PAPER Ghrelin Increases Food Intake in Obese As Well As Lean Subjects

Home , Appetite

International Journal of Obesity (2005) 29, 1130–1136 & 2005 Nature Publishing Group All rights reserved 0307-0565/05 $30.00 www.nature.com/ijo PAPER Ghrelin increases food intake in obese as well as lean subjects

MR Druce1, AM Wren1, AJ Park1, JE Milton1, M Patterson1, G Frost1, MA Ghatei1, C Small1 and SR Bloom1*

1Department of Metabolic Medicine, Imperial College, London, Hammersmith Hospital Campus, London, UK

OBJECTIVE: To investigate whether effects on food intake are seen in obese subjects receiving exogenous administration of ghrelin. DESIGN: Randomised, double-blind, placebo-controlled study of intravenous ghrelin at doses 1 pmol/kg/min and 5 pmol/kg/ min. SUBJECTS: In all, 12 healthy lean subjects (mean body mass index (BMI) 20.570.17 kg/m2) and 12 healthy overweight and obese subjects (mean BMI 31.971.02 kg/m2). MEASUREMENTS: Food intake, appetite and palatability of food, ghrelin and other obesity-related hormones, growth hormone. RESULTS: Low-dose infusion of ghrelin increased ad libitum energy intake at a buffet meal in the obese group only (mean increase 36.679.4%, Po0.01.) High-dose ghrelin infusion increased energy intake in both groups (mean increase 20.1710.6% in the lean and 70.1715.5% in the obese, Po0.01 in both cases.) Ghrelin infusion increased palatability of food in the obese group. CONCLUSION: Ghrelin increases food intake in obese as well as lean subjects. Obese people are sensitive to the appetite- stimulating effects of ghrelin and inhibition of circulating ghrelin may be a useful therapeutic target in the treatment of obesity. International Journal of Obesity (2005) 29, 1130–1136. doi:10.1038/sj.ijo.0803001; published online 24 May 2005

Keywords: appetite; ghrelin; food intake; growth hormone secretagogues

Introduction altered by the same volume of water.7 Chronic ghrelin Ghrelin is a hormone synthesised in the stomach1 and is the administration induces adiposity7 without attenuation of endogenous ligand for the growth hormone secretagogue the effects on food intake.6 In addition to its effects on receptor,2,3 which is expressed in hypothalamic nuclei appetite, ghrelin stimulates gastric emptying and increases including the arcuate nucleus and in brain stem nuclei.4 gastric acid secretion in rodents.9 Gastric expression and circulating levels of ghrelin are Human data also support a role for ghrelin in appetite upregulated by fasting.5–7 regulation.10 As in rodents, ghrelin is principally secreted by In rodents, ghrelin is a potent stimulus to feeding with the stomach and levels in postgastrectomy patients are maximum effects observed within an hour of peripheral around 35% of those of age-matched controls.11 Plasma administration.8 The resultant plasma levels are comparable levels rise preprandially and fall within 1 h of eating,12 to those observed after a 24-h fast,6 suggesting physiological suggesting that ghrelin may play a role in meal initiation as a relevance. Ghrelin levels fall after feeding. The mechanism hunger signal.13 We have previously shown that exogenous has not been identified, but the fall in level appears to be infusion of ghrelin at a dose of 5.0 pmol/kg/min increased influenced by gut nutrients rather than gastric distension as food intake at a buffet meal by 28% in normal weight human plasma ghrelin is suppressed by an oral glucose load but not subjects, compared to a saline control day.10 This was associated with a preprandial increase in appetite, while satiety after the meal was the same in both groups. *Correspondence: Professor SR Bloom, Department of Metabolic Medi- Fasting levels of several of the circulating hormones that cine, Imperial College, London, Hammersmith Hospital Campus, London, influence appetite regulation are different in obese people UK. compared with their lean counterparts. For example, leptin E-mail: [email protected] 14 Received 11 October 2004; revised 21 February 2005; accepted 18 April is higher in obese people compared to lean and PYY levels 2005; published online 24 May 2005 are lower in the obese.15 Some studies have observed Ghrelin in obese and lean subjects MR Druce et al 1131 significantly lower basal levels of ghrelin in obese subjects Square design. Each subject’s study days were at least 48 h compared to lean16,17 and ghrelin changes with alterations apart. in body mass index (BMI), increasing after weight loss18,19 Food intake was standardised and recorded for the 24 h and falling with weight gain.20 prior to the study. All subjects fasted from 2100 h and drank Sensitivity to appetite-regulating hormones may also differ only water from midnight on the evening prior to each between lean and obese groups. For example, while it was study. Subjects refrained from alcohol and strenuous exercise hoped that leptin could be used to promote weight loss, it for 24 h before and after each study day. proved ineffective in simple polygenic obesity, which is Subjects arrived at 0830 h on each study day. Cannulae associated with leptin resistance.14 In contrast, obese people were inserted into veins in both forearms, one for the appear to retain sensitivity to the appetite-inhibiting infusion of ghrelin or saline and the other for the collection hormone PYY.15 There is evidence that the effect of ghrelin of blood samples. After venous cannulation, the subjects on growth hormone secretion differs in lean and obese relaxed for 30 min before the start of the study protocol. individuals with a smaller response in the obese.21 Basal blood samples were taken at À30 and 0 min. All time It is not known whether obese individuals are responsive cues were removed from the study room and subjects were to the appetite-stimulating effects of ghrelin. To establish encouraged to relax by reading and watching films. Infusions this would be a step towards greater understanding of the commenced at 0900 h (t ¼ 0) and continued for a total of role of ghrelin in the aetiology of obesity. In addition, this 75 min. Further blood samples were taken at 15, 30, 45 and may provide an insight into the potential utility of 75 min of infusion, plus 30 min after the end of the infusion. modification of ghrelin as an approach to the treatment of Blood samples were collected into plastic heparin-coated obesity. tubes (LIP Ltd, UK) containing 5000 kallikrein inhibitor units (0.2 ml aprotinin) (Bayer). Plasma was separated immediately by centrifugation at 41C and stored at À201C until assayed. At 45 min, while the infusions continued, a buffet meal Research design and methods was served. All subjects had completed their meal by 75 min Subjects at which time the infusion was terminated. Visual analogue Healthy lean and obese subjects were recruited by advertis- scales of 100 mm were used for rating hunger, nausea and ing in newspapers and on the Imperial College Campus in fullness at baseline and prebreakfast, and for assessment of London. Ethical approval was obtained from the local ethics meal palatability after breakfast.26 The subjects left the study committee (project registration number 2000/5941) and the room 30 min after cessation of the infusion and completed study was performed in accordance with the principles of the food diaries until 1300 h the following day to allow Declaration of Helsinki. In all, 24 subjects were recruited, 12 assessment of subsequent food intake. The food diaries were were lean and 12 were obese, with four males and eight analysed by a dietitian unaware of study assignments and females in each group. The mean (7s.e.m.) BMI was energy intake was calculated with the aid of Dietplan 20.570.17 kg/m2 in the lean group and 31.971.02 kg/m2 (Forestfield Software Ltd, West Sussex, UK). in the obese group. All subjects were between the ages of 18 and 50 y (mean 24.771.34 for the lean group and 33.472.0 for the obese group) and all had had a stable body weight for Dose at least 3 months. The criteria for exclusion included Subjects received infusion of either ghrelin 5.0 pmol/kg/min smoking more than five cigarettes a day, substance abuse, in saline (high dose), or ghrelin 1.0 pmol/kg/min in saline pregnancy, use of medications except the oral contraceptive, (low dose), or saline alone, in random order. These doses medical or psychiatric illnesses and abnormalities detected were chosen on the basis of previous work carried out by the on physical examination, electrocardiography or blood tests Department of Metabolic Medicine at Imperial College, (full blood count, electrolytes, liver function, random London, in which an an intravenous infusion of ghrelin at glucose and thyroid function.) The subjects were screened a dose of 5.0 pmol/kg/min led to a significant increase in by a dietitian who assessed their eating behaviour with the food intake in normal weight subjects at a test meal.10 The Dutch Eating Behaviour Questionnaire22 and the Eating lower dose of 1.0 pmol/kg/min was chosen in order to Attitudes Test Questionnaire.23 They also completed food highlight any difference in sensitivity to ghrelin between diaries to assess their usual dietary habits. the lean and obese groups. Infusion time was based on previous ghrelin infusion protocols.

Protocol The protocol design was based on previously published Meal studies.10,24,25 Each subject was studied on three occasions, Subjects were instructed to eat as much as they wanted from and received three infusions – saline, low-dose ghrelin two standardised types of sandwiches, with fillings prear- (1.0 pmol/kg/min) and high-dose ghrelin (5.0 pmol/kg/ ranged with subjects to ensure a choice suitable for them, min), in a double-blinded, randomised, crossover, Latin and water to drink as required. Each subject received the

International Journal of Obesity Ghrelin in obese and lean subjects MR Druce et al 1132 same meal on each study visit, and thus macronutrient software.) Throughout, P-values less than 0.05 were con- composition of the meal was constant. Food was offered in sidered significant. sufficient excess to satisfy all appetites and subjects were informed they could take away any uneaten food with them if desired. The meal was completed at a single sitting, all subjects having finished eating within 30 min while the Results infusion continued. The amounts of food and water were Subject characteristics quantified preprandially and postprandially by a blinded Subject characteristics are summarised in Table 1. The mean 2 investigator and the caloric intake calculated. BMI for the lean group was 20.5 kg/m with a range of 19.6– 21.5 kg/m2, selected to fall at the lean end of the range of normal BMI. The mean BMI for the overweight and obese 2 2 Materials subjects was 31.9 kg/m , range 27.6–39 kg/m . The mean 7 7 Human ghrelin was obtained from Bachem (Merseyside, ages were 24.7 4.6 y in the lean group and 33.4 6.9 y in UK.). The Limulus Amoebocyte Lysate Assay test for pyrogen the obese group. Leptin levels were lower in the lean group 7 was negative and the peptide was sterile on culture. Peptide (mean 6.21 4.8 pmol/l) than in the obese group 7 was dissolved in 0.9% saline (Bayer, Haywards Heath, UK) (18.68 7.8 pmol/l). Leptin levels differed markedly between containing Haemaccel (Beacon, Kent, UK) (10% by volume) males and females. In the lean group, mean leptin was 7 7 to reduce adsorption to the syringe and tubing. 1.85 0.5 pmol/l for males and 8.39 4.4 pmol/l for females. In the obese group, mean leptin was 10.777.0 pmol/l for males and 22.6674.9 pmol/l for females. Cholesterol was 7 Assays higher in the obese group (5.59 1.1 mmol/l) than in the 7 Ghrelin-like immunoreactivity was measured with an estab- lean group (4.46 0.9 mmol/l). lished specific and sensitive radioimmunoassay.27–29 The assay crossreacts fully (100%) with both octanoyl and des octanoyl ghrelin and did not crossreact with any known Plasma ghrelin gastrointestinal or pancreatic peptide hormones. The anti- Mean basal ghrelin in the lean group was 459.6745.2 pmol/l sera (SC-10368) was obtained from Santa Cruz biotechnology and in the obese group was 440.8749.1 pmol/l. There was no 125 and used at a final dilution of 1:50 000. The I ghrelin was statistically significant difference between the lean and obese prepared with Bolton & Hunter reagent (Amersham Inter- groups or between levels in males and females. national, UK) and purified by RP-HPLC using a linear Ghrelin levels achieved are shown in Figure 1. Peak gradient from 10 to 40% acetonitrile, and 0.05% TFA over premeal ghrelin levels achieved (at 45 min infusion) were 90 min. The specific activity of ghrelin label was 48 Bq/fmol. comparable in lean and obese groups for the low dose of The assay was performed in total volume of 0.7 ml of 0.06 M ghrelin. The peak in the lean group was 725.7741.3 pmol/l phosphate buffer pH 7.2 containing 0.3% BSA and was and in the obese group 745.5740.4 pmol/l. These levels 1 incubated for 3 days at 4 C before separation of free and correspond to 1.6- and 1.7-fold increases compared to the bound antibody by charcoal absorption. The assay detected ghrelin concentration in these groups at the same point on changes of 25 pmol/l of plasma ghrelin with 95% confidence the saline infusion day. The peak ghrelin level in the lean limit, with an intra-assay coefficient of variation 5.5%. All group following high-dose ghrelin infusion was lower than samples were measured in one assay to avoid interassay in the obese group with a value of 1598.27143.5 pmol/l as variation. Results were duplicated using a commercially compared to 2118.97139.9 pmol/l. With the high-dose available kit (Phoenix Pharmaceuticals, Belmont, CA, USA.) infusion, the peak ghrelin level reached was 3.6-fold greater Insulin was measured using an established in-house radioimmunoassay.30 Plasma leptin was measured with the Linco Research kit (Missouri, USA). Growth hormone was quanti- Table 1 Characteristics of lean and obese subjects taking part in the study fied in the clinical reference laboratory by chemilumines- cence immunoassay using the Nichols Advantage machine. Lean Obese P-Value

Age (y) 24.774.6 33.476.9 0.002 BMI (kg/m2)20.570.6 31.973.5 o0.001 Statistical analysis Fasting glucose (mmol/l) 4.2570.4 4.7370.3 o0.001 Subject characteristics are expressed as means 7s.d. Plasma Cholesterol (mmol/l) 4.4670.9 5.5971.1 0.01 Chol/HDL ratio 3.270.4 4.1270.6 0.001 ghrelin hormone levels are expressed as means7s.e.m. The o Basal leptin (ng/ml) 6.2174.8 18.6877.8 0.0001 integrated area under the curve was calculated using the Fasting ghrelin (pmol/l) 459.67156.6 441.77171.0 0.78 trapezoid rule. Caloric intake is expressed as absolute kilo- Fasting insulin (pmol/l) 32.7711.7 46.2722.6 0.002 calories and as means7s.e.m., and as percentage change Figures shown are mean and s.d. from the mean. Reference ranges for compared to baseline saline day. Comparison within groups biochemical data: fasting glucose o6.2 mmol/l, fasting cholesterol 3.6– having different treatments used paired t-tests (Sigmastat 6.5 mmol/l and chol/HDL ratio 3–5.

International Journal of Obesity Ghrelin in obese and lean subjects MR Druce et al 1133 a saline low dose high dose ghrelin ghrelin ghrelin ghrelin 1 pmol/kg/min 5 pmol/kg/min 3000 200

2000 ** 180

1000

ghrelin pmol/l 160

0 ** -300 1530456075105 140 time (minutes) *

b saline low dose high dose food intake (% of baseline) food 120 ghrelin ghrelin 3000 100 2000 lean obese

Figure 2 Mean food intake in all groups as a percentage of intake on the 1000 saline infusion day. Error bars show s.e.m. *P ¼ o0.05, **Po0.01. Baseline mean intake on the saline control day was 750.67106.5 kcal in the lean group ghrelin pmol/l and 754.07136.9 kcal in the obese group. 0 -30 0 15 30 45 60 75 105 time (minutes) groups. In the obese group, the mean increase was Figure 1 Mean ghrelin levels achieved over the course of infusions. Error 380.57100.0 kcal (1593.17418.7 kJ), where P ¼ 0.003 when bars represent s.e.m. (a) Levels in the lean group and (b) Levels in the obese compared to saline. In the lean group, the mean increase was group. High-dose ghrelin is 5.0 pmol/kg/min and low dose is 1.0 pmol/kg/ 118.1746.8 kcal (4944.67195.9 kJ), where P ¼ 0.004 when min. Infusion started at 0 min, meal served at 45 min and meal and infusion ended at 75 min. compared to saline. In percentage terms, the increases in energy intake during the buffet meal following ghrelin infusion, as shown in than at the same point during saline infusion in the lean Figure 2, were as follows. In the obese group, the mean group, and five-fold greater in the obese group. increases were 36.679.4% on the low-dose day and 70.17 The areas under the curve over the 45 min infusion 15.5% on the high-dose day. In the lean group, the mean showed no statistically significant difference for the low increases were 20.2710.8% on the low-dose infusion day dose, with values of 7.870.8 and 9.170.9 nmol/l/min for and 20.1710.6% on the high-dose day. the lean and obese groups, respectively. However, following There was no effect on food intake in any group during the the high-dose infusion, the area under the curve was lower 24 h following the infusion, that is, there was no sustained for the lean than obese group with values of 31.872.2 and increase in food intake, nor was there a rebound reduction in 42.872.6 nmol/l/min, respectively (P ¼ 0.02). calories consumed following the cessation of infusion (data not shown.)

Effects of Ghrelin infusion on food intake On the saline day, mean food intake for the obese group was Effects of ghrelin infusion on appetite and food 754.07136.9 kcal (3156.87573.2 kJ.) Mean food intake for palatability the lean group was 750.67106.5 kcal (3142.67445.9 kJ.) In Infusion of ghrelin at low dose did not result in any obese subjects, low-dose ghrelin infusion significantly significant changes in scores of appetite or hunger, in either increased the energy intake on the ghrelin day compared lean or obese subjects, as measured by visual analogue scales. to the saline day: mean increase 165.8745.9 kcal Infusion of high-dose ghrelin resulted in an increase in score (694.27192.2 kJ), P ¼ 0.004. Low-dose exogenous ghrelin for the question ‘how hungry do you feel’ in both lean and infusion tended to increase energy intake in the lean group obese subjects (P ¼ 0.004 and 0.03, respectively) and a fall in compared to intake on the saline day; however, this was not score for the question ‘how full do you feel’ in both lean and statistically significant: mean increase 128.7767.1 kcal obese subjects (P ¼ 0.02 and 0.03, respectively) at 30 min (538.87280.9 kJ), P ¼ 0.08. High-dose exogenous ghrelin after injection. No subjects developed nausea during infusion increased energy intake in both the lean and obese infusion.

International Journal of Obesity Ghrelin in obese and lean subjects MR Druce et al 1134 ghrelin ghrelin 250 1 pmol/kg/min 5 pmol/kg/min lean

140 obese * 200 130 *

120 150

110 100 ** 100

90 50 peak GH increment (mIU/l)

palatability (% of saline day) 80 0 lean obese 1 pmol/kg/min 5 pmol/kg/min Figure 3 Subjects ratings of palatability of test meal as a percentage of Ghrelin dose palatability on the saline control day. Error bars represent s.e.m. *Po 0.05. Figure 4 Effect of ghrelin infusion on growth hormone secretion. Peak growth hormone increment at 45 min (compared to change in growth hormone on saline day) is expressed in mIU/l. Error bars represent s.e.m. For the low dose *P ¼ 0.06 and for the high dose **P ¼ 0.01. Visual analogue scores for the palatability of the food offered at the buffet breakfast (Figure 3) showed no significant differences between infusion days in the lean different responsiveness to the appetite-stimulating effects of group. In the obese group, palatability was increased by a ghrelin. Furthermore, there is some evidence to suggest that mean of 17.577.4% compared to saline on the low-dose the normal postprandial fall in ghrelin is attenuated in the infusion day and 21.378.7% compared to saline on the obese.31 This could result in decreased postprandial satiety, high-dose infusion day (P ¼ 0.04 in both cases) (Figure 3). further increasing food intake. The mechanism underlying the decreased ghrelin associated with obesity remains unclear. Effects of ghrelin infusion on growth hormone In this study, we found no significant difference in basal secretion ghrelin levels between the lean and obese individuals. This Peak growth hormone was achieved at 45 min after the start finding was somewhat surprising. The ghrelin measurements of the infusion of ghrelin. After low dose, the mean growth were repeated with a widely available commercial assay hormone increment from baseline to peak (compared to that (Phoenix Pharmaceuticals, Belmont, CA, USA), which con- 7 on the saline day) was 40.3 8.8 mIU/l in the lean group and firmed the initial findings with our assay. However, the lean 7 20.8 4.7 mIU/l in the obese group (P ¼ 0.06). At high dose, and obese groups clearly differed with respect to BMI, leptin, 7 the increment in the lean group was 201.2 43.7 mIU/l in insulin and cholesterol levels as expected. The observation 7 the lean group and 69.3 17.8 mIU/l in the obese group that circulating ghrelin is inversely proportional to BMI (P ¼ 0.01.) This is shown in Figure 4. comes from investigations of subjects including more individuals at the extremes of the BMI range.16 Other studies have focused on such extremes, for example, many subjects Discussion with BMI o18.5 due to anorexia nervosa in one17 and Ghrelin is a circulating hormone shown to promote feeding severely obese subjects with BMI 33–65 kg/m2 in another.32 and adiposity following administration in rodents.6,7 In The relationship between BMI and ghrelin is nonlinear.33 A addition, ghrelin stimulates appetite and food intake in significant difference may be more difficult to demonstrate humans.10 In this current investigation, we aimed to in our cohort of volunteers, who were purposely chosen with investigate whether ghrelin acts to increase appetite and a narrower BMI range (19.6–39 kg/m2) in order to more food intake in obese individuals. This finding may provide closely reflect the spectrum of body weight in healthy insight into the aetiology of obesity and would also help to normal weight and overweight individuals. Indeed, other elucidate whether ghrelin would be a suitable candidate as a published studies have also used cohorts of healthy lean and therapeutic target in the treatment of obesity. Other groups obese subjects with no difference in basal ghrelin.34 have observed a difference in fasting plasma ghrelin Although mean fasting ghrelin levels do not differ between concentrations between lean and obese humans, with higher the two groups, other classical markers of obesity are altered ghrelin levels in lean individuals.16,17 These observations as expected in our cohort. Thus basal leptin and insulin have led to the hypothesis that ghrelin acts as a counter- levels and fasting cholesterol are all higher in the obese regulatory hormone, rising in lean people to stimulate group. appetite and weight gain and falling in obesity to limit Low-dose ghrelin infusion led to similar plasma levels in energy intake. The different basal levels may be reflected in the lean and obese groups. Overall the circulating ghrelin

International Journal of Obesity Ghrelin in obese and lean subjects MR Druce et al 1135 level reached at the time of presentation of the buffet meal growth hormone secretion. It has been postulated that the was similar (1.6 and 1.7 times the level seen at the same effect could be due to the lower endogenous levels of growth point on the saline infusion day) with a comparable hormone secretagogues such as ghrelin. However, in our cumulative elevation of plasma ghrelin in each case (as study, the growth hormone response in the obese group is estimated by area under the curve.) In comparison the high blunted at both doses of ghrelin administered despite dose resulted in a significant difference in plasma concen- endogenous fasting ghrelin concentrations, which did not tration at the time of presentation of the meal and also in differ significantly from those of the lean group. This total area under the curve between the lean and obese contrasts with the maintained effect of ghrelin on food groups. Therefore, we have applied statistical calculations to intake in the obese group. At the high dose in particular, the both doses administered (reflecting the experimental de- obese group had a significantly higher mean ghrelin level sign), but for the purposes of this discussion, we will focus (1.3 times that of the lean group) yet the growth hormone mainly on the food intake results obtained following low- response was almost three times as great in the lean group. dose infusion. This will enable comparison of the effects of This corroborates the impairment in growth hormone equivalent plasma levels of ghrelin achieved in the two response to ghrelin in the obese observed by other groups.21 groups. Interestingly, the growth hormone response to ghrelin is also The low-dose infusion of ghrelin, which led to a peak blunted in patients with anorexia nervosa who have high ghrelin level of 1.6–1.7 times that observed on the saline ghrelin levels.36 In our cohort, the effect may also be control day, resulted in a mean increase of 20.2% in food enhanced by the greater age in the obese group, as age also intake in the lean group and 36.6% in the obese group. This results in a blunting of the growth hormone response to change was statistically significant for the obese group ghrelin.37 (Po0.05), and thus it is clear that obese people retain In summary, we have observed that the feeding response sensitivity to exogenous infusion of ghrelin. It may be noted to ghrelin is maintained in obesity. In contrast, the growth that after high-dose infusion, there was a mean increase in hormone response to ghrelin is attenuated. Ghrelin is a food intake of 20.1% in the lean group and 70.1% in the stimulus to food intake and its effects may contribute to the obese group. In both groups, the high dose led to a aetiology of obesity. Furthermore this highlights ghrelin as a statistically significant increase in food intake compared to potential therapeutic target in strategies to treat obesity. the saline control day (Po0.05 and o0.01, respectively). Further work with ghrelin-blocking agents is indicated to However, the differences in plasma levels achieved after high elucidate this further. dose in the lean and obese groups preclude further analysis. It is helpful to examine the food intake response following intravenous ghrelin in published work, although direct comparison would not be valid due to differences in infusion Acknowledgements duration and experimental paradigm. Wren et al10 reported We Dr M Donaldson and Mr JH Meek for growth hormone that an intravenous infusion of ghrelin of 5 pmol/kg/min assays, Professor M Allison for peptide testing, and the Sir resulted in a 28% increase in food intake. John McMichael Centre for providing the facility for studies Visual analogue scores of hunger and fullness did not show with human volunteers. MD, AW and AP are funded by statistically significant differences compared to saline in the Wellcome Trust Clinical Research Training Fellowships, MP is lean group following administration of ghrelin. However, in funded by a grant from the BBSRC. This work is supported by the obese group, low-dose ghrelin increased the palatability a programme grant from the Wellcome Trust. of food at the buffet meal by a mean of 17.5% and high dose by 21.3% (Po0.05 in both cases). It could be that enhance- ment of food palatability following ghrelin administration References resulted in the increased food intake. 1 Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K. These findings indicate that ghrelin sensitivity is retained Ghrelin is a growth-hormone-releasing acylated peptide from in obese individuals and therefore ghrelin receptor antago- stomach. Nature 1999; 402: 656–660. nists may prove to be useful appetite-reducing agents. 2 Kojima M, Hosoda H, Kangawa K. Purification and distribution of ghrelin: the natural endogenous ligand for the growth hormone Further studies of the effects of ghrelin antagonists in both secretagogue receptor. Horm Res 2001; 56 (Suppl 1): 93–97. normal weight and obese humans are now warranted to test 3 Bednarek MA, Feighner SD, Pong SS, McKee KK, Hreniuk DL, Silva this further. There is some evidence of reduced food intake in MV, Warren VA, Howard AD, Van der Ploeg LH, Heck JV. Structure–function studies on the new growth hormone-releasing rodents following administration of ghrelin antagonists,35 peptide, ghrelin: minimal sequence of ghrelin necessary for but the effects of these agents in human subjects have not activation of growth hormone secretagogue receptor 1a. JMed been investigated. Chem 2000; 43: 4370–4376. A further interesting observation in our study is the effect 4 Guan XM, Yu H, Palyha OC, McKee KK, Feighner SD, Sirinath- of ghrelin administration on the growth hormone response. singhji DJ, Smith RG, Van der Ploeg LH, Howard AD. Distribution of mRNA encoding the growth hormone secretagogue receptor in Obesity is characterised by several endocrine abnormalities brain and peripheral tissues. Brain Res Mol Brain Res 1997; 48: including reduction of both spontaneous and stimulated 23–29.

International Journal of Obesity Ghrelin in obese and lean subjects MR Druce et al 1136 5 Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Ueno N, Makino 21 Tassone F, Broglio F, Destefanis S, Rovere S, Benso A, Gottero C, S, Fujimiya M, Niijima A, Fujino MA, Kasuga M. Ghrelin is an Prodam F, Rossetto R, Gauna C, Van Der Lely AJ, Ghigo E, appetite-stimulatory signal from stomach with structural resem- Maccario M. Neuroendocrine and metabolic effects of acute blance to motilin. Gastroenterology 2001; 120: 337–345. ghrelin administration in human obesity. J Clin Endocrinol Metab 6 Wren AM, Small CJ, Abbott CR, Dhillo WS, Seal LJ, Cohen MA, 2003; 88: 5478–5483. Batterham RL, Taheri S, Stanley SA, Ghatei MA, Bloom SR. 22 Dewberry C, Ussher JM. Restraint and perception of body weight Ghrelin causes hyperphagia and obesity in rats. Diabetes 2001; 50: among British adults. J Soc Psychol 1994; 134: 609–619. 2540–2547. 23 Van Strien T, Rookus MA, Bergers GP, Frijters JE, Defares PB. Life 7 Tschop M, Smiley DL, Heiman ML. Ghrelin induces adiposity in events, emotional eating and change in body mass index. Int J rodents. Nature 2000; 407: 908–913. Obes 1986; 10: 29–35. 8 Wren AM, Small CJ, Ward HL, Murphy KG, Dakin CL, Taheri S, 24 Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Kennedy AR, Roberts GH, Morgan DG, Ghatei MA, Bloom SR. The Dakin CL, Wren AM, Brynes AE, Low MJ, Ghatei MA, Cone RD, novel hypothalamic peptide ghrelin stimulates food intake and Bloom SR. Gut hormone PYY(3–36) physiologically inhibits food growth hormone secretion. Endocrinology 2000; 141: 4325–4328. intake. Nature 2002; 418: 650–654. 9 Masuda Y, Tanaka T, Inomata N, Ohnuma N, Tanaka S, Itoh Z, 25 Batterham RL, Le Roux CW, Cohen MA, Park AJ, Ellis SM, Hosoda H, Kojima M, Kangawa K. Ghrelin stimulates gastric acid Patterson M, Frost GS, Ghatei MA, Bloom SR. Pancreatic secretion and motility in rats. Biochem Biophys Res Commun 2000; polypeptide reduces appetite and food intake in humans. J Clin 276: 905–908. Endocrinol Metab 2003; 88: 3989–3992. 10 Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, 26 Raben A, Tagliabue A, Astrup A. The reproducibility of subjective Dhillo WS, Ghatei MA, Bloom SR. Ghrelin enhances appetite and appetite scores. Br J Nutr 1995; 73: 517–530. increases food intake in humans. J Clin Endocrinol Metab 2001; 27 Tan TM, Vanderpump M, Khoo B, Patterson M, Ghatei MA, 86: 5992. Goldstone AP. Somatostatin infusion lowers plasma ghrelin 11 Ariyasu H, Takaya K, Tagami T, Ogawa Y, Hosoda K, Akamizu T, without reducing appetite in adults with Prader–Willi syndrome. Suda M, Koh T, Natsui K, Toyooka S, Shirakami G, Usui T, J Clin Endocrinol Metab 2004; 89: 4162–4165. Shimatsu A, Doi K, Hosoda H, Kojima M, Kangawa K, Nakao K. 28 Martin NM, Small CJ, Sajedi A, Patterson M, Ghatei MA, Bloom Stomach is a major source of circulating ghrelin, and feeding SR. Pre-obese and obese agouti mice are sensitive to the anorectic state determines plasma ghrelin-like immunoreactivity levels in effects of peptide YY(3–36) but resistant to ghrelin. Int J Obes Relat humans. J Clin Endocrinol Metab 2001; 86: 4753–4758. Metab Disord 2004; 28: 886–893. 12 Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse BE, 29 Le Roux CW, Patterson M, Vincent R, Hunt C, Ghatei M, Bloom S. Weigle DS. A preprandial rise in plasma ghrelin levels suggests a Postprandial plasma ghrelin is suppressed proportional to meal role in meal initiation in humans. Diabetes 2001; 50: 1714–1719. calorie content in normal weight but not obese subjects. J Clin 13 Arvat E, Maccario M, Di Vito L, Broglio F, Benso A, Gottero C, Endocrinol Metab 2004; 90: 1068–1071. Papotti M, Muccioli G, Dieguez C, Casanueva FF, Deghenghi R, 30 Edwards CM, Todd JF, Mahmoudi M, Wang Z, Wang RM, Ghatei Camanni F, Ghigo E. Endocrine activities of ghrelin, a natural MA, Bloom SR. Glucagon-like peptide 1 has a physiological role growth hormone secretagogue (GHS), in humans: comparison in the control of postprandial glucose in humans: studies with and interactions with hexarelin, a nonnatural peptidyl GHS, the antagonist exendin 9–39. Diabetes 1999; 48: 86–93. and GH-releasing hormone. J Clin Endocrinol Metab 2001; 86: 31 English PJ, Ghatei MA, Malik IA, Bloom SR, Wilding JP. Food fails 1169–1174. to suppress ghrelin levels in obese humans. J Clin Endocrinol 14 Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens Metab 2002; 87: 2984. TW, Nyce MR, Ohannesian JP, Marco CC, McKee LJ, Bauer TL. 32 Holdstock C, Engstrom BE, Ohrvall M, Lind L, Sundbom M, Serum immunoreactive-leptin concentrations in normal-weight Karlsson FA. Ghrelin and adipose tissue regulatory peptides: and obese humans. N Engl J Med 1996; 334: 292–295. effect of gastric bypass surgery in obese humans. J Clin Endocrinol 15 Batterham RL, Cohen MA, Ellis SM, Le Roux CW, Withers DJ, Metab 2003; 88: 3177–3183. Frost GS, Ghatei MA, Bloom SR. Inhibition of food intake in 33 Muccioli G, Tschop M, Papotti M, Deghenghi R, Heiman M, obese subjects by peptide YY3–36. N Engl J Med 2003; 349: Ghigo E. Neuroendocrine and peripheral activities of ghrelin: 941–948. implications in metabolism and obesity. Eur J Pharmacol 2002; 16 Tschop M, Weyer C, Tataranni PA, Devanarayan V, Ravussin E, 440: 235–254. Heiman ML. Circulating ghrelin levels are decreased in human 34 Zahorska-Markiewicz B, Mizia-Stec K, Olszanecka-Glinianowicz obesity. Diabetes 2001; 50: 707–709. M, Janowska J. Effect of weight reduction on serum ghrelin and 17 Shiiya T, Nakazato M, Mizuta M, Date Y, Mondal MS, Tanaka M, TNFalpha concentrations in obese women. Eur J Intern Med 2004; Nozoe S, Hosoda H, Kangawa K, Matsukura S. Plasma ghrelin 15: 172–175. levels in lean and obese humans and the effect of glucose on 35 Asakawa A, Inui A, Kaga T, Katsuura G, Fujimiya M, Fujino MA, ghrelin secretion. J Clin Endocrinol Metab 2002; 87: 240–244. Kasuga M. Antagonism of ghrelin receptor reduces food intake 18 Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger and body weight gain in mice. Gut 2003; 52: 947–952. EP, Purnell JQ. Plasma ghrelin levels after diet-induced weight 36 Broglio F, Gianotti L, Destefanis S, Fassino S, Abbate DG, Mondelli loss or gastric bypass surgery. N Engl J Med 2002; 346: 1623–1630. V, Lanfranco F, Gottero C, Gauna C, Hofland L, Van Der Lely AJ, 19 Hansen TK, Dall R, Hosoda H, Kojima M, Kangawa K, Christian- Ghigo E. The endocrine response to acute ghrelin administration sen JS, Jorgensen JO. Weight loss increases circulating levels of is blunted in patients with anorexia nervosa, a ghrelin hyperse- ghrelin in human obesity. Clin Endocrinol (Oxf) 2002; 56: cretory state. Clin Endocrinol (Oxf) 2004; 60: 592–599. 203–206. 37 Broglio F, Benso A, Castiglioni C, Gottero C, Prodam F, Destefanis 20 Otto B, Cuntz U, Fruehauf E, Wawarta R, Folwaczny C, Riepl RL, S, Gauna C, Van Der Lely AJ, Deghenghi R, Bo M, Arvat E, Ghigo Heiman ML, Lehnert P, Fichter M, Tschop M. Weight gain E. The endocrine response to ghrelin as a function of gender in decreases elevated plasma ghrelin concentrations of patients with humans in young and elderly subjects. J Clin Endocrinol Metab anorexia nervosa. Eur J Endocrinol 2001; 145: 669–673. 2003; 88: 1537–1542.

International Journal of Obesity