J C E M B r i e f
R e p o r t — E n d o c r i n e
O N L I N E
R e s e a r c h
Less Myostatin and More Lean Mass in Large-Born Infants From Nondiabetic Mothers Francis de Zegher,* Míriam Pérez-Cruz,* Marta Díaz, María Dolores Gómez-Roig, Abel López-Bermejo, and Lourdes Ibáñez Department of Pediatric Endocrinology (F.d.Z.), University Hospital Gasthuisberg, 3000 Leuven, Belgium; Endocrinology Unit (M.D., L.I.), and Department of Obstetrics & Gynecology (M.P.-C., M.D.G.-R.), Hospital Sant Joan de Déu, University of Barcelona, 08950 Esplugues, Barcelona, Spain; Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas (M.D., L.I.), Instituto de Salud Carlos III, 28029 Madrid, Spain; and Department of Pediatrics (A.L.-B.), Dr Josep Trueta Hospital, and Girona Institute for Biomedical Research, 17007 Girona, Spain
Context and Objective: Sexagenarians born large are at lower risk for type 2 diabetes than those born small, a key feature of their body composition being a higher muscle mass, which explains their higher body mass index and also their lower fat-to-lean-mass ratio. Myogenesis is completed in early infancy under the inhibitory control of myostatin. We tested whether large-born infants from nondiabetic mothers develop an early surplus of lean mass while having a lower myostatinemia. Design, Methods, Study Participants, and Main Outcomes: In a longitudinal study (0 – 4 mo), we compared the body composition and endocrine markers (fasting glucose, insulin, IGF-1, high molecular weight adiponectin) of breast-fed appropriate- vs large-for-gestational-age infants (n ⫽ 125) from nondiabetic mothers. Circulating myostatin concentrations were assayed after collection of the above-mentioned data. Setting: The study was conducted at the University Hospital for Women and Children. Intervention: There were no interventions. Results: Between 0 – 4 months, large-for-gestational-age infants switched from an adipose to a lean body composition (due to a nearly 20% excess of lean mass) and to an insulin-sensitive and hyperadiponectinemic state while having low IGF-1 concentrations and the lowest myostatinemia hitherto reported in the human (all between P ⱕ .01 and P ⱕ .0001). Conclusion: Large-born infants from nondiabetic mothers were found to combine a low myostatinemia with an excess of lean mass. The fetal-neonatal control of myostatinemia deserves further attention because it could become a target of interventions that aim at reducing the risk for diabetes in later life by augmenting myogenesis in early life. (J Clin Endocrinol Metab 99: E2367–E2371, 2014)
exagenarians born large are at lower risk for type 2 diabetes than those born small (1), a key feature of their body composition being a higher muscle mass (2, 3) that explains their higher body mass index (BMI) (4) and also their lower fat to lean mass ratio (3). large-for-ges-
S
tational-age (LGA) infants from obese/diabetic mothers are markedly adipose at birth (4), and they may be at higher, rather than lower, risk for later type 2 diabetes (5). Myogenesis is completed in early infancy and influenced negatively by myostatin (6), a conserved myokine of
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received May 13, 2014. Accepted August 6, 2014. First Published Online August 20, 2014
* F.d.Z. and M.P.-C. contributed equally to this work. Abbreviations: AGA, appropriate for gestational age; BMI, body mass index; CV, coefficient of variation; HMW, high molecular weight; LGA, large for gestational age.
doi: 10.1210/jc.2014-2334
J Clin Endocrinol Metab, November 2014, 99(11):E2367–E2371
jcem.endojournals.org
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:28 For personal use only. No other uses without permission. . All rights reserved.
E2367
E2368
de Zegher et al
Lean Mass and Myostatin in Breast-Fed LGA Infants J Clin Endocrinol Metab, November 2014, 99(11):E2367–E2371
the TGF- superfamily (7). Myostatin inhibits the differentiation of myoblasts and of brown, but not white, adipocytes (8, 9). Early lowering of myostatin actions is thus a potential strategy to reduce the risk for later diabetes (10, 11). We tested whether large-born infants from nondiabetic mothers develop an early surplus of lean mass while also having a lower myostatinemia.
Study Population, Design, and Methods Longitudinal study population (infants aged 0 – 4 mo) The study population consisted of 125 infants [95 born appropriate-for-gestational-age (AGA) and 30 born LGA enrolled between January 2007 and December 2011 into a longitudinal study of body composition and the endocrine state during early infancy (12) or in an extension thereof (Supplemental Figure 1). The inclusion criteria for the present analysis were as follows: 1) birth at Hospital Table 1.
Sant Joan de Déu (Barcelona, Spain) after an uncomplicated, term (37– 42 wk), singleton pregnancy, with no hypertension, no alcohol or drug abuse, no diabetes mellitus [normal O’Sullivan test in the second trimester (AGA) or in the second and the third trimesters (LGA)]; 2) birth weight Z-score between ⫺1 and ⫹1 for gestational age [AGA birth weight range was 2.9 –3.8 kg] or a Z-score above ⫹2 for gestational age [LGA birth weight range was 3.9 – 4.9 kg]; 3) exclusive breast-feeding for 4 months; 4) serum available for endocrine assessments at 0 and 4 months; 5) body composition assessments available, by absorptiometry (10), at 2 weeks and 4 months; and 6) written informed consent by at least one parent in the Spanish or Catalan language. Exclusion criteria were complications at birth (resuscitation or parenteral nutrition) or evidence of a congenital malformation. All AGA and LGA infants were studied with the same methods. Blood was sampled at birth (from the umbilical
Characteristics of Control Infants Born AGA and Infants Born LGA AGA
n Girls, % Mother Pregestational weight, kg Pregestational BMI, kg/m2 Gestational weight gain, kg Primigravida, n, % Birth Gestational age, wk Birth weight, kg Birth length, cm Body composition Age, d Length, cm Weight, kg BMC, g Fat mass, kg Lean mass, kg Fat to lean mass ratio Endocrinology Glucose, mg % Insulin, mIU/L HOMA-IR IGF-1, ng/mL HMW-adiponectin, mg/L Myostatin, ng/mLe
LGA
0 mo
4 mo
0 – 4 mo
0 mo
4 mo
0 – 4 mo
95 52
95 52
95 52
30 50
30 50
30 50
62 ⫾ 1 24.3 ⫾ 0.4 12 ⫾ 1 66 (70)
– – – –
– – – –
64 ⫾ 2 27.7 ⫾ 1.0a 13 ⫾ 1 20 (67)
– – – –
– – – –
39.8 ⫾ 0.1 3.4 ⫾ 0.1 49.8 ⫾ 0.2
– – –
– – –
40.2 ⫾ 0.2 4.3 ⫾ 0.1b 52.4 ⫾ 0.3b
– – –
– – –
14 ⫾ 1 51.0 ⫾ 0.2 3.8 ⫾ 0.1 109 ⫾ 2 0.8 ⫾ 0.1 3.0 ⫾ 0.1 0.26 ⫾ 0.01
129 ⫾ 1 63.4 ⫾ 0.3 7.2 ⫾ 0.1 193 ⫾ 4 2.7 ⫾ 0.1 4.2 ⫾ 0.1 0.67 ⫾ 0.02
115 ⫾ 1 12.4 ⫾ 0.2 3.3 ⫾ 0.1 84 ⫾ 4 2.0 ⫾ 0.1 1.3 ⫾ 0.1 0.41 ⫾ 0.02
15 ⫾ 1 54.6 ⫾ 0.4b 4.8 ⫾ 0.1b 128 ⫾ 5b 1.2 ⫾ 0.1b 3.5 ⫾ 0.1b 0.34 ⫾ 0.02b
132 ⫾ 2 65.9 ⫾ 0.3b 8.0 ⫾ 0.1b 186 ⫾ 9 2.8 ⫾ 0.1 5.0 ⫾ 0.1b 0.57 ⫾ 0.03c
117 ⫾ 1 11.3 ⫾ 0.4 3.1 ⫾ 0.1 58 ⫾ 7a 1.6 ⫾ 0.1a 1.4 ⫾ 0.1 0.23 ⫾ 0.02b
– – – 54 ⫾ 3 36 ⫾ 2 22 ⫾ 1
88 ⫾ 1 5.3 ⫾ 0.6 1.2 ⫾ 0.2 46 ⫾ 2 31 ⫾ 1 31 ⫾ 2
– – – ⫺8 ⫾ 3 ⫺5 ⫾ 2 9⫾2
– – – 56 ⫾ 5 31 ⫾ 2 16 ⫾ 2a
87 ⫾ 1 2.5 ⫾ 0.5d 0.5 ⫾ 0.1a 33 ⫾ 2b 42 ⫾ 4a 21 ⫾ 3a
– – – ⫺22 ⫾ 5c 11 ⫾ 4b 5⫾2
Abbreviation: BMC, bone mineral content; HOMA-IR, homeostasis model assessment index of insulin resistance. AGA and LGA infants were from nondiabetic mothers and were exclusively breast-fed for 4 months. Values are mean ⫾ SEM. a
P ⱕ .01 vs AGA.
b
P ⱕ .0001 vs AGA.
c
P ⱕ .05 vs AGA.
d
P ⱕ .001 vs AGA.
e
AGA (n ⫽ 60) and LGA (n ⫽ 23).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:28 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-2334
jcem.endojournals.org
2.0
Z-socre aat 4 months
***
**
**
*
**
**
1.5 1.0 0.5 0.0 -0.5 -1.0 HOMA-IR
IGF-I Myo
BMC Fat/Lean
HMWadipo Fat Mass
Lean Mass
Figure 1. Z-score values for endocrine and body composition features of 30 breast-fed LGA infants at 4 months. Z-score values were calculated from mean and SD results in 95 breast-fed AGA infants at 4 months. BMC, bone mineral content; Fat/Lean, fat to lean mass ratio; HMWadipo, HMW-adiponectin; Myo, myostatin. *, P ⬍ .05; **, P ⬍ .01; ***, P ⬍ .001 vs breast-fed AGA controls.
cord) and at age 4 months in the early morning, prior to breast-feeding. Body composition was assessed by dual X-ray absorptiometry with a Lunar Prodigy, coupled to Lunar software (Lunar Corp), adapted for assessment of infants. Body fat and lean mass were assessed during natural sleep. Coefficients of variation (CVs) were less than 3% for lean and fat mass (10). The original studies and the present analysis were approved by the Institutional Review Board of Barcelona University, Hospital Sant Joan de Déu. Part of the body composition and endocrine results from a fraction of AGA controls was reported (12). Assays After blood sampling and centrifugation, serum was deep frozen at ⫺80°C. Circulating glucose was measured by the glucose oxidase method. Serum concentrations of insulin and IGF-1 were measured by immunochemiluminescence (Immulite 2000; Diagnostic Products Corp), with the intraassay CVs less than 10%. High-molecularweight (HMW) adiponectin was assessed by an ELISA (Linco). Finally, total myostatin was measured (60 AGA infants, 23 LGA infants) with a competitive ELISA (Immunodiagnostik) detecting the different forms of circulating myostatin with polyclonal antimyostatin antibodies having no cross-reactivity with structurally related compounds; the detection limit was 0.8 ng/mL; intra- and interassay CVs were 2.3% and 5.0%. The same assay disclosed recently that adults have an average myostatinemia of approximately 30 ng/mL with an approximate SD of 10 ng/mL (13). Statistics Statistical analyses were performed using SPSS version 12.0 (SPSS Inc). Skewed data were log transformed before comparison. General linear models for repeated measure-
E2369
ments were used to detect longitudinal differences in continuous variables and differences in these variables at consecutive time points. Independent associations between myostatinemia and birth weight were sought by step-wise multiple regression analysis, adjusting for gender, gestational age, maternal weight, parity, and LGA status. Values of P ⬍ .05 were considered statistically significant.
Results
Mothers of LGA infants had an average pregestational BMI in the overweight range and had a gestational weight gain similar to that of the mothers of AGA infants. Postnatal gains of length and weight were similar in AGA and LGA infants, but LGA infants partitioned their weight gain so that their fat to lean ratio evolved from an adipose to a lean level (Table 1). At 4 months, LGA infants had the same fat mass as AGA infants but strikingly more lean mass; in addition, LGA infants had low insulin and IGF-1 concentrations and high concentrations of HMW adiponectin (Figure 1). The IGF-1 data become more relevant when embedded in a broader context (Figure 2). All the above-mentioned AGA vs LGA differences were similar in girls and boys. Myostatin was detectable in all samples (60 AGA, 23 LGA), and concentrations were similar in girls and boys. At birth, myostatin concentrations were low in AGA infants [compared with indicative values in adults (13, 14)], and they were even lower in LGA infants. Myostatin concentrations were inversely associated with birth weight ( ⫽ ⫺.40, P ⫽ .001, adjusted for gender, gestational age, and maternal weight and parity;  ⫽ ⫺.10, P ⫽ .02, after further adjustment for LGA status at birth). Between 0 and 4 months, the myostatin levels rose similarly in AGA and LGA infants, the mean increment in pooled AGA-LGA infants being 7.5 ng/mL (95% confidence interval 4.3– 11.0 ng/mL; P ⬍ .0001). At 4 months, myostatin concentrations of AGA infants were in the adult range, with those of LGA infants still being lower (Figure 1).
Discussion Breast-fed LGA infants from nondiabetic mothers were found to develop a marked surplus of lean mass by the end of early infancy while maintaining low levels of circulating myostatin. By age 4 months, LGA infants had low fasting
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:28 For personal use only. No other uses without permission. . All rights reserved.
E2370
de Zegher et al
Lean Mass and Myostatin in Breast-Fed LGA Infants J Clin Endocrinol Metab, November 2014, 99(11):E2367–E2371
∆ 0-4 mo IGF-I (ng/mL)
4 mo IGF-I (ng/mL)
p
R e p o r t — E n d o c r i n e
O N L I N E
R e s e a r c h
Less Myostatin and More Lean Mass in Large-Born Infants From Nondiabetic Mothers Francis de Zegher,* Míriam Pérez-Cruz,* Marta Díaz, María Dolores Gómez-Roig, Abel López-Bermejo, and Lourdes Ibáñez Department of Pediatric Endocrinology (F.d.Z.), University Hospital Gasthuisberg, 3000 Leuven, Belgium; Endocrinology Unit (M.D., L.I.), and Department of Obstetrics & Gynecology (M.P.-C., M.D.G.-R.), Hospital Sant Joan de Déu, University of Barcelona, 08950 Esplugues, Barcelona, Spain; Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas (M.D., L.I.), Instituto de Salud Carlos III, 28029 Madrid, Spain; and Department of Pediatrics (A.L.-B.), Dr Josep Trueta Hospital, and Girona Institute for Biomedical Research, 17007 Girona, Spain
Context and Objective: Sexagenarians born large are at lower risk for type 2 diabetes than those born small, a key feature of their body composition being a higher muscle mass, which explains their higher body mass index and also their lower fat-to-lean-mass ratio. Myogenesis is completed in early infancy under the inhibitory control of myostatin. We tested whether large-born infants from nondiabetic mothers develop an early surplus of lean mass while having a lower myostatinemia. Design, Methods, Study Participants, and Main Outcomes: In a longitudinal study (0 – 4 mo), we compared the body composition and endocrine markers (fasting glucose, insulin, IGF-1, high molecular weight adiponectin) of breast-fed appropriate- vs large-for-gestational-age infants (n ⫽ 125) from nondiabetic mothers. Circulating myostatin concentrations were assayed after collection of the above-mentioned data. Setting: The study was conducted at the University Hospital for Women and Children. Intervention: There were no interventions. Results: Between 0 – 4 months, large-for-gestational-age infants switched from an adipose to a lean body composition (due to a nearly 20% excess of lean mass) and to an insulin-sensitive and hyperadiponectinemic state while having low IGF-1 concentrations and the lowest myostatinemia hitherto reported in the human (all between P ⱕ .01 and P ⱕ .0001). Conclusion: Large-born infants from nondiabetic mothers were found to combine a low myostatinemia with an excess of lean mass. The fetal-neonatal control of myostatinemia deserves further attention because it could become a target of interventions that aim at reducing the risk for diabetes in later life by augmenting myogenesis in early life. (J Clin Endocrinol Metab 99: E2367–E2371, 2014)
exagenarians born large are at lower risk for type 2 diabetes than those born small (1), a key feature of their body composition being a higher muscle mass (2, 3) that explains their higher body mass index (BMI) (4) and also their lower fat to lean mass ratio (3). large-for-ges-
S
tational-age (LGA) infants from obese/diabetic mothers are markedly adipose at birth (4), and they may be at higher, rather than lower, risk for later type 2 diabetes (5). Myogenesis is completed in early infancy and influenced negatively by myostatin (6), a conserved myokine of
ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received May 13, 2014. Accepted August 6, 2014. First Published Online August 20, 2014
* F.d.Z. and M.P.-C. contributed equally to this work. Abbreviations: AGA, appropriate for gestational age; BMI, body mass index; CV, coefficient of variation; HMW, high molecular weight; LGA, large for gestational age.
doi: 10.1210/jc.2014-2334
J Clin Endocrinol Metab, November 2014, 99(11):E2367–E2371
jcem.endojournals.org
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:28 For personal use only. No other uses without permission. . All rights reserved.
E2367
E2368
de Zegher et al
Lean Mass and Myostatin in Breast-Fed LGA Infants J Clin Endocrinol Metab, November 2014, 99(11):E2367–E2371
the TGF- superfamily (7). Myostatin inhibits the differentiation of myoblasts and of brown, but not white, adipocytes (8, 9). Early lowering of myostatin actions is thus a potential strategy to reduce the risk for later diabetes (10, 11). We tested whether large-born infants from nondiabetic mothers develop an early surplus of lean mass while also having a lower myostatinemia.
Study Population, Design, and Methods Longitudinal study population (infants aged 0 – 4 mo) The study population consisted of 125 infants [95 born appropriate-for-gestational-age (AGA) and 30 born LGA enrolled between January 2007 and December 2011 into a longitudinal study of body composition and the endocrine state during early infancy (12) or in an extension thereof (Supplemental Figure 1). The inclusion criteria for the present analysis were as follows: 1) birth at Hospital Table 1.
Sant Joan de Déu (Barcelona, Spain) after an uncomplicated, term (37– 42 wk), singleton pregnancy, with no hypertension, no alcohol or drug abuse, no diabetes mellitus [normal O’Sullivan test in the second trimester (AGA) or in the second and the third trimesters (LGA)]; 2) birth weight Z-score between ⫺1 and ⫹1 for gestational age [AGA birth weight range was 2.9 –3.8 kg] or a Z-score above ⫹2 for gestational age [LGA birth weight range was 3.9 – 4.9 kg]; 3) exclusive breast-feeding for 4 months; 4) serum available for endocrine assessments at 0 and 4 months; 5) body composition assessments available, by absorptiometry (10), at 2 weeks and 4 months; and 6) written informed consent by at least one parent in the Spanish or Catalan language. Exclusion criteria were complications at birth (resuscitation or parenteral nutrition) or evidence of a congenital malformation. All AGA and LGA infants were studied with the same methods. Blood was sampled at birth (from the umbilical
Characteristics of Control Infants Born AGA and Infants Born LGA AGA
n Girls, % Mother Pregestational weight, kg Pregestational BMI, kg/m2 Gestational weight gain, kg Primigravida, n, % Birth Gestational age, wk Birth weight, kg Birth length, cm Body composition Age, d Length, cm Weight, kg BMC, g Fat mass, kg Lean mass, kg Fat to lean mass ratio Endocrinology Glucose, mg % Insulin, mIU/L HOMA-IR IGF-1, ng/mL HMW-adiponectin, mg/L Myostatin, ng/mLe
LGA
0 mo
4 mo
0 – 4 mo
0 mo
4 mo
0 – 4 mo
95 52
95 52
95 52
30 50
30 50
30 50
62 ⫾ 1 24.3 ⫾ 0.4 12 ⫾ 1 66 (70)
– – – –
– – – –
64 ⫾ 2 27.7 ⫾ 1.0a 13 ⫾ 1 20 (67)
– – – –
– – – –
39.8 ⫾ 0.1 3.4 ⫾ 0.1 49.8 ⫾ 0.2
– – –
– – –
40.2 ⫾ 0.2 4.3 ⫾ 0.1b 52.4 ⫾ 0.3b
– – –
– – –
14 ⫾ 1 51.0 ⫾ 0.2 3.8 ⫾ 0.1 109 ⫾ 2 0.8 ⫾ 0.1 3.0 ⫾ 0.1 0.26 ⫾ 0.01
129 ⫾ 1 63.4 ⫾ 0.3 7.2 ⫾ 0.1 193 ⫾ 4 2.7 ⫾ 0.1 4.2 ⫾ 0.1 0.67 ⫾ 0.02
115 ⫾ 1 12.4 ⫾ 0.2 3.3 ⫾ 0.1 84 ⫾ 4 2.0 ⫾ 0.1 1.3 ⫾ 0.1 0.41 ⫾ 0.02
15 ⫾ 1 54.6 ⫾ 0.4b 4.8 ⫾ 0.1b 128 ⫾ 5b 1.2 ⫾ 0.1b 3.5 ⫾ 0.1b 0.34 ⫾ 0.02b
132 ⫾ 2 65.9 ⫾ 0.3b 8.0 ⫾ 0.1b 186 ⫾ 9 2.8 ⫾ 0.1 5.0 ⫾ 0.1b 0.57 ⫾ 0.03c
117 ⫾ 1 11.3 ⫾ 0.4 3.1 ⫾ 0.1 58 ⫾ 7a 1.6 ⫾ 0.1a 1.4 ⫾ 0.1 0.23 ⫾ 0.02b
– – – 54 ⫾ 3 36 ⫾ 2 22 ⫾ 1
88 ⫾ 1 5.3 ⫾ 0.6 1.2 ⫾ 0.2 46 ⫾ 2 31 ⫾ 1 31 ⫾ 2
– – – ⫺8 ⫾ 3 ⫺5 ⫾ 2 9⫾2
– – – 56 ⫾ 5 31 ⫾ 2 16 ⫾ 2a
87 ⫾ 1 2.5 ⫾ 0.5d 0.5 ⫾ 0.1a 33 ⫾ 2b 42 ⫾ 4a 21 ⫾ 3a
– – – ⫺22 ⫾ 5c 11 ⫾ 4b 5⫾2
Abbreviation: BMC, bone mineral content; HOMA-IR, homeostasis model assessment index of insulin resistance. AGA and LGA infants were from nondiabetic mothers and were exclusively breast-fed for 4 months. Values are mean ⫾ SEM. a
P ⱕ .01 vs AGA.
b
P ⱕ .0001 vs AGA.
c
P ⱕ .05 vs AGA.
d
P ⱕ .001 vs AGA.
e
AGA (n ⫽ 60) and LGA (n ⫽ 23).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:28 For personal use only. No other uses without permission. . All rights reserved.
doi: 10.1210/jc.2014-2334
jcem.endojournals.org
2.0
Z-socre aat 4 months
***
**
**
*
**
**
1.5 1.0 0.5 0.0 -0.5 -1.0 HOMA-IR
IGF-I Myo
BMC Fat/Lean
HMWadipo Fat Mass
Lean Mass
Figure 1. Z-score values for endocrine and body composition features of 30 breast-fed LGA infants at 4 months. Z-score values were calculated from mean and SD results in 95 breast-fed AGA infants at 4 months. BMC, bone mineral content; Fat/Lean, fat to lean mass ratio; HMWadipo, HMW-adiponectin; Myo, myostatin. *, P ⬍ .05; **, P ⬍ .01; ***, P ⬍ .001 vs breast-fed AGA controls.
cord) and at age 4 months in the early morning, prior to breast-feeding. Body composition was assessed by dual X-ray absorptiometry with a Lunar Prodigy, coupled to Lunar software (Lunar Corp), adapted for assessment of infants. Body fat and lean mass were assessed during natural sleep. Coefficients of variation (CVs) were less than 3% for lean and fat mass (10). The original studies and the present analysis were approved by the Institutional Review Board of Barcelona University, Hospital Sant Joan de Déu. Part of the body composition and endocrine results from a fraction of AGA controls was reported (12). Assays After blood sampling and centrifugation, serum was deep frozen at ⫺80°C. Circulating glucose was measured by the glucose oxidase method. Serum concentrations of insulin and IGF-1 were measured by immunochemiluminescence (Immulite 2000; Diagnostic Products Corp), with the intraassay CVs less than 10%. High-molecularweight (HMW) adiponectin was assessed by an ELISA (Linco). Finally, total myostatin was measured (60 AGA infants, 23 LGA infants) with a competitive ELISA (Immunodiagnostik) detecting the different forms of circulating myostatin with polyclonal antimyostatin antibodies having no cross-reactivity with structurally related compounds; the detection limit was 0.8 ng/mL; intra- and interassay CVs were 2.3% and 5.0%. The same assay disclosed recently that adults have an average myostatinemia of approximately 30 ng/mL with an approximate SD of 10 ng/mL (13). Statistics Statistical analyses were performed using SPSS version 12.0 (SPSS Inc). Skewed data were log transformed before comparison. General linear models for repeated measure-
E2369
ments were used to detect longitudinal differences in continuous variables and differences in these variables at consecutive time points. Independent associations between myostatinemia and birth weight were sought by step-wise multiple regression analysis, adjusting for gender, gestational age, maternal weight, parity, and LGA status. Values of P ⬍ .05 were considered statistically significant.
Results
Mothers of LGA infants had an average pregestational BMI in the overweight range and had a gestational weight gain similar to that of the mothers of AGA infants. Postnatal gains of length and weight were similar in AGA and LGA infants, but LGA infants partitioned their weight gain so that their fat to lean ratio evolved from an adipose to a lean level (Table 1). At 4 months, LGA infants had the same fat mass as AGA infants but strikingly more lean mass; in addition, LGA infants had low insulin and IGF-1 concentrations and high concentrations of HMW adiponectin (Figure 1). The IGF-1 data become more relevant when embedded in a broader context (Figure 2). All the above-mentioned AGA vs LGA differences were similar in girls and boys. Myostatin was detectable in all samples (60 AGA, 23 LGA), and concentrations were similar in girls and boys. At birth, myostatin concentrations were low in AGA infants [compared with indicative values in adults (13, 14)], and they were even lower in LGA infants. Myostatin concentrations were inversely associated with birth weight ( ⫽ ⫺.40, P ⫽ .001, adjusted for gender, gestational age, and maternal weight and parity;  ⫽ ⫺.10, P ⫽ .02, after further adjustment for LGA status at birth). Between 0 and 4 months, the myostatin levels rose similarly in AGA and LGA infants, the mean increment in pooled AGA-LGA infants being 7.5 ng/mL (95% confidence interval 4.3– 11.0 ng/mL; P ⬍ .0001). At 4 months, myostatin concentrations of AGA infants were in the adult range, with those of LGA infants still being lower (Figure 1).
Discussion Breast-fed LGA infants from nondiabetic mothers were found to develop a marked surplus of lean mass by the end of early infancy while maintaining low levels of circulating myostatin. By age 4 months, LGA infants had low fasting
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2015. at 04:28 For personal use only. No other uses without permission. . All rights reserved.
E2370
de Zegher et al
Lean Mass and Myostatin in Breast-Fed LGA Infants J Clin Endocrinol Metab, November 2014, 99(11):E2367–E2371
∆ 0-4 mo IGF-I (ng/mL)
4 mo IGF-I (ng/mL)
p
