Lack of an acute effect of ghrelin on markers of bone turnover in healthy controls and post-gastrectomy subjects

406–413www.elsevier.com/locate/bone

Bone 41 (2007)

Lack of an acute effect of ghrelin on markers of bone turnover in healthycontrols and post-gastrectomy subjects☆

M.S.B. Huda a, B.H. Durham d, S.P. Wong a, T.M. Dovey c, P. McCulloch b, D. Kerrigan b,J.H. Pinkney a, W.D. Fraser d, J.P.H. Wilding a,⁎

a Clinical Sciences Centre, University of Liverpool Diabetes and Endocrinology Research Group, University Hospital Aintree,Longmoor lane, Liverpool L9 7AL, UK

b Department of Surgery, University Hospital Aintree, Liverpool L9 7AL, UKc School of Psychology, University of Liverpool, UK

d Department of Clinical Chemistry, Royal Liverpool University Hospital, Prescott Street, Liverpool, UK

Received 3 March 2007; revised 9 May 2007; accepted 16 May 2007Available online 31 May 2007

Abstract

Background: Ghrelin is a gut-brain peptide that powerfully stimulates appetite and growth hormone secretion and is also known to directlyregulate osteoblast cell function in vitro and in animal models. Little is known about the effects of ghrelin on bone turnover in humans. As thestomach is the main site of ghrelin synthesis, gastrectomy patients are deficient in ghrelin; they are also prone to osteopenia and osteomalacia.Hypothesis: Ghrelin may play a role in bone regulation in humans; ghrelin deficiency following gastrectomy is associated with the disruptedregulation of bone turnover seen in these subjects.Subjects and methods: In a randomised, double-blind, placebo-controlled study 8 healthy controls and 8 post-gastrectomy subjects were infusedwith intravenous ghrelin (5 pmol/kg/min) or saline over 240 min on different days. Subjects were given a fixed energy meal during the infusion.Ghrelin, GH, type-1 collagen β C-telopeptide (βCTX), a marker of bone resorption, and procollagen type-1 amino-terminal propeptide (P1NP), amarker of bone formation, were measured.Results: Fasting ghrelin was significantly lower in the gastrectomy group during the saline infusion (226.1±62.0 vs. 762±71.1 ng/l pb0.001).Growth hormone was significantly higher at 90 min after the ghrelin infusion, compared to saline in both healthy controls (61.1±8.8 vs. 1.4±0.6mIU/l pb0.001) and gastrectomy subjects (61.1±11.8 vs. 0.9±0.2 mIU/l pb0.001) confirming the ghrelin was bioactive. Gastrectomy subjectswere significantly older and had significantly higher plasma βCTX than healthy controls at all time points (ANOVA p=0.009). After adjustment forage and BMI ghrelin was found to be a significant predictor of baseline plasma βCTX and was inversely correlated with baseline plasma βCTX(β=−0.54 p=0.03 R2=26%). However, there was no significant effect of the ghrelin infusion on plasma βCTX or P1NP in either subject group.Conclusions: Ghrelin infusion has no acute effect on markers of bone turnover in healthy controls and post-gastrectomy subjects, but is inverselycorrelated with bone resorption.© 2007 Elsevier Inc. All rights reserved.

Keywords: Ghrelin; Bone resorption; Bone formation

Introduction

Ghrelin is a unique gut-brain peptide that stimulates appetiteand GH release [1] and circulating concentrations of ghrelinincrease progressively in the pre-prandial period and decrease to

☆ This work was supported by an International Endocrine Research Prize fromPharmacia awarded to JHP and JPHW.⁎ Corresponding author. Fax: +44 151 529 5888.E-mail address: [email protected] (J.P.H. Wilding).

8756-3282/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.bone.2007.05.006

a nadir within 1 h of eating. [2] Ghrelin was discovered as thenatural ligand for the growth hormone secretagogue receptorGHS-R1a [3,4] which has subsequently been identified inseveral different human tissues including the stomach, heart,lung, pancreas, intestine, gonads, adrenal glands, adipose tissue,bone, T cells, pituitary and hypothalamus [5].

In addition to effects on appetite and GH secretion, ghrelinhas been found in a recent study to influence bone turnover [6].GHS-R1a receptors were found in rat osteoblast cells and invitro, ghrelin stimulated osteoblast proliferation—this effect

Table 1Details of surgery in gastrectomy subjects

Subject Surgery Time period since operation(years)

1 Total gastrectomy 102 Total gastrectomy 133 Oesophago-gastrectomy a 94 Radical oesophago-gastrectomy b 75 Radical oesophago-gastrectomy 66 Total gastrectomy 47 Total gastrectomy 18 Oesophago-gastrectomy 3

a Lower portion of oesophagus removed.b Associated abdominal lymph node clearance.

407M.S.B. Huda et al. / Bone 41 (2007) 406–413

was blocked by a GHS-R1a antagonist, suggesting that ghrelinacts via this receptor [6]. Ghrelin given intraperitoneally over 4weeks also increased bone mineral density (BMD) in ratscompared to controls [6]. Other studies have confirmed the invitro effects of ghrelin to stimulate rodent and human osteoblastproliferation [7,8].

Two-thirds of circulating plasma ghrelin is produced fromthe stomach and hence post-gastrectomy, ghrelin levelsdecrease by 65% [9,10]. Post-gastrectomy osteopenia in rodentswas prevented by preserving 10–30% of the oxyntic gland areain the stomach. This area of the stomach is rich in ghrelin-producing cells, as well as other endocrine cells producinghistamine and pancreatastatin [11]. However, daily subcuta-neous ghrelin injections for 8 weeks did not reverse BMD lossin gastrectomised rodents [12]. Gastrectomy in humans leads toa loss in body weight of around 10% within the first 6 months ofsurgery, mainly of body fat [13]. Impaired bone quality withboth osteopenia and osteomalacia is recognised to occur inpatients following gastrectomy [14,15].

Bone tissue is continuously remodelled and to maintain bonemass, and osteoclastic bone resorption and osteoblastic boneformation are coupled [16,17]. Bone resorption shows a wellcharacterised circadian rhythm with a peak in the night/earlymorning and a nadir in the late afternoon [18,19]. Foodconsumption may influence this and fasting seems to diminishthe circadian rhythm of bone resorption [20]. Fasting also altersthe circadian rhythm of parathyroid hormone [21].

Recent studies have suggested that feeding influences boneturnover. Bone turnover appears to be suppressed after feedingand, specifically, bone resorption is suppressed by 50% after aglucose load [22,23]. Insulin and other gut hormones such asglucose-dependant insulinotropic polypeptide (GIP) and gluca-gon-like-pepide-1 (GLP-1) have not been found to influence thepost-prandial suppression of bone resorption but GLP-2 doesappear to have an effect [24,25]. At present, the mechanismbehind the alteration in bone turnover markers after feeding isnot well understood. It is therefore possible that ghrelin may playa role in the post-prandial regulation of bone turnover as ghrelinincreases acutely before and decreases sharply after meals.

Currently there are little data regarding ghrelin and markersof bone turnover in humans. The important clinical question ofwhether replacement of ghrelin after gastrectomy affects boneturnover has not been answered.

The aim of this study was to determine whether ghrelin playsan important role in bone regulation in humans. To test thishypothesis we studied the effects of a continuous ghrelininfusion on healthy subjects and post-gastrectomy subjectsaround a fixed energy meal.

Subjects and methods

Healthy volunteers and gastrectomy subjects were recruited by advertise-ment and from a surgical database. The study was approved by the South SeftonResearch Ethics Committee (project registration number EC.05.03) and wasperformed in accordance with the principles of the Declaration of Helsinki.Volunteers gave written informed consent and had a normal physicalexamination and electrocardiogram. Urea and electrolytes were normal in allsubjects.

Subjects

We studied 8 healthy volunteers aged 31–62 years (mean±standard error ofthe mean (SEM), 41.2±4.2), body mass index (BMI) 18.0–27.2 (22.9±0.9) kg/m2 and 8 post-gastrectomy volunteers aged 39–75 years (60.1±4.5), BMI 17.0–26.0 (22.4±1.0) kg/m2. Formal power calculations were not undertaken as thereare little existing data on the effects of ghrelin on bone markers in humans.Subject numbers were hence based on recruitment and previous published workshowing that a minimum of seven subjects were needed to demonstrate theeffects of ghrelin on appetite and energy intake [26], and a minimum of eightsubjects to show the effects of gut peptides on bone markers [25].

All gastrectomy subjects underwent total gastrectomy for gastric carcinomaat least 12 months previously and were in remission (none had any evidence ofrecurrence by gastroscopy within the last 12 months). All subjects had a truncalvagotomy as part of the procedure. The time between surgery and the study was6.7±1.4 years. Details of surgery are in Table 1. All subjects were mobile andphysically active with no known recent bony fractures or trauma. Gastrectomysubjects with significant bile reflux (assessed by clinical history andgastroscopy) and symptoms of dumping syndrome were excluded. Subjects inboth groups with ischaemic heart disease and aged over 75 years were excluded.

Protocol

Subjects were asked to fast and drink only water from 21:00 on the nightprior to the study and to refrain from alcohol or strenuous exercise in the 24 hpreceding the study day. Each subject was studied on two occasions (at least aweek apart) and received two infusions −0.9% saline and ghrelin (5 pmol/kg/min), in a randomised, double-blind, crossover design. The dose of 5 pmol/kg/min was chosen as previous studies have shown that an intravenous infusion ofghrelin at this dose leads to a significant increase in food intake in lean andmoderately obese subjects [27,28]. Human ghrelin was supplied in the acylated(octanoylated) form and produced by Clinalfa Products, Merck Biosciences AG,Weidenmattweg 4, CH-4448 Laufelfingen, Switzerland.

Subjects attended at 07:30 on each study day. After arrival, they wereblinded to external time cues until the end of the study. Two intravenouscannulae were inserted into a vein in both forearms. One was used for bloodsampling, and the other was used for the ghrelin or saline infusion. After venouscannulation the subjects relaxed for 30 min before the start of the study protocol.The infusion was started at 08:30 (t=0) and lasted for 240 min. Previous studieshave shown that plasma ghrelin levels during an infusion reach a steady statewithin 60 min [28]. A fixed energy breakfast was therefore given to all subjectsat 60 min following the start of the infusion to allow the effects of feeding to beassessed whilst plasma ghrelin was at a steady state. This consisted of 40 gcornflakes, 250 ml whole milk, 3.5 g sugar (1 sachet) and 100 ml fresh orangejuice (1550 kJ, 62% carbohydrate, 13% protein and 25% fat). This was a typicalEuropean style breakfast that most subjects would find palatable. Subjectsconsumed the test breakfast within 15 min. Blood samples were taken atbaseline, 60, 90, 120 min and then hourly until the end of the study. Serum GHwas measured at baseline and at 90 min. The study was completed at 12.30.Samples were collected into plastic EDTA tubes containing 0.07 mg aprotonin(500 Kallikrein Inactivator Units (KIU)) and centrifuged immediately. Serum

408 M.S.B. Huda et al. / Bone 41 (2007) 406–413

samples were collected in serum separator tubes (SST) and allowed to stand for15 min at −4 °C before centrifugation. All samples were then stored at −80 °Cuntil assayed.

Assays

Plasma total (acylated and deacylated) ghrelin was determined in duplicateby a commercially available radioimmunoassay (Phoenix Pharmaceuticals,Belmont, CA). Intra-assay and inter-assay coefficients of variation were lessthan 10% across the working range of the assay.

Plasma concentration of type-1 collagen β C-telopeptide (βCTX), a markerof bone resorption, and procollagen type-1 amino-terminal propeptide (P1NP), amarker of bone formation, were measured using electrochemiluminescenceassays (ECLIA) (Roche Diagnostics, Lewes, UK). The intra-assay CVand inter-assay CV for βCTX were less than 4% and 5% respectively across the workingrange, with a detection limit of 0.01 μg/l. The intra-assay CVand inter-assay CVfor P1NP were less than 2% and 2.5% respectively, across the working range,with a detection limit of 4 μg/l. Growth hormone (GH) was measured using asolid phase, two site, chemiluminescent enzyme-labelled immunometric assayrun on an Immulite 2000 automated analyser (Diagnostic Products Corporation-UK, Llanberis, Gwyynedd UK). The sensitivity was 0.03 IU/l with a within runprecision 3.4% CV at 12.7 IU/l.

Statistical analysis

The Statistical Package for the Social Sciences version 14.0 (SPSS,Chicago, IL, USA) was used for data analysis. Values are expressed asgeometric means±SEM. Comparison between groups was performed usingpaired t-tests or the equivalent non-parametric tests (Mann–Whitney orWilcoxon Signed Ranks). Within-group and group interactions with timewere analysed using ANOVA for repeated measures. Where violations inparametric assumptions were found within the data set, the within group andgroup interactions were measured using a multivariate ANOVA (MANOVA).Total area under the curve (AUC) was calculated using the trapezoid rule.Statistical association was calculated using Pearson (r) correlation coefficients.Multiple regression analysis was performed using stepwise regression. Twotailed p values were considered significant when pb0.05.

Results

Baseline characteristics

Baseline characteristics are summarised in Table 2. Therewas no significant difference in gender, BMI and weightbetween the two groups. The gastrectomy group was howeversignificantly older (60.1±4.5 vs. 41.2±4.2 pb0.05).

Table 2Baseline characteristics of lean, obese and gastrectomy subjects a

Lean Gastrectomy

n 8 8Sex (male) 50% 38%Age (years) 41.2±4.2 60.1±4.5 ⁎

BMI (kg/m2) 22.3±0.9 22.4±1.0Weight (kg) 63.6±3.9 62.9±3.0Fasting ghrelin (ng/l) 762±71.1 226.1±62.0 ⁎

Mean steady state ghrelin (ng/l) 5526.8±65.2 5559.6±209.3Mean ghrelin AUC ⁎⁎ (ng/ml/min) 1458.9±180.2 1496.5±143.3a Plus–minus values are means±SEM.⁎ pb0.05 when compared to lean values.⁎⁎ AUC area under the curve.

Plasma ghrelin

Fasting ghrelin was significantly lower in the gastrectomygroup during the saline infusion (226.1±62.0 vs. 762±71.1 ng/lpb0.001). Plasma ghrelin levels during the ghrelin infusionreached steady state after 60 min (Fig. 1) and there was nosignificant difference in the mean steady state plasma ghrelin orthe area under the curve (AUC) between the two groups (Table 1).

Growth hormone

Serum GH at baseline was similar in controls and gastrec-tomy subjects (5.6±1.8 vs. 5.2±2.4 mIU/l p=0.7). GH wassignificantly higher at 90 min after the ghrelin compared to thesaline infusion in healthy controls (61.1±8.8 vs. 1.4±0.6 mIU/lpb0.001) and in gastrectomy subjects (61.1±11.8 vs. 0.9±0.2mIU/l pb0.001) confirming that the ghrelin infusion signifi-cantly increased GH in both groups. There was no differencebetween the groups however, in percentage increase in GHbetween 0 and 90 min during the ghrelin infusion (controlsincreased by 1113% and gastrectomy increased by 1170%p=NS).

Bone turnover markers

βCTXBaseline plasma βCTX was similar on both infusion days in

healthy controls (0.30±0.1 μg/l vs. 0.33±0.1 μg/l p=0.4) andgastrectomy subjects (0.55±0.1 μg/l vs. 0.59±0.1 μg/lp=0.09). Plasma βCTX decreased significantly with timeafter the test meal in both groups (ANOVA F(5,75)=28.7pb0.001) (Figs. 2A and B). There was no significant effect ofghrelin infusion on plasma βCTX (ANOVA p=NS) in eithergroup. Gastrectomy subjects had significantly higher plasmaβCTX than healthy controls (ANOVA F(1,32)=7.7 p=0.009)at time points 8.30 (p=0.02), 9.30 (pb0.01), 10.00 (p=0.01),10.30 (p=0.04), 11.30 (p=0.05) and 12.30 (p=0.01) (Fig. 3).There was a highly significant inverse correlation betweenbaseline plasma ghrelin and baseline plasma βCTX levels (r=

Fig. 1. Plasma ghrelin levels during the ghrelin infusion (5 mol/kg/min) inhealthy controls and gastrectomy subjects.

Fig. 3. Plasma βCTX in healthy controls and gastrectomy subjects during salineand ghrelin infusions.

Fig. 2. Percentage change of baseline in plasma βCTX in (A) healthy controlsand (B) in gastrectomy subjects.

409M.S.B. Huda et al. / Bone 41 (2007) 406–413

−0.5 p=0.006) (Fig. 4A). It can be seen in Fig. 4A that thereappears to be two distinct populations which represent thehealthy control group and the gastrectomy group. We also foundcorrelations between age and βCTX (r=0.50 p=0.05) (Fig. 4B)and between age and ghrelin (r=−0.52 p=0.04) (Fig. 4C).When baseline plasma ghrelin, age and BMI were entered into astepwise multiple regression model, only ghrelin was found tobe a significant predictor of baseline plasma βCTX and wasinversely correlated with baseline plasma βCTX (β=−0.54p=0.03 R2 =26%).

P1NPPercentage change in baseline plasma P1NP during the

saline and ghrelin infusions in healthy controls and gastrectomysubjects is shown in Figs. 5A and B respectively. There was nosignificant difference in baseline P1NP between the ghrelin andsaline infusion days in either group (p=NS). There was asignificant interaction with time (MANOVA F(5,10)=5.23p=0.01) and between the groups over time (MANOVAF(5,10)=6.8 p=0.005). Post-hoc non-parametric tests showeda significant change from baseline in the gastrectomy group attime points 10:30 (p=0.001) and 11:30 (p=0.004). There was

no significant difference in P1NP between controls andgastrectomy subjects. Similarly, there was no significant effectof ghrelin on plasma P1NP in either group and also nosignificant correlation between plasma ghrelin and plasmaP1NP.

Discussion

The mobilisation of energy stores and minerals is stimulatedby bone resorption, and conversely bone resorption candecrease as nutrients are ingested. Bone resorption undergoessignificant circadian variation [29] and appears to be particu-larly responsive to food intake, although mechanisms are stillunclear. In healthy individuals, bone resorption estimated byplasma βCTX is known to decrease by 50% after ingestion of75 g of glucose [30]. Our study confirms these findings with asignificant decrease in βCTX in both healthy controls andgastrectomy subjects after a fixed energy meal.

We have shown that in healthy controls a continuous ghrelininfusion has no significant effect on bone resorption. Ghrelintends to be suppressed after a meal, and it therefore seemsunlikely that ghrelin is involved in the short-term physiologicalcontrol of bone resorption around feeding. An alternativehypothesis would be that satiety hormones released postprandially may suppress βCTX. Indeed, Henriksen andcolleagues found that plasma βCTX was decreased by highdoses of GLP-2, confirming that plasma βCTX can beinfluenced acutely, but glucose-dependant insulinotropic poly-peptide (GIP) and glucagon-like peptide-1 (GLP-1) had noeffect [25]. It is also possible that insulin may exert an effect asplasma βCTX is suppressed after an insulin tolerance test [31].The effects of other satiety hormones such as peptide YY (PYY)on bone resorption have not been studied. Our study used aninfusion that achieved a significantly supra-physiologicalplasma concentration of ghrelin, and hence our study designshould have ensured that an acute effect of ghrelin, if present,

Fig. 5. Percentage change of baseline in plasma P1NP in (A) healthy controls,(B) gastrectomy subjects during the saline and ghrelin infusions. ⁎ indicates asignificant difference from baseline (pb0.05).

Fig. 4. Graphs showing relationships between (A) baseline plasma ghrelin andbaseline plasma βCTX, (B) baseline plasma βCTX and age and (C) baselineplasma ghrelin and age.

410 M.S.B. Huda et al. / Bone 41 (2007) 406–413

would have been observed. It is, of course, possible that βCTXchanges may be seen with more chronic administration ofghrelin or with a higher dose of ghrelin infusion, and our studydoes not exclude this.

The Rancho Bernado study looked at the association ofghrelin and bone metabolism in a large older population andfound that ghrelin was inversely associated with the boneresorption marker urine N-telopeptide (NTX) [32]. In keepingwith their findings we also found a strong inverse correlationbetween baseline ghrelin and plasma βCTX. However, our

scatter graph shows two distinct groups which represent our twosubject groups. The gastrectomy group was significantly older,and we did find an expected weak correlation between age andbaseline βCTX, as well as an inverse relationship between ageand ghrelin. The relationship between aging and ghrelin has notbeen fully established, and human and rodent studies haveshown contradictory results [33–36]. Our multiple regressionmodel does suggest that despite age being a confounder,baseline ghrelin was the strongest predictor of baseline βCTX.We would accept that our study was underpowered to performrobust multiple regression analyses as this was not the primaryaim of the study, but it is of interest that our findings confirmthose of the much larger Rancho Bernado population study [32].Our study does not however support an acute causal relationshipbetween change in ghrelin and bone resorption.

Ghrelin did not affect bone resorption in gastrectomysubjects. Subjects with previous gastrectomy are known tohave a significant reduction in plasma ghrelin [9,10], and rodentstudies indicate that preserving the ghrelin-producing region ofthe stomach prevents post-gastrectomy osteopenia [11]. Ourstudy did not show any acute effect of ghrelin on boneresorption, but again a long-term effect after chronic ghrelinadministration cannot be excluded. However, of particular note,gastrectomised rodents did not have an increase in BMD afterghrelin administration [12].

411M.S.B. Huda et al. / Bone 41 (2007) 406–413

Gastrectomy subjects have significantly higher plasmaβCTX than control subjects, indicating a higher rate of boneresorption. This in part will be due to age but also confirms arecent study which reported increased urinary markers of boneresorption post-gastrectomy [37]. Gastrectomy results in a highprevalence of bone disorders including osteopenia, osteomala-cia and increased vertebral fractures [38]. The higher rates ofbone resorption observed may lead to an increase in bonedisorders after gastrectomy.

In vitro experiments have demonstrated that ghrelinstimulates the proliferation and differentiation of osteoblastcell lines [6,7]. It is well established that ghrelin hasproliferative effects on other cell lines such as prostate,hepatoma and adrenal, as well as anti-proliferative effects onthyroid and breast cell lines [39–43]. Fukushima and colleaguesshowed that ghrelin also promotes bone formation in vivo byadministering peripheral ghrelin to rodents over a 4 week periodand demonstrating increased BMD [6].

Our study indicates that in humans in the short term, ghrelindoes not alter plasma P1NP. Two studies have not found anyassociation between ghrelin and BMD whereas another foundthat ghrelin predicts BMD in healthy adolescent girls [32,44,45].Prospective long-term studies would be needed to determinewhether chronic ghrelin administration affects bone formation inhumans.

In contrast to bone resorption, markers of bone formationstudied to date, such as the bone matrix protein osteocalcin,are not influenced by feeding or fasting [46]. Our studyshowed a statistical interaction between P1NP and time in bothgroups, and a significant decrease in P1NP 60 min after thetest meal in gastrectomy subjects. The significance of thisfinding in gastrectomy subjects is unclear. One rodent studyfound that serum osteocalcin and urinary hydroxyl prolinewere increased in the presence of normal total ALP ingastrectomised rats [47] but other studies have found contra-dictory results [48,49] and the mechanism for changes in bonemetabolism after gastrectomy remains uncertain. It is possiblethat these changes from baseline occurred independently of themeal and further 24 h profiles of P1NP in gastrectomy subjectsmay be useful.

We have shown that the synthetic ghrelin used in ourexperiment was biologically active by demonstrating asignificant increase in GH in both groups. This also confirmsprevious work that suggests that, despite low ghrelin levels, theghrelin–GH axis is intact in gastrectomy subjects [50]. We didnot measure insulin-like growth factor-1 (IGF-1) levels duringthe infusion as previous published work has shown no effect ofghrelin infusion (at higher doses) on IGF-1 [51]. GH and IGF-1play an important role in bone formation during adolescenceand early adult life and are important in maintaining bone mass.Growth hormone deficient patients have an increased incidenceof osteoporosis and long-term GH replacement can improveBMD in these patients [52]. It is interesting to note that, despitethe gastrectomy subjects being significantly older than thecontrols, their fasting GH levels were similar; age is normallyassociated with a decrease in GH levels [53]. These findings arein contrast with a study by Popovic and colleagues who found

that baseline IGF-1 and GH levels were lower in gastrectomysubjects compared to age-matched controls [54]. We did notmeasure IGF-1 levels in our two subject groups which wouldhave confirmed this, but nevertheless the reason for thisdiscrepancy is not clear.

GH secretagogues (GHS) are also known to affect boneturnover markers and increase bone mineral content in rodents[55]. However, unlike GH, administration of GHS does notincrease bone resorption, suggesting that mechanisms linked toGH release do not account for the effects of these compounds[56]. Fukushima and colleagues also showed that peripheralghrelin given to genetically GH-deficient spontaneous dwarfrats (these rodents lack GH and thus the GH-IGF-1 axis)increases BMD, suggesting that ghrelin can act on bone,independent of GH [6,57].

The main limitation of our study was that in common withother groups who have used a synthetic ghrelin infusion[27,28,58], we did not measure acylated ghrelin which mayhave added valuable information. However, the levels of totalghrelin achieved by the infusion were sufficiently high enoughto pick up any subtle effect on bone markers. Our gastrectomysubjects were also older which reflects the nature of thisparticular surgical population. Studies have shown that age iscorrelated with serum βCTX and urinary βCTX [59,60], andthis is clearly a confounder. The difference in age will make acontribution to the higher rates of bone resorption in thegastrectomy subjects observed in our study.

In conclusion, we have shown that exogenous ghrelin has noacute effects on bone resorption or bone formation in healthycontrols and post-gastrectomy subjects, but there was asignificant inverse relationship between baseline ghrelin andβCTX as a marker of bone resorption.

Acknowledgments

The authors would like to thank Barbara Martin, JanetCashen and Jackie Cleator for all of their hard work on thestudy, as well as all of the volunteers who took part. We wouldalso like to thank Pharmacia for sponsorship.

References

[1] Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K.Ghrelin is a growth-hormone-releasing acylated peptide from stomach.Nature 1999;402(6762):656–60.

[2] Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse BE, WeigleDS. A preprandial rise in plasma ghrelin levels suggests a role in mealinitiation in humans. Diabetes 2001;50(8):1714–9.

[3] Howard AD, Feighner SD, Cully DF, Arena JP, Liberator PA, RosenblumCI, et al. A receptor in pituitary and hypothalamus that functions in growthhormone release. Science 1996;273(5277):974–7.

[4] Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K.Ghrelin is a growth-hormone-releasing acylated peptide from stomach.Nature 1999;402(6762):656–60.

[5] Gnanapavan S, Kola B, Bustin SA, Morris DG, McGee P, Fairclough P, etal. The tissue distribution of the mRNA of ghrelin and subtypes of itsreceptor, GHS-R, in humans. J Clin Endocrinol Metab 2002;87(6):2988.

[6] Fukushima N, Hanada R, Teranishi H, Fukue Y, Tachibana T, Ishikawa H,et al. Ghrelin directly regulates bone formation. J Bone Miner Res 2005;20(5):790–8.

412 M.S.B. Huda et al. / Bone 41 (2007) 406–413

[7] Kim SW, Her SJ, Park SJ, Kim D, Park KS, Lee HK, et al. Ghrelinstimulates proliferation and differentiation and inhibits apoptosis inosteoblastic MC3T3-E1 cells. Bone 2005;37(3):359–69.

[8] Maccarinelli G, Sibilia V, Torsello A, Raimondo F, Pitto M, Giustina A,et al. Ghrelin regulates proliferation and differentiation of osteoblasticcells. J Endocrinol 2005;184(1):249–56.

[9] Ariyasu H, Takaya K, Tagami T, Ogawa Y, Hosoda K, Akamizu T, et al.Stomach is a major source of circulating ghrelin, and feeding statedetermines plasma ghrelin-like immunoreactivity levels in humans. J ClinEndocrinol Metab 2001;86(10):4753–8.

[10] Jeon TY, Lee S, Kim HH, Kim YJ, Son HC, Kim DH, et al. Changes inplasma ghrelin concentration immediately after gastrectomy in patientswith early gastric cancer. J Clin Endocrinol Metab 2004;89(11):5392–6.

[11] Lehto-Axtelius D, Chen D, Surve VV, Hakanson R. Post-gastrectomyosteopenia in the rat: bone structure is preserved by retaining 10%–30% ofthe oxyntic gland area. Scand J Gastroenterol 2002;37(4):437–43.

[12] Dornonville de la CC, Lindqvist A, Egecioglu E, Tung YC, Surve V,Ohlsson C, et al. Ghrelin treatment reverses the reduction in weight gainand body fat in gastrectomised mice. Gut 2005;54(7):907–13.

[13] Liedman B, Andersson H, Bosaeus I, Hugosson I, Lundell L. Changes inbody composition after gastrectomy: results of a controlled, prospectiveclinical trial. World J Surg 1997;21(4):416–20.

[14] Liedman B. Symptoms after total gastrectomy on food intake, bodycomposition, bonemetabolism, and quality of life in gastric cancer patients—Is reconstruction with a reservoir worthwhile? Nutrition 1999;15(9):677–82.

[15] Tovey FI, Hall ML, Ell PJ, Hobsley M. A review of postgastrectomy bonedisease. J Gastroenterol Hepatol 1992;7(6):639–45.

[16] Manolagas SC. Birth and death of bone cells: basic regulatory mechanismsand implications for the pathogenesis and treatment of osteoporosis.Endocr Rev 2000;21(2):115–37.

[17] Parfitt AM. Osteonal and hemi-osteonal remodeling: the spatial andtemporal framework for signal traffic in adult human bone. J Cell Biochem1994;55(3):273–86.

[18] Logue FC, Fraser WD, O'Reilly DS, Beastall GH. The circadian rhythm ofintact parathyroid hormone (1–84) and nephrogenous cyclic adenosinemonophosphate in normal men. J Endocrinol 1989;121(1):R1–3.

[19] Schlemmer A, Hassager C, Jensen SB, Christiansen C. Marked diurnalvariation in urinary excretion of pyridinium cross-links in premenopausalwomen. J Clin Endocrinol Metab 1992;74(3):476–80.

[20] Schlemmer A, Hassager C. Acute fasting diminishes the circadian rhythmof biochemical markers of bone resorption. Eur J Endocrinol 1999;140(4):332–7.

[21] Fraser WD, Logue FC, Christie JP, Cameron DA, O'Reilly DS, BeastallGH. Alteration of the circadian rhythm of intact parathyroid hormonefollowing a 96-hour fast. Clin Endocrinol (Oxf) 1994;40(4):523–8.

[22] Bjarnason NH, Henriksen EE, Alexandersen P, Christgau S, HenriksenDB, Christiansen C. Mechanism of circadian variation in bone resorption.Bone 2002;30(1):307–13.

[23] Clowes JA, Hannon RA, Yap TS, Hoyle NR, Blumsohn A, Eastell R.Effect of feeding on bone turnover markers and its impact on biologicalvariability of measurements. Bone 2002;30(6):886–90.

[24] Clowes JA, Robinson RT, Heller SR, Eastell R, Blumsohn A. Acutechanges of bone turnover and PTH induced by insulin and glucose:euglycemic and hypoglycemic hyperinsulinemic clamp studies. J ClinEndocrinol Metab 2002;87(7):3324–9.

[25] Henriksen DB, Alexandersen P, Bjarnason NH, Vilsboll T, Hartmann B,Henriksen EE, et al. Role of gastrointestinal hormones in postprandialreduction of bone resorption. J Bone Miner Res 2003;18(12):2180–9.

[26] Neary NM, Small CJ, Wren AM, Lee JL, Druce MR, Palmieri C, et al.Ghrelin increases energy intake in cancer patients with impaired appetite:acute, randomized, placebo-controlled trial. J Clin Endocrinol Metab2004;89(6):2832–6.

[27] Druce MR, Wren AM, Park AJ, Milton JE, Patterson M, Frost G, et al.Ghrelin increases food intake in obese as well as lean subjects. Int J Obes(Lond) 2005;29(9):1130–6.

[28] Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, et al.Ghrelin enhances appetite and increases food intake in humans. J ClinEndocrinol Metab 2001;86(12):5992.

[29] Wichers M, Schmidt E, Bidlingmaier F, Klingmuller D. Diurnal rhythm ofCrossLaps in human serum. Clin Chem 1999;45(10):1858–60.

[30] Bjarnason NH, Henriksen EE, Alexandersen P, Christgau S, HenriksenDB, Christiansen C. Mechanism of circadian variation in bone resorption.Bone 2002;30(1):307–13.

[31] Bjarnason NH, Henriksen EE, Alexandersen P, Christgau S, HenriksenDB, Christiansen C. Mechanism of circadian variation in bone resorption.Bone 2002;30(1):307–13.

[32] Weiss LA, Langenberg C, Barrett-Connor E. Ghrelin and bone: is there anassociation in older adults?: the Rancho Bernardo study. J Bone Miner Res2006;21(5):752–7.

[33] Di FV, Zamboni M, Zoico E, Mazzali G, Dioli A, Omizzolo F, et al.Unbalanced serum leptin and ghrelin dynamics prolong postprandialsatiety and inhibit hunger in healthy elderly: another reason for the“anorexia of aging”. Am J Clin Nutr 2006;83(5):1149–52.

[34] Rigamonti AE, Pincelli AI, Corra B, Viarengo R, Bonomo SM, GalimbertiD, et al. Plasma ghrelin concentrations in elderly subjects: comparison withanorexic and obese patients. J Endocrinol 2002;175(1):R1–5.

[35] Sturm K, MacIntosh CG, Parker BA, Wishart J, Horowitz M, ChapmanIM. Appetite, food intake, and plasma concentrations of cholecystokinin,ghrelin, and other gastrointestinal hormones in undernourished olderwomen and well-nourished young and older women. J Clin EndocrinolMetab 2003;88(8):3747–55.

[36] Sun Y, Garcia JM, Smith RG. Ghrelin and growth hormone secretagoguereceptor (GHS-R) expression in mice during aging. Endocrinology2007;148(3):1323–9.

[37] Kwon SJ, Hahm JS, Cho YJ, Ahn Y, Shin DI. The influence of gastrectomyon the change of bone metabolism and bone density. Korean J Intern Med2000;15(1):25–31.

[38] Zittel TT, Zeeb B, Maier GW, Kaiser GW, Zwirner M, Liebich H, et al.High prevalence of bone disorders after gastrectomy. Am J Surg 1997;174(4):431–8.

[39] Cassoni P, Papotti M, Catapano F, Ghe C, Deghenghi R, Ghigo E, et al.Specific binding sites for synthetic growth hormone secretagogues in non-tumoral and neoplastic human thyroid tissue. J Endocrinol 2000;165(1):139–46.

[40] Cassoni P, Papotti M, Ghe C, Catapano F, Sapino A, Graziani A, et al.Identification, characterization, and biological activity of specific receptorsfor natural (ghrelin) and synthetic growth hormone secretagogues andanalogs in human breast carcinomas and cell lines. J Clin EndocrinolMetab 2001;86(4):1738–45.

[41] Jeffery PL, Herington AC, Chopin LK. Expression and action of thegrowth hormone releasing peptide ghrelin and its receptor in prostatecancer cell lines. J Endocrinol 2002;172(3):R7–R11.

[42] Murata M, Okimura Y, Iida K, Matsumoto M, Sowa H, Kaji H, et al.Ghrelin modulates the downstream molecules of insulin signaling inhepatoma cells. J Biol Chem 2002;277(7):5667–74.

[43] Volante M, Allia E, Gugliotta P, Funaro A, Broglio F, Deghenghi R, et al.Expression of ghrelin and of the GH secretagogue receptor by pancreaticislet cells and related endocrine tumors. J Clin Endocrinol Metab 2002;87(3):1300–8.

[44] Misra M, Miller KK, Stewart V, Hunter E, Kuo K, Herzog DB, et al.Ghrelin and bone metabolism in adolescent girls with anorexia nervosa andhealthy adolescents. J Clin Endocrinol Metab 2005;90(9):5082–7.

[45] Oh KW, Lee WY, Rhee EJ, Baek KH, Yoon KH, Kang MI, et al. Therelationship between serum resistin, leptin, adiponectin, ghrelin levels andbone mineral density in middle-aged men. Clin Endocrinol (Oxf) 2005;63(2):131–8.

[46] Christgau S, Bitsch-Jensen O, Hanover BN, Gamwell HE, Qvist P,Alexandersen P, et al. Serum CrossLaps for monitoring the response inindividuals undergoing antiresorptive therapy. Bone 2000;26(5):505–11.

[47] Fries W, Rumenapf G, Schwille PO. Disturbances of mineral and bonemetabolism following gastric antrectomy in the rat. Bone Miner 1992;19(3):245–56.

[48] Rumenapf G, Schwille PO, Erben RG, Schreiber M, Fries W, Schmiedl A,et al. Osteopenia following total gastrectomy in the rat-state of mineralmetabolism and bone histomorphometry. Eur Surg Res 1997;29(3):209–21.

413M.S.B. Huda et al. / Bone 41 (2007) 406–413

[49] Rumenapf G, Schwille PO, Erben RG, Schreiber M, Berge B, Fries W,et al. Gastric fundectomy in the rat: effects on mineral and bonemetabolism, with emphasis on the gastrin–calcitonin–parathyroidhormone–vitamin D axis. Calcif Tissue Int 1998;63(5):433–41.

[50] Le Roux CW, Neary NM, Halsey TJ, Small CJ, Martinez-Isla AM, GhateiMA, et al. Ghrelin does not stimulate food intake in patients with surgicalprocedures involving vagotomy. J Clin Endocrinol Metab 2005;90(8):4521–4.

[51] Lucidi P, Murdolo G, Di LC, Parlanti N, De CA, Fatone C, et al. Metabolicand endocrine effects of physiological increments in plasma ghrelinconcentrations. Nutr Metab Cardiovasc Dis 2005;15(6):410–7.

[52] Doga M, Bonadonna S, Gola M, Mazziotti G, Nuzzo M, Giustina A. GHdeficiency in the adult and bone. J Endocrinol Invest 2005;28(Suppl8):18–23.

[53] Rudman D, Kutner MH, Rogers CM, Lubin MF, Fleming GA, Bain RP.Impaired growth hormone secretion in the adult population: relation to ageand adiposity. J Clin Invest 1981;67(5):1361–9.

[54] Popovic V, Miljic D, Pekic S, Pesko P, Djurovic M, Doknic M, et al. Lowplasma ghrelin level in gastrectomized patients is accompanied byenhanced sensitivity to the ghrelin-induced growth hormone release.J Clin Endocrinol Metab 2005;90(4):2187–91.

[55] Svensson J, Lall S, Dickson SL, Bengtsson BA, Romer J, hnfelt-Ronne I,et al. Effects of growth hormone and its secretagogues on bone. Endocrine2001;14(1):63–6.

[56] Cocchi D, Maccarinelli G, Sibilia V, Tulipano G, Torsello A, Pazzaglia UE,et al. GH-releasing peptides and bone. J Endocrinol Invest 2005;28(Suppl8):11–4.

[57] IshikawaM, Hiroi N, Kamioka T, Tanaka T, Tachibana T, Ishikawa H, et al.Metabolic effects of 20 kDa and 22 kDa human growth hormones on adultmale spontaneous dwarf rats. Eur J Endocrinol 2001;145(6):791–7.

[58] Damjanovic SS, Lalic NM, Pesko PM, PetakovMS, Jotic A,Miljic D, et al.Acute effects of ghrelin on insulin secretion and glucose disposal rate ingastrectomized patients. J Clin Endocrinol Metab 2006;91(7):2574–81.

[59] Pi YZ, Wu XP, Liu SP, Luo XH, Cao XZ, Xie H, et al. Age-related changesin bone biochemical markers and their relationship with bone mineraldensity in normal Chinese women. J Bone Miner Metab 2006;24(5):380–5.

[60] Liu SP, Liao EY, Wu XP, Cao XZ, Shan PF, Su X. Age-related andmenopause-related changes of urinary excretion of C- and N-terminalcross-linked telopeptides of type I collagen and the relationships thereofwith menopause-related bone loss. Zhonghua Yixue Zazhi 2006;86(6):371–5.