrelin-containing cells in the stomach, resulting in a decline in ghrelin levels, may be the signal for PYY levels to increase, the later peak showed for ghrelin but not PYY does not support the hypothesis that PYY is directly regulated by ghrelin. Conversely, our data would better support regulation of ghrelin by PYY, given the nadir in ghrelin 30 minutes after the peak in PYY levels. Although these observational data cannot prove causality, an interventional study involving PYY3–36 infusion showed a decrease in ghrelin levels after PYY3–36 administration in lean and obese subjects (6). Activation of vagal afferent neurons, which densely innervate the gastrointestinal tract and send information to the central nervous system on factors such as luminal distention and changes in hormone levels (e.g., cholecystokinin) (13) through the nucleus tractus solitarius, plays an important role in short-term regulation of feeding. In this study, the vagus nerve was left intact during GBP, and, thus regulation of PYY by neural inputs cannot be excluded based on the rapidity of nutrient-induced changes in PYY.
Vertical banded gastroplasty, another surgical procedure that restricts gastric volume but permits the normal passage of alimentary contents without reconfiguring the anatomy, has relatively lower long-term success rates compared with GBP (1). In contrast to our findings of an exaggerated PYY response after GBP, postprandial PYY levels are similar to those of lean controls 1 year after gastroplasty (14), suggesting that the anatomic alterations in the Roux-en-Y GBP procedure are important in regulating PYY. Likewise, plasma ghrelin levels are higher after gastric banding compared with GBP (15, 16). Together, these data suggest that anatomic changes of the bypass procedure are necessary to reduce appetite and, thus, contribute to sustainable weight loss.
In summary, we showed that Roux-en-Y GBP, one of the most effective treatment options for morbid obesity, results in an exaggerated PYY response to oral carbohydrate ingestion, as well as a decrease in ghrelin, which through regulation of appetite may mechanistically explain why the procedure achieves long-term success. Although factors other than PYY and ghrelin could certainly (and most likely do) play a role in the appetite suppression occurring after Roux-en-Y GBP, our finding of an increased PYY response to nutrient intake after successful GBP supports the hypothesis that these hormones may be relevant in appetite regulation after the procedure. These physiology data have direct relevance for the treatment of obesity, which is often limited by difficulty in controlling appetite and is associated with a high recidivism rate after dietary intervention. Our data support ongoing efforts to determine whether PYY3–36 replacement represents an effective therapy for obesity.
This study was supported by NIH Grants MO1-RR01032 and R01–58, 785 and an Amgen grant (C.S.M.), NIH Grant K23 RR018860 (J.L.C.), and NIH Grant K08-DK02604 (E.C.M.). We thank the General Clinical Research Center nurses at Joslin for collecting the samples for this research.
Nonstandard abbreviations: GBP, gastric bypass; PYY, peptide YY; OGTT, oral glucose tolerance test; AUC, area under the curve.