Main

The mechanism of the regulation of fat mass content and energy consumption by leptin via feedback from adipocytes to satiety centers in the hypothalamus has been the subject of intense investigation(1–3). Increased leptin concentrations in obese adults and children suggest that obesity may be associated with changes in this feedback system(4,5). In addition, correction of a sterility defect in female Lepob/Lepob mice by human recombinant leptin and the production of leptin from placental tissue have also been reported(6–9). These findings suggest that leptin may play a role in fertility, pregnancy, and fetal growth.

In earlier studies, we and others(10,11) showed that leptin exists at rather high concentrations in cord blood and correlates positively with birth weight, which suggests that body fat content is a major factor in the regulation of leptin levels even in the fetus. However, to estimate if leptin has any functional role during the fetal and neonatal periods, the leptin levels at different gestational ages and after birth must be investigated. In addition, the fact that serum leptin levels in cord blood vary considerably for subjects with similar birth weight suggests that there may be differences in the genetic control of leptin production at birth.

In the present study, we investigated perinatal changes in serum leptin concentrations from fetuses to neonates as well as the changes in pregnant women. We examined how leptin behaves throughout the perinatal period, in which energy intake changes dramatically from passive trans-placental uptake to active suckling and digestion. In addition, we examined the correlation between the relative leptin level in cord blood and the rate of body weight gain after birth.

MATERIALS AND METHODS

Subjects

Newborns. Venous cord blood was obtained from 139 term healthy newborns (72 males and 67 females) and 20 preterm newborns (12 males and 8 females) whose characteristics are presented in Table 1. The birth weights of all of the newborns were appropriate for their gestational age. Among these term newborns, 24 (11 males and 13 females) were reexamined at approximately 6 d of life (4 to 9 d of life, mean 6.0 ± 1.3 d). The rate of body weight loss (body weight over the period of sampling minus birth weight divided by birth weight multiplied by 100) during the sampling period was -2.4 to 8.3% (mean -2.8 ± 2.9%). The mean calorie intake per day after birth until sampling was 102.5 to 214.1 Kcal/d (mean 158.3 ± 30.5 Kcal/d). Calorie intake during the sampling period was 76.9 to 122.1 Kcal/kg/d (mean 98.8 ± 14.0 Kcal/kg/d). Furthermore, two newborns were longitudinally examined from birth to 17 d.

Table 1 Characteristics of the newborns and serum leptin concentrations
Full size table

Pregnant women. One-hundred forty-three healthy pregnant women were included in the study. All pregnancies were without complications. The mean maternal body mass index (BMI) before pregnancy and at the sample-collecting period during pregnancy was 20.8 ± 3.5 kg/m2 and 23.1 ± 3.5 kg/m2, respectively. The gestational week of the pregnancy ranged from 5 to 41 wk (mean 25.6 ± 11.1 wk), and included 35 first trimester (0-13 wk, mean 9.8 ± 2.5 wk), 35 second trimester (14-27 wk, 22.0 ± 4.2 wk), and 73 third trimester (28-41 wk, 34.9 ± 4.1 wk).

Sampling and Ethics

After sample collection, the serum was immediately separated and frozen at -80°C until the analysis. All of the parents of the newborns and the pregnant women gave their written informed consent before enrollment.

Leptin Assay

Serum leptin concentrations were measured in duplicate by radioimmunoassay as previously described(12). The minimal detectable concentration of leptin was 0.5 ng/mL. The intraassay and interassay coefficients of variation (CVs) were 7.8 and 5.6%, respectively.

Follow-Up of Body Weight Gain after Birth

In 112 infants (56 males and 56 females), we were able to follow up their body weight gain for up to 1 month after birth. The relationship between the serum leptin concentration in the cord blood of these newborns and body weight gain was statistically analyzed.

Statistical Analysis

Differences between groups were evaluated by t test. Pearson correlation and multivariate regression analyses were performed. A log transformation of serum leptin concentrations was performed to normalize the distribution of values. A value of p < 0.05 was considered to reflect significance. Data are presented as the mean and ±SD. All analyses were conducted with SPSS software (version 6.1 for Macintosh, SPSS Inc., Chicago).

RESULTS

Changes in serum leptin levels during the perinatal period. All of the newborns had detectable leptin concentrations except two (a 23-week-gestation male and a 25-week-gestation female). Overall, the serum leptin concentrations in these newborns ranged from 0 (under the limit of detection) to 79.3 ng/mL (mean 17.7 ± 14.4 ng/mL). Serum leptin concentrations in males (mean 13.8 ± 13.1 ng/mL) were significantly lower (p = 0.0001) than those in females (mean 22.4 ± 14.5 ng/mL), as it has been previously observed(10). The leptin concentrations in cord blood at 21-41 wk gestation correlated positively with gestational age (r = 0.598, p < 0.0001) (Fig. 1B) and birth weight (r = 0.635, p < 0.0001). Because the body fat mass of a fetus tends to increase rapidly after 32 wk of gestation(13,14), the linear regression lines between log leptin and gestational week at gestational ages of 21-32 wk and 33-41 wk were compared. The linear regression line between log leptin and gestational week at a gestational age of 21-32 wk was y = 0.005x - 0.208. That in newborns at a gestational age of 33-41 wk was y = 0.105x - 3.012 (r = 0.451, p < 0.0001). The slopes of the linear regression lines were different. The linear regression lines between leptin and gestational week at gestational ages of 21-32 wk and 33-41 wk are shown in Figure 1B.

Figure 1
figure 1

Serum leptin concentrations in pregnant women and cord blood. (A) Serum leptin concentrations and gestational weeks for 143 pregnant women and 29 nonpregnant women. *, p < 0.0001 vs the values in the pregnant women. The closed circles and error bars indicate mean ± SD of the serum leptin concentrations in nonpregnant women and pregnant women in the first, second, and third trimesters. (B) Serum leptin concentrations in cord blood and gestational weeks for 139 newborns. The solid line indicates the linear regression line for newborns at a gestational age of 21-32 wk, y = 0.038x - 0.398. The dotted line indicates the linear regression line for newborns at a gestational age of 33-41 wk, y = 2.804x - 91.487.

Full size image

The serum leptin concentrations of 24 newborns at approximately 6 d of life ranged from 0 (under the limit of detection) to 4.45 ng/mL (mean 1.9 ± 1.1 ng/mL), which is considerably lower than those in corresponding cord blood (mean 20.0 ± 14.7 ng/mL, p < 0.0001) (Fig. 2). These values positively correlated with gestational week (r = 0.598, p < 0.0001), birth weight (r = 0.635, p < 0.0001), Kaup index (weight in g divided by the square of height in centimeters, multiplied by 10) (r = 0.507, p < 0.0001), and body weight/body height (r = 0.613, p < 0.0001). They also inversely correlated with the mean calorie intake per day after birth until sampling (r = -0.522, p = 0.0096). However, they did not correlate significantly with percent body weight loss during the sampling period (r = -0.061, p = 0.7790) or the calorie intake during the sampling period (r = -0.102, p = 0.6560). The sequential analysis of the serum leptin concentration after birth in two newborns showed a rapid decline after birth. This low level was maintained for more than 1 wk and gradually increased thereafter (Fig. 2).

Figure 2
figure 2

Sequential change in leptin concentration in newborns after birth. Leptin concentrations in cord blood and at approximately 6 d after birth were examined in 24 term newborns. The closed circles and error bars indicate the mean ± SD of serum leptin concentrations in cord blood and at approximately 6 d after birth. *, p < 0.0001 vs the value at around 6 d after birth. The sequential analysis of the serum leptin concentration after birth in two newborns is indicated by open or closed squares.

Full size image

Serum leptin concentrations in 143 pregnant women ranged from 2.29 to 102.65 ng/mL (mean 37.2 ± 20.9 ng/mL). These increased from the first trimester and remained significantly higher than those in nonpregnant women (n = 29) (mean 16.7 ± 11.6 ng/mL, range 2.0 to 45.2 ng/mL) even after controlling for BMI (p < 0.0001). These values positively correlated with body weight (r = 0.484, p < 0.0001) and BMI (r = 0.482, p < 0.0001) at the time of sampling. There was a weak negative correlation between gestational week and serum leptin concentration (r = -0.175, p = 0.0359) (Fig. 1A).

The multivariate regression analysis for leptin in term newborns including gender, Kaup index, gestational week, and placental weight as independent variables showed a significant association with gender [standardized regression coefficient (β) = 0.346, p < 0.0001], Kaup index (β = 0.324, p < 0.0001), and gestational week (β = 0.193, p = 0.0107), whereas placental weight had no additional influence (β = 0.080, p = 0.3032). The level of R2 (multiple determination coefficient) observed in this analysis was 0.27 (p < 0.0001). The multivariate regression analysis for leptin in pregnant women including BMI and gestational week as independent variables showed a significant association with BMI (β = 0.581, p < 0.0001) and gestational week (β = -0.343, p < 0.0001). The level of R2 observed in this analysis was 0.34 (p < 0.0001).

Correlation between leptin concentration in cord blood and body weight gain after birth. In female infants, serum leptin concentrations tended to inversely correlate to weight gain (r = -0.326, p = 0.014), whereas this correlation was not significant after controlling for the Kaup index (r = -0.244, p = 0.070) (Table 2). In total and male infants, there was no significant correlation between serum leptin concentration and body weight gain (r = -0.108, p = 0.259; r = 0.195, p = 0.154, respectively).

Table 2 Characteristics of 112 newborns whose body weight gains were followed up and correlations between serum leptin concentrations and various parameters for each group (total, male and female)
Full size table

DISCUSSION

In our study, the serum leptin concentrations in the cord blood of 20 preterm newborns at gestational ages of 21-36 wk were somewhat lower than those of term newborns. Fetal growth does not occur at a uniform rate, and fetal body fat mass tends to increase rapidly after 32 wk of gestation(13,14). Considering the present results from this perspective, the slope of the linear regression line of the correlation between gestational week during 21-32 wk and leptin concentration was quite different from that during 33-41 wk (Fig. 1B). These findings suggest that the serum leptin concentration could increase in correlation with the absolute increase in body fat mass during the fetal period. In addition, the maternal and fetal discordance of circulating leptin concentrations during gestation suggests the possibility that the fetus produces its own leptin at least in part, even though the supply of leptin from the placenta may exist as it has previously been suggested(8,9,11,15). Furthermore, markedly low leptin concentrations during the early gestational stage may support the negative speculation about a critical effect of leptin on fetal growth at least during this stage, in addition to the fact that Lepob/Lepob mouse and human subjects with leptin mutations do not present any abnormalities with respect to fetal growth(16,17). The significance of the expression of leptin receptor in the fetus in the early stage of gestation remains an enigma, and the possibility that leptin can be transported from the maternal circulation to the fetus, functioning as an essential factor during the early gestational stage, must still be clarified.

The serum leptin concentrations in pregnant women increased significantly from the first trimester of the pregnancy, which remained the similar levels throughout the pregnancy in our study. They correlated significantly with BMI. These results show some discrepancies between the report by Masuzaki et al.(8) that emphasized the contribution of placenta for maternal circulating leptin level. However, other reports suggested that the leptin level is positively correlated with fat mass even in pregnancy, as we have observed, and that the effect of pregnancy to the maternal leptin level is mediated not only by the placental supply(15,18,19). In addition, the difference of the way of sample collection (our cross-sectional collection versus their longitudinal one) may contribute to these discrepancies.

The leptin concentration has been said to be influenced by various hormones, especially gonadal steroids. Testosterone lowers leptin production, and estrogen seems to increase leptin production(20,21). Apparent gender differences in leptin concentrations at birth may be attributed to the gonadal steroid status in utero. In addition, the primary genetic differences between males and females may also have an influence as we had previously discussed(10). During the perinatal period, these hormonal conditions change dramatically, especially owing to separation from the placenta. Starvation prevents leptin mRNA expression in neonates, as has been reported for adipose tissue of adult rodents(22,23). During the perinatal period, the energy balance changes dramatically from passive trans-placental uptake to active suckling and digestion after birth. Accordingly, the relatively high concentration of leptin at birth and rapid decrease after birth might be a response to these changes in hormonal and nutritional factors that influence the leptin concentration(23,24). The high concentration of leptin at birth may be favorable to increase heat production in brown adipose tissue during nonshivering thermogenesis for reactions to cold stress(22,25,26).

In a recent study of adult Pima Indians, subjects whose leptin levels were relatively lower were prone to weight gain, suggesting that a relatively low leptin level might be a consequence of the genetic control of leptin production(27). The serum leptin level in cord blood varied considerably in subjects with similar birth weight and Kaup index as in adults(10). We examined the possibility that this variability might be the result of the genetic control of leptin production in each newborn and might, in part, control growth during the neonatal period. As a result, the relationship between the leptin concentration in cord blood and body weight gain after birth is not clear. Further studies of body weight gain in these newborns during childhood will be important in finding the ultimate truth.