In the present study, we observed negative associations between the relative umbilical cord gene expression of the SNURF-SNRPN/UBE3A cluster and several parameters at birth, which were more pronounced in girls. Postnatally, we observed positive associations between the relative expression of this cluster and several postnatal growth parameters, with the associations being more significant in boys. Overall, these results suggest that gene expression within the PWS imprinted domain exhibits a sexual dimorphic association with both prenatal and postnatal growth in apparently healthy children. The differential associations observed at distinct time points (at birth and postnatally) are consistent with the nutritional phases seen in PWS (as described in the Introduction) [6, 7, 24] and may reflect differential effects of PWS genes on nutrition and growth, depending on the developmental stage and sex of the individual.

It is widely accepted that paternally expressed imprinted genes enhance growth, particularly fetal growth [2], and thus, it is expected that the paternally expressed SNURF-SNRPN/UBE3A cluster influences prenatal growth. Furthermore, fetuses have the ability to adapt to their external environment and make anticipatory responses due to their developmental plasticity [25]. In our study of healthy infants, we observed that smaller infants at birth had increased expression of growth-related genes. This may indicate a compensatory mechanism to improve prenatal growth, whereby a paternal line drives maternal resources to maximize fetal nutrition and optimize growth in utero and in infancy prior to weaning [26]. This proposed mechanism of compensation provides the fetus with a survival advantage [27], since fetal growth is a crucial factor that determines health and disease risk later in childhood and adulthood [28].

MAGEL2 is known to be highly expressed in the hypothalamus [29] and plays a role in the regulation of whole-body energy metabolism [30]. It is closely associated with circadian rhythms [5], changes in leptin sensitivity [5, 31, 32] and the regulation of muscle mass, among others [32]. In this study, we found that MAGEL2 expression in boys positively correlated with infant growth from the first months of life to 6 years, highlighting a previously unknown relationship between MAGEL2 expression and growth in apparently healthy infants. SNORD116 is thought to have a crucial role in PWS [5], since different alterations in this gene, including microdeletions, have been related to PWS-like phenotypes [33]. Furthermore, Zhang et al. [34] showed differences in SNORD116 expression at birth compared to postnatal stages such as weaning and adulthood, indicating that its expression may be regulated throughout development. Our results also show that SNORD116 correlations at birth and in the following years move in opposite directions, supporting the idea that this gene could have an important role in the PWS growth phenotype and is developmentally regulated. Finally, SNORD115 has been also linked to the obese phenotype observed in PWS, because it is involved in the regulation of alternative splicing of the serotonin receptor 2C [35], which is known to influence appetite regulation and energy balance [36,37,38]. Stronger associations in infancy were found in the present study, where the expression SNORD115 showed a positive correlation with infant growth at various time points.

We found that the gene expression of the SNURF-SNRPN/UBE3A cluster was positively associated with several parameters of postnatal growth, both during the first year of life and until 6 years of age. This suggests that the expression of the studied genes may influence postnatal growth beyond birth and into early childhood. Notably, infants with PWS typically develop obesity later in life, suggesting a strong postnatal developmental component to PWS-associated obesity. Some authors have hypothesized that the delayed onset of weight gain in early infancy observed in PWS may be attributed to the loss of MAGEL2, where its expression may progressively favor the development of leptin insensitivity postnatally, thereby contributing to the obesity phenotype associated with PWS [29, 32], as demonstrated in animal models [39]. Others have reported that abnormalities in oxytocin secretion contribute to the development of obesity in PWS patients, as oxytocin regulates food intake, energy expenditure, thermogenesis, and muscle tone and mass at different developmental stages in these individuals [40].

We identified sexually dimorphic differences in the associations of the SNURF-SNRPN/UBE3A domain with prenatal and postnatal growth. In girls, the associations at birth were found to be stronger, while in boys, the associations with postnatal growth were more evident. Although many imprinted genes are differentially expressed according to sex [14], we did not find any sex differences in the expression within our population. Our results support the notion that the phenotype of PWS is mainly acquired via hypothalamic dysfunction, including alterations in growth hormone (GH) secretion, abnormal body composition [41], and impaired thermoregulation [42]. The hypothalamus is a region of the brain that is highly dimorphic according to sex, which could partially explain the sexually dimorphic associations observed in our study [34]. In addition, the secretion of GH in adult rats has also been found to be sexually dimorphic, and both neonatal and adult steroid environments can influence the adult GH secretory pattern [43]. Fluctuations in sex hormone levels that lead to a different regulation of the hypothalamic–pituitary axis may potentially account for the differences observed between boys and girls during infancy, coinciding with the activation of the hypothalamic–pituitary–gonadal axis, a phenomenon known as minipuberty [44]. During this period, boys experience a significant increase in testosterone, which may impact their somatic development, while no similar effect has been observed in girls [45]. This testosterone surge could also influence the role of the studied genes in somatic growth. Sex-specific associations of imprinted genes with weight parameters at birth and postnatally have also been reported. In a study focused on methylation of imprinted genes in cord blood leukocytes rather than their expression, differences were found between boys and girls when examining associations with parameters of weight status, such as birth weight, weight for length at 1 year of age, and BMI at 3 years of age [46].

It was also observed that the relative expression of the studied genes in the umbilical cord was associated with postnatal growth curves during infancy and up to 4 years of age. Specifically, higher expression of MAGEL2, SNORD116, and SNORD115 in boys was associated with greater postnatal growth. This indicates that the expression of these genes in the umbilical cord is not only associated with early-postnatal growth but can also help predict weight gain during the first years of life. This observation aligns with the theory postulated by Campbell [47], which states that paternal genes enhance postnatal growth by promoting appetite and suckling ability. Further, as mentioned previously, patients with PWS experience dysfunction of the hypothalamic–pituitary axis [48], which regulates the secretion of GH, a potent stimulator of growth [49]. Thus, it is not surprising that genes implicated in PWS affect postnatal growth, potentially acting as important growth regulators at the hypothalamic level.

Finally, we also found that the cord expression of MAGEL2, SNORD116, and SNORD115 contributed to explaining the sexually dimorphic pattern observed in early-postnatal growth, with divergent postnatal growth curves between boys and girls being apparent in infants with gene expression levels above the median. However, in multivariate models that included all infants studied, gene expression, as a dichotomous variable, was not directly associated with growth parameters, likely because such associations were only evident the group with higher gene expression. Notwithstanding these findings, gene expression did show persistent significant interactions with sex in its association with growth parameters. These results indicate that gene expression in the PWS imprinted domain may promote the well-known sexual dimorphism in humans, where boys grow faster at an earlier age. We suggest that higher expression levels of these imprinted genes interact with the testosterone surge during minipuberty in boys, which may help explain, at least in part, the dimorphic pattern of early human postnatal growth. However, further research is required to validate these hypotheses.

This study is the first of its kind to establish an association between genes related to PWS and normal prenatal and postnatal growth in healthy children. Addressing this knowledge gap is crucial for understanding how growth is regulated, which can help prevent potential complications later in life. Furthermore, the umbilical cord, a transitory tissue that is easily accessible for sampling, can possess prognostic value, as indicated in this study by the prediction of postnatal growth trajectories based on cord gene expression [50].

It should be noted that one limitation of this study is that we did not assess either DNA methylation status or protein levels of the studied genes due to limited resources. Although DNA methylation plays a key regulatory role in imprinted genes, expression is downstream of DNA regulatory factors and therefore gene expression is potentially more functionally relevant. In addition, while birth weight was used as a proxy for prenatal growth, ultrasound data collected throughout pregnancy would provide a more accurate quantification of prenatal growth. In addition, given the small sample size of our study, further longitudinal studies are needed to corroborate our findings.

To summarize, we report that the expression levels of paternally expressed genes from the PWS domain in the umbilical cord were negatively associated with prenatal growth but positively with early-postnatal growth in apparently healthy infants (Supplementary Table 6). This suggests that these genes may provide an advantage during the perinatal period and early-postnatal life, first to the mother–fetus pair and then to the young infant. In boys, the expression of these genes in the umbilical cord could reflect early effectors of growth patterns and may help predict weight gain during the first years of postnatal life. Lastly, gene expression levels of MAGEL2, SNORD116, and SNORD115 may contribute to the well-known sexual dimorphism in humans, whereby boys grow faster at an earlier age.