Psychoneuroendocrinology (2009) 34, 413—419

Ghrelin/obestatin ratio in two populations with lowbodyweight: Constitutional thinness and anorexianervosa

Natacha Germain a, Bogdan Galusca a, Dominique Grouselle b,Delphine Frere c, Virginie Tolle b, Philippe Zizzari b, Francois Lang d,Jacques Epelbaum b, Bruno Estour a,*

ava i lab le at www.sc ienced i rect .com

journa l homepage: www.e l sev ie r.com/locate/psyneuen

a Service d’Endocrinologie de CHU Saint Etienne, FrancebUMR 894 INSERM, Faculte de Medecine, Universite Paris Descartes, IFR Broca-Sainte Anne, 75014 Paris, Francec Laboratoire de Medecine Nucleaire de CHU Saint Etienne, Franced Service de Psychiatrie de CHU Saint Etienne, France

Received 25 July 2008; received in revised form 1 September 2008; accepted 1 October 2008

KEYWORDSObestatin;Acylated grelin;Constitutional thinness;Anorexia nervosa

Summary Constitutional thinness (CT) and anorexia nervosa (AN) are two categories ofseverely underweight subjects. Some appetite-regulating hormones display opposite levels inAN and CT. While levels of ghrelin, an orexigenic hormone, fit with the normal food intake in CT,the lack of efficacy of increased ghrelin levels in AN is not clear. Obestatin is a recently describedpeptide derived from the preproghrelin gene, reported to inhibit appetite in contrast to ghrelin.The aim of this study was to determine whether the circadian profile of obestatin, total andacylated ghrelin levels is different in CT subjects when compared with AN patients.

Six-points circadian profiles of plasma obestatin, acylated ghrelin, total ghrelin and otherhormonal and nutritional parameters were evaluated in four groups of young women: 10 CT, 15restricting-type AN, 7 restored from AN and 9 control subjects.

Obestatin circadian levels were significantly higher in AN ( p < 0.0001) while no difference wasfound between CT and control subjects. Acylated and total ghrelin were found increased in AN.Acylated ghrelin/obestatin and total ghrelin/obestatin were found decreased in AN compared toCTor C subjects ( p < 0.05). The percentage of acylated ghrelin was found decreased in CT group(p < 0.05).

The decreased ghrelin/obestatin ratio found in AN might participate in the restraint innutriment intake of these patients. In contrast, in CT a lower percentage of acylated over totalghrelin might be considered in the aetiology of this condition.# 2008 Elsevier Ltd. All rights reserved.

* Corresponding author at: Endocrinology Department, CHU Saint Etienne, 42055 Saint Etienne, Cedex 2, France.Tel.: +33 4 77 12 77 27; fax: +33 4 77 12 04 93.

E-mail address: bruno.estour@chu-st-etienne.fr (B. Estour).

0306-4530/$ — see front matter # 2008 Elsevier Ltd. All rights reserved.doi:10.1016/j.psyneuen.2008.10.001

414 N. Germain et al.

1. Introduction

Obestatin is a recently discovered 23 amino acid peptidecharacterized from rat stomach (Zhang et al., 2005). Itderives from the preproghrelin gene and C-terminal amida-tion is a prerequisite for its biological activity. In contrast toghrelin which positively affects food intake and energy bal-ance (Horvath et al., 2001), an anorexigenic action wasinitially reported for obestatin (Zhang et al., 2005). Afterthis first report, many conflicting results surged (Chartrelet al., 2007) and inhibitory effects of obestatin on food intakeare currently disputed (Zizzari et al., 2007; Depoortereet al., 2008).

Obestatin levels have been assessed in human subjectswith high or low body weight in order to determine a patho-physiological role of this putative hormone. High levels werefound in restrictive anorexia nervosa (AN) women who pre-sent with a low fat mass (Harada et al., 2008; Nakahara et al.,2008). Conversely, a low level of obestatin was reported inobese people (Gao et al., 2008), during pregnancy (Fontenotet al., 2007) and in children before and after overweightreduction (Reinehr et al., 2008). The obestatin/ghrelin ratiocould monitor the opposite effects of these two hormones.Indeed, a low obestatin/ghrelin ratio was found in obesity(Vicennati et al., 2007), leading these authors to confer tothe two preproghrelin gene-derived peptides a pathophysio-logical role on weight gain.

We recently characterized a condition: constitutionalthinness (CT), in which young, severely thin (body mass index(BMI) between 14.5 and 16.5 kg/m2) women continue todisplay a close-to-normal fat mass ratio, regular physiologicalmenstrual cycles and energy metabolism (Bossu et al., 2007).The difference between these subjects and anorexia nervosapatients is also reflected in the differing secretory patterns inghrelin and PYY, two appetite-regulating hormones. Highorexigenic ghrelin and low anorexigenic levels of PYY aredetected in AN and considered as adaptative. Oppositeprofile of these two hormones was found in CT (Germainet al., 2007). The contrasting situation, between high PYYlevels in CTand low levels in obesity, suggests a physiopatho-logic role in body weight control for this hormone (Batterhamet al., 2003; Germain et al., 2007). Obestatin has not beenevaluated as yet in CT women.

The aim of this study was to determine whether thecircadian profile of obestatin, total and acylated ghrelinlevels is different in CT subjects when compared with ANpatients and normal BMI controls.

2. Patients and methods

The local research and ethics committee of Saint Etienne,France approved the study and all subjects gave writteninformed consent.

2.1. Subjects

The study included three groups of Caucasian women: con-stitutionally thin (CT), anorexia nervosa (AN) and controlsubjects (C). The CT and AN subjects were BMI matched.

Ten CT subjects were recruited from our outpatient clinicamong the patients evaluated for leanness, using the follow-

ing criteria: BMI < 16.5 kg/m2, stable throughout the growthperiod, presence of physiological menstruations withoutoestro-progestative treatment and with a stated desire forweight gain.

Fifteen subjects with restrictive-type AN were recruitedduring first hospitalization before any therapeutic interven-tion. They met the criteria for AN of the Diagnostic andStatistical Manual of Mental Disorders (American PsychiatricAssociation, 1994), and had a BMI< 16.5 kg/m2. None of themused oral contraceptives and all presented with secondaryamenorrhea for more than 6 months. Among these patientsseven were retested after weight gain (BMI > 18.5 kg/m2) andmenses restoration.

Nine normal weight controls (mean BMI 20.4 � 0.9 kg/m2)were matched by age (18—27 years) with AN and CTsubjects.

In CT and controls all data were collected during thefollicular phase of cycle.

None of the subjects had documented chronic or conge-nital disease, and none of them were taking any medication.

2.2. Sampling

Venous blood samples were collected on dry glass tubes,centrifuged and plasma was aliquoted and kept frozen at�80 8C before the assay. After an overnight fast IGF-1, 17b-oestradiol, free T3, SHBG, were measured at 08.00 h. Duringthe circadian study, samples were collected every 4 h for aperiod of 24 h (08.00—12.00—16.00—20.00—24.00—04.00 h)in order to measure, leptin, growth hormone (GH), cortisol,obestatin, total and acylated ghrelin. Standardized mealwere proposed at 08.15 h (400 kcal), 12.15 h (800 kcal) and19.15 h (800 kcal) and certainly eaten only in control groupand constitutional thinness.

2.3. Assays

Obestatin levels were determined with a commercial RIAobestatin (human, monkey) kit (Phoenix, Belmont, CA). Pre-proghrelin 52—85 and 86—117 (Phoenix, Belmont, CA) do notcross-react up to the concentration of 3973 pmol/l. Thesensitivity of this assay is 4 pmol/l and the intra-assay coeffi-cient of variation is 8%.

Plasma ghrelin isoforms were assayed by an enzymeimmunoassay (EIA) in a competitive solid phase method usingtwo rabbit polyclonal antibodies raised against the 15—28 C-terminus fragment for total ghrelin (gift from C. Tomasetto,Strasbourg, France) and the 1—11 N-terminus N-octanoylatedfragment for acylated ghrelin (gift from M. Kojima, Fukuoka,Japon). Human n-octanoylated ghrelin was used as standard(Phœnix, Belmont, USA) and coupled to acetylcholinesteraseas tracer (SPI-bio, Montigny le Bretonneux, France). The totalghrelin antibody recognized all isoforms and the acylatedghrelin antibody was specific of the acylated form. Antibo-dies were used at a final dilution of 1/105 for C-terminus and1/107 for N-terminus. Their limit of detection was 50 pg/ml(50—4000) and 10 pg/ml (10—400), respectively. Out of eightconsecutive assays, their intra- and inter-assay variabilitieswere 4% and 7%, 5% and 6% at 50% binding, respectively (Tolleet al., 2003).

The following techniques were used to measure: plasmacortisol–—RIA (Cortisol, Immunotech; intra- and interassay

Ghrelin/obestatin ratio in lean subjects 415

CVs, 7% and 8%, respectively; detection limit 10 nmol/24 h;normal range 107—310 ng/ml), GH-immunotech IRMA (Beck-man Coulter) manufacturer’s reference level <5 mU/l), IGF-I-immunotech IRMA (Beckman Coulter) intra and interassayCVs, 7% and 16%, respectively, manufacturer’s referencerange 107—310 mg/l, 17b estradiol—RIA (Dia Sorin, France)manufacturer’s reference range during the follicular phase30—50 ng/l, Leptin-RIA (Nichols Institute Diagnostics, SanJuan Capistrano, CA, USA, manufacturer’s reference rangefor a normal BMI (18—25) 3.7—11.1 mg/l), SHBG-IRMA(Biomerieux) manufacturer’s reference range 20—85 nmol/l, free T3-RIA (Beckman) manufacturer’s reference range2.5—5.8 pmol/l.

2.4. Body composition measurements

Dual-energy X-ray absorptiometry (DXA) allowed the quanti-fication of the percentage of total body fat mass (FM) and fat-free mass (FFM) expressed in kilograms (LUNAR, DPX-L,CV < 1%) (Mazess et al., 1990; Fuller et al., 1992).

2.5. Statistical analysis

All values are presented asmean � S.E.M. Intergroup compar-ison was performed by ANOVA. Fisher’s PLSD (protected leastsignificant difference), a post hoc ANOVA test for multiplemeasures was used to evaluate inter-group differences andintra-group circadian variation of ghrelin, leptin, GH, andcortisol since six circadian determinations were performedfor these parameters. Levene testwas also applied to compareintra-group variances. In order to evaluate several interac-tions, we calculated Pearson correlation indices in total andeach group. The ratios between acylated or total ghrelin andobestatin were calculated in each spared group. Paired t testwere used to evaluate the impact of weight gain on appetite-regulating hormones in recovered-from-AN subjects. Statisti-cal significance was set at p < 0.05. To compensate for multi-ple tests per variable, Bonferroni’s correction was applied. All

Table 1 Nutritional, hormonal and body composition parameters incontrols (mean values � S.E.M).

Anthropometry and body composition Age (years)Height (m)BMI (kg/m2)Fat mass (%)

Hormonal parameters Leptin (mg/l)GH (mIU/l)IGF-1 (mg/l)Cortisol (ng/l)17b-Estradiol (ng/l)SHBG (nmol/l)Free T3 (pmol/l)

Appetite-regulating hormones Mean acylated ghrelin (pgMean total ghrelin (pg/mlMean obestatin (pg/ml)

* Significantly different (p < 0.05) vs. control group.§ Significantly different (p < 0.05) vs. AN group.

statistical analyses were performed with StatView 4.5 soft-ware (Abacus Concepts, Inc., Palo Alto, CA).

3. Results

Table 1 shows body weight composition and baseline hor-mones characteristics for the three groups. The mean agewas similar. Despite significant differences in BMI ( p < 0.001)and percentage of fat mass ( p < 0.05) between CT andcontrol groups, cortisol, GH, free T3, IGF I, estradiol andleptin remained in the same range.

The circadian profiles for obestatin and acylated and totalghrelin are shown in Figure 1. The area under the curve forobestatincircadiancyclewasequivalentbetweenCTandCandreached significantly higher levels in AN (p < 0.0001). Totalghrelin levels were the highest in subjects with AN and pro-gressively lower in CT (p < 0.003) and controls (p < 0.0001)subjects. Acylated ghrelin levelswere also higher in AN than inCT and control subjects (p < 0.002). In seven AN subjectspositively responding to therapy, obestatin, acylated or totalghrelin decreased significantly (p < 0.001) (Figure 2).

Acylated ghrelin/obestatin and total ghrelin/obestatinratios were found decreased in AN compared to CT or Csubjects ( p < 0.05) (Figure 3).

Acylated ghrelin/total ghrelin ratio was lower in CTwhencompared to AN or control groups ( p < 0.001). Meanwhileintragroup variability was the highest in the AN group accord-ing to Levenne test ( p < 0.0001) (Figure 4).

Obestatin was not correlated with acylated ghrelin ortotal ghrelin in each separate group or in total group. Acy-lated ghrelin correlated positively with total ghrelin in eachgroup, and in a non-linear manner in total group (r = 0.809,p < 0.0001) (Figure 5). The acylated/total ghrelin ratio cor-related positively with absolute values of acylated ghrelin(r = 0.815, p < 0.0001).

Overall, inverse linear correlation occurred between obe-statin and nutritional markers: BMI (r = �0.512, p = 0.001),FT3 (r = �0.525, p = 0.0007), IGF1 (r = �0.524, p = 0.0007), %

lean anorexia nervosa (AN), constitutionally thin women (CT) and

Lean AN (n = 15) CT (n = 9) Controls (n = 10)

20.4 � 1.3 24.1 � 1.2 23.1 � 1.41.63 � 0.1 1.61 � 0.02 1.63 � 0.0114.8 � 0.1 * 16.1 � 0.1 * 20.5 � 0.410.9 � 1.0 * 16.9 � 0.7*§ 28.1 � 1.2

2.8 � 0.4 * 6.6 � 0.8*§ 12.2 � 1.911.6 � 4.2 * 4.8 � 0.6§ 4.7 � 0.6151 � 19* 295 � 34§ 283 � 20396 � 41* 236 � 20§ 271 � 5815.0 � 3.2 * 64.9 � 19.9§ 61.4 � 21.0

123.4 � 12.1 * 91.5 � 14.2§ 112.2 � 27.42.8 � 0.1 * 4.1 � 0.1§ 3.8 � 0.1

/ml) 1123 � 209 * 781 � 167 838 � 85) 4181 � 533 * 3518 � 536§ 2998 � 223

393 � 25* 260 � 26 288 � 26

Figure 1 Six-points circadian levels of obestatin, acylated ghre-lin and total ghrelin in anorexia nervosa (AN) patients, constitu-tionally thin (CT) subjects and controls. Area under curve wassignificantly increased in AN group for all these peptides. Figure 2 Mean circadian levels of obestatin, total and acylated

ghrelin decrease after weight gain recovery in anorexia nervosa.

416 N. Germain et al.

fat mass (r = �0.573, p = 0.0007) and a negative logarithmiccorrelation with leptin (r = �0.8, p < 0.0001) (Figure 6).Conversely there is no correlation between total or acylatedghrelin and the same nutritional markers, leptin included.

4. Discussion

In this study different circadian profiles of the obestatin andghrelin plasma levels, two gastric hormones transcribedby the preproghrelin gene, were observed in two groups of

women presenting with similarly low BMI: constitutionallythin women and AN patients.

Despite low BMI values <16.5 kg/m2, CT women do notexhibit abnormal feeding behaviour and caloric deficit socharacteristic for restrictive AN. However, and in contrast toAN patients, CT women present with a close-to-normal fatmass and leptin (Bossu et al., 2007) and normal levels of T3and IGF1, two nutritional factors known to be influenced bycaloric restriction (Chopra and Smith, 1975; Isley et al., 1983;

Figure 3 Decreased values of acylated ghrelin/obestatin andtotal ghrelin/obestatin ratios in anorexia nervosa patients com-pared with constitutionally thin subjects or controls.

Figure 4 Decreased levels of acylated/total ghrelin ratio werefound in constitutionally thin subjects.

Ghrelin/obestatin ratio in lean subjects 417

Caregaro et al., 2001; Bossu et al., 2007). Moreover, mostendocrine parameters (17b-estradiol, cortisol, GH, and IGF-I)in CT subjects (Tolle et al., 2003) are not different fromcontrols, confirming their phenotypic differences fromrestrictive AN patients which exhibit hypercortisolism (Boyaret al., 1977), alteration of the GH-IGFI axis (Stoving et al.,1999) and amenorrhea (American Psychiatric Association,

Figure 5 A non-linear positive correlation was found between acywas taken into account when parameters with such measurements

1994). Thus CT seems to be a close-to-normal situation butfor a high risk of low bone mass (Galusca et al., 2008).

This study completes the description of food intake hor-mones in CT, reporting for the first time normal circadianlevels of obestatin and acylated ghrelin in CT women. Ourprevious reports presented conflicting levels of total ghrelinwhen compared to controls but overall significantly lowerthan in AN patients (Tolle et al., 2003; Germain et al., 2007).Secondly, in the current study acylated ghrelin/total ghrelinratio was found lower than in controls. It is now well docu-mented that the acylation of ghrelin is required for theorexigenic action of ghrelin (Gauna et al., 2004) and there-fore its relative decrease accounted in constitutional thin-ness could play a pathophysiological role along with theincrease of PYY, an anorexigenic hormone, previouslydescribed in these subjects (Germain et al., 2007). On theother hand, genetic factors influencing thinness remain to beexplored in these subjects.

In anorexia nervosa, high obestatin levels were previouslyreported in fasting conditions (Harada et al., 2008; Monte-

lated ghrelin and total ghrelin. Every point from circadian cyclewere correlated.

Figure 6 Non-linear negative relationship between mean cir-cadian levels of obestatin and leptin in total group.

418 N. Germain et al.

leone et al., 2008; Nakahara et al., 2008). While high totalghrelin in AN has been consistently observed, acylated ghre-lin was found elevated in one study only (Harada et al., 2008).The circadian evaluation in the current study showed thatincreased obestatin plasma levels in restrictive anorexianervosa patients were associated to increased total andacylated ghrelin all over the day. The circulating levels ofthe three preproghrelin gene-derived peptides decreasedsignificantly after refeeding. In AN, high ghrelin, low leptinand low PYY levels (Germain et al., 2007) account for anadaptive orexigenic profile of appetite regulation that, the-oretically, should increase nutriment consumption. It is dif-ficult to understand why AN patients present with a ‘‘ghrelinresistance’’ situation. Since the first report from Zhang pre-senting obestatin as a functional ghrelin antagonist (Zhanget al., 2005), some authors developed the orexigenic/anor-exigenic ratio concept between ghrelin and obestatin. A highpreprandial ghrelin/obestatin ratio was found in obesitysuggesting that this ratio was involved in the pathophysiologyof obesity (Guo et al., 2007). Opposite to obesity, we foundthe ghrelin/obestatin ratio lower in AN than in CTor controls.Such chronic imbalance could facilitate the lack in nutrimentintake if obestatin own inhibitory effects on food regulationand body weight were more validated (Bresciani et al., 2006;Seoane et al., 2006; Sibilia et al., 2006; Chartrel et al., 2007;Depoortere et al., 2008; Lagaud et al., 2007; Zizzari et al.,2007). Food-induced vagal stimulation induces a drasticincreased of this ghrelin/obestatin ratio in AN (Monteleoneet al., 2008) that accounts for an amplified signal of hunger,explaining why food contacts are voluntarily avoided inrestriction type of AN.

In accordance with previous data a strong relationship wasfound between nutritional markers and obestatin (Nakaharaet al., 2008). Oppositely to earlier studies (Tolle et al., 2003;Misra et al., 2005) a close relationship between leptin andghrelin, either total or acylated was not found. However,several recent data suggest distinct pathways of control andaction for these two hormones (Cummings and Foster, 2003;Chan et al., 2004; Schmid et al., 2005). Taken together, theseand our observations suggest that obestatin is at least anutritional marker reflecting body adiposity.

In fed normal weight subjects the percentage of acylatedghrelin accounts for 22% of total ghrelin and drops during 3-

day fasting to about 9% (Liu et al., 2008) without any changesin total ghrelin levels. In anorexia nervosa group of our study,the percentage of acylated ghrelin increased up to 50%related to total ghrelin increase. Similar variation of activeto total ghrelin ratio were found in cancer related cachexia(Garcia et al., 2005). These findings are likely to reflectdifferent ghrelin regulation between acute and chronic fast-ing. In obesity acylated ghrelin levels appear comparativelymore reduced at baseline and more responsive to feedingthan total ghrelin levels (Ariyasu et al., 2002; Marzullo et al.,2004). Such changes could be explained by the impact ofnutrition status on the activity of ghrelin O-acyltransferase(Yang et al., 2008).

The lack of correlation we found in total group betweenghrelin and obestatin in the current study could be in agree-ment with recent demonstration of putative obestatin and C-ghrelin specific transcripts suggesting that these ghrelingene-derived peptides may be produced independently ofpreproghrelin (Seim et al., 2007).

In conclusion, we report a normal obestatin profile inconstitutional thinness, along with a low percentage ofacylated ghrelin that should be taken into account togetherwith gastrointestinal elevated PYY levels in the physiopathol-ogy of this entity. Decreased ghrelin/obestatin ratio we foundin restrictive AN could facilitate the food intake restriction inthese subjects.

Conflict of interest

All authors declare that they have no conflicts of interest.

References

American Psychiatric Association, 1994. Diagnostic and StatisticalManual of Mental Disorders. American Psychiatric Association,Washington, DC.

Ariyasu, H., Takaya, K., Hosoda, H., Iwakura, H., Ebihara, K., Mori,K., Ogawa, Y., Hosoda, K., Akamizu, T., Kojima, M., Kangawa, K.,Nakao, K., 2002. Delayed short-term secretory regulation ofghrelin in obese animals: evidenced by a specific RIA for theactive form of ghrelin. Endocrinology 143, 3341—3350.

Batterham, R.L., Cohen, M.A., Ellis, S.M., Le Roux, C.W., Withers,D.J., Frost, G.S., Ghatei, M.A., Bloom, S.R., 2003. Inhibition offood intake in obese subjects by peptide YY3-36. N. Engl. J. Med.349, 941—948.

Bossu, C., Galusca, B., Normand, S., Germain, N., Collet, P., Frere,D., Lang, F., Laville, M., Estour, B., 2007. Energy expenditureadjusted for body composition differentiates constitutional thin-ness from both normal subjects and anorexia nervosa. Am. J.Physiol. Endocrinol. Metab. 292, E132—E137.

Boyar, R.M., Hellman, L.D., Roffwarg, H., Katz, J., Zumoff, B.,O’Connor, J., Bradlow, H.L., Fukushima, D.K., 1977. Cortisolsecretion and metabolism in anorexia nervosa. N. Engl. J. Med.296, 190—193.

Bresciani, E., Rapetti, D., Dona, F., Bulgarelli, I., Tamiazzo, L.,Locatelli, V., Torsello, A., 2006. Obestatin inhibits feeding butdoes not modulate GH and corticosterone secretion in the rat. J.Endocrinol. Invest. 29, RC16—RC18.

Caregaro, L., Favaro, A., Santonastaso, P., Alberino, F., Di Pascoli, L.,Nardi, M., Favaro, S., Gatta, A., 2001. Insulin-like growth factor 1(IGF-1), a nutritional marker in patients with eating disorders.Clin. Nutr. 20, 251—257.

Chan, J.L., Bullen, J., Lee, J.H., Yiannakouris, N., Mantzoros, C.S.,2004. Ghrelin levels are not regulated by recombinant leptin

Ghrelin/obestatin ratio in lean subjects 419

administration and/or three days of fasting in healthy subjects. J.Clin. Endocrinol. Metab. 89, 335—343.

Chartrel, N., Alvear-Perez, R., Leprince, J., Iturrioz, X., Reaux-LeGoazigo, A., Audinot, V., Chomarat, P., Coge, F., Nosjean, O.,Rodriguez, M., Galizzi, J.P., Boutin, J.A., Vaudry, H., Llorens-Cortes, C., 2007. Comment on ‘‘Obestatin, a peptide encoded bythe ghrelin gene, opposes ghrelin’s effects on food intake’’.Science 315, 766 author reply 766.

Chopra, I.J., Smith, S.R., 1975. Circulating thyroid hormones andthyrotropin in adult patients with protein-calorie malnutrition. J.Clin. Endocrinol. Metab. 40, 221—227.

Cummings, D.E., Foster, K.E., 2003. Ghrelin—leptin tango in body-weight regulation. Gastroenterology 124, 1532—1535.

Depoortere, I., Thijs, T., Moechars, D., De Smet, B., Ver Donck, L.,Peeters, T.L., 2008. Effect of peripheral obestatin on food intakeand gastric emptying in ghrelin-knockout mice. Br. J. Pharmacol.153, 1550—1557.

Fontenot, E., DeVente, J.E., Seidel, E.R., 2007. Obestatin andghrelin in obese and in pregnant women. Peptides 28, 1937—1944.

Fuller, N.J., Jebb, S.A., Laskey, M.A., Coward, W.A., Elia, M., 1992.Four-component model for the assessment of body composition inhumans: comparison with alternative methods, and evaluation ofthe density and hydration of fat-free mass. Clin. Sci. (Lond.) 82,687—693.

Galusca, B., Zouch, M., Germain, N., Bossu, C., Frere, D., Lang, F.,Lafage-Proust, M.H., Thomas, T., Vico, L., Estour, B., 2008.Constitutional thinness: unusual human phenotype of low bonequality. J. Clin. Endocrinol. Metab. 93, 110—117.

Gao, X.Y., Kuang, H.Y., Liu, X.M., Wang, X.Y., Pan, Y.H., Ma, X.X.,2008. Decreased obestatin in plasma in metabolically obese,normal-weight men with normal glucose tolerance. DiabetesRes. Clin. Pract. 79, e5—e6.

Garcia, J.M., Garcia-Touza, M., Hijazi, R.A., Taffet, G., Epner, D.,Mann, D., Smith, R.G., Cunningham, G.R., Marcelli, M., 2005.Active ghrelin levels and active to total ghrelin ratio in cancer-induced cachexia. J. Clin. Endocrinol. Metab. 90, 2920—2926.

Gauna, C., Meyler, F.M., Janssen, J.A., Delhanty, P.J., Abribat, T., vanKoetsveld, P., Hofland, L.J., Broglio, F., Ghigo, E., van der Lely,A.J., 2004. Administration of acylated ghrelin reduces insulinsensitivity, whereas the combination of acylated plus unacylatedghrelin strongly improves insulin sensitivity. J. Clin. Endocrinol.Metab. 89, 5035—5042.

Germain, N., Galusca, B., Le Roux, C.W., Bossu, C., Ghatei, M.A.,Lang, F., Bloom, S.R., Estour, B., 2007. Constitutional thinnessand lean anorexia nervosa display opposite concentrations ofpeptide YY, glucagon-like peptide 1, ghrelin, and leptin. Am.J. Clin. Nutr. 85, 967—971.

Guo, Z.F., Zheng, X., Qin, Y.W., Hu, J.Q., Chen, S.P., Zhang, Z., 2007.Circulating preprandial ghrelin to obestatin ratio is increased inhuman obesity. J. Clin. Endocrinol. Metab. 92, 1875—1880.

Harada, T., Nakahara, T., Yasuhara, D., Kojima, S., Sagiyama, K.,Amitani, H., Laviano, A., Naruo, T., Inui, A., 2008. Obestatin, acylghrelin, and des-acyl ghrelin responses to an oral glucose toler-ance test in the restricting type of anorexia nervosa. Biol.Psychiatry 63, 245—247.

Horvath, T.L., Diano, S., Sotonyi, P., Heiman, M., Tschop, M., 2001.Minireview: ghrelin and the regulation of energy balance–—ahypothalamic perspective. Endocrinology 142, 4163—4169.

Isley, W.L., Underwood, L.E., Clemmons, D.R., 1983. Dietary com-ponents that regulate serum somatomedin-C concentrations inhumans. J. Clin. Invest. 71, 175—182.

Lagaud, G.J., Young, A., Acena, A., Morton, M.F., Barrett, T.D.,Shankley, N.P., 2007. Obestatin reduces food intake and sup-presses body weight gain in rodents. Biochem. Biophys. Res.Commun. 357, 264—269.

Liu, J., Prudom, C.E., Nass, R., Pezzoli, S.S., Oliveri, M.C., Johnson,M.L., Veldhuis, P., Gordon, D.A., Howard, A.D., Witcher, D.R.,Geysen, H.M., Gaylinn, B.D., Thorner, M.O., 2008. Novel ghrelin

assays provide evidence for independent regulation of ghrelinacylation and secretion in healthy youngmen. J. Clin. Endocrinol.Metab. 93, 1980—1987.

Marzullo,P.,Verti,B.,Savia,G.,Walker,G.E.,Guzzaloni,G.,Tagliaferri,M., Di Blasio, A., Liuzzi, A., 2004. The relationship between activeghrelin levels and human obesity involves alterations in restingenergy expenditure. J. Clin. Endocrinol. Metab. 89, 936—939.

Mazess, R.B., Barden, H.S., Bisek, J.P., Hanson, J., 1990. Dual-energyX-ray absorptiometry for total-body and regional bone-mineraland soft-tissue composition. Am. J. Clin. Nutr. 51, 1106—1112.

Misra, M., Miller, K.K., Kuo, K., Griffin, K., Stewart, V., Hunter, E.,Herzog, D.B., Klibanski, A., 2005. Secretory dynamics of ghrelin inadolescent girls with anorexia nervosa and healthy adolescents.Am. J. Physiol. Endocrinol. Metab. 289, E347—E356.

Monteleone, P., Serritella, C., Martiadis, V., Maj, M., 2008. Derangedsecretion of ghrelin and obestatin in the cephalic phase of vagalstimulation in women with anorexia nervosa. Biol. Psychiatry(Epub ahead of print).

Nakahara, T., Harada, T., Yasuhara, D., Shimada, N., Amitani, H.,Sakoguchi, T., Kamiji, M.M., Asakawa, A., Inui, A., 2008. Plasmaobestatin concentrations are negatively correlatedwith bodymassindex, insulin resistance index, and plasma leptin concentrations inobesity and anorexia nervosa. Biol. Psychiatry 64, 252—255.

Reinehr, T., de Sousa, G., Roth, C.L., 2008. Obestatin and ghrelinlevels in obese children and adolescents before and after reduc-tion of overweight. Clin. Endocrinol. (Oxf.) 68, 304—310.

Schmid, D.A., Held, K., Ising, M., Uhr, M., Weikel, J.C., Steiger, A.,2005. Ghrelin stimulates appetite, imagination of food, GH,ACTH, and cortisol, but does not affect leptin in normal controls.Neuropsychopharmacology 30, 1187—1192.

Seim, I., Collet, C., Herington, A.C., Chopin, L.K., 2007. Revisedgenomic structure of the human ghrelin gene and identification ofnovel exons, alternative splice variants and natural antisensetranscripts. BMC Genomics 8, 298.

Seoane, L.M., Al-Massadi, O., Pazos, Y., Pagotto, U., Casanueva, F.F.,2006. Central obestatin administration does not modify eitherspontaneous or ghrelin-induced food intake in rats. J. Endocrinol.Invest. 29, RC13—RC15.

Sibilia, V., Bresciani, E., Lattuada, N., Rapetti, D., Locatelli, V., DeLuca, V., Dona, F., Netti, C., Torsello, A., Guidobono, F., 2006.Intracerebroventricular acute and chronic administration ofobestatin minimally affect food intake but not weight gain inthe rat. J. Endocrinol. Invest. 29, RC31—RC34.

Stoving, R.K., Veldhuis, J.D., Flyvbjerg, A., Vinten, J., Hangaard, J.,Koldkjaer, O.G., Kristiansen, J., Hagen, C., 1999. Jointly amplifiedbasal and pulsatile growth hormone (GH) secretion and increasedprocess irregularity in women with anorexia nervosa: indirect evi-dence for disruption of feedback regulation within the GH-insulin-like growth factor I axis. J. Clin. Endocrinol. Metab. 84, 2056—2063.

Tolle, V., Kadem, M., Bluet-Pajot, M.T., Frere, D., Foulon, C., Bossu,C., Dardennes, R., Mounier, C., Zizzari, P., Lang, F., Epelbaum, J.,Estour, B., 2003. Balance in ghrelin and leptin plasma levels inanorexia nervosa patients and constitutionally thin women. J.Clin. Endocrinol. Metab. 88, 109—116.

Vicennati, V., Genghini, S., De Iasio, R., Pasqui, F., Pagotto, U.,Pasquali, R., 2007. Circulating obestatin levels and the ghrelin/obestatin ratio in obesewomen. Eur. J. Endocrinol. 157, 295—301.

Yang, J., Brown, M.S., Liang, G., Grishin, N.V., Goldstein, J.L., 2008.Identification of the acyltransferase that octanoylates ghrelin, anappetite-stimulating peptide hormone. Cell 132, 387—396.

Zhang, J.V., Ren, P.G., Avsian-Kretchmer, O., Luo, C.W., Rauch, R.,Klein, C., Hsueh, A.J., 2005. Obestatin, a peptide encoded by theghrelin gene, opposes ghrelin’s effects on food intake. Science310, 996—999.

Zizzari, P., Longchamps, R., Epelbaum, J., Bluet-Pajot, M.T., 2007.Obestatin partially affects ghrelin stimulation of food intakeand growth hormone secretion in rodents. Endocrinology 148,1648—1653.