Effect of obesity on total and free insulin-likegrowth factor (IGF)-1, and their relationship toIGF-binding protein (BP)-1, IGFBP-2, IGFBP-3,insulin, and growth hormone

SY Nam, EJ Lee, KR Kim, BS Cha, YD Song, SK Lim, HC Lee and KB Huh

Division of Endocrinology, Department of Internal Medicine, Yong Dong Severance Hospital, Yonsei University, College of Medicine,Seoul, Korea

OBJECTIVES: We investigated the effect of obesity on the serum levels of total and free IGF-1 and their relationship tothe circulating levels of insulin and IGF binding proteins (IGFBPs) in age and sex-matched groups.SUBJECTS: The study included 43 obese subjects (ideal body weight; IBW> 120%) and 45 controls (IBW< 100%). All ofthe subjects were male.MEASUREMENT: Total IGF-1, free IGF-1, IGFBP-1, IGFBP-2, IGFBP-3, and insulin were measured in obese subjects andnormal control subjects.RESULTS: No signicant differences in the circulating levels of total and IGFBP-3 were observed between the obeseand control groups. In contrast to total IGF-1, free IGF-1 in obese subjects was signicantly increased compared tonormal controls (P< 0.05). Serum total and free IGF-1 were inversely correlated with age (r0.42, P 0.001, and0.44, P 0.001). Fasting serum insulin concentrations were elevated in all the obese subjects (P< 0.05) and positivelycorrelated with IBW (r 0.57, P 0.001). The levels of serum GH and IGFBP-1 were suppressed in all the obese subjects(P< 0.05). IGFBP-1 was inversely correlated with IBW (r0.51, P 0.001) and serum insulin concentrations(r0.48, P 0.001). The IGFBP-2 concentrations were also suppressed in obese subjects and inversely related tofree IGF-1 (r0.48, P 0.001). Using multiple linear regression analysis, total IGF-1 and insulin concentrations werepositively correlated (r 0.58, P 0.001) and free IGF-1 and IGFBP-1 concentrations were negatively correlated(r0.57, P 0.001).CONCLUSION: We conrmed that total IGF-1 and IGFBP-3 concentrations were not signicantly different between theobese and control groups, despite GH hyposecretion in obesity. We also found that free IGF-1 concentrations werehigher in obese subjects than in normal controls. It seems likely that overnutrition and chronic hyperinsulinaemia inobesity may alter this regulated growth response by insulin stimulation of IGF-1 production and suppression ofhepatic IGFBP-1 and IGFBP-2 production, which may inhibit IGF-1 bioactivity.

Keywords: obesity; total IGF-1; free IGF-1; IGFBPs

Introduction

Obesity is associated with the impairment of normalGH (growth hormone) secretion and bluntedresponses to all stimuli such as GH-releasing hormone(GHRH), hypoglycaemia, L-dopa, arginine, glucagon,physical exercise, and sleep.14 Based on the wellknown GH hyposecretion in obesity, GH-dependentinsulin-like growth factor (IGF)-1 concentrationswould be expected to be subnormal in obesity. How-ever, the results have been conflicting,57 due toclinical differences in age, sex and degree and typeof obesity.

Circulating IGF-1 is predominantly bound in a

ternary 150 kDa complex, consisting of IGF-1, IGF-binding protein (IGFBP)-3, and an acid-labile sub-unit.8,9 Thus IGFBP-3 and IGF-1 have relatively longhalf-lives in the circulation,10 and since they are GHdependent they represent a stable index of long-termchanges in GH availability and action.11 IGFBP-2 isthe second most abundant IGF-binding protein inserum. The serum concentration of IGFBP-2 isknown to be inversely related to GH secretion,being high in states of GH deficiency and low instates of GH excess.12 These reports have suggestedthat measurement of IGFBP-2 might be useful indetecting GH deficiency. In contrast, the serum con-centrations of IGFBP-1 are lower than those ofIGFBP-3 and IGFBP-2, and acutely regulated byfood intake and fluctuations in serum insulin andglucose concentrations.1315 Because of this uniqueproperty, IGFBP-1 is thought to play an important rolein acute regulation of IGF-1 availability.16,17 In pre-vious reports, IGFBP-1 sequestered free IGF-1 and

Correspondence: Dr SY Nam, Division of Endocrinology,Department of Internal Medicine, Yong Dong SeveranceHospital, Young Dong PO Box 1217, Seoul, Korea.Received 21 August 1996; revised 3 January 1997; accepted 16January 1997

International Journal of Obesity (1997) 21, 355359 1997 Stockton Press All rights reserved 03070565/97 $12.00

inhibited its metabolic actions.18,19 A small portion ofcirculating IGF-1 is detected in the free or readilydissociable state, which is thought to be the metabo-lically active form. The amount of free/dissociableIGF-1 in serum is dependent on a complex interplaybetween the production rate and concentrations ofIGF-1 and IGFBPs.10

In this study, we investigated the effect of obesityon the serum concentrations of total and free IGF-1and their relationship to the concentrations of IGF-binding proteins and insulin in age and sex-matchedgroups.

Subjects and methods

Subjects and protocol

Forty-three obese Korean males aged 2049 y (mean - s.e.m., 34.9 7.9) and 45 normal Korean malesubjects aged 2049 y (33.9 8.3) were studied. Thenormal control subjects were within 10% of their idealbody weight (IBW), and the obese subjects weighedmore than 120% of ideal body weight, as determinedby the Fogarty Center Conference on Obesity. Allsubjects were recruited from subjects attending themedical-checkup programs at the health-care centre ofYong Dong Severance Hospital.

An oral glucose tolerance test was performed oneach subject to screen out impaired glucose toleranceor diabetes mellitus. Apart from obesity, no abnorm-alities were detected. No subjects had used anyhormonal preparations or drugs within the 60 d priorto the study.

The study was approved by the Hospital EthicsCommittee, and informed consent was obtained fromeach subject.

Blood samples were obtained from each subject byvenepuncture between 08:00 am and 10:00 am after anovernight fast. Serum was separated by centrifugation,and stored at 20C until analysis.

Analytical methods

Serum glucose level was measured by the glucose-oxidase method and insulin level by enzyme-linkedimmunoadsorbent assay (ELISA). This ELISA usestwo monoclonal antibodies (Novo Nordisk A/S, Den-mark) directed against human insulin and does notcross-react with human proinsulin. Serum triglycerideand cholesterol concentrations were determined in theovernight fasting state with semi-automated methods(Boehringer, Mannheim, Germany). Serum GH wasmeasured by an immunoradiometric assay (IRMA)from Daiichi (Tokyo, Japan); the sensitivity was0.1mg/L and its intra and interassay coefficients ofvariations were 1.3 and 1.4%, respectively. SerumIGFBP-1, IGFBP-2, and IGFBP-3 were measured byan immunoradiometric assay (IRMA) kit (DiagnosticSystem Laboratories, Inc. Webster, TX); the sensitiv-

ities were 0.01 ng/mL, 0.5 ng/mL, 0.5 ng/mL, respec-tively and their intra and inter-assay coefficients ofvariations were less than 10%. Serum total IGF-1 wasmeasured by an IRMA kit (Diagnostic SystemLaboratories, Inc. Webster, TX); the sensitivity was0.3 mg/L and its intra and interassay coefficients ofvariations were less than 10%. Free/dissociable IGF-1concentrations were measured by a highly sensitivetwo-site IRMA kit (Diagnostic System Laboratories,Inc. Webster, TX) according to the manufacturerssuggestions. Assay characteristics have been reportedpreviously.20 This is a direct detection assay in which100 mL of serum sample are added to tubes containinga dense coating of high affinity anti-IGF-1 antibody,which binds IGF-1 that is not bound to or is easilydissociable from IGFBPs. Samples were then incu-bated for two hours at 28C, washed, and incubatedwith an 125I-labeled anti-IGF-1 antibody directed to asecond epitope of IGF-1 for two hours at roomtemperature. After decanting and washing 3 times,the tubes were counted in a g-counter. Assay standardswere recombinant human (rh)IGF1 (0.1320mg/L).The minimal detection limit was 0.05mg/L or 5 pg/tube. According to the manufacturers specificationsand previous characterization of the assay, there wasno cross-reactivity with IGF-II, and no residualIGFBP-1 or IGFBP-3 was detectable in the assaytube after the first wash.20 The interassay coefficientsof variations were 7.7, 3.6 and 10.7% at 0.26, 5.52 and13.87 ng/mL, and the intraassay coefficients of varia-tions were 10.3, 5.1, and 3.3% at 0.29, 6.26, and14.20 ng/mL. All samples from each subject wereanalyzed in duplicate at the same time.

Statistical analysis

The results were expressed as the mean s.e.m.Statistical comparisons were made using the Studentsunpaired t-Test between obese and control groups.Correlation between different parameters was calcu-lated by linear regression analysis. Factors related tototal and free IGF-1 were analyzed using stepwisemultiple regression. P< 0.05 was accepted as thesignificance level.

Results

Percentage of ideal body weight (138.2 11.7%,mean s.e.m.), BMI (30.0 2.5 kg/m2), and waist tohip ratio (0.93 0.2) in obese subjects were signifi-cantly (P< 0.001) higher than those of normal con-trols (97.6 6.4%, 21.3 1.4 kg/m2, and 0.84 0.0).Fasting serum glucose (FSG), total cholesterol, trigly-ceride, insulin and GH concentrations are summarizedin Table 1.

The FSG, triglyceride and insulin concentrations inobese subjects were significantly greater than those ofnormal controls. In all obese subjects, fasting GH

IGF-1 and IGFBPs in obesitySY Nam et al

356

concentrations were suppressed as compared withcontrol subjects (P< 0.05).

No significant differences in the circulating con-centrations of the total IGF-1 and IGFBP-3 wereobserved between the control and obese subjects(Figure 1). Serum total IGF-1 was weakly correlatedwith IGFBP-3 (r 0.28, P 0.05). The free IGF-1concentrations were significantly elevated in obesesubjects (1.46 1.1mg/L vs 0.91 0.9mg/L, P< 0.05;Figure 1). The ratio of free to total IGF-1 also washigher in obese subjects (0.63% vs 0.39%, P< 0.05;Figure 1).

The IGFBP-1 concentrations were significantlylower in the obese subjects (Figure 1) and negativelycorrelated with percent of IBW (r0.51, P 0.001)and serum insulin (r0.48, P 0.001).

The IGFBP-2 concentrations were also suppressedin obese subjects (P< 0.05; Figure 1) and inverselyrelated to free IGF-1 (r0.48, P 0.001). Addi-tionally, the ratio of IGFBP-2 to IGF-1 was decreased(1.6 0.5 vs 2.2 1.4, P< 0.005) in obese subjects.

Using multiple linear regression analysis, the totalIGF-1 and free IGF-1 were analyzed by age, IBW,WHR, fasting serum insulin, GH, IGFBP-1, IGFBP-2,and IGFBP-3 in all subjects. Of all variables, age andserum insulin concentrations were associated withserum total IGF-1. The total IGF-1 was inverselycorrelated with age (r0.42, P< 0.001) and posi-tively correlated with serum insulin (r 0.58,P< 0.001). The free IGF-1 was negatively correlatedwith age (r0.44, P< 0.001) and IGFBP-1(r0.57, P< 0.001).

Discussion

Using age and sex matched subjects, we confirmedthat GH levels were reduced in the obese while GH-dependent total IGF-1 levels were not significantlydifferent between obese and control subjects. Similarresults have been reported previously,2123 but thecontradictory findings of increased or decreasedIGF-1 levels found by others, although due in partto important clinical differences in age, sex anddegree of obesity in the different studies, could beprofoundly influenced by nutritional factors.24,25 Inaddition, the discordant relationship between GH andIGF-1 in obesity may be also explained by the

hyperinsulinaemia present in obesity:26 insulin mayincrease hepatic IGF-1 production.27 We found apositive correlation between the serum total IGF-1and insulin concentrations.

A small portion of circulating IGF-1 is detected inthe free or readily dissociable state, which is thoughtto be the metabolically active form.28,29 In the presentstudy, free/dissociable IGF-1 was assessed by a two-site immunoradiometric assay. The absolute values forfree/dissociable IGF-1 in both obese and controlgroups were higher than those reported by Frystyket al30 using RIA after separation by centrifugalultrafiltration. The difference between our resultsand theirs are mostly likely due to the ages ofsubjects; older subjects have lower total and freeIGF-1.30 However, the ratios of free to total IGF-1in both groups are similar between the two assaysystems.

The amount of free/dissociable IGF-1 in serum isdependent on a complex interplay between the pro-duction rate and the concentrations of IGF-1 andIGFBPs. IGFBP-3 serves as major stabilizer of thecirculating IGF pool. Circulating IGFBP-3 mightundergo limited proteolysis, which results in smaller

Table 1 Metabolic and hormonal data in control and obesegroups

Control Obesity

Fasting serum glucose (mmol/L) 5.0 0.5 5.30.6*Total cholesterol (mg/dL) 185.834.5 191.8 39.1Triglyceride (mg/dL) 114.849.3 176.3 38.1*Insulin (pmol/L) 23.5 12.0 49.2 23.8*GH (ng/mL) 2.0 0.8 0.81.1**P< 0.05 control vs obesity.

Figure 1 Levels of total IGF-1, free IGF-1, ratio of free to totalIGF-1, IGFBP-1, IGFBP-2, and IGFBP-3 in control (u) and obesesubjects (j). *P

molecular weight IGFBP-3 fragments that have loweraffinity for IGF-1 than the intact binding protein.Consequently, the levels of free/dissociable IGF-1are elevated in situations where IGFBP-3 proteolysisis increased.31 Bereket et al32 showed that acute meal-related hyperinsulinaemia did not increase IGFBP-3proteolysis. However, it is not known whether IGFBP-3 proteolysis increases in obese subjects.

In contrast, IGFBP-1 exhibits marked diurnal var-iation in relation to meals.32 Insulin is the principalregulator of IGFBP-1 and is reported to inhibit hepaticIGFBP-1 synthesis.33 Because of this property,IGFBP-1 is thought to play an important role in theregulation of IGF bioactivity. Therefore, the obesity-related hyperinsulinaemia may decrease circulatingconcentrations of IGFBP-1, which can result in ele-vated free IGF-1 in obesity. In our study, circulatingIGFBP-1 concentrations in obese subjects were lowerthan normal controls. There were also negative corre-lations between free IGF-1 and IGFBP-1.

IGFBP-2 is the second most abundant IGF-bindingprotein in serum.34 Several reports have shown thatIGFBP-2 is differentially regulated by GH and IGF-1.12,34 Administration of GH causes a reduction inIGFBP-2, while the serum concentration of IGFBP-2is high in states of GH deficiency.34 In normal adults,the infusion of IGF-1 results in suppression of GH andan increase in IGFBP-2.38 The dissociation of theeffect of GH and IGF-1 in regulating IGFBP-2 sug-gests that IGFBP-2 and the ratio of IGFBP-2 to IGF-1may be useful in detecting GH deficiency. Thus, basedon the well-known GH hyposecretion in obesity, thelevels of IGFBP-2 and the ratio of IGFBP-2 to IGF-1would be expected to be high. However, we found thatIGFBP-2 concentrations and the ratios of IGFBP-2 toIGF-1 were suppressed in obese subjects. These find-ings indicate that factors other than GH may beregulating serum IGFBP-2 in obesity. Clemmons etal34 showed that acute stimulation of insulin did notsuppress IGFBP-2, but IGFBP-2 concentrations wereelevated after prolonged fasting. This suggests thatprolonged changes in insulin secretion may result in asignificant change of IGFBP-2. It appears, therefore,that the obesity-related chronic hyperinsulinaemiamay result in suppression of IGFBP-2 in obesity.We also found a negative correlation betweenIGFBP-2 and free IGF-1 concentrations. This indi-cates that IGFBP-2 can quantitatively influence IGF-1transport and may have some role in regulating IGF-1bioavailability in obesity, because serum IGFBP-2 aremore abundant than IGFBP-1 in adults.34

Conclusions

We confirmed that GH-dependent IGF-1 andIGFBP-3 concentrations were not significantly differ-ent between obese and control groups, despite the GH

hyposecretion in obesity. We also found that free IGF-1 concentrations were increased in obesity comparedto normal controls. It seems likely that overnutritionand chronic hyperinsulinaemia in obesity may alterthis regulated growth response: insulin may stimulateproduction of IGF-1 and suppress production ofIGFBP-1 and IGFBP-2, which are thought to beinhibitory IGFBPs.

Acknowledgement

We thank the Ewon Reference Laboratory for themeasurement of serum IGF-1, IGFBPs, GH and insu-lin. This paper was presented at the 15th annualmeeting of the Korean Association of Endocrinologyheld in Seoul 1011th May, 1996.

References1 Barbarino A, De Marinis L, Troncone L. Growth hormone

response to propranolol and L-dopa in obese subjects. Meta-bolism 1978; 27: 275278.

2 William T, Berelowitz M, Joffe SN, Thorner MO, Rivier J,Vale W, Frohman LA. Impaired growth hormone responses togrowth hormone-releasing factor in obesity. A pituitary defectreversed with weight reduction. N Engl J Med 1984; 311:14031407.

3 Lee EJ, Kim KR, Lee KM, Lim SK, Lee HC, Lee JH, Lee DJ,Huh KB. Reduced growth hormone response to L-dopa andpyridostigmine in obesity. Int J Obes 1984; 18: 465468.

4 Lee EJ, Nam SY, Kim KR, Lee HC, Cho JH, Nam MS, SongYD, Lim SK, Huh KB. Acipimox potentiates growth hormone(GH) response to GH-releasing hormone with or withoutpyridostigmine by lowering serum free fatty acid in normaland obese subjects. J Clin Endocrinol Metab 1995; 80: 24952498.

5 Rudman D, Kutner MH, Rogers CM, Lubin MF, Fleming GA,Bain RP. Impaired growth hormone secretion in the popula-tion: Relation to age and adiposity. J Clin Invest 1981; 67:13611369.

6 Copeland KC, Colletti RB, Devlin JT, McAuliffe TL. Therelationship between insulin-like growth factor-1, adiposity,and aging. Metabolism 1990; 39: 584587.

7 Minuto F, Barreca A, Del Monte P, Fortini P, Resentini M,Morabito F. Spontaneous growth hormone and somatomedin-C/insulin-like growth factor-1 secretion in obese subjectsduring puberty. J Endocrinol Invest 1988; 11: 489495.

8 Lamson G, Giudice LC, Rosenfeld RG. Insulin-like growthfactor binding proteins: structural and molecular relationship.Growth Factor 1991; 5: 1928.

9 Baxter WF, Martin JL. Characterization of the acid-labilesubunit of the growth hormone-dependent insulin-likegrowth factor binding protein complex. J Clin EndocrinolMetab 1988; 67: 265272.

10 Mohan S, Baylink DJ (Editorial). Insulin-like growth factor(IGF)-binding proteins in serumDo they have additionalroles besides modulating the endocrine IGF actions? J ClinEndocrinol Metab 1996; 81: 38173820.

11 Blum WF, Ranke MB, Kietzmann K, Gauggel E, Zeisel HJ,Bierich JR. A specific radioimmunoassay for growth hormone(GH)-dependent somatomedin binding protein: its use fordiagnosis of GH deficiency. J Clin Endocrinol Metab 1990;70: 12921298.

12 Smith WJ, Nam TJ, Underwood LE, Busby WH, Celnicker A,Clemmons DR. Use of insulin-like growth factor-bindingprotein-2 (IGFBP-2), IGFBP-3, and IGF-1 for assessinggrowth hormone status in short children. J Clin EndocrinolMetab 1993; 77: 12941299.

IGF-1 and IGFBPs in obesitySY Nam et al

358

13 Suikkari AM, Koivisto VA, Rutanen EM, Yki-Jarvinen H,Karonen SL, Seppala M. Insulin regulates the serum levels oflow molecular weight insulin-like growth factor bindingproteins. J Clin Endocrinol Metab 1988; 66: 266272.

14 Holly JM, Biddlecombe RA, Dunger DB, Edge JA, Amiel SA,Howell R, Chard T, Rees LH, Wass JA. Circardian variationof GH-independent IGF-binding protein in diabetes mellitusand its relationship to insulin. A new role for insulin? ClinEndocrinol (Oxford) 1988; 29: 667675.

15 Busby WH, Snyder DK, Clemmons DR. Radioimmunoassayof a 26,000-dalton plasm an insulin-like growth factor bindingprotein: control by nutritional factors. J Clin EndocrinolMetab 1988; 67: 12251230.

16 Baxter RC, Cowell CT. Diurnal variation rhythm of growthhormone independent binding protein for insulin-like growthfactors in human plasma. J Clin Endocrinol Metab 1987; 65:432440.

17 Conover CA, Lee PD, Kanaley JA, Clarkson JT, Jensen MD.Insulin regulation of insulin-like growth factor binding pro-tein-1 in obese and nonobese humans. J Clin EndocrinolMetab 1992; 74: 13551360.

18 Miell JP, Taylor AM, Jones J, Holly JM, Gaillard RC, PralongFP, Ross RJ, Blum WF. The effect of dexamethasone treat-ment on immunoreactive and bioactive insulin-like growthfactors (IGFs) and IGF-binding proteins in normal malevolunteers. J Endocrinol 1993; 136: 525533.

19 Jones JI, Clemmons DR. Insulin-like growth factors and theirbinding proteins: biological actions. Endocr Rev 1995; 16: 334.

20 Lee PDK, Powell DR, Baker B, Liu F, Belsha CW, BrewerED, Hintz RL. Characterization of a direct non-extractionimmunoradiometric assay for free IGF-1 (Abstract 939).Proceedings of the 76th Annual Meeting of the EndocrineSociety 1994.

21 Caufriez A, Golstein J, Lebrun P, Herchuelz A, Fulanetto R,Copinschi G. Relation between immunoreactive somatome-din-C, insulin, and T3 patterns during fasting in obese sub-jects. Clin Endocrinol 1984; 20: 6570.

22 Phillips L, Unterman T. Somatomedin activity in disorder ofnutrition and metabolism. Clin Endocrinol Metab 1984; 13:145189.

23 Gryzwa M. Serum somatomedin activity and growth-hormonelevel in obese men: Dependence on degree of obesity andhyperlipidemia. Exp Clin Endocrinol 1986; 88: 325330.

24 Soliman A, Hassan A, Aref M, Hintz R, Rosenfeld R, RogolA. Serum IGF-1 and IGF-II concentrations and growth hor-mone and insulin responses to arginine infusion in children

with protein energy malnutrition before and after nutritionalrehabilitation. Pediatr Res 1986; 20: 11221130.

25 Clemmons DR, Klibanski A, Underwood LE, McArthur JW,Ridgway EC, Betins IZ, Van Wyk JJ. Reduction of plasmaimmunoreactive somatomedin-C during fasting in humans.J Clin Endocrinol Metab 1981; 53: 12471251.

26 DeFronzo RA, Ferrannini E. Insulin resistance. A multifacetedsyndrome responsible for NIDDM, obesity, hypertension,dyslipidemia, and athersclerotic cardiovascular disease. Dia-betes Care 1991; 14: 173194.

27 Froesch ER, Guler HP, Schmid C, Zapf J. Therapeuticpotential of insulin-like growth factor 1. Trends EndocrinolMetab 1990; 1: 254260.

28 Cascieri MA, Seperatcin R, Hayes NS, Green BG, ChicchiGG, Applebaum J, Bayne ML. Serum half-life and biologicalactivity of human insulin-like growth factor I which do notbind to serum binding proteins. Endocrinology 1988; 123:373381.

29 Lewitt MS. Role of the insulin-like growth factors in theendocrine control of glucose homeostasis. Diabetes Res ClinPract 1994; 23: 315.

30 Frystyk J, Vestbo E, Skjrbk C, Mogensen CE, rskov H.Free insulin-like growth factors in human obesity. Metabolism1995; 44: 3744.

31 Binoux M, Lalou C, Lasarre C, Blat C, Hossenlopp P. Limitedproteolysis of insulin-like growth factor binding protein-3(IGFBP-3): a physiological mechanism in the regulation ofIGF bioavailability. Current directions in insulin-like growthfactor research. Plenum Press: New York. 1994 pp 293300.

32 Bereket A, Wilson TA, Blethen SL, Fan J, Frost RA, GelatoMC, Lang CH. Effect of short-term fasting on free/dissociableinsulin-like growth factor 1 concentrations in normal humanserum. J Clin Endocrinol Metab 1996; 81: 43794384.

33 Brismar K, Fernqvist-Forbes E, Wahren J, Wahren J, Hall K.Effect of insulin on the hepatic production of insulin-likegrowth factor binding protein (IGFBP)-I, IGFBP-3 and IGF-Iin insulin dependent diabetes. J Clin Endocrinol Metab 1994;79: 872878.

34 Clemmons DR, Snyder DK, Busby WH. Variable controllingthe secretion of insulin-like growth factor binding protein-2 innormal human subjects. J Clin Endocrinol Metab 1991; 73:727233.

35 Young SCJ, Smith-Banks A, Underwood LE, Clemmons DR.Effects of recombinant IGF-1 and GH treatment upon serumIGF binding proteins in calorically restricted adults. J ClinEndocrinol Metab 1992; 75: 603608.

IGF-1 and IGFBPs in obesitySY Nam et al

359