|Home | About | Journals | Submit | Contact Us | Français|
Reductions in levels of the hunger-stimulating hormone ghrelin have been proposed to mediate part of the effects of vertical sleeve gastrectomy (VSG) and Roux-en-Y gastric bypass surgeries for obesity. We studied circulating levels of acyl and desacyl ghrelin in rats after these surgeries. We found that blood levels of ghrelin were reduced after VSG, but not after Roux-en-Y gastric bypass, based on enzyme-linked immunosorbent assay and mass-spectrometry analyses. We compared the effects of VSG in ghrelin-deficient mice and wild-type mice on food intake, body weight, dietary fat preference, and glucose tolerance. We found that VSG produced comparable outcomes in each strain. Reduced ghrelin signaling therefore does not appear to be required for these effects of VSG.
Bariatric surgery is currently the most effective treatment for obesity and several related metabolic diseases.1 In one such procedure termed vertical sleeve gastrectomy (VSG), the stomach is resected to approximately 10%–20% of its original size.2 Although VSG is termed a restrictive procedure, it also involves the removal of the gastric epithelium, which is the primary source of ghrelin, a 28–amino acid peptide that circulates in biologically active (acyl) and inactive (desacyl) forms, and is the endogenous ligand for the growth hormone secretagogue receptor.3
Given pharmacologically, acyl ghrelin increases food intake in human beings and rodents, increases ratings for palatable foods, and the neural responses to food cues.4,5 In addition to effects on energy balance, a wide range of data link acyl ghrelin to reduced hepatic insulin sensitivity and glucose-induced insulin secretion.6 Given that major effects of VSG and Roux-en-Y gastric bypass (RYGB) include reduced food intake, body weight, and increased hepatic insulin sensitivity and insulin secretion,2 it is plausible to hypothesize that reduced secretion of acyl ghrelin could contribute to the potent effects of bariatric surgery.
However, despite the obvious nature of this hypothesis, the effect of various bariatric procedures on ghrelin levels has remained controversial (Supplementary Table 1). To that end, we used rat models of both VSG and RYGB (Supplementary Figure 1) in which we could carefully control the conditions under which we collected blood samples and used both enzyme-linked immunosorbent assay and mass spectrometry to determine if acyl and desacyl ghrelin were altered after these surgeries. Acyl and desacyl ghrelin levels from VSG and RYGB rats are depicted in Figure 1. The first sample was taken in ad libitum–fed rats. In the subsequent sample, blood was collected from 6-hour–fasted rats, just before the onset of the dark cycle, at which time acyl and desacyl ghrelin levels were increased in sham, RYGB, and pair-fed rats, but were not increased in VSG rats (P < .05). The ratio of acyl:desacyl ghrelin was increased in VSG (2.0 ± 0.2) rats compared with pair-fed (1.3 ± 0.1), bypass (0.8 ± 0.1), and ad libitum–(1.3 ± 0.1) fed rats at this time, but absolute levels of the peptide were reduced. Re-feeding rapidly suppressed acyl ghrelin levels in sham and RYGB rats within 30 minutes (P < .05).
Given the controversy over methodologies to measure multiple circulating forms of ghrelin (Supplementary Table 1), we measured the same samples using mass spectrometry. Not surprisingly, the absolute values from this assay were not the same as those from the enzyme-linked immunosorbent assay. However, the pattern of changes between time points and among surgical groups was nearly identical. Fasting levels of acyl ghrelin were reduced significantly in VSG (471 ± 61 pg/mL) compared with the other treatment groups (sham, 799 ± 81.7 pg/mL; RYGB, 1130 ± 562 pg/mL; pair-fed, 716 ± 82 pg/mL), as were circulating levels of desacyl ghrelin (VSG, 315 ± 25.6 pg/mL; sham, 719 ± 83.5 pg/mL; RYGB, 816 ± 192.4 pg/mL; and pair-fed, 797 ± 72.3 pg/mL; P < .05). There were no significant differences among groups after re-feeding.
Thus, results from 2 different yet complementary methodologic approaches, clearly showed that circulating levels of acyl or desacyl ghrelin were unaffected by RYGB, but were reduced in VSG. Because we failed to see changes in ghrelin levels after RYGB, we then went on to directly test the necessity of reduced ghrelin signaling for bariatric surgery’s benefits by measuring a number of effects of VSG in a mouse with targeted ghrelin gene disruption. We reasoned that if these reductions contributed to the profound effects of VSG on food intake, food preference,7 body weight, and glucose tolerance, that VSG would be less effective in mice with targeted genetic disruption of the ghrelin gene.
We therefore assigned wild-type and ghrelin-deficient (knockout [KO]) mice to sham or VSG surgical groups, which were matched for lean and fat tissue mass. Geno-typing was reconfirmed with a commercially available enzyme-linked immunosorbent assay using an antibody directed against the entire ghrelin peptide (Figure 2A). Mice were exposed to a high-fat diet (41%) for 10 weeks before surgery. Consistent with previous reports,8 ghrelin KO mice weighed significantly less than wild-type mice before surgery (P < .05). Nevertheless, VSG reduced food intake (Figure 2B) (P < .05), and body weight (Figure 2C and D) comparably in both genotypes (P< .05). The effects of surgery on oral glucose tolerance (Figure 2E) (P < .05) and fat preference (Figure 2F) also were maintained in ghrelin KO mice. Thus, we conclude that changes in circulating ghrelin are not necessary for the primary metabolic benefits of the surgery.
The physiologic regulation of ghrelin is governed by factors related to nutritional status, such as the time of day, the length of time between meals, and anticipatory cues. Circulating levels of acyl and desacyl-ghrelin in our study were unaltered in a rat model of RYGB, but were dramatically reduced after VSG. The idea that reducing circulating levels of ghrelin produces some, or all, of the metabolic benefits of bariatric surgery is a credible and popular mechanistic hypothesis. However, we found that there were no differences in any of the effects of VSG that we studied in ghrelin-deficient mice. These mice experienced the same degree of weight loss, showed similar improvements in glucose tolerance, and displayed a reduction in fat preference that was consistent with the surgery’s effect in wild-type mice. It is important to acknowledge that developmental compensations in ghrelin-deficient mice may result in underestimating the contribution of ghrelin to the surgical outcomes of VSG. Nevertheless, these data support the conclusion that the sizeable reductions in circulating ghrelin produced by VSG are not necessary for the potent effects of VSG across a number of behavioral and physiologic variables. Rather, VSG must have other effects that eclipse, or exceed, the outcome produced by reductions in circulating ghrelin levels after this surgery.
Supported by a grant from Ethicon Endo-Surgery, Inc, and a Canadian Institutes of Health Research fellowship (A.P.C.).
Conflicts of interest
This author discloses the following: Matthias Tschöp is a consultant to Roche Pharmaceuticals and Ethicon Endo-Surgery. The remaining authors disclose no conflicts.