Introduction

Nutrition is the most important factor that influences growth during early childhood. The importance of nutrition during early childhood growth is further emphasized by the marked differences in the growth rates of breastfed (BF) and formula-fed (FF) infants [1, 16]. Current studies show that different growth patterns have been observed depending on the feeding practices. More pronounced growth rates have been observed in FF infants than in BF infants after the first 3 months of life. In the second 6 months of life, BF infants exhibit a progressive decline in their growth rates, while FF infants exhibit a continuous increase [1, 16, 19, 24, 35]. This pattern highlights the potential importance of breastfeeding in the prevention of obesity [3, 30, 34, 41]. The causes of the differences in the growth patterns of BF and FF infants still remain poorly understood. The main determinant of growth in this period is the feeding preference. The ingredients of formulas are not similar to those of breast milk, and excessive oral intake has been observed in FF infants [2, 24, 30, 34]. On the contrary, BF infants self-regulate their energy intake to match their energy needs on the basis of their hunger and satiety. This self-regulating mechanism might be associated with the changing content of human milk during breastfeeding and its effects on energy metabolism. The stable content of formulas during nutrition might also be associated with a lack of self-regulation during feeding and increased nutrition.

Human milk contains many bioactive substances, such as immunoglobulins, enzymes, hormones, growth factors, and other bioactive peptides. Some of these proteins and all of these peptides are secreted in biologically active forms. Fat, leptin, ghrelin, insulin, adiponectin, obestatin, resistin, epidermal growth factor, platelet-derived growth factor, and insulin-like growth factor 1 are bioactive substances that play roles in energy intake and the regulation of body composition [13, 20, 30, 38, 41]. Fat levels are increased at the end of the breastfeeding period and prevent overfeeding via a satiety signal [18, 32, 40]. Ghrelin, a hormone also found in human milk, is produced in the gastrointestinal tract and acts on the central nervous system to regulate appetite and weight [41, 45]. A portion of ghrelin has a unique fatty acid modification (n-octanoylation) at Ser 3 [22, 28]. Of the two forms of ghrelin, acylated and deacylated ghrelin, the acylated form (known as active ghrelin) is thought to be essential for binding to the growth hormone secretagogue receptor 1a [15, 22, 28]. Deacylated ghrelin, however, is not totally inactive. It has influences on both cell proliferation and adipogenesis and counteracts the metabolic effects of active ghrelin [6, 10]. The total ghrelin level is the sum of the acylated and deacylated ghrelin levels. Leptin is a signaling protein that informs the brain about body fat stores and, at the hypothalamic level, is a hormone that increases the metabolic rate and decreases appetite [8, 37]. The aim of this study was to evaluate the foremilk and hindmilk leptin, ghrelin, and fat levels, the changes in these levels with age, a possible association between these levels and growth, and a possible association between feeding preferences and the self-regulation of feeding in term healthy newborns.

Materials and methods

Subjects and study design

This study was performed in a cohort of 62 healthy infants (from 30 to 91 days old) and 46 nursing mothers in Eskisehir, Turkey. The study procedure was approved by the local ethics committee of Eskisehir Osmangazi University, and informed consent was received from all of the participants. This study was financially endorsed by the Eskisehir Osmangazi University Scientific Study Research Commission (200711033).

All of the participating infants were evaluated twice. The first visit was performed at 1 to 3 months of age (rapid growth period), and the second visit was performed at 4 to 6 months of age (the period of decreased growth and before the start of complementary feeding). The infants were subdivided (according to their nutrition) into breastfed (BF), formula-fed (FF), and BF plus FF (BF + FF) groups. Detailed physical examinations and anthropometric evaluations were performed by the same physician (Z.K.) at the first and second visits. Weights and heights were evaluated using CDC 2000 data. The body mass indices were calculated using the formula \( {\hbox{weight }}\left( {\hbox{kg}} \right){ }/{\left[ {{\hbox{height }}\left( {\hbox{m}} \right)} \right]^2} \).

The exclusion criteria were as follows: deliveries after complicated pregnancies, infants that were small for their gestational age, preterm infants, infants with any existing disease, and infants with feeding practices that changed from the first to the second visit.

Collection of milk samples

Breast milk samples (6 ml) were obtained after 2 h of fasting at 9–10 a.m. at the beginning (foremilk) and at the end of nursing (hindmilk) at both the first and second visits. Each sample (2 ml) was stored in an Eppendorf tube to measure its triglyceride and total cholesterol levels. The remaining 4-ml human milk samples were stored at −80°C to measure their active ghrelin, total ghrelin, and leptin levels. Before storage, all of the samples were centrifuged at 4,000 rpm for 15 min; the supernatant lipid layers were removed, and the samples were centrifuged again. One set of milk samples was acidified by adding 1 N HCl (10% of volume; pH 3–4) to stabilize the labile side chain of active ghrelin and prevent rapid ghrelin deacylation, which has been described previously [21]. To protect the peptides from proteolysis, 500 IU (10 μL) of aprotinin was added per milliliter of sample [2].

Hormone assays

The total and active ghrelin levels were studied using a radioimmune assay (RIA) using Linco Ghrelin RIA kits. The leptin levels were studied using ELISA with the DRG Human Leptin kit. The breast milk triglyceride and total cholesterol levels were analyzed after homogenization (with vortexing) and 1:30 dilution using enzymatic colorimetric methods using a Shimadzu UV-VIS Spectrophotometer and the Triglycerides GPO and Cholesterol CHOD-PAP Biolabo kits.

Statistics

Statistical analysis was performed using SPSS version 13.0. The appropriately distributed (parametric) data are given as arithmetic means ± standard deviation (SD), and the inappropriately distributed (nonparametric) data are given as medians (minimum–maximum). The comparisons between groups were performed using the Student’s t test or Mann–Whitney U test for parametric and nonparametric data, respectively. Changes in the concentrations of the parameters in the milk with time (e.g., at the first visit vs. the second visit) were examined using paired samples test or the Mann–Whitney U test for parametric and nonparametric data, respectively. Correlations were assessed with Pearson’s test or Spearman’s test for parametric and nonparametric data, respectively. A p value less than 0.05 was considered to be significant in all of the tests.

Results

Sixty-two healthy term newborns (26 girls and 36 boys, 30 to 91 days old) were enrolled. Forty-two percent (n = 26) of the infants were fully BF, 25.8% (n = 16) were FF, and 32.2% (n = 20) were in the BF + FF group. The gender differences and gestational ages were similar (p > 0.05) across the groups. The birth weights were lower in the FF group than in the BF group (p < 0.05). The ages at the first and second visits were similar across the groups. On the first visit, the BMI was higher in the BF group than in the BF + FF group (p < 0.05); however, their BMIs were similar on the second visit.

The weight gain and the weight gain percentage were higher in the FF infants than in the BF + FF infants between birth and the first visit (p < 0.05). However, the weight gain and the weight gain percentage were similar in all of the study groups between the first and second visits. The weight gain percentage was higher in the FF infants than in the BF and BF + FF infants between birth and the second visit (p < 0.01 and p < 0.05, respectively). The BMIs were significantly higher in the FF and BF + FF infants than in the BF infants between the first and second visits (p < 0.01; Table 1 and Fig. 1).

Table 1 Demographic and anthropometric measures of study groups
Full size table
Fig. 1
figure 1

Weight (a) and body mass index (BMI) (b) changes between first and second visit according to nutrition

Full size image

On the first visit, the total ghrelin, active ghrelin, and total cholesterol levels were higher in the foremilk, while higher triglyceride levels were observed in the hindmilk (p < 0.001) of the BF group. The leptin levels were also higher in the hindmilk, but this difference was not statistically significant (p > 0.05). Like in the BF group, higher levels of total ghrelin, active ghrelin, and total cholesterol were observed in the foremilk of the BF + FF group than in the hindmilk (p < 0.001 for all), while higher leptin and total cholesterol levels were observed in the hindmilk (p < 0.001 and p < 0.01, respectively; Table 2, Fig. 2).

Table 2 Foremilk and hindmilk hormone and lipid levels according to nutrition at first and second visit
Full size table
Fig. 2
figure 2

Comparison between foremilk and hindmilk tG-HH, leptin, TG and TC levels in BF and BF + FF groups at first visit

Full size image

At the second visit, as at the first visit, the total and active ghrelin and total cholesterol levels in the foremilk were higher than in the hindmilk in the BF and BF + FF groups (p < 0.001 for all of the parameters in the BF group; p < 0.01 for the total ghrelin levels and p < 0.001 for the active ghrelin and total cholesterol levels in the BF + FF group). The triglyceride levels in the hindmilk were higher than in the foremilk in both the BF and BF + FF groups (p < 0.001). The leptin levels in the hindmilk were higher than in the foremilk in the BF + FF group, but this difference was not statistically significant (p > 0.05; Table 2). The foremilk leptin levels were lower in the BF + FF group than in the BF group at both the first and second visits (p < 0.01 and p < 0.05, respectively; Table 3).

Table 3 Comparison of foremilk hormone and lipid levels according to groups in first and second visit
Full size table

At the second visit, decreased total ghrelin (p < 0.01), leptin (p < 0.05), triglyceride (p < 0.001), and total cholesterol (p < 0.01) levels and increased active ghrelin levels (p < 0.001) were observed in the foremilk of the BF group compared to the first visit. In the foremilk of the BF + FF group, decreased total ghrelin, total cholesterol, and triglyceride levels were observed (p < 0.01 for all). Decreased leptin and increased active ghrelin levels were also observed, but their changes were not statistically significant (p > 0.05; Table 4).

Table 4 Foremilk hormone and lipid levels in BF and BF + FF groups at first and second visit
Full size table

The active ghrelin levels were positively correlated with the triglyceride levels in both the foremilk and the hindmilk of the BF group at the first visit (p < 0.05). The foremilk leptin levels were negatively correlated with the total ghrelin and total cholesterol levels in the BF and BF + FF groups (p < 0.05 and p < 0.01, respectively). There was also no correlation between the active and total ghrelin levels. The total ghrelin levels were negatively correlated with the leptin levels in the BF group (p < 0.05).

Discussion

The growth patterns of BF and FF infants differ. One proposed mechanism is that breastfeeding, compared with bottlefeeding, is more responsive to the infant cues of hunger and satiety. Breastfed infants learn and develop better control of their milk intake, whereas bottlefeeding may decrease the infants’ abilities to self-regulate milk intake [30]. The duration and amount of bottlefeeding might depend on the caregivers’ decisions on the basis of visual cues regarding the amount of milk remaining in the bottle [17, 34]. The current results also indicate that bottlefeeding may be more important than the type of the milk in the bottle in limiting the infants’ ability to self-regulate milk intake [30]. Milk composition and taste vary with each breastfeeding, between feeds, and over time, whereas they remain constant during bottlefeeding. While the self-control mechanism is not clearly defined, it may be associated with the changing amounts of human milk that are available in each breastfeeding period. In our study, we evaluated the growth patterns and levels of ghrelin, leptin, and fat levels in the foremilk and hindmilk of BF infants.

In our study, the birth weights were lower in the FF group than in the BF group. At the first visit, the weight gain was higher in the FF group than in the BF and BF + FF groups, but this difference was not statistically significant. The weight gain percentage between birth and the first visit was 41.2% in the FF group, 35.4% in the BF group, and 31.5% in the BF + FF group. Although the weight gain of the fully BF babies was larger than that of the FF babies in the first 3 months, few differences in weight gain have been reported [16, 19, 35]. Additionally, the older age at the first visit and the lower case number in the FF group (compared with the BF group) may affect the weight gain between birth and the first visit, although these effects were not statistically significant. The weight gain between birth and the first visit was significantly higher in the FF group than in the BF + FF group.

At the first visit, the calculated BMIs were higher in the BF group than in the BF + FF and FF groups, but these differences were not statistically significant. At the second visit, the BMIs were similar across the study groups. There was a significant increase in the BMIs of the FF and BF + FF groups (14.9% and 11.8%, respectively), and a small increase was observed in the BF group (3.5%). These findings indicate that weight gain is more prominent than increase in height in the FF group, and these growth patterns (height and weight) are proportional in BF infants. These findings also indicate that FF infants exhibit significant weight gain during early infancy, and this gain continues from the second to the fifth month of life. The rate of weight gain decreased in the BF infants after the first 3 months of life.

Previous studies have shown that the fat content of human milk is much higher toward the end of a feeding episode than at the start of the episode. This might signal to the infant that the feeding episode is ending [18, 32, 40]. To the best of our knowledge, we are the first to demonstrate decreased total cholesterol levels despite increased triglyceride levels in the hindmilk of BF infants at the first and second visits. Aside from the fat content of human milk, current studies have shown that some hormones in human milk have roles in energy metabolism, appetite, and food intake [41]. The presence of ghrelin in human milk was first described by Aydin et al. [2] in 2006. However, the foremilk and hindmilk ghrelin levels have not been studied. Based on our results, the total and active ghrelin levels in the hindmilk were significantly decreased on the first (declines of 36.9% and 27.6%, respectively) and second visits (declines of 29.3% and 20.1%, respectively). The decreased total and active ghrelin levels in the hindmilk have the potential to result in satiety and decreased appetite and signal the completion of breastfeeding.

This decline in breast milk during feeding is not clearly defined. Some studies have shown that an infusion of glucose and elevated fat levels decrease ghrelin levels [5, 7, 26, 32, 45]. One potential explanation is that decreased high-density lipoprotein cholesterol (HDL-C) levels parallel the decreased total cholesterol levels, resulting in decreased levels of total ghrelin, which uses HDL-C as a transporter [4, 45]. We were unable to determine the HDL-C levels in breast milk; thus, it is difficult to estimate this association according to our results. Current reports have demonstrated that the potential sources of ghrelin in human milk are the maternal plasma or the tissue in the mother’s breast [25, 27, 45]. Based on our study results, lower hindmilk ghrelin levels on the first and second visits might indicate that human milk ghrelin is synthesized or regulated in breast tissue during breastfeeding. A few published papers have reported that leptin may be produced by different cell types in mammary tissue. In addition to leptin synthesis, epithelial cells may transfer leptin from the blood, and these two mechanisms may account for the presence of leptin in milk [9, 12].

Contradictory published results have been published regarding the presence of leptin in formula. Although several studies [29, 39, 44] have reported the presence of leptin in formula, O’Connor et al. [36] did not report this. Leptin levels may be low in formula that is prepared using pasteurized low-fat milk. Additionally, breast milk nutrients are lost during tube feeding [42]. Thus, leptin and/or ghrelin pass through the baby (via the formula) and remain low. To the best of our knowledge, there have been no reports regarding the presence of ghrelin in formula. Even if ghrelin were present in formula, it is stable during feeding and may not affect appetite control. Similarly, the leptin and fat concentrations in formula are stable during feeding. Gastric leptin is involved in the short-term regulation of digestion; in contrast, the leptin that is secreted by white adipocytes primarily acts on the hypothalamus for the long-term regulation of food intake [8, 11, 37]. Different studies have shown either increased or unchanged leptin levels after meals [7, 14, 33]. Some studies have shown higher or similar leptin levels in the hindmilk compared with the foremilk [31, 43]. Leptin and ghrelin are found at higher levels in whole milk than in skim milk, probably because of its association with milk fat globules [23, 27, 39]. In our study, increased hindmilk leptin and triglyceride levels in the BF and BF + FF groups at both visits support the positive association between them. Additionally, we found that the active ghrelin levels in breast milk positively correlate with the triglyceride levels. Increased hindmilk leptin levels at the end of breastfeeding coupled with decreased excessive food intake resulted in BF infants with lower weights than FF infants. The study found decreased levels of total ghrelin, triglycerides, and total cholesterol in the foremilk of both groups at the second visit compared with levels at the first visit. However, the foremilk leptin levels only decreased in the BF group. We thought that the decreased leptin and total ghrelin levels were associated with decreased levels of triglycerides and total cholesterol at the second visit. There was no significant decrease in the foremilk leptin levels at the second visit compared with levels at the first visit in the BF + FF group. Additionally, we also observed lower foremilk leptin levels in the BF + FF group than in BF group at both visits. These findings might be related to the larger increase in BMI between the first and second visit in the BF + FF group than in the BF group. The greater level of breast milk leptin in fully BF babies might have played a role in their faster growth in the first 3 months of age and might have protected against later obesity.

A number of limitations must be noted in the present study. In particular, the sample size was small, and the birth weights were lower in the FF group than in the BF group. The ghrelin and HDL-C levels should be evaluated in the foremilk and hindmilk. There are many hormones that affect energy metabolism other than the ghrelin and leptin in breast milk. Long-term prospective studies are needed to reveal the effects of various energy metabolism-related hormones on feeding patterns and growth.

Conclusions

Weight gain is more prominent in FF children than in BF children. Changing the levels of ghrelin, leptin, triglycerides, and total cholesterol in the foremilk and hindmilk showed that ghrelin and fats have potential roles in the self-control of feeding. The stable content of formulas during nutrition might also be associated with the lack of self-regulation during feeding and increased nutrition. The study also demonstrated that human milk ghrelin, leptin, and fat levels change with age in fully BF infants. These physiological states are disturbed by formula feeding. In the future, formulas with variable levels of hormones and fat could be developed using new technologies and potentially prevent obesity in FF infants. Additionally, a better understanding of self-regulation in response to human milk may result in increased breastfeeding and potentially prevent obesity.