Defining the lipid profiles of human milk, infant formula, and animal milk: implications for infant feeding

Alexandra D George; Sudip Paul; Tingting Wang; Kevin Huynh; Corey Giles; Natalie Mellett; Thy Duong; Anh Nguyen; Donna Geddes; Toby Mansell; Richard Saffery; Peter Vuillermin; Anne-Louise Ponsonby; David Burgner; Satvika Burugupalli; Peter J Meikle; Barwon Infant Study Investigator Team

doi:10.3389/fnut.2023.1227340

Defining the lipid profiles of human milk, infant formula, and animal milk: implications for infant feeding

Front Nutr. 2023 Aug 30:10:1227340. doi: 10.3389/fnut.2023.1227340. eCollection 2023.

Authors

Alexandra D George^{1

2

3}, Sudip Paul^{1

2

3}, Tingting Wang^{1

3}, Kevin Huynh^{1

2

3}, Corey Giles^{1

2

3}, Natalie Mellett¹, Thy Duong¹, Anh Nguyen¹, Donna Geddes^{4

5}, Toby Mansell^{5

6}, Richard Saffery^{5

6}, Peter Vuillermin^{5

7

8}, Anne-Louise Ponsonby^{5

9}, David Burgner^{5

6}, Satvika Burugupalli^{1

3}, Peter J Meikle^{1

2

3

10}; Barwon Infant Study Investigator Team

Collaborators

Barwon Infant Study Investigator Team:
Anne-Louise Ponsonby, David Burgner, Fiona Collier, John Carlin, Katie Allen, Mimi Tang, Peter Sly, Peter Vuillermin, Richard Saffery, Sarath Ranganathan, Terry Dwyer

Affiliations

¹ Metabolomics Laboratory, Baker Heart and Diabetes Institute, Melbourne, VIC, Australia.
² Baker Department of Cardiometabolic Health, University of Melbourne, Parkville, VIC, Australia.
³ Department of Cardiovascular Research, Translation and Implementation, La Trobe University, Bundoora, VIC, Australia.
⁴ School of Molecular Sciences, The University of Western Australia, Perth, WA, Australia.
⁵ Murdoch Children's Research Institute, Parkville, VIC, Australia.
⁶ Department of Pediatrics, University of Melbourne, Parkville, VIC, Australia.
⁷ School of Medicine, Deakin University, Melbourne, VIC, Australia.
⁸ Child Health Research Unit, Barwon Health, Geelong, VIC, Australia.
⁹ The Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia.
¹⁰ Department of Diabetes, Central Clinical School, Monash University, Clayton, VIC, Australia.

Abstract

Background: Breastfed infants have lower disease risk compared to formula-fed infants, however, the mechanisms behind this protection are unknown. Human milk has a complex lipidome which may have many critical roles in health and disease risk. However, human milk lipidomics is challenging, and research is still required to fully understand the lipidome and to interpret and translate findings. This study aimed to address key human milk lipidome knowledge gaps and discuss possible implications for early life health.

Methods: Human milk samples from two birth cohorts, the Barwon Infant Study (n = 312) and University of Western Australia birth cohort (n = 342), were analysed using four liquid chromatography-mass spectrometry (LC-MS) methods (lipidome, triacylglycerol, total fatty acid, alkylglycerol). Bovine, goat, and soy-based infant formula, and bovine and goat milk were analysed for comparison. Composition was explored as concentrations, relative abundance, and infant lipid intake. Statistical analyses included principal component analysis, mixed effects modelling, and correlation, with false discovery rate correction, to explore human milk lipidome longitudinal trends and inter and intra-individual variation, differences between sample types, lipid intakes, and correlations between infant plasma and human milk lipids.

Results: Lipidomics analysis identified 979 lipids. The human milk lipidome was distinct from that of infant formula and animal milk. Ether lipids were of particular interest, as they were significantly higher, in concentration and relative abundance, in human milk than in formula and animal milk, if present in the latter samples at all. Many ether lipids were highest in colostrum, and some changed significantly through lactation. Significant correlations were identified between human milk and infant circulating lipids (40% of which were ether lipids), and specific ether lipid intake by exclusively breastfed infants was 200-fold higher than that of an exclusively formula-fed infant.

Conclusion: There are marked differences between the lipidomes of human milk, infant formula, and animal milk, with notable distinctions between ether lipids that are reflected in the infant plasma lipidome. These findings have potential implications for early life health, and may reveal why breast and formula-fed infants are not afforded the same protections. Comprehensive lipidomics studies with outcomes are required to understand the impacts on infant health and tailor translation.

Keywords: DOHaD (development origins of health and disease); breastfeeding; breastmilk; fat; metabolomics.