New components of human breast milk (BM), especially regulatory hormones (RH) of food intake and glucose/lipid metabolism (adiponectin, leptin, ghrelin, adipocyte fatty acid binding protein [AFABP], and other), are considered to influence nutritional status and possibly play a role in the development of components of metabolic syndrome later in adulthood. The positive role of breast-feeding in decreasing incidence of various civilization diseases is mentioned in many scientific articles (1); however, the components of BM responsible for this effect have not been identified so far. The concept of programming is based on the hypothesis that factors that influence organisms in the critical developmental period determine the risk of diseases later in adulthood. It is presumed that these risks may be avoided by targeted specific nutrition during pregnancy and the early postnatal period (effect of specific nutrients—eg, minerals, trace metals, vitamins, antioxidants, fatty acids, RHs).
Adiponectin is a circulating hormone produced by adipocytes that plays an important role in the regulation of glucose metabolism, increases insulin sensitivity, slows down progression of atherosclerotic changes, and is considered to be a biomarker of metabolic syndrome. Serum adiponectin concentrations inversely correlate with obesity and type 2 diabetes mellitus (2). Presence of adiponectin in human BM was published in 2006 (3,4). During fetal growth, adiponectin is inversely related to insulin, leptin, and body weight. Both hormones are also present in cord blood and placenta. Adiponectin and leptin plasma concentrations were reported to be positively related with birth weight and with adiposity and the body mass index (BMI) at birth. This suggests that adipokines may play a key role in fetal development. In neonates, adiponectin levels are 2 to 3 times higher than in adults, but these levels seem to be largely independent of maternal influences (5).
Fatty acid–binding proteins regulate intracellular transport of fatty acids to various metabolic pathways and participate in the pathogenesis of metabolic syndrome (6). AFABP and epidermal fatty acid–binding proteins were also detected in human BM in 2006 (4).
Leptin is an adipocyte-derived hormone that decreases food intake and reflects the amount of body fat. Its presence in colostrum and mature BM was reported in 1997 (7,8), and the concentration in BM correlates with the concentration in serum of both mothers and sucklings (9,10) and maternal BMI (4). Fully breast-fed infants have higher serum levels of leptin than formula-fed infants (11). Leptin receptors have been identified in gastric epithelial cells and in the absorptive cells of mouse and human small intestine (12,13). These findings suggest that leptin could pass from milk to infant blood and may play a role in the short-term regulation of feeding acting as a satiety signal and could also exert a long-term effect on energy balance and body weight (14,15).
On the basis of the above-mentioned facts, we hypothesize that RH present in BM may influence the organism of infants. The aim of our project was to characterize precisely the intraindividual changes of concentrations of RH in BM during 12 months of lactation and to describe the relation of RH concentrations to anthropometric data and concentrations of other RH in BM.
SUBJECTS AND METHODS
Mothers delivering their children at the Department of Gynaecology and Obstetrics of the Motol University Hospital during the period autumn 2006—summer 2008 were asked to participate in the study. Only mothers with noncomplicated vaginal delivery of a full-term infant were included. The mothers were selected from a large cohort of mothers who collected BM samples within another prospective study concerning allergy incidence in infants. Only BM samples from mothers who breast-fed their child at least up to 6 months were selected for our study. A total of 327 samples were collected from 72 lactating mothers enrolled in the study. Thirty of them gave birth to a boy and 42 of them to a girl. We collected samples from all of the mothers at 1 month (M1), 3 months (M3), and 6 months (M6), and 39 samples at 12 months (M12). We obtained additional data (maternal prepregnancy BMI, complications of pregnancy, infants' birth weight, birth length, sex, gestational age, body weight at M1, M3, M6, and M12) from questionnaires and medical records. There were only infants born from noncomplicated pregnancies included in the study. All of the newborns were born at term, their average birth weight was 3472 ± 42 g, birth length was 50.2 ± 0.3 cm, and average BMI of mothers before pregnancy was 21.9 ± 0.4 kg/m2 (all mean ± standard error of the mean [SEM]). All of the mothers declared that they did not smoke during lactation. None of the mothers was an alcohol addict. We have no detailed data on occasional alcohol intake or detailed dietary habits during pregnancy and lactation. The average body weight of infants at M1, M3, M6, and M12 was 4045 ± 53 g, 5911 ± 86 g, 7621 ± 110 g, and 9590 ± 125 g, respectively (all mean ± SEM). The study was approved by the ethical committee of University Hospital Motol, and informed consent was obtained from all of the participants.
Collection of Samples
Five milliliters of BM was collected into sterile tubes (containing ethylenediaminetetraacetic acid and protease inhibitor) by manual expression or with a breast pump, divided into aliquots, and frozen at −20°C for further analysis. Colostrum was collected 48 hours after the beginning of lactation—either on the third or the fourth day after delivery (day 0, D0) and mature milk at M1, M3, M6, and M12 of lactation. All of the samples were collected after the first morning breast-feeding (after 7:00 AM) from the same breast used for breast-feeding of the infant.
Whole BM samples were stored frozen until analysis. We assayed adiponectin in whole milk and leptin and AFABP in skim milk. Before analysis, the samples were thawed at 4° to 6°C overnight and centrifuged at 2500g at 4°C for 20 minutes to separate the fat milk. We removed the fat layer with a spatula and used the liquid for assays. The centrifugation step was repeated for turbid samples. We performed all of the determinations of adiponectin, leptin, and AFABP with commercially available enzyme-linked immunosorbent assay (ELISA) kits (BioVendor, Modrice, Czech Republic) as described previously (4).
Statistical analysis was performed using Prism 5.0 statistical software (GraphPad Software, San Diego, CA). Values were tested for normality of distribution. Results are reported as mean ± SEM or as median in the case of non-Gaussian distribution. The differences within groups were tested using the Kruskal-Wallis test with the Dunn multiple comparison test. Correlations are expressed by the Spearman correlation coefficient. A P value of <0.05 was considered statistically significant.
Adiponectin levels in BM on D0 were 22.8 ± 0.8 (mean ± SEM); in M1, 22.0 ± 0.6; in M3, 20.5 ± 0.6; in M6, 21.4 ± 0.8; and in M12, 25.7 ± 1.4 ng/mL (Fig. 1). None of the samples was under the detection limit of the ELISA method. We found significantly higher levels of adiponectin in M12 in comparison with M3 and M6 (overall P = 0.0026).
AFABP levels were not normally distributed. AFABP levels on D0 were 5.5; in M1, 2.3; in M3, 0.4; in M6, 0.03; and in M12, 0.3 ng/mL (median). Mean levels on D0 were 12.3 ± 2.0; in M1, 6.2 ± 1.3; in M3, 1.3 ± 0.2; in M6, 2.5 ± 1.0; and in M12, 4.6 ± 1.9 ng/mL (Fig. 2). On D0, 5% of samples were under the detection limit; in M1, 12%; in M3, 37%; in M6, 50%; and in M12, 50%. We found significantly higher levels of AFABP in D0 and M1 when compared with M3, M6, and M12 (P < 0.0001).
Leptin levels were not normally distributed. Leptin levels on D0 were 0.3; in M1, 0.1; in M3, 0.04; in M6, 0.02; and in M12, 0.05 ng/mL (median). Mean levels on D0 were 0.3 ± 0.04; in M1, 0.2 ± 0.03; in M3, 0.1 ± 0.01; in M6, 0.1 ± 0.02; and in M12, 0.2 ± 0.04 ng/mL (Fig. 3). On D0, 7% of samples were under the detection limit; in M1, 31%; in M3, 37%; in M6, 44%; and in M12, 41%. We found significantly higher levels of leptin on D0 than in M1, M3, M6, and M12 (P < 0.0001).
There was no correlation throughout lactation between body weight of infants and BM levels of measured parameters except for AFABP at M1 (r = −0.2738, P = 0.0228). AFABP (r = −0.2644, P = 0.0469) and leptin (r = −0.2666, P = 0.0450) correlated negatively with body length on D0. None of the parameters correlated with prepregnancy BMI of mothers. There was no correlation found between body weight gain during the first year of life and BM levels of any RH at any time point up to 12 months of lactation except for adiponectin at M6 (Spearman r = 0.2774, P = 0.0488). We did not find any difference in BM concentrations of any RH between mothers who delivered boys versus girls at any time point up to 12 months of lactation. We found positive correlation between all 3 parameters (adiponectin, AFABP, and leptin) on D0, in M6, and in M12, except for adiponectin versus leptin in M6. Correlation between adiponectin and AFABP/leptin, respectively, on D0 and M12 is shown in Figure 4.
In our study, to our knowledge, we show for the first time dynamic intraindividual changes of adiponectin, leptin, and AFABP in human BM during 12 months of lactation and the relation of their BM levels to each other and to anthropometrical parameters of the infants. All of the analytes were well detectable in colostrum, and adiponectin was also well detectable in all of the samples throughout lactation up to M12. However, leptin and AFABP concentrations seem to vary greatly since M1, and in many of the samples of mature BM these hormones are under the detection limit of the ELISA method. In spite of this fact, when all of the measurable samples are taken into account, all 3 proteins show U-shaped dynamics of concentrations during lactation with a decreasing trend up to M3 and higher levels in M6 and M12, respectively. This fact has not been previously described in the literature and may be explained by introduction of complementary feeding with subsequent longer intervals between breast-feeding. Thus, morning samples provided for analysis at M6/M12 are possibly more concentrated than morning samples up to M3 during exclusive breast-feeding, including night dose for the child. In 2006, Martin et al (3) reported negative association of adiponectin and month of lactation up to 7 months postpartum with approximately 6.9 ng/mL lower adiponectin in 7 months when compared with 1 week (5.72%/mo decrease in concentration). We did not confirm these findings and, on the contrary, we show significantly higher adiponectin levels at M12; however, Martin et al (3) used the radioimmunoassay method and skim milk samples for detection and different method of BM collection (draining an entire breast with the use of an electric pump). Moreover, the time of day of collection may play a role because our samples were collected during the first morning breast-feeding, and Martin et al (3) collected samples from 10 AM to 1 PM. There were no studies performed on the diurnal variability of concentrations of these proteins in human BM. Lack of measurements of adiponectin in fore- and hindmilk is a common limitation of all of the studies reported to date. Uçar et al (10) report no such difference for leptin concentrations. Weyermann et al (16) measured adiponectin and leptin in 6 weeks postpartum and in a randomly selected small cohort also at 6 months of lactation using the ELISA method from skim milk. They show higher median concentrations of both hormones at 6 months when compared with samples collected at 6 weeks (16).
Our findings that adiponectin BM levels are several times higher than leptin levels are in accordance with studies of other authors, and absolute values of adiponectin and leptin BM concentrations correspond to previously reported data (3,16). It seems that concentrations measured by ELISA in whole-milk samples are comparable with levels detected by radioimmunoassay in skim milk at least in colostrum. During lactation, other factors may play a role in differences reported using various methods.
In our study, there was no correlation throughout lactation between the body weight of infants and the BM levels of measured parameters except for AFABP at M1. Savino et al (17) have shown that serum, but not BM, leptin levels correlate positively with infant weight and BMI. Weyermann et al (16) did not show any relation between adiponectin BM concentrations and birth weight except for a cohort of infants with birth weight <3000 g, in whom adiponectin BM levels were significantly higher. In small-for-gestational age infants, adiponectin levels were significantly lower. Because we included only healthy term infants with a normal range of birth weight, we could not confirm the above-mentioned findings.
There was a borderline positive correlation found between body weight gain during the first year of life and adiponectin BM concentration at M6. It seems that adiponectin may be predictive for body weight gain of the infant, but longer follow-up of children is probably necessary (at least up to 2 years of age) to confirm this hypothesis.
A positive correlation between adiponectin and leptin BM levels in the first several months of lactation was reported in other studies (3,16). We confirm these findings and extend it up to 12 months of lactation. Correlation with AFABP levels in BM has not been reported yet to our knowledge; thus, we cannot make a comparison of our data with other studies.
It is questionable whether the presence of RH in BM may have physiological implications for the development of infants. The effects of BM leptin and adiponectin on the human neonate are not known. BM leptin is related to maternal plasma leptin and adiposity (7,9) and to serum leptin in infants (10). Savino et al (11,17) described higher serum leptin levels in breast-fed infants than in formula-fed, suggesting that additional leptin may be transferred from mothers through BM. There is a study available showing leptin transfer from mothers' milk to infants' stomachs, and serum in experimental animals (8) and leptin receptors were identified in the human digestive tract (13). Oral administration of leptin to neonatal rats decreases food intake and lowers leptin production in the stomach and subcutaneous adipose tissue. Leptin concentrations influence hypothalamic neurons involved in regulation of food intake and thus play a role in the programming of appetite control during early postnatal life.
Adiponectin levels remain well detectable throughout the whole course of lactation. Therefore, prolonged breast-feeding may be an additional source of this adipokine in addition to adipokine production by infant adipose tissue. Higher levels of adiponectin in BM were associated with overweight at 2 years of age in infants who were breast-fed for at least 6 months of age (18). We could not confirm these findings because the follow-up period in our study was too short. In a recent article, Woo et al (19) described association of higher milk adiponectin with lower infant weight-for-age z score and weight-for-length z score. The same study group assessed the structural form of adiponectin in BM by Western blot showing predominantly the biologically active high-molecular-weight form, which may suggest physiological effect in the neonate.
In our study, we demonstrated intraindividual variability of adiponectin, leptin, and AFABP BM levels in a large study group. All of the hormones were detectable in BM up to 12 months of lactation with a decreasing trend in levels until M3 and a subsequent increase to M12. Higher levels in the second half of the lactation period may be caused by longer intervals between breast-feeding due to the introduction of complementary food. We speculate that these hormones may play a role in nutritional programming of infants, probably during the whole lactation period. Additional prospective studies are needed to confirm this hypothesis.
1. Savino F, Fissore MF, Liguori SA, et al
. Can hormones contained in mothers' milk account for the beneficial effect of breast-feeding on obesity in children? Clin Endocrinol (Oxf) 2009; 71:757–765.
2. Nedvidkova J, Smitka K, Kopsky V, et al
. Adiponectin, an adipocyte-derived protein. Physiol Res 2005; 54:133–140.
3. Martin LJ, Woo JG, Geraghty SR, et al
. Adiponectin is present in human milk and is associated with maternal factors. Am J Clin Nutr 2006; 83:1106–1111.
4. Bronsky J, Karpisek M, Bronska E, et al
. Adiponectin, adipocyte fatty acid binding protein, and epidermal fatty acid binding protein: proteins newly identified in human breast milk. Clin Chem 2006; 52:1763–1770.
5. Savino F, Petrucci E, Nanni G. Adiponectin: an intriguing hormone for paediatricians. Acta Paediatr 2008; 97:701–705.
6. Stejskal D, Karpisek M, Bronsky J. Serum adipocyte-fatty acid binding protein discriminates patients with permanent and temporary body weight loss. J Clin Lab Anal 2008; 22:380–382.
7. Houseknecht KL, McGuire MK, Portocarrero CP, et al
. Leptin is present in human milk and is related to maternal plasma leptin concentration and adiposity. Biochem Biophys Res Commun 1997; 240:742–747.
8. Casabiell X, Piñeiro V, Tomé MA, et al
. Presence of leptin in colostrum and/or breast milk from lactating mothers: a potential role in the regulation of neonatal food intake. J Clin Endocrinol Metab 1997; 82:4270–4273.
9. Uysal FK, Onal EE, Aral YZ, et al
. Breast milk leptin: its relationship to maternal and infant adiposity. Clin Nutr 2002; 21:157–160.
10. Uçar B, Kirel B, Bör O, et al
. Breast milk leptin concentrations in initial and terminal milk samples: relationships to maternal and infant plasma leptin concentrations, adiposity, serum glucose, insulin, lipid and lipoprotein levels. J Pediatr Endocrinol Metab 2000; 13:149–156.
11. Savino F, Liguori SA, Fissore MF, et al
. Looking for a relation between serum leptin concentration and body composition parameters in healthy term infants in the first 6 months of life. J Pediatr Gastroenterol Nutr 2008; 46:348–351.
12. Barrenetxe J, Villaro AC, Guembe L, et al
. Distribution of the long leptin receptor isoform in brush border, basolateral membrane, and cytoplasm of enterocytes. Gut 2002; 50:797–802.
13. Aparicio T, Kermorgant S, Darmoul D, et al
. Leptin and Ob-Rb receptor isoform in the human digestive tract during fetal development. J Clin Endocrinol Metab 2005; 90:6177–6184.
14. Stocker CJ, Cawthorne MA. The influence of leptin on early life programming of obesity. Trends Biotechnol 2008; 26:545–551.
15. Palou A, Picó C. Leptin intake during lactation prevents obesity and affects food intake and food preferences in later life. Appetite 2009; 52:249–252.
16. Weyermann M, Beermann C, Brenner H, et al
. Adiponectin and leptin in maternal serum, cord blood, and breast milk. Clin Chem 2006; 52:2095–2102.
17. Savino F, Liguori SA, Petrucci E, et al
. Evaluation of leptin in breast milk, lactating mothers and their infants. Eur J Clin Nutr 2010; 64:972–977.
18. Weyermann M, Brenner H, Rothenbacher D. Adipokines in human milk and risk of overweight in early childhood: a prospective cohort study. Epidemiology 2007; 18:722–729.
19. Woo JG, Guerrero ML, Altaye M, et al
. Human milk adiponectin is associated with infant growth in two independent cohorts. Breastfeed Med 2009; 4:101–109.