The validation of biomarkers of metabolic efficacy in infant nutrition

Breastfeeding is the best start for a baby's life. The World Health Organisation (WHO) recommends that all women should exclusively breastfeed their babies until six months of age. However not all women have the opportunity to exclusively breastfeed their babies. In the UK it is reported that whilst 81% of mothers start with breast feeding only 17% of infants are exclusively breast-fed until 3 months, and 1% until 6 months of age. This does not have to be the mother's choice, already in the first weeks 13% of the breast feeding mothers get advice to provide the baby with extra feeding. Thus there is a need to design breast-milk alternatives that mimic the nutritional quality of breast milk as closely as possible.

Recent evidence suggests that non breast-fed infants have an increased risk of becoming obese and an increased risk of high blood pressure in later life. It is currently unclear why infant feeding practices have such a big impact on health outcomes later in life. However, breastfeeding seems to be metabolised differently by infants than formula milk; our research showed that babies that are breast fed have a significantly different blood lipid (fat) profile compared to bottle fed babies.

The aim of this project is to develop novel methods to help us understand the mechanism(s) of the differences in the metabolism of bottle fed babies and breast fed babies. We want to identify specific lipids that can be used as markers that reflect the benefits of breastfeeding. We will do this by exploring changes in the lipid profiles of babies in the UK during their first year of life. By comparing blood markers with patterns of infant growth we will be able to identify markers that can be used to assess the metabolic response to breast-milk alternatives. We will also include mothers and infants from rural and urban sub-Saharan Africa to explore between population differences in the association between mode of feeding, blood lipid levels, and infant health outcomes'.

One possible way in which mode of feeding may impact on the baby's metabolism is via bacteria in the gut (the gut microbiome). It is believed that, compounds in food, affect which bacteria live in the gut and that the different combinations of bacteria in the gut leads to different metabolic products that will be taken up into the blood stream. We do not know if these differences in the gut microbiome also affect the lipids in the blood. To explore this potential link, we will also study what the effect is of the gut microbiome of the babies on their metabolism.

Recently we developed a method using high resolution mass spectrometry to determine a detailed lipid profile covering over a 100 lipids from one dried blood spot. This is a blood sample obtained from a heel prick and only requires a drop of blood spotted on paper. This is the most suitable method to study the metabolism of babies. We will adapt this dried blood spot based method to captures both lipids and metabolites dependent on gut microbiome.

The identified markers, and the methods to measure these, will then be available for use by scientist and industry to study the effect of new formula or new feeding methods for infants who do not have the chance to be exclusively breast fed and give these babies the best possible start in life.

The study of endogenous metabolism in healthy infants has remained largely unexplored. One of the main reasons is that repeated drawing of blood in healthy new-borns is regarded as over invasive. Dried blood spots (DBS) from heel pricks have long been established as the most appropriate sample format, to screen infants for inborn errors of metabolism. We recently developed and validated of a lipid profiling method using DBS from infants. This method provides data on approximately 100 endogenous lipids from a single 3.2 mm DBS disc. The process is automated, fast, robust, inexpensive and therefore suitable for studying large cohorts. We have already applied this method to a subset of DBS samples collected in the Cambridge Baby Growth Study and showed that lipid profiles of breast-fed infants are significantly different to formula-fed infants and that specific lipids are associated with growth rates. This work provides the basis for our hypothesis that the metabolism of breast-fed and formula-fed infants are differentially conditioned, which is reflected in clear differences in the lipid profiles at an early age. We hypothesize that these early detectable differences in metabolism are, at least partially, responsible for the phenomenon known as metabolic programming. This proposal aims to validate lipid biomarkers that reflect healthy metabolism associated with optimal neonatal growth. These biomarkers will deliver a robust indicator of metabolic efficacy of infant nutrition and will provide a practical tool that can assist the study of feeding strategies and development of new infant formulae for infants that cannot or should not be exclusively breastfed. We aim to make the research directly translational, through the delivery of standardised protocols, which allow for classification of a healthy or sub-optimal metabolic response. This can be implemented in studies of the role infant nutrition on metabolism and long term health.

Exec Summary: The proposed project will provide: (a) a novel method, using a minimally-invasive sampling protocol, for the lipidomic assessment of the effects of early nutrition on metabolism; (b) the definitions of healthy and sub-optimal metabolic phenotypes; and (c) enable the development of formula milks that more closely mimic the metabolic responses observed in breast fed infants.

The market: According to the summary of the Euromonitor International report on baby food (http://www.euromonitor.com/baby-food-in-the-united-kingdom/report) in the UK the compound annual growth rate for baby food will remain constant at 5% over the next years rising to an estimated £1,046 million in 2018. The main reason for it is the rise in unit prices due to substantial product innovation. This clearly shows that companies involved in infant formula want to innovate and further develop better infant formula.

Need for innovation: It is essential for the development and innovation of new infant formula that there will be better tools to determine the metabolic effects of formula. Trials with new formula compositions are well designed but have to rely on gross phenotypic data like growth rate and BMI. The lack of biochemical markers that can help to assess the direct effect on metabolism and the lack of makers for aberrant growth, makes this research costly and slow.
Metabolic markers can therefore revolutionize study and trials of new formula. This can help to develop new feeding strategies for infants that cannot or should not be exclusively breastfed. Better feeding strategies for these infants will give them a healthier start of life and should reduce their risk of obesity and related disease in later life. This could have a significant impact on future healthcare costs throughout the life course by reducing the incidence of metabolic diseases.

Aims and outcomes: Our aim is to study the longitudinal changes in the metabolism of breastfed infants and compile an independent body of evidence on the metabolic effect of breast-feeding in optimal and suboptimal situation within the first year of life. It is not our aim to develop new formula or test current formula. The project will provide the tools and know-how for other scientist or industry to do these assessments. All data will be made open access within Metabolights and other relevant international repositories and will provide companies with sufficient background knowledge to test the metabolic efficacy of any new formula.

The final result of the proposed work will be a fully validated and translatable methodology that can be used to by other groups (science or industry). The methodology will give the ability to annotate the lipidomic profile from a DBS sample of an infant as normal or associated with sub-optimal growth. It should also be able to give evidence if the changes in diet lead to changes in the lipid profile that are associated with a healthy outcome. Each year we will organise practical workshops at MRC HNR on the methodology, which will be open to researchers from academia and industry. The objective of the workshops will be to train researchers to independently use dried blood spots in their lipid profiling studies using the SOP and to analyse their data.

This technology would for the first time give science community and industry the opportunity to determine if change to the formula lead to positive changes in the metabolism without having to wait for several years before the effect becomes measurable in the gross phenotypic data. The lipid profile also has the potential to be more sensitive than gross phenotypic data, which would yield a reduction of the number of infants necessary in a trial and thus a significant reduction of time and costs to test new formula.