Postprandial metabolism of infant formulas containing different fat sources : digestion, absorption, and metabolic responses

Human milk is the best nutrition available for infants. However, in some situations breastfeeding is not an option. In that case, infant formula (IF) is the best alternative. An important component of IF is fat, which delivers about 50% of the energy for an infant. With the digestion of fat, free fatty acids are released. Those fatty acids are absorbed into the epithelial cells and are either oxidized, transported to the liver via the portal vein or packed into chylomicrons to be transported to peripheral tissues and the liver. In the peripheral tissues, the fatty acids can be used for beta-oxidation or storage.

Different fat sources are used in IF; mostly a combination of vegetable fats is used. Nowadays bovine milk fat is used more often in IF. Because of the preferred level of linoleic acid, maximal 67% bovine milk fat can be used in a fat blend for IF. When using bovine milk fat a wider variety of fatty acids is introduced, including short- and medium-chain fatty acids (SCFA and MCFA). Furthermore, the triacylglycerol (TAG) structure of bovine milk fat differs from that of the in IF common used vegetable fats. Research focused on specific lipids has shown that a difference in fatty acid length may impact the activity of gastrointestinal lipases and therefore lipolysis. Furthermore, medium-chain TAGs have been reported to increase energy metabolism and satiety. The TAG structure has been reported to be important for absorption of fat and may also affect postprandial metabolism. Whether bovine milk fat in IF exerts different effects on digestion, absorption, and metabolism, compared to vegetable fats, is not known yet. Therefore, the first aim of the thesis was to study the digestion, absorption, and metabolism of different fat sources used in infant formula; anhydrous bovine milk fat and vegetable fats.

Volatile organic compounds (VOCs) in exhaled air can reflect metabolic aspects of physiology, such as specific diseases. Some indications exist that VOCs may also reflect metabolic effects of nutrition. Since it is difficult to perform invasive studies in infants, analysis of VOCs in exhaled air could provide a non-invasive approach to study effects on metabolism of the addition of bovine milk fat to IF. This was explored in the second aim, which was to study whether the analysis of exhaled air can be used to assess metabolism of fat.

A clear overview of differences and similarities between human milk fat and fat sources used for IF is lacking. Therefore, in chapter 2, we reviewed the fatty acid composition and triglyceride structure of human milk fat, vegetable fats, and bovine milk fat. Furthermore, an overview is provided of the known and potential health effects of bovine milk fat components that have been described in literature.

Whether using bovine milk fat in a fat blend for IF, and therefore introducing more SCFA and MCFA, affect lipolysis in a different manner than a fat blend containing solely vegetable fat affects lipolysis is unknown. Therefore, we studied the release of free fatty acids of an IF containing 100% vegetable fat with an IF containing 67% bovine milk fat and 33% vegetable fat, and human milk as described in chapter 3. A static two-phase in vitro digestion model with infant conditions was used. We did not find any differences in the total level of free fatty acids released from the three different products. During lipolysis more SCFA and MCFA were released from the IF with bovine milk fat, compared to the IF with only vegetable fats, which is in line with the fatty acid profile of the two fat blends. In the IF with bovine milk fat less palmitic acid was released compared to from the one containing vegetable fat only. From human milk even less long-chain saturated fatty acids (LCSFA) were released. So the difference in fat source in IF did not change the total amount of lipolysis, but it did change the profile of free fatty acids released. The type of fatty acids released is important for their absorption, as LCSFA have the ability to form insoluble complexes with calcium in the lumen, which will be excreted.

SCFA and MCFA have been reported to increase energy expenditure, since they can directly be absorbed from the lumen, independent of transportations via chylomicrons, and transported to the liver via the portal vein, where they can be used for oxidation. Whether the usage of bovine milk fat, and thus the inclusion of more SCFA and MCFA, in IF affects lipid and energy metabolism is not known yet. We studied the effects of an IF containing a fat blend with 67% of bovine milk fat and an IF with a fat blend containing 100% vegetable fat on lipid and energy metabolism. Since invasive methods are needed to get insight in metabolic alterations we performed a proof-of-concept study in healthy young adult males.

In chapter 4 the results of our study on the effects of the different fat sources on lipid metabolism are described. No differences in absorption, lipoprotein metabolism or substrate utilisation were found. Chylomicron size was found to be slightly increased after consumption of IF with bovine milk fat. However, this was not caused by an increased lipid content. The fatty acid profile of chylomicrons formed postprandial reflected that of the different fat sources consumed. Directly after consumption of the IF with bovine milk fat an increase in circulating ketone bodies was observed. Chapter 5 reports our finding that using bovine milk fat in the fat blend of an IF prolongs satiety. This was not caused by differences in diet-induced thermogenesis. However, this might be caused by an increased postprandial secretin response, possibly via stimulation of vagal afferent pathways.

With the absorption and oxidation of fatty acids several metabolites are formed. Some of these metabolites are excreted via exhaled air; VOCs. Analysis of these VOCs is being used in clinical settings to determine biomarkers for several diseases, like lung cancer and asthma. This method has also made its way to nutrition studies. The effect of several long- and short-term nutrition intervention studies on VOCs has been explored. One of the challenges for using VOCs analysis in these type of studies is the inter- and intra-subject variability. The day-to-day variation in response to a meal was not studied before. In this thesis we explored the intra-individual variation of VOCs in exhaled air after a meal, to determine whether the measurement of VOCs in exhaled air with the online method PTR-MS could be a suitable method for nutritional intervention studies. Chapter 6 describes several interconnected human trials that were performed. The first study showed that consumption of a meal leads to a different VOC profile in the five postprandial hours examined. However, it also showed quantitative differences between study days, although trends over time were similar. Therefore, in a second study the effects of two different meals, a low-fat and a high-fat dairy drink, were determined on several days. This study indicated that consumption of different meals resulted in a different VOC profile in exhaled air. VOCs that were affected could be linked to lipid metabolism.

In chapter 7 the results of this thesis are discussed and put in perspective. The translation of the results obtained in adults to infants is discussed. Furthermore, the complexities of the method of breath analysis are discussed. Recommendations for future research are provided.

The first aim of the thesis was to study the digestion, absorption, and metabolism of different fat sources used in infant formula; anhydrous bovine milk fat and vegetable fats. We answered the research questions by conducting in vitro lipolysis experiments and in vivo proof-of-concept trials in adults. We observed that there is no difference in total lipolysis of IF containing a mixture of bovine milk fat and vegetable fat and IF with vegetable fats. However, the type of fatty acids released were different, which was in line with the differences in fatty acid composition of the IFs. These differences in fatty acid release did not result in a difference in fat absorption rate or energy expenditure. The difference in fatty acid profile was reflected in the fatty acid profile of chylomicrons. Using 67% anhydrous bovine milk fat in IF did prolong satiety compared to an IF with a fat blend consisting for 100% of vegetable fats. Prolongation of satiety might reduce energy intake and eating behaviour in infants. The difference in fatty acid profile in chylomicrons might have different effects on peripheral tissues to which they are presented. More studies are needed to examine those effects.

The second aim was to study whether the analysis of exhaled air can be used to assess metabolism of fat. With our studies we provide some insight in the inter-and intra-subject variation. Despite these variations, a different VOC profile could be observed between two different meals differing in fat content. Although more research is needed to reduce the intra-subject variability, the method of breath analysis seems to be a promising method to use in nutritional intervention studies in order to get more insight in metabolism.