Abstract :
[en] The lipid components of human milk play a critical role in supporting infant growth and development. Current research on human milk fatty acids (FAs) has primarily focused on changes during different lactation periods, whereas the effects of geographic regions on FA composition remain less explored. Moreover, human milk phospholipids (HMPLs) play an important role in supporting infant brain development. However, there are rare studies which comprehensively analyze the differences between HMPLs and their substitutes from multiple dimensions, including phospholipid concentration, phospholipid classes, phospholipid fatty acids, and phospholipid molecules. Furthermore, there is currently no standardized similarity evaluation model specifically designed for HMPLs.
To address these gaps, total FAs, Sn-2 FAs, and phospholipid fatty acids (PLFAs) were analyzed in human milk samples collected from seven cities across China. Gas chromatography (GC) was employed for quantitative analysis of FAs after methylation, while thin-layer chromatography (TLC) was used for separation of lipid classes, and gradient elution was applied for targeted PLFA analysis. The results showed that the lactation region was an important factor affecting the comprehensive fatty acid profile of human milk. The higher docosahexaenoic acid (DHA, C22:6n3) content was observed in coastal cities (Weihai), while higher linoleic acid (C18:3n3) was detected in Lanzhou. This might be related to the daily diet of volunteer mothers. Additionally, palmitic acid (C16:0), oleic acid (C18:1n9c), and linoleic acid (C18:2n6c) were important fatty acid species in human milk, accounting for over 70 % of the total fatty acids. These data contribute to the establishment of a comprehensive Chinese human milk lipid database.
Next, commercially available natural phospholipids, including bovine, goat, camel, and mare milk, as well as soybean, sunflower, and egg yolk phospholipids, were compared with HMPLs. A multidimensional similarity evaluation model based on the Bray-Curtis index was applied to assess differences in phospholipid concentration, classes, fatty acids, and molecularspecies. The results showed that there were significant differences between HMPLs and its natural substitutes. Furthermore, compared with plant-derived phospholipids, animal-derived phospholipids were more similar to HMPLs in chemical composition, with camel milk phospholipid having the highest similarity score of 0.75, while the scores for soybean phospholipids and sunflower phospholipids were 0.22 and 0.24, respectively. These data further deepen our understanding of HMPLs and further clarified the differences between HMPLs and its natural substitutes.
In addition, phospholipid profiles of human colostrum were compared with those of other mammalian colostrum using liquid chromatography-mass spectrometry (LC-MS). In bovine colostrum, the highest phospholipid concentration was detected, which was 39.80 mg/100mL and higher than that of human colostrum (37.01 mg/100mL). Across all colostrum samples, C16:0, C18:1n9c, and C18:2n6c were the predominant FA species in phospholipids. However, the distribution of specific phospholipid molecular species differed: in human colostrum, PC34:2 was most abundant, whereas PC34:1 predominated in other mammalian colostrum. These findings provide valuable insights into the potential use of mammalian colostrum in infant nutrition and support the rational development of colostrum-enriched products.
Finally, thirty commercial infant formulas, including 18 cow-milk-based and 12 goat-milk-based, were analyzed for phospholipid composition and compared with HMPLs using the similarity evaluation model. The results showed that the total phospholipid concentration of human milk (19.53 mg/100mL) was significantly higher than that of infant formulas (cow-based (10.88 mg/100mL) and goat-based (4.74 mg/100mL)). Across all infant formulas, C16:0, C18:1n9c, and C18:2n6c were the major PLFAs, but the molecular species composition differed markedly. In human milk, PC36:2 and PC34:1 were the most abundant, whereas infant formulas were dominated by PC34:1, PC36:4, and other species. Based on TOPSIS method, although the cow-milk-based infant formulas (65.68) had a higher score than that of goat-milk-based infant formulas (53.62), there was still a gap between HMPLs and infant formula phospholipids. Therefore, these results not only revealed the current phospholipid difference between infant formulas and human milk, but also provided basic data for improving the quality of infant formulas
