Exploring role of maternal and infant determinants on glucocorticoid composition of human milk

Abstract

Human milk (HM) is a dynamic and multifaceted complex mammary secretion, optimised for the nourishment and immune protection of infants. The exact nutritional composition of HM is highly variable. The health, stages of development, physiological and psychological status of both mother and infant are major contributors to this variability, and may coincide with the needs of the feeding infant. HM is enriched with non-nutritive maternally originated hormones. Their biological roles in infants are not well defined, although they may alter infant physiological and psychological functions. Emerging evidence shows maternal originated milk-borne glucocorticoids play critical roles in affecting offspring metabolic programming and behavioural development. To date, studies have focused on the impact of maternal social, biological and environmental factors on macronutrient content of milk such as lipids, proteins and carbohydrates. Despite the importance of HM bioactive compounds, studies investigating milk glucocorticoids are quite limited, with focus on a small number of factors influencing its levels in mother’s milk. The object of this thesis was to investigate glucocorticoids composition of HM with respect to maternal biological and social factors. Given the current void in understanding the roles of these glucocorticoids on infant health, some analysis of the relationship between infant characteristics and HM glucocorticoids was also undertaken. The first study involved a large collaborative analysis of selected maternal factors, potentially important in determining the concentration of HM glucocorticoids. This study analysed the impact of a wide range of maternal social, biological, and environmental related factors on milk glucocorticoids composition in HM samples, collected at 3 months of established lactation. From 650 milk samples, we demonstrated that glucocorticoid concentrations were influenced by maternal weight, maternal educational status and preterm birth. Interestingly, HM glucocorticoids composition did not differ based on the individual preference of mothers feeding their infants exclusively or partially. The second study examined the glucocorticoid concentrations in a cohort of women who gave birth to premature infants. This study also demonstrated that gestational age has a significant impact on milk cortisone concentration, but showed no relation to infant postnatal age. With further method development, it was possible to separate the free versus conjugated (or bound) glucocorticoids. It was demonstrated that the majority of glucocorticoids (cortisol 76% and cortisone 84.2%) in these samples were in the free (and hence biologically available) form. Given the results of the above studies, we may conclude that there are many factors that affect the variation of milk glucocorticoids. And as far as concentration of cortisol is concerned, the time of the day (circadian rhythms) could be a major predictor. Therefore, a third study was done to examine the HM glucocorticoid concentration over a period of 24-hours, using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Both predominant glucocorticoids, cortisol (1.60 ± 0.71 ng/ml) and cortisone (3.40 ± 1.18 ng/ml), exhibited a pronounced 24 hour pattern. This was characterised by the rapid increase in the cortisol and cortisone levels in the early morning hours followed by a gradual decline throughout the day, mirroring the well-established plasma glucocorticoids circadian pattern. Furthermore, duration of a feed, and difference between feeding breast (breast sides) seemed to have no impact on glucocorticoid concentration in the milk samples collected prior to and again immediately upon the cessation of infant feeding. The last study of this thesis investigated the glucocorticoids composition of HM during the first 12 months of lactation, with a further analysis of the relationship between glucocorticoids and infant growth. Milk samples were collected at 2, 5, 9 and 12 months of infants’ age. We saw no significant relationship between HM glucocorticoids (cortisol and cortisone) and maternal characteristics (maternal weight, height, BMI and percentage fat mass) and infant growth outcomes (infant height, weight, BMI head circumference and percentage fat mass). These studies collectively demonstrated the fluctuating profile of HM glucocorticoids. Presence of glucocorticoids in milk demonstrated a circadian pattern, implying that infant feeding time might be an important factor. Additionally, the glucocorticoids concentration in HM was not influenced by lactational stage. Interestingly, maternal and infant related biological and social factors, including preterm birth contributed to the alteration of milk glucocorticoids levels. Cortisone was shown to consistently be the predominant glucocorticoid in HM. This is in contrast to what is found in maternal plasma, suggesting a selective modification or concentration in the human breast milk. Only few studies have measured cortisone levels and hence little is known of its biological function. Findings of this thesis emphasise the need for further investigation on HM hormonal composition and how it may impact infant growth and development, in later life.