- Email: [email protected]
Neurobiologic changes in the hypothalamus associated with weight loss after gastric bypass
SURGICAL FORUM
Neurobiologic Changes in the Hypothalamus Associated with Weight Loss after Gastric Bypass Irina V Romanova, PhD, Eduardo JB Ramos, MD, Yuan Xu, MD, Robert Quinn, DVM, Chung Chen, PhD, Zachariah M George, BSc, Akio Inui, MD, PhD, Undurti Das, MD, Michael M Meguid, MD, PhD, FACS Effects of Roux-en-Y gastric bypass (RYGB) on hypothalamic food intake regulation have not been investigated. The hypothalamic arcuate nucleus (ARC) and the magnocellular (m) and parvocellular (p) parts of the paraventricular nucleus (PVN) regulate hunger and satiety, and are under control of the orexigenic neuropeptide Y (NPY), and the anorexigenic ␣-melanocyte stimulating hormone (␣-MSH) and serotonin (5-HT). We hypothesized that after RYGB, weight loss is associated with hypothalamic down regulation of NPY and up regulation of 5-HT and ␣-MSH. STUDY DESIGN: Obesity was induced in 12 Sprague Dawley rats using a high-energy diet for 7 weeks, and then the rats were divided into three groups (n ⫽ 4/group): RYGB, sham-operated pair-fed (PF), and shamoperated ad libitum (obese control). Ten days after operation, immunohistochemical quantification of NPY, ␣-MSH, and 5-HT1B-receptors in ARC and PVN was performed. Data were analyzed using ANOVA and Tukey’s test. RESULTS: Body weight decreased in RYGB (417 ⫾ 21g; mean ⫾ SE) and in PF (436 ⫾ 14g) rats 10 days after operation compared with obese control rats (484 ⫾ 15g; p ⬍ 0.05 for each comparison). NPY in ARC, pPVN, and mPVN decreased by 43%, 43%, and 61%, respectively in RYGB and by 55%, 42%, and 71% in PF, respectively, compared with obese controls (p ⬍ 0.05 for each pairwise comparison). RYGB versus PF did not show differences. ␣-MSH in ARC, pPVN and mPVN increased by 35%, 175%, and 67%, respectively in RYGB and by 29%, 162%, and 116% in PF, respectively, compared with obese controls (each p ⬍ 0.05). In mPVN, ␣-MSH significantly decreased by 23% in RYGB versus PF (p ⬍ 0.05). 5-HT-1B-receptor in pPVN increased by 58% in RYGB and by 26% in PF, compared with obese controls (p ⬍ 0.05). Compared with obese controls, 5HT-1B-receptor in mPVN increased by 39% in RYGB (p ⬍ 0.05) and by 9% in PF (p ⬎ 0.05). An increase of 5-HT-1B-receptor in pPVN and mPVN occurred in RYGB versus PF (p ⬍ 0.05). CONCLUSIONS: Obese rats that undergo weight loss after RYGB demonstrate changes in hypothalamic down regulation of NPY and up regulation of ␣-MSH and serotonin. (J Am Coll Surg 2004;199:887–895. © 2004 by the American College of Surgeons) BACKGROUND:
Using body mass index (BMI), defined as a person’s body weight in kg divided by their height in meters squared, as
the criterion for obesity, it is estimated that 25% to 33% of people in the US have a BMI of 25 to 30 kg/m2 and are considered overweight; another 30% have a BMI ⬎ 30 kg/m2 and are considered obese. Approximately 8 million people are morbidly obese. Obesity increases the incidence of type 2 diabetes mellitus, hypertension, hypercholesterolemia, and stroke, and decreases life expectancy.1,2 Obesity has been treated by diet, behavior modification, and through pharmacologic approaches with limited success. In morbid obesity, the most effective method to achieve a sustained weight loss is by surgical intervention.3 Besides weight loss, Roux-en-Y gastric bypass and biliopancreatic diversion effectively control type 2 diabetes mellitus in morbidly obese individuals by unknown mechanisms. It has been postulated that exclusion of the proximal gut contributes to improvement of diabetes.4 Obesity occurs as a result of biochemical changes in
No competing interests declared.
This work was supported in part by the Hendrick’s Fund #130230-30, by the Scientist Exchange Program of the Office of the International Affairs, NCI award to IVR (2002–2003), by an NIH/NCI Grant Ca-70239 to MMM, and by an Educational Grant from the Department of Surgery, University Hospital. Presented at the American College of Surgeons 89th Annual Clinical Congress, Surgical Forum, Chicago, IL, October 2003. Received December 17, 2003; Revised June 28, 2004; Accepted July 12, 2004. From the Surgical Metabolism and Nutrition Laboratory, Neuroscience Program, Department of Surgery, University Hospital, SUNY Upstate Medical University (Romanova, Ramos, Xu, Quinn, George, Das, Meguid), Syracuse, NY; the Department of Management Information and Decision, Whitman School of Management, Syracuse University (Chen), Syracuse, NY; and the Division of Diabetes, Digestive and Kidney Diseases, Department of Clinical Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan (Inui). Correspondence address: Michael M Meguid MD, PhD, Department of Surgery, University Hospital, 750 Adams St, Syracuse, NY 13210.
© 2004 by the American College of Surgeons Published by Elsevier Inc.
887
ISSN 1072-7515/04/$30.00 doi:10.1016/j.jamcollsurg.2004.07.013
888
Romanova et al
Hypothalamus Changes after Gastric Bypass
Abbreviations and Acronyms
ARC ⫽ ␣-MSH ⫽ m ⫽ NPY ⫽ p ⫽ PF ⫽ PVN ⫽ RYGB ⫽ 5-HT ⫽
arcuate nucleus ␣-melanocyte stimulating hormone magnocellular neuropeptide Y parvocellular pair-fed paraventricular nucleus Roux-en-Y gastric bypass serotonin
peripheral organs that lead to behavioral responses such as increased appetite and food intake, which play a contributory role in increasing peripheral fat and body mass. Under normal circumstances, daily food intake is modulated by the interaction of the gastrointestinal tract and the hypothalamus (the gut-brain-gut axis). In the hypothalamus, there is a complex neuronal network involving a series of anatomically and functionally related nuclei receiving afferent input from the periphery that modulate the efferent biologic responses through the hypothalamic-pituitary adrenal axis and gastrointestinal function.5,6 The most important food intake-regulating nuclei are the arcuate nucleus (ARC) and paraventricular nucleus (PVN), situated along the third ventricle of the brain. Anatomic and functional connections between these brain structures play a major role in the neurobiologic basis modulating the efferent brain responses in food intake; neuronal connections between the PVN and both the anterior and posterior pituitary activate the hypothalamic-pituitary adrenal axis, providing the afferent loop to regulate metabolism. In the ARC, several gastrointestinal hormones including insulin, leptin, and ghrelin regulate synthesis of neuropeptide Y (NPY) or ␣-melanocyte-stimulating hormone (␣-MSH), which are potent orexigenic and anorexigenic neuropeptides, respectively.5 Neuronal processes of cells producing NPY and ␣-MSH project from the ARC to the PVN (which is traditionally divided into two anatomic parts: the parvocellular [p] and the magnocellular [m]). At the same time, axons from cells of the PVN project to the pituitary gland and are involved in efferent regulation of peripheral organs by anterior and posterior pituitary hormones.6-8 Serotonin has a suppressive effect on appetite, food intake, and body weight gain,9,10 and the 5-HT1B-receptors are located in these neurons mediating satiety-inducing effects.11 Immunofluorescent double labeling shows the
J Am Coll Surg
interaction between neurons expressing 5-HT-1Breceptors and NPY in the PVN, supporting their dual co-existence in food intake regulatory nuclei.12 The Roux-en-Y gastric bypass (RYGB) operation induces longterm weight loss by a small gastric pouch and probably by a degree of subclinical malabsorption, secondary to the bypassed small bowel.13-15 But no data exist on the neurobiologic changes in the hypothalamus, which occur with the resultant decrease in food intake after RYGB that leads to considerable weight loss. We hypothesize that after RYGB in diet-induced obese rats, weight loss is associated with hypothalamic down regulation of the orexigenic peptide NPY and up regulation of the anorexigenic peptide ␣-MSH and 5-HT-1Breceptors. Such changes would contribute to persistent suppression of appetite and weight loss. The aim of this study was to examine changes in NPY, ␣-MSH, and 5-HT-1B-receptors in diet-induced obese rats 10 days after RYGB and in sham-operated pair-fed rats that are food intake restricted, and to compare such changes with those observed in sham-operated obese control rats. METHODS This study was approved by the Committee for the Humane Use of Animals at State University of New York Upstate Medical University and was performed in accordance with the guidelines established by the National Institutes of Health. Twelve 3 to 4-week-old male Sprague Dawley rats (Taconic Farms), weighing 50.9 ⫾ 1.3g (mean ⫾ SD), were acclimated to constant study environmental conditions: 12-12-hours light-dark cycle (lights on at 06:00), 26 ⫾ 1°C room temperature, and 45% humidity. After acclimation, rats were fed a high-energy diet (D12266, Research Diets) for 7 weeks together with a highly palatable liquid diet (Boost Plus, Mead Johnson). The high-energy diet consisted of 4.5 kcal/g, of which 21% of metabolizable energy content was protein, 31% fat, and 48% carbohydrate (consisting of 50% sucrose). Boost Plus (Mead Johnson) provided 1.5 kcal/mL, of which 16.7% is metabolizable energy as protein, 30% as fat, and 47.3% as carbohydrate. The rats were allowed ad libitum diets and municipal tap water. Experimental protocol
After 7 weeks on the high-energy diet, the diet-induced obese rats, weighing 492.9 ⫾ 8.4g, were stratified according to body weight and randomly assigned within
Vol. 199, No. 6, December 2004
Romanova et al
Hypothalamus Changes after Gastric Bypass
889
tric bypass procedures. The celiotomy incision was closed in layers. After both surgical procedures, depending on the hydration status of the rats, 20 to 30 mL/day of normal saline solution was injected subcutaneously into each axilla and groin for 3 consecutive days to prevent clinical dehydration. Liquid diet was provided for the first 4 days followed by the coarsely ground high-energy diet. The rats were followed for 10 days after operation and were then euthanized. No deaths occurred in the present study. Tissue preparation
Figure 1. Roux-en-Y bypass as performed in obese rats.
each strata to each of the three groups. This stratified randomization technique ensures that the mean weight of each group was approximately identical before the surgical procedures. The three groups were: RYGB, sham-operated pairfed (PF) rats that were fed postoperatively the same amount of food consumed by the RYGB group; and sham-operated rats, which continued on the high-energy diet ad libitum (obese control). Before operation, rats were food deprived for 12 hours. The procedure for the RYGB and sham-fed operation in rats has been previously described in detail.13-15 Briefly, as shown in Figure 1, two double rows of titanium staple lines (TRH30-4.8, Ethicon) were placed between the lesser and greater curvature of the stomach, creating a 20% gastric pouch. The gastric staple line was reinforced with multiple interrupted 4-0 polyglactin sutures (Ethicon), placed between the double rows of staple lines and hand sewn to provide a secure gastric partition. The jejunum was divided 16 cm below the ligament of Treitz and a gastrojejunostomy was hand sewn using interrupted 5-0 polyglactin sutures. The stump of proximal jejunum was closed with running suture. A side-to-side jejunostomy was also hand sewn at a distance of 10 cm below gastrojejunostomy. The sham operation consisted of a midline incision in which the stomach and distal esophagus were exposed. The stomach and the small bowel were manipulated for the same duration required for the gas-
Rats were anesthetized with isoflurane anesthesia (Baxter) and perfused through the heart with two solutions: phosphate buffer 0.02 M containing 0.9% saline solution (PBS; pH ⫽ 7.4) and then with 4% paraformaldehyde in 0.1 M phosphate buffer. Brains removed from the skull, and the forebrain, containing the hypothalamus, were dissected and also fixed in 4% paraformaldehyde for 4 hours at 4°C. Tissue blocks were cryoprotected in 30% sucrose for 48 hours, embedded in Tissue-Tek (Sakura Finetek USA, Inc), frozen in dry ice, and stored for 1 week at ⫺70°C. Coronal free-floating sections of 20 m-thickness of the forebrain including the hypothalamic area were prepared on a cryocut and collected in PBS. Immunocytochemical detection of NPY, ␣-MSH, and 5-HT-1B-receptors
Immunocytochemical reaction of each section with each antibody was repeated four times. The level of the sections of the ARC and PVN nuclei was verified by Nissle staining of the sections, and the anatomic sites observed under the microscope were compared with the anatomic sites in a brain atlas.16 Equal levels of nuclei were used for immunocytochemical visualization of NPY, ␣-MSH, or 5-HT1Breceptors by the peroxidase-anti-peroxidase (PAP) method17 and specific polyclonal primary antibodies. NPY was detected by rabbit anti-NPY RGG-7180 (Peninsula Laboratory, 1:1,500), ␣-MSH by rabbit anti-␣-MSH RGG-7251 (Peninsula Laboratory, 1:1,500), and 5HT-1B-receptors by guinea pig anti-5-HT1B-receptors AB 5410 (Chemicon, 1:2,000). All solutions were based on PBS, containing 0.3% Triton X-100 (Fisher Biotech), and the protocols previously described in our laboratory.18 Visualization of reactions were simultaneously carried out with 3,3=-diaminobenzidine tetra hydrochloride (Sigma). Control for specificity of the immunostaining was done by
890
Romanova et al
Hypothalamus Changes after Gastric Bypass
J Am Coll Surg
omission of primary or secondary antibodies and preabsorption test (2.5 g/mL). No immunoreactivity was found on the control sections. Hypothalamic nuclei sections from RYGB, PF, and obese control rats were mounted on slides, air-dried, and cover slipped and were examined under a light microscope using Micromaster (Fisher Scientific), and digital images were obtained with a Nikon CF160 microscope. Image analysis of NPY, ␣-MSH, and 5-HT-1Breceptor immunostaining
For image analysis, sections from equivalent brain levels of RYGB, PF, and obese control groups were obtained as described in the preceding text. ARC or PVN stained for NPY, ␣-MSH, or 5-HT-1B-receptors were photographed under the same conditions using a digital camera HC-300Z FUJIX mounted on a Leitz-Ortholux microscope. Because the PVN is not a homogeneous structure, but consists of recognized m and p areas, these were examined and analyzed separately. High-resolution digital photomicrographs with ⫻10 magnification were processed with the NIH Scion Image analysis program. This program allows measurement of changes in ARC, p, and m parts of PVN by subtracting nonspecific background density points and measuring the optical density of NPY, ␣-MSH, or 5-HT-1B-receptor immunostaining. This was repeated 30 to 50 times in each group of rats for each nucleus and for each type of immunoreactivity. Differences in optical density in RYGB and PF rats were expressed as percent of obese control rats. Statistical analysis
Densities of NPY, ␣-MSH, or 5-HT-1B-receptor immunostaining and changes in body weight were analyzed by one-way ANOVA to investigate the effect among groups of RYGB, PF, and obese control rats. When the ANOVA test rejected the hypothesis of no difference, Tukey’s pairwise multiple comparison procedure was applied to ensure that the family (ie, tests of RYGB versus obese control, PF versus obese control, and RYGB versus PF) error rate was less than or equal to 0.05 jointly. Graphic data are presented showing comparisons of RYGB versus obese control, PF versus obese control, and RYGB versus PF. Data are reported as mean ⫾ SE and graphically as box plots of 50% of the data in each group. The horizontal line in each box plot represents the mean, and asterisks represent the extreme data of analyzed points.
Figure 2. Degree of weight loss after Roux-en-Y gastric bypass (RYGB) and pair-fed (PF) compared with obese controls on 10th postoperative day (POD #10) and weight of obese rats preoperatively (pre-op). *p ⬍ 0.05 compared with obese controls.
RESULTS Clinical results
By the 10th postoperative day, RYGB rats versus obese controls ate 57% less and had lost 17% of body weight (Fig. 2). Body weight in RYGB rats was 417 ⫾ 21g, in PF rats it was 436 ⫾ 14g, and in obese control rats it was 484 ⫾ 15g. There was no difference in body weight between RYGB and PF (p ⫽ 0.45). Ten rat days was selected as the end point for our study because it approximates the equivalent of 1 human year, so it would be expected to demonstrate marked changes in the hypothalamus. NPY immunostaining
In Figure 3A and 3B NPY immunoreactivity in the ARC was decreased by 43% in the RYGB (0.340 ⫾ 0.036) and by 55% in the PF (0.270 ⫾ 0.026) groups, compared with the obese control group (0.600 ⫾ 0.048), p ⬍ 0.05. Comparison of RYGB and PF rats was not statistically significant. Figures 4A and 4B show a photograph of the decrease of NPY immunoreactivity in the ARC after gastric bypass. In Figure 3B, in the p part of PVN, NPY immunoreactivity was decreased by 43% in RYGB (0.323 ⫾ 0.190) and by 42% in PF (0.329 ⫾ 0.024), compared with obese control rats (0.564 ⫾ 0.040), p ⬍ 0.05. There was no significant difference (p ⬎ 0.05) in NPY expression in RYGB versus PF. In the m part of PVN, NPY immunoreactivity was decreased by 61% in RYGB (0.121 ⫾ 0.009) and by 71% in PF (0.089 ⫾ 0.009) compared with obese controls (0.307 ⫾ 0.026), p ⬍ 0.05, as shown in Figure 3C. In the m part of PVN there was no difference in NPY immunoreactivity between RYGB and PF, as in p. Morphologic changes between RYGB and obese control in PVN rats are represented in Figures 5A and 5B.
Vol. 199, No. 6, December 2004
Romanova et al
Hypothalamus Changes after Gastric Bypass
891
Figure 4. (A, B) Distribution of neuropeptide Y (NPY) (C, D) and alpha-melanocyte stimulating hormone (␣-MSH) immunoreactivity fibers in arcuate nucleus (ARC) of control and Roux-en-Y gastric bypass (RYGB) rats. A considerable decrease in NPY and increase in ␣-MSH-immunoreactivity in ARC occurred after RYGB. v, third ventricle; Bar ⫽ 100 m (shows scale of magnificaiton).
␣-MSH immunostaining
Figure 3. (A) Neuropeptide Y (NPY) optical density (OD) in arcuate nucleus, (B) parvocellular part of the paraventricular nucleus (PVN), and (C) magnocellular part of PVN in Roux-en-Y gastric bypass (RYGB), pair-fed (PF), and obese control rats. The box plots show a decrease of NPY (OD) in RYGB and PF rats versus controls (p ⬍ 0.05). The horizontal line in each box plot is the mean and asterisks represent the extreme data of analyzed points.
␣-MSH-immunoreactivity in the ARC was increased by 35% in RYGB (0.152 ⫾ 0.013) and by 29% in PF (0.146 ⫾ 0.013), compared with obese controls (0.113 ⫾ 0.008), p ⬍ 0.05, as shown schematically in Figure 6A. No significant difference was measured in ␣-MSH expression in ARC between RYGB and PF rats (p ⬎ 0.05). The photographs in Figures 4C and 4D show the increase of ␣-MSH immunoreactivity in the ARC after gastric bypass. ␣-MSH immunoreactivity in the p part of the PVN was increased by 175% in RYGB (0.204 ⫾ 0.018) and by 162% in PF (0.194 ⫾ 0.012), compared with obese controls (0.074 ⫾ 0.005), p ⬍ 0.05, as shown in Figures 6B and 6C. No significant difference was found between RYGB and PF rats. In the m part of the PVN, ␣-MSH immunoreactivity was increased by 67% in RYGB (0.050 ⫾ 0.003) and by 116% in PF (0.065 ⫾ 0.008), compared with obese controls (0.030 ⫾ 0.002), p ⬍ 0.05. ␣-MSH immunoreactivity was significantly decreased by 23% in RYGB versus PF (p ⬍ 0.05). Morphologic changes in PVN comparing RYGB versus obese control rats are demonstrated in Figures 5C and 5D. 5-HT-1B-receptor immunostaining
There was no measurable staining of 5-HT-1B-receptors in the ARC. Figure 7 shows a 58% increase in 5-HT-1B-
892
Romanova et al
Hypothalamus Changes after Gastric Bypass
J Am Coll Surg
significant increase was measured in 5-HT-1B-receptorimmunoreactivity in p and m parts of PVN after RYGB (p ⬍ 0.05). Morphologic changes in PVN between RYGB versus obese control rats are demonstrated photographically in Figures 5E and 5F.
Figure 5. (A, B) Distribution of neuropeptide Y (NPY) (C, D) alphamelanocyte stimulating hormone (␣-MSH) and (E, F) serotonin (5HT)-1B-receptor immunoreactivity in the magnocellular (m) and parvocellular (p) parts of paraventricular nucleus (PVN) in control and Roux-en-Y gastric bypass (RYGB) rats. A significant decrease in NPY and increase in ␣-MSH and 5-HT-1B-receptor immunoreactivity in paraventricular nucleus occurred following RYGB. v, third ventricle, Bar ⫽ 100 m (shows scale of magnification).
receptor-immunoreactivity in the p part of RYGB (0.049 ⫾ 0.002) and a 26% increase in PF (0.039 ⫾ 0.002), compared with obese controls (0.031 ⫾ 0.001), p ⬍ 0.05. In the m part of PVN, a 39% increase in RYGB (0.370 ⫾ 0.038) and a 9% increase in PF (0.292 ⫾ 0.018) occurred, compared with obese controls (0.270 ⫾ 0.026), p ⬍ 0.05 and p ⬎ 0.05, respectively. Comparing RYGB and PF, a
DISCUSSION A subset of patients has been recognized as being morbidly obese as part of the national epidemic of obesity. These patients have metabolic comorbidities including type 2 diabetes and heart disease, which lead to premature death. To reduce their weight and comorbidities, and therefore their risk factors, some patients undergo gastric bypass operations. The RYGB operation reduces the size of the gastric reservoir, reducing caloric intake and inducing early satiety, and probably induces a subclinical form of malabsorption, with both factors contributing to weight loss. But the effect of RYGB on the brain and its contributory role in weight reduction has not been previously described and is explored using our RYGB rat model. Young Sprague Dawley rats were fed a high-energy diet to induce obesity and then subjected to RYGB or sham operation. After sham operation, rats either continued on the high-energy diet (obese control) or were PF with the mean of the daily food intake consumed by RYGB rats. Ten days later, changes in the most potent orexigenic peptide (NPY) were measured in an important food intake hypothalamic region, then ARC. A similar procedure was followed with the most potent anorexigenic peptide, ␣-MSH, and also with the 5HT-1Breceptor, a monoamine known to be involved in food intake inhibition. Because second order neurons project from the ARC to the PVN, which is involved in the regulation of efferent visceral activity, changes in these neuromodulators were also measured in the PVN. The main focus in our study was to compare RYGB in obese rats versus sham-operated obese rats, in changes of hypothalamic orexigenic peptide NPY and in the anorexigenic peptide ␣-MSH and 5-HT-11B-receptors. Comparison of RYGB and PF in many parameters did not show significance, which may be a type II error. A more likely explanation is that the concept of PF as a model in this type of study needs to be reconsidered. These rats are given one meal a day, providing only food that is equivalent to that eaten by the RYGB rats throughout the day, and the PF rats eat this meal immediately, ie, the PF rats are on a diet. These rats are frisky
Vol. 199, No. 6, December 2004
Romanova et al
Hypothalamus Changes after Gastric Bypass
893
Figure 7. Serotonin (5-HT)-1B-receptor optical density (OD) in the (A) parvocellular part of the paraventricular nucleus (PVN) and (B) the magnocellular part of PVN in Roux-en-Y gastric bypass (RYGB), pair-fed (PF), and obese control rats. The box plots show an increase of 5-HT-1B OD in RYGB versus obese controls and versus PF. The horizontal line in each box plot is the mean and asterisks represent the extreme data of analyzed points.
and they are more aggressive because they are constantly hungry, and this behavioral difference alters the hypothalamic response to food intake mediators. Although both groups versus the sham obese controls lose weight to the same degrees because of the stressful nature of the model, the neurochemical changes may not reflect parallel directional changes, giving the appearance of a type
Figure 6. Alpha-melanocyte stimulating hormone (␣-MSH) optical density (OD) in (A) arcuate nucleus, (B) the parvocellular part of paraventricular nucleus (PVN) and the magnocellular part of PVN (C) in Roux-en-Y gastric bypass (RYGB), pair-fed (PF), and obese control
rats. The box plots show an increase of ␣-MSH OD in RYGB and PF rats versus controls (p ⬍ 0.05). In the magnocellular part of PVN, a significant decrease occurred in RYGB versus PF (p ⬍ 0.05). The horizontal line in each box plot is the mean and asterisks represent the extreme data of analyzed points.
894
Romanova et al
Hypothalamus Changes after Gastric Bypass
II error. This raises questions concerning interpretation of physiologic data with the use of a PF model. In current studies we are considering eliminating this group. So based on our hypothesis, we conclude that the observed neurobiologic changes are from weight loss. The RYGB led to weight loss (because of decreased food intake) and to the neurobiologic changes; the PF model also led to weight loss but introduced a stress variable. In obese control rats, NPY is substantially elevated in the ARC and in the two foci of the PVN. These are markedly reduced after weight loss, either in RYGB or PF groups. The degree of difference in the reduction between the RYGB and PF is not striking, suggesting that caloric restriction is a major stimulus affecting NPY expression. The same decrease of NPY immunoreactivity in ARC and PVN was shown in anorectic tumorbearing rats.18 This also may support the importance of NPY-dependent stimuli on feeding behavior. In contrast, ␣-MSH is considerably lower in obese controls and increases with RYGB and PF. This finding suggests a role of this anorexigenic peptide as an eating inhibitor.5 The greater increase in ␣-MSH in the ARC correlated with changes in both parts of the PVN.19 Our data suggest that ␣-MSH, through regulation of several efferent systems in the PVN, similar to stress regulatory systems in magnocellular PVN, is involved in the mechanism of weight loss in RYGB and PF rats. Serotonin has a suppressive effect on food intake.6,9,10 Drugs that increase the activity of central 5-HT have been widely used as appetite suppressants.11 5-HT innervation is widely distributed in the hypothalamus by different types of receptors. We included 5-HT evaluation in our study because it has been previously shown that 5-HT innervates NPY neurons at the level of cell bodies in the ARC and at distal axons in the PVN,20 and dense 5-HT-1B-receptor immunoreactivity has been shown in the m part of PVN, where they are compactly localized.12 Fluorescent double labeling supports the coexistence of NPY fibers with 5-HT-1B-receptor immunopositive neurons in PVN, although a greater degree of overlap exists in the ARC.12 Recently, using an immunohistochemical dual-labeling technique, it was shown that a substantial increase in c-FOS-immunoreactivity in ␣-MSH neurons occurs in ARC.21 Our data demonstrate the relative changes observed in NPY, ␣-MSH, and 5HT-1B-receptor concentrations in neuronal tissue at the time that weight loss occurred after RYGB, relative to obese controls. Our current data
J Am Coll Surg
are valuable in providing information about the changes that occur and the magnitude and direction of these changes, permitting the design of future studies to enhance our understanding of the neurobiologic changes in the hypothalamus associated with RYGB-induced weight loss and to also determine its role in weight loss. Author Contributions Morphological study with hypothalamic analyses: Romanova Execution of procedures and/or operation: Ramos, Xu Acquisition of samples: Xu, George Drafting of manuscript: Meguid Critical revision: Inui, Das, Meguid Statistical expertise: Chen Obtaining funding: Ramos, Meguid Supervision: Meguid Supervision of animals: Quinn Animal handling: George Acknowledgment: We thank Ms Karen Hughes for her expert technical assistance.
REFERENCES 1. Hill JO, Wyatt HR, Reed GW, Peters JC. Obesity and the environment: where do we go from here? Science 2003;299:853– 856. 2. Mokdad AH, Bowman BA, Ford ES. The continuing epidemic of obesity and diabetes in the United States. JAMA 2001;286: 1195–1200. 3. Brolin RE. Bariatric surgery and long-term control of morbid obesity. JAMA 2002;288:2793–2796. 4. Pories WJ, Albrecht RJ. Etiology of type II diabetes mellitus: role of the foregut. World J Surg 2001;25:527–531. 5. Schwartz MW, Morton GJ. Keeping hunger at bay. Nature 2002;418:595–597. 6. Meguid MM, Fetissov SO, Varma M, et al. Hypothalamic dopamine and serotonin in the regulation of food intake. Nutrition 2000;16:843–857. 7. Legradi G, Emerson CH, Ahima RS, et al. Leptin prevents fasting-induced suppression of prothyrotropin-releasing hormone messenger ribonucleic acids in neurons of the hypothalamic paraventricular nucleus. Endocrinology 1997;138:2569– 2576. 8. Flier JS, Maratos-Flier E. Obesity and the hypothalamus: novel peptides for new pathway. Cell 1998;92:437–440. 9. Samanin R, Garattini S. Pharmacology of ingestive behavior. Terapie 1996;51:107–115. 10. Laviano A, Meguid MM. Serotonin and obesity. Curr Med Chem-Central Nervous System Agents 2003;3:89–100. 11. Curzon G, Gibson EL, Oluyomi AO. Appetite suppression by commonly used drugs depends on 5-HT receptors but not 5-HT availability. Trends Pharmacol Sci 1997;18:21–25.
Vol. 199, No. 6, December 2004
Romanova et al
12. Makarenko IG, Meguid MM, Ugrumov MV. Distribution of serotonin 5-hydroxytriptamine 1B (5-HT-1B) receptors in the normal rat hypothalamus. Neuroscience Lett 2002;328:155– 159. 13. Xu Y, Ohinata K, Meguid MM, et al. Gastric bypass model in the obese rat to study metabolic mechanism of weight loss. J Surg Res 2002;107:56–63. 14. Ramos EJ, Xu Y, Romanova I, et al. Is obesity an inflammatory disease? Surgery 2003;134:329–335. 15. Meguid MM, Ramos EJB, Suzuki S, et al. A surgical rat model of human Roux-en-Y gastric bypass. J Gastrointest Surg 2004;8: 621–630. 16. Paxinos G, Watson C, eds. The rat brain in stereotaxic coordinates, New York: Academic Press; 1997: [figures 25 and 26]. 17. Sternberger L, ed. Immunochemistry. New York: Wiley; 1979: 122.
Hypothalamus Changes after Gastric Bypass
895
18. Makarenko IG, Meguid MM, Gatto L, et al. Decreased NPY innervation of the hypothalamus nuclei in rats with cancer anorexia. Brain Res 2003;24:100–108. 19. Sarkar S, Lechan RM. Central administration of neuropeptide Y reduces alpha-melanocyte-stimulating hormone-induced cyclic adenosine 5=-monophosphate response element binding protein (CREB) phosphorylation in pro-thyrotropin-releasing hormone neurons and increases CREB phosphorylation in corticotropin-releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology 2003;144:281–291. 20. Guy J, Pelletier G, Bosler O. Serotonin innervation of neuropeptide Y containing neurons in the arcuate nucleus. Neurosci Lett 1988;85:9–13. 21. Heisler LK, Cowley MA, Tecott LH, et al. Activation of central melanocortin pathways by fenfluramine. Science 2002;297: 1609–1611.
ANNOUNCEMENT Beginning September 2004 JACS editorial office will test an all electronic editorial system.
JANUARY 2005 All manuscripts will be submitted on the electronic system ONLY. WATCH FOR GUIDELINES All new electronic editorial office for JACS manuscripts. JACS ONLINE CME-1 PROGRAM CME-1 COURSE EVERY MONTH 2 CREDITS AVAILABLE EACH MONTH 48 CREDITS AVAILABLE AT ALL TIMES www.jacsme.facs.org
Neurobiologic Changes in the Hypothalamus Associated with Weight Loss after Gastric Bypass Irina V Romanova, PhD, Eduardo JB Ramos, MD, Yuan Xu, MD, Robert Quinn, DVM, Chung Chen, PhD, Zachariah M George, BSc, Akio Inui, MD, PhD, Undurti Das, MD, Michael M Meguid, MD, PhD, FACS Effects of Roux-en-Y gastric bypass (RYGB) on hypothalamic food intake regulation have not been investigated. The hypothalamic arcuate nucleus (ARC) and the magnocellular (m) and parvocellular (p) parts of the paraventricular nucleus (PVN) regulate hunger and satiety, and are under control of the orexigenic neuropeptide Y (NPY), and the anorexigenic ␣-melanocyte stimulating hormone (␣-MSH) and serotonin (5-HT). We hypothesized that after RYGB, weight loss is associated with hypothalamic down regulation of NPY and up regulation of 5-HT and ␣-MSH. STUDY DESIGN: Obesity was induced in 12 Sprague Dawley rats using a high-energy diet for 7 weeks, and then the rats were divided into three groups (n ⫽ 4/group): RYGB, sham-operated pair-fed (PF), and shamoperated ad libitum (obese control). Ten days after operation, immunohistochemical quantification of NPY, ␣-MSH, and 5-HT1B-receptors in ARC and PVN was performed. Data were analyzed using ANOVA and Tukey’s test. RESULTS: Body weight decreased in RYGB (417 ⫾ 21g; mean ⫾ SE) and in PF (436 ⫾ 14g) rats 10 days after operation compared with obese control rats (484 ⫾ 15g; p ⬍ 0.05 for each comparison). NPY in ARC, pPVN, and mPVN decreased by 43%, 43%, and 61%, respectively in RYGB and by 55%, 42%, and 71% in PF, respectively, compared with obese controls (p ⬍ 0.05 for each pairwise comparison). RYGB versus PF did not show differences. ␣-MSH in ARC, pPVN and mPVN increased by 35%, 175%, and 67%, respectively in RYGB and by 29%, 162%, and 116% in PF, respectively, compared with obese controls (each p ⬍ 0.05). In mPVN, ␣-MSH significantly decreased by 23% in RYGB versus PF (p ⬍ 0.05). 5-HT-1B-receptor in pPVN increased by 58% in RYGB and by 26% in PF, compared with obese controls (p ⬍ 0.05). Compared with obese controls, 5HT-1B-receptor in mPVN increased by 39% in RYGB (p ⬍ 0.05) and by 9% in PF (p ⬎ 0.05). An increase of 5-HT-1B-receptor in pPVN and mPVN occurred in RYGB versus PF (p ⬍ 0.05). CONCLUSIONS: Obese rats that undergo weight loss after RYGB demonstrate changes in hypothalamic down regulation of NPY and up regulation of ␣-MSH and serotonin. (J Am Coll Surg 2004;199:887–895. © 2004 by the American College of Surgeons) BACKGROUND:
Using body mass index (BMI), defined as a person’s body weight in kg divided by their height in meters squared, as
the criterion for obesity, it is estimated that 25% to 33% of people in the US have a BMI of 25 to 30 kg/m2 and are considered overweight; another 30% have a BMI ⬎ 30 kg/m2 and are considered obese. Approximately 8 million people are morbidly obese. Obesity increases the incidence of type 2 diabetes mellitus, hypertension, hypercholesterolemia, and stroke, and decreases life expectancy.1,2 Obesity has been treated by diet, behavior modification, and through pharmacologic approaches with limited success. In morbid obesity, the most effective method to achieve a sustained weight loss is by surgical intervention.3 Besides weight loss, Roux-en-Y gastric bypass and biliopancreatic diversion effectively control type 2 diabetes mellitus in morbidly obese individuals by unknown mechanisms. It has been postulated that exclusion of the proximal gut contributes to improvement of diabetes.4 Obesity occurs as a result of biochemical changes in
No competing interests declared.
This work was supported in part by the Hendrick’s Fund #130230-30, by the Scientist Exchange Program of the Office of the International Affairs, NCI award to IVR (2002–2003), by an NIH/NCI Grant Ca-70239 to MMM, and by an Educational Grant from the Department of Surgery, University Hospital. Presented at the American College of Surgeons 89th Annual Clinical Congress, Surgical Forum, Chicago, IL, October 2003. Received December 17, 2003; Revised June 28, 2004; Accepted July 12, 2004. From the Surgical Metabolism and Nutrition Laboratory, Neuroscience Program, Department of Surgery, University Hospital, SUNY Upstate Medical University (Romanova, Ramos, Xu, Quinn, George, Das, Meguid), Syracuse, NY; the Department of Management Information and Decision, Whitman School of Management, Syracuse University (Chen), Syracuse, NY; and the Division of Diabetes, Digestive and Kidney Diseases, Department of Clinical Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan (Inui). Correspondence address: Michael M Meguid MD, PhD, Department of Surgery, University Hospital, 750 Adams St, Syracuse, NY 13210.
© 2004 by the American College of Surgeons Published by Elsevier Inc.
887
ISSN 1072-7515/04/$30.00 doi:10.1016/j.jamcollsurg.2004.07.013
888
Romanova et al
Hypothalamus Changes after Gastric Bypass
Abbreviations and Acronyms
ARC ⫽ ␣-MSH ⫽ m ⫽ NPY ⫽ p ⫽ PF ⫽ PVN ⫽ RYGB ⫽ 5-HT ⫽
arcuate nucleus ␣-melanocyte stimulating hormone magnocellular neuropeptide Y parvocellular pair-fed paraventricular nucleus Roux-en-Y gastric bypass serotonin
peripheral organs that lead to behavioral responses such as increased appetite and food intake, which play a contributory role in increasing peripheral fat and body mass. Under normal circumstances, daily food intake is modulated by the interaction of the gastrointestinal tract and the hypothalamus (the gut-brain-gut axis). In the hypothalamus, there is a complex neuronal network involving a series of anatomically and functionally related nuclei receiving afferent input from the periphery that modulate the efferent biologic responses through the hypothalamic-pituitary adrenal axis and gastrointestinal function.5,6 The most important food intake-regulating nuclei are the arcuate nucleus (ARC) and paraventricular nucleus (PVN), situated along the third ventricle of the brain. Anatomic and functional connections between these brain structures play a major role in the neurobiologic basis modulating the efferent brain responses in food intake; neuronal connections between the PVN and both the anterior and posterior pituitary activate the hypothalamic-pituitary adrenal axis, providing the afferent loop to regulate metabolism. In the ARC, several gastrointestinal hormones including insulin, leptin, and ghrelin regulate synthesis of neuropeptide Y (NPY) or ␣-melanocyte-stimulating hormone (␣-MSH), which are potent orexigenic and anorexigenic neuropeptides, respectively.5 Neuronal processes of cells producing NPY and ␣-MSH project from the ARC to the PVN (which is traditionally divided into two anatomic parts: the parvocellular [p] and the magnocellular [m]). At the same time, axons from cells of the PVN project to the pituitary gland and are involved in efferent regulation of peripheral organs by anterior and posterior pituitary hormones.6-8 Serotonin has a suppressive effect on appetite, food intake, and body weight gain,9,10 and the 5-HT1B-receptors are located in these neurons mediating satiety-inducing effects.11 Immunofluorescent double labeling shows the
J Am Coll Surg
interaction between neurons expressing 5-HT-1Breceptors and NPY in the PVN, supporting their dual co-existence in food intake regulatory nuclei.12 The Roux-en-Y gastric bypass (RYGB) operation induces longterm weight loss by a small gastric pouch and probably by a degree of subclinical malabsorption, secondary to the bypassed small bowel.13-15 But no data exist on the neurobiologic changes in the hypothalamus, which occur with the resultant decrease in food intake after RYGB that leads to considerable weight loss. We hypothesize that after RYGB in diet-induced obese rats, weight loss is associated with hypothalamic down regulation of the orexigenic peptide NPY and up regulation of the anorexigenic peptide ␣-MSH and 5-HT-1Breceptors. Such changes would contribute to persistent suppression of appetite and weight loss. The aim of this study was to examine changes in NPY, ␣-MSH, and 5-HT-1B-receptors in diet-induced obese rats 10 days after RYGB and in sham-operated pair-fed rats that are food intake restricted, and to compare such changes with those observed in sham-operated obese control rats. METHODS This study was approved by the Committee for the Humane Use of Animals at State University of New York Upstate Medical University and was performed in accordance with the guidelines established by the National Institutes of Health. Twelve 3 to 4-week-old male Sprague Dawley rats (Taconic Farms), weighing 50.9 ⫾ 1.3g (mean ⫾ SD), were acclimated to constant study environmental conditions: 12-12-hours light-dark cycle (lights on at 06:00), 26 ⫾ 1°C room temperature, and 45% humidity. After acclimation, rats were fed a high-energy diet (D12266, Research Diets) for 7 weeks together with a highly palatable liquid diet (Boost Plus, Mead Johnson). The high-energy diet consisted of 4.5 kcal/g, of which 21% of metabolizable energy content was protein, 31% fat, and 48% carbohydrate (consisting of 50% sucrose). Boost Plus (Mead Johnson) provided 1.5 kcal/mL, of which 16.7% is metabolizable energy as protein, 30% as fat, and 47.3% as carbohydrate. The rats were allowed ad libitum diets and municipal tap water. Experimental protocol
After 7 weeks on the high-energy diet, the diet-induced obese rats, weighing 492.9 ⫾ 8.4g, were stratified according to body weight and randomly assigned within
Vol. 199, No. 6, December 2004
Romanova et al
Hypothalamus Changes after Gastric Bypass
889
tric bypass procedures. The celiotomy incision was closed in layers. After both surgical procedures, depending on the hydration status of the rats, 20 to 30 mL/day of normal saline solution was injected subcutaneously into each axilla and groin for 3 consecutive days to prevent clinical dehydration. Liquid diet was provided for the first 4 days followed by the coarsely ground high-energy diet. The rats were followed for 10 days after operation and were then euthanized. No deaths occurred in the present study. Tissue preparation
Figure 1. Roux-en-Y bypass as performed in obese rats.
each strata to each of the three groups. This stratified randomization technique ensures that the mean weight of each group was approximately identical before the surgical procedures. The three groups were: RYGB, sham-operated pairfed (PF) rats that were fed postoperatively the same amount of food consumed by the RYGB group; and sham-operated rats, which continued on the high-energy diet ad libitum (obese control). Before operation, rats were food deprived for 12 hours. The procedure for the RYGB and sham-fed operation in rats has been previously described in detail.13-15 Briefly, as shown in Figure 1, two double rows of titanium staple lines (TRH30-4.8, Ethicon) were placed between the lesser and greater curvature of the stomach, creating a 20% gastric pouch. The gastric staple line was reinforced with multiple interrupted 4-0 polyglactin sutures (Ethicon), placed between the double rows of staple lines and hand sewn to provide a secure gastric partition. The jejunum was divided 16 cm below the ligament of Treitz and a gastrojejunostomy was hand sewn using interrupted 5-0 polyglactin sutures. The stump of proximal jejunum was closed with running suture. A side-to-side jejunostomy was also hand sewn at a distance of 10 cm below gastrojejunostomy. The sham operation consisted of a midline incision in which the stomach and distal esophagus were exposed. The stomach and the small bowel were manipulated for the same duration required for the gas-
Rats were anesthetized with isoflurane anesthesia (Baxter) and perfused through the heart with two solutions: phosphate buffer 0.02 M containing 0.9% saline solution (PBS; pH ⫽ 7.4) and then with 4% paraformaldehyde in 0.1 M phosphate buffer. Brains removed from the skull, and the forebrain, containing the hypothalamus, were dissected and also fixed in 4% paraformaldehyde for 4 hours at 4°C. Tissue blocks were cryoprotected in 30% sucrose for 48 hours, embedded in Tissue-Tek (Sakura Finetek USA, Inc), frozen in dry ice, and stored for 1 week at ⫺70°C. Coronal free-floating sections of 20 m-thickness of the forebrain including the hypothalamic area were prepared on a cryocut and collected in PBS. Immunocytochemical detection of NPY, ␣-MSH, and 5-HT-1B-receptors
Immunocytochemical reaction of each section with each antibody was repeated four times. The level of the sections of the ARC and PVN nuclei was verified by Nissle staining of the sections, and the anatomic sites observed under the microscope were compared with the anatomic sites in a brain atlas.16 Equal levels of nuclei were used for immunocytochemical visualization of NPY, ␣-MSH, or 5-HT1Breceptors by the peroxidase-anti-peroxidase (PAP) method17 and specific polyclonal primary antibodies. NPY was detected by rabbit anti-NPY RGG-7180 (Peninsula Laboratory, 1:1,500), ␣-MSH by rabbit anti-␣-MSH RGG-7251 (Peninsula Laboratory, 1:1,500), and 5HT-1B-receptors by guinea pig anti-5-HT1B-receptors AB 5410 (Chemicon, 1:2,000). All solutions were based on PBS, containing 0.3% Triton X-100 (Fisher Biotech), and the protocols previously described in our laboratory.18 Visualization of reactions were simultaneously carried out with 3,3=-diaminobenzidine tetra hydrochloride (Sigma). Control for specificity of the immunostaining was done by
890
Romanova et al
Hypothalamus Changes after Gastric Bypass
J Am Coll Surg
omission of primary or secondary antibodies and preabsorption test (2.5 g/mL). No immunoreactivity was found on the control sections. Hypothalamic nuclei sections from RYGB, PF, and obese control rats were mounted on slides, air-dried, and cover slipped and were examined under a light microscope using Micromaster (Fisher Scientific), and digital images were obtained with a Nikon CF160 microscope. Image analysis of NPY, ␣-MSH, and 5-HT-1Breceptor immunostaining
For image analysis, sections from equivalent brain levels of RYGB, PF, and obese control groups were obtained as described in the preceding text. ARC or PVN stained for NPY, ␣-MSH, or 5-HT-1B-receptors were photographed under the same conditions using a digital camera HC-300Z FUJIX mounted on a Leitz-Ortholux microscope. Because the PVN is not a homogeneous structure, but consists of recognized m and p areas, these were examined and analyzed separately. High-resolution digital photomicrographs with ⫻10 magnification were processed with the NIH Scion Image analysis program. This program allows measurement of changes in ARC, p, and m parts of PVN by subtracting nonspecific background density points and measuring the optical density of NPY, ␣-MSH, or 5-HT-1B-receptor immunostaining. This was repeated 30 to 50 times in each group of rats for each nucleus and for each type of immunoreactivity. Differences in optical density in RYGB and PF rats were expressed as percent of obese control rats. Statistical analysis
Densities of NPY, ␣-MSH, or 5-HT-1B-receptor immunostaining and changes in body weight were analyzed by one-way ANOVA to investigate the effect among groups of RYGB, PF, and obese control rats. When the ANOVA test rejected the hypothesis of no difference, Tukey’s pairwise multiple comparison procedure was applied to ensure that the family (ie, tests of RYGB versus obese control, PF versus obese control, and RYGB versus PF) error rate was less than or equal to 0.05 jointly. Graphic data are presented showing comparisons of RYGB versus obese control, PF versus obese control, and RYGB versus PF. Data are reported as mean ⫾ SE and graphically as box plots of 50% of the data in each group. The horizontal line in each box plot represents the mean, and asterisks represent the extreme data of analyzed points.
Figure 2. Degree of weight loss after Roux-en-Y gastric bypass (RYGB) and pair-fed (PF) compared with obese controls on 10th postoperative day (POD #10) and weight of obese rats preoperatively (pre-op). *p ⬍ 0.05 compared with obese controls.
RESULTS Clinical results
By the 10th postoperative day, RYGB rats versus obese controls ate 57% less and had lost 17% of body weight (Fig. 2). Body weight in RYGB rats was 417 ⫾ 21g, in PF rats it was 436 ⫾ 14g, and in obese control rats it was 484 ⫾ 15g. There was no difference in body weight between RYGB and PF (p ⫽ 0.45). Ten rat days was selected as the end point for our study because it approximates the equivalent of 1 human year, so it would be expected to demonstrate marked changes in the hypothalamus. NPY immunostaining
In Figure 3A and 3B NPY immunoreactivity in the ARC was decreased by 43% in the RYGB (0.340 ⫾ 0.036) and by 55% in the PF (0.270 ⫾ 0.026) groups, compared with the obese control group (0.600 ⫾ 0.048), p ⬍ 0.05. Comparison of RYGB and PF rats was not statistically significant. Figures 4A and 4B show a photograph of the decrease of NPY immunoreactivity in the ARC after gastric bypass. In Figure 3B, in the p part of PVN, NPY immunoreactivity was decreased by 43% in RYGB (0.323 ⫾ 0.190) and by 42% in PF (0.329 ⫾ 0.024), compared with obese control rats (0.564 ⫾ 0.040), p ⬍ 0.05. There was no significant difference (p ⬎ 0.05) in NPY expression in RYGB versus PF. In the m part of PVN, NPY immunoreactivity was decreased by 61% in RYGB (0.121 ⫾ 0.009) and by 71% in PF (0.089 ⫾ 0.009) compared with obese controls (0.307 ⫾ 0.026), p ⬍ 0.05, as shown in Figure 3C. In the m part of PVN there was no difference in NPY immunoreactivity between RYGB and PF, as in p. Morphologic changes between RYGB and obese control in PVN rats are represented in Figures 5A and 5B.
Vol. 199, No. 6, December 2004
Romanova et al
Hypothalamus Changes after Gastric Bypass
891
Figure 4. (A, B) Distribution of neuropeptide Y (NPY) (C, D) and alpha-melanocyte stimulating hormone (␣-MSH) immunoreactivity fibers in arcuate nucleus (ARC) of control and Roux-en-Y gastric bypass (RYGB) rats. A considerable decrease in NPY and increase in ␣-MSH-immunoreactivity in ARC occurred after RYGB. v, third ventricle; Bar ⫽ 100 m (shows scale of magnificaiton).
␣-MSH immunostaining
Figure 3. (A) Neuropeptide Y (NPY) optical density (OD) in arcuate nucleus, (B) parvocellular part of the paraventricular nucleus (PVN), and (C) magnocellular part of PVN in Roux-en-Y gastric bypass (RYGB), pair-fed (PF), and obese control rats. The box plots show a decrease of NPY (OD) in RYGB and PF rats versus controls (p ⬍ 0.05). The horizontal line in each box plot is the mean and asterisks represent the extreme data of analyzed points.
␣-MSH-immunoreactivity in the ARC was increased by 35% in RYGB (0.152 ⫾ 0.013) and by 29% in PF (0.146 ⫾ 0.013), compared with obese controls (0.113 ⫾ 0.008), p ⬍ 0.05, as shown schematically in Figure 6A. No significant difference was measured in ␣-MSH expression in ARC between RYGB and PF rats (p ⬎ 0.05). The photographs in Figures 4C and 4D show the increase of ␣-MSH immunoreactivity in the ARC after gastric bypass. ␣-MSH immunoreactivity in the p part of the PVN was increased by 175% in RYGB (0.204 ⫾ 0.018) and by 162% in PF (0.194 ⫾ 0.012), compared with obese controls (0.074 ⫾ 0.005), p ⬍ 0.05, as shown in Figures 6B and 6C. No significant difference was found between RYGB and PF rats. In the m part of the PVN, ␣-MSH immunoreactivity was increased by 67% in RYGB (0.050 ⫾ 0.003) and by 116% in PF (0.065 ⫾ 0.008), compared with obese controls (0.030 ⫾ 0.002), p ⬍ 0.05. ␣-MSH immunoreactivity was significantly decreased by 23% in RYGB versus PF (p ⬍ 0.05). Morphologic changes in PVN comparing RYGB versus obese control rats are demonstrated in Figures 5C and 5D. 5-HT-1B-receptor immunostaining
There was no measurable staining of 5-HT-1B-receptors in the ARC. Figure 7 shows a 58% increase in 5-HT-1B-
892
Romanova et al
Hypothalamus Changes after Gastric Bypass
J Am Coll Surg
significant increase was measured in 5-HT-1B-receptorimmunoreactivity in p and m parts of PVN after RYGB (p ⬍ 0.05). Morphologic changes in PVN between RYGB versus obese control rats are demonstrated photographically in Figures 5E and 5F.
Figure 5. (A, B) Distribution of neuropeptide Y (NPY) (C, D) alphamelanocyte stimulating hormone (␣-MSH) and (E, F) serotonin (5HT)-1B-receptor immunoreactivity in the magnocellular (m) and parvocellular (p) parts of paraventricular nucleus (PVN) in control and Roux-en-Y gastric bypass (RYGB) rats. A significant decrease in NPY and increase in ␣-MSH and 5-HT-1B-receptor immunoreactivity in paraventricular nucleus occurred following RYGB. v, third ventricle, Bar ⫽ 100 m (shows scale of magnification).
receptor-immunoreactivity in the p part of RYGB (0.049 ⫾ 0.002) and a 26% increase in PF (0.039 ⫾ 0.002), compared with obese controls (0.031 ⫾ 0.001), p ⬍ 0.05. In the m part of PVN, a 39% increase in RYGB (0.370 ⫾ 0.038) and a 9% increase in PF (0.292 ⫾ 0.018) occurred, compared with obese controls (0.270 ⫾ 0.026), p ⬍ 0.05 and p ⬎ 0.05, respectively. Comparing RYGB and PF, a
DISCUSSION A subset of patients has been recognized as being morbidly obese as part of the national epidemic of obesity. These patients have metabolic comorbidities including type 2 diabetes and heart disease, which lead to premature death. To reduce their weight and comorbidities, and therefore their risk factors, some patients undergo gastric bypass operations. The RYGB operation reduces the size of the gastric reservoir, reducing caloric intake and inducing early satiety, and probably induces a subclinical form of malabsorption, with both factors contributing to weight loss. But the effect of RYGB on the brain and its contributory role in weight reduction has not been previously described and is explored using our RYGB rat model. Young Sprague Dawley rats were fed a high-energy diet to induce obesity and then subjected to RYGB or sham operation. After sham operation, rats either continued on the high-energy diet (obese control) or were PF with the mean of the daily food intake consumed by RYGB rats. Ten days later, changes in the most potent orexigenic peptide (NPY) were measured in an important food intake hypothalamic region, then ARC. A similar procedure was followed with the most potent anorexigenic peptide, ␣-MSH, and also with the 5HT-1Breceptor, a monoamine known to be involved in food intake inhibition. Because second order neurons project from the ARC to the PVN, which is involved in the regulation of efferent visceral activity, changes in these neuromodulators were also measured in the PVN. The main focus in our study was to compare RYGB in obese rats versus sham-operated obese rats, in changes of hypothalamic orexigenic peptide NPY and in the anorexigenic peptide ␣-MSH and 5-HT-11B-receptors. Comparison of RYGB and PF in many parameters did not show significance, which may be a type II error. A more likely explanation is that the concept of PF as a model in this type of study needs to be reconsidered. These rats are given one meal a day, providing only food that is equivalent to that eaten by the RYGB rats throughout the day, and the PF rats eat this meal immediately, ie, the PF rats are on a diet. These rats are frisky
Vol. 199, No. 6, December 2004
Romanova et al
Hypothalamus Changes after Gastric Bypass
893
Figure 7. Serotonin (5-HT)-1B-receptor optical density (OD) in the (A) parvocellular part of the paraventricular nucleus (PVN) and (B) the magnocellular part of PVN in Roux-en-Y gastric bypass (RYGB), pair-fed (PF), and obese control rats. The box plots show an increase of 5-HT-1B OD in RYGB versus obese controls and versus PF. The horizontal line in each box plot is the mean and asterisks represent the extreme data of analyzed points.
and they are more aggressive because they are constantly hungry, and this behavioral difference alters the hypothalamic response to food intake mediators. Although both groups versus the sham obese controls lose weight to the same degrees because of the stressful nature of the model, the neurochemical changes may not reflect parallel directional changes, giving the appearance of a type
Figure 6. Alpha-melanocyte stimulating hormone (␣-MSH) optical density (OD) in (A) arcuate nucleus, (B) the parvocellular part of paraventricular nucleus (PVN) and the magnocellular part of PVN (C) in Roux-en-Y gastric bypass (RYGB), pair-fed (PF), and obese control
rats. The box plots show an increase of ␣-MSH OD in RYGB and PF rats versus controls (p ⬍ 0.05). In the magnocellular part of PVN, a significant decrease occurred in RYGB versus PF (p ⬍ 0.05). The horizontal line in each box plot is the mean and asterisks represent the extreme data of analyzed points.
894
Romanova et al
Hypothalamus Changes after Gastric Bypass
II error. This raises questions concerning interpretation of physiologic data with the use of a PF model. In current studies we are considering eliminating this group. So based on our hypothesis, we conclude that the observed neurobiologic changes are from weight loss. The RYGB led to weight loss (because of decreased food intake) and to the neurobiologic changes; the PF model also led to weight loss but introduced a stress variable. In obese control rats, NPY is substantially elevated in the ARC and in the two foci of the PVN. These are markedly reduced after weight loss, either in RYGB or PF groups. The degree of difference in the reduction between the RYGB and PF is not striking, suggesting that caloric restriction is a major stimulus affecting NPY expression. The same decrease of NPY immunoreactivity in ARC and PVN was shown in anorectic tumorbearing rats.18 This also may support the importance of NPY-dependent stimuli on feeding behavior. In contrast, ␣-MSH is considerably lower in obese controls and increases with RYGB and PF. This finding suggests a role of this anorexigenic peptide as an eating inhibitor.5 The greater increase in ␣-MSH in the ARC correlated with changes in both parts of the PVN.19 Our data suggest that ␣-MSH, through regulation of several efferent systems in the PVN, similar to stress regulatory systems in magnocellular PVN, is involved in the mechanism of weight loss in RYGB and PF rats. Serotonin has a suppressive effect on food intake.6,9,10 Drugs that increase the activity of central 5-HT have been widely used as appetite suppressants.11 5-HT innervation is widely distributed in the hypothalamus by different types of receptors. We included 5-HT evaluation in our study because it has been previously shown that 5-HT innervates NPY neurons at the level of cell bodies in the ARC and at distal axons in the PVN,20 and dense 5-HT-1B-receptor immunoreactivity has been shown in the m part of PVN, where they are compactly localized.12 Fluorescent double labeling supports the coexistence of NPY fibers with 5-HT-1B-receptor immunopositive neurons in PVN, although a greater degree of overlap exists in the ARC.12 Recently, using an immunohistochemical dual-labeling technique, it was shown that a substantial increase in c-FOS-immunoreactivity in ␣-MSH neurons occurs in ARC.21 Our data demonstrate the relative changes observed in NPY, ␣-MSH, and 5HT-1B-receptor concentrations in neuronal tissue at the time that weight loss occurred after RYGB, relative to obese controls. Our current data
J Am Coll Surg
are valuable in providing information about the changes that occur and the magnitude and direction of these changes, permitting the design of future studies to enhance our understanding of the neurobiologic changes in the hypothalamus associated with RYGB-induced weight loss and to also determine its role in weight loss. Author Contributions Morphological study with hypothalamic analyses: Romanova Execution of procedures and/or operation: Ramos, Xu Acquisition of samples: Xu, George Drafting of manuscript: Meguid Critical revision: Inui, Das, Meguid Statistical expertise: Chen Obtaining funding: Ramos, Meguid Supervision: Meguid Supervision of animals: Quinn Animal handling: George Acknowledgment: We thank Ms Karen Hughes for her expert technical assistance.
REFERENCES 1. Hill JO, Wyatt HR, Reed GW, Peters JC. Obesity and the environment: where do we go from here? Science 2003;299:853– 856. 2. Mokdad AH, Bowman BA, Ford ES. The continuing epidemic of obesity and diabetes in the United States. JAMA 2001;286: 1195–1200. 3. Brolin RE. Bariatric surgery and long-term control of morbid obesity. JAMA 2002;288:2793–2796. 4. Pories WJ, Albrecht RJ. Etiology of type II diabetes mellitus: role of the foregut. World J Surg 2001;25:527–531. 5. Schwartz MW, Morton GJ. Keeping hunger at bay. Nature 2002;418:595–597. 6. Meguid MM, Fetissov SO, Varma M, et al. Hypothalamic dopamine and serotonin in the regulation of food intake. Nutrition 2000;16:843–857. 7. Legradi G, Emerson CH, Ahima RS, et al. Leptin prevents fasting-induced suppression of prothyrotropin-releasing hormone messenger ribonucleic acids in neurons of the hypothalamic paraventricular nucleus. Endocrinology 1997;138:2569– 2576. 8. Flier JS, Maratos-Flier E. Obesity and the hypothalamus: novel peptides for new pathway. Cell 1998;92:437–440. 9. Samanin R, Garattini S. Pharmacology of ingestive behavior. Terapie 1996;51:107–115. 10. Laviano A, Meguid MM. Serotonin and obesity. Curr Med Chem-Central Nervous System Agents 2003;3:89–100. 11. Curzon G, Gibson EL, Oluyomi AO. Appetite suppression by commonly used drugs depends on 5-HT receptors but not 5-HT availability. Trends Pharmacol Sci 1997;18:21–25.
Vol. 199, No. 6, December 2004
Romanova et al
12. Makarenko IG, Meguid MM, Ugrumov MV. Distribution of serotonin 5-hydroxytriptamine 1B (5-HT-1B) receptors in the normal rat hypothalamus. Neuroscience Lett 2002;328:155– 159. 13. Xu Y, Ohinata K, Meguid MM, et al. Gastric bypass model in the obese rat to study metabolic mechanism of weight loss. J Surg Res 2002;107:56–63. 14. Ramos EJ, Xu Y, Romanova I, et al. Is obesity an inflammatory disease? Surgery 2003;134:329–335. 15. Meguid MM, Ramos EJB, Suzuki S, et al. A surgical rat model of human Roux-en-Y gastric bypass. J Gastrointest Surg 2004;8: 621–630. 16. Paxinos G, Watson C, eds. The rat brain in stereotaxic coordinates, New York: Academic Press; 1997: [figures 25 and 26]. 17. Sternberger L, ed. Immunochemistry. New York: Wiley; 1979: 122.
Hypothalamus Changes after Gastric Bypass
895
18. Makarenko IG, Meguid MM, Gatto L, et al. Decreased NPY innervation of the hypothalamus nuclei in rats with cancer anorexia. Brain Res 2003;24:100–108. 19. Sarkar S, Lechan RM. Central administration of neuropeptide Y reduces alpha-melanocyte-stimulating hormone-induced cyclic adenosine 5=-monophosphate response element binding protein (CREB) phosphorylation in pro-thyrotropin-releasing hormone neurons and increases CREB phosphorylation in corticotropin-releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology 2003;144:281–291. 20. Guy J, Pelletier G, Bosler O. Serotonin innervation of neuropeptide Y containing neurons in the arcuate nucleus. Neurosci Lett 1988;85:9–13. 21. Heisler LK, Cowley MA, Tecott LH, et al. Activation of central melanocortin pathways by fenfluramine. Science 2002;297: 1609–1611.
ANNOUNCEMENT Beginning September 2004 JACS editorial office will test an all electronic editorial system.
JANUARY 2005 All manuscripts will be submitted on the electronic system ONLY. WATCH FOR GUIDELINES All new electronic editorial office for JACS manuscripts. JACS ONLINE CME-1 PROGRAM CME-1 COURSE EVERY MONTH 2 CREDITS AVAILABLE EACH MONTH 48 CREDITS AVAILABLE AT ALL TIMES www.jacsme.facs.org