Thomas Staessen

For several fish species, it has been demonstrated that the dietary replacement of fishmeal and fish oil by plant ingredients can lead to a reduction in nutrient digestibility. Besides the presence of various anti-nutritional factors, hampered fat digestion has been linked to disruptions of the bile acid metabolism (i.e., increased faecal bile acid loss and/or decreased bile acid synthesis) in fish fed plant-based diets.

Bile acids are efficiently conserved within the body of most vertebrates by means of enterohepatic circulation. As in mammals, enterohepatic circulation of bile acid is believed to exist in fish and evidence for this has been provided by several studies. However, in contrast to mammalians studies, quantitative data on how nutritional factors affect the bile acid metabolism of fish is largely lacking.

This thesis aimed to provide insight into whether, and to what degree, nutritional factors affect the bile acid metabolism of rainbow trout (Oncorhynchus mykiss). Several nutritional factors were investigated throughout this thesis, including bile acid precursor level, non-starch polysaccharide level, bile acid supplementation, fat level, type of non-protein energy source and feeding level. Disturbances of the bile acid metabolism were studied by quantifying, for the first time in fish, changes in faecal bile acid loss, bile acid synthesis and/or the total body bile acid pool. Furthermore, these disturbances of the bile acid metabolism were investigated in relation to nutrient digestibility, with a focus on fat digestion.

Chapter 1 describes in general the process of fat digestion and different aspects of the bile acid metabolism (e.g., bile acid biochemistry and functions, enterohepatic circulation of bile acids and steps in bile acid synthesis). The information in this chapter is intended to serve as a supplement in support to the information given in the experimental chapters. Chapter 1 also gives a brief overview of the current state of knowledge on the bile acid metabolism of fish and indicates where knowledge is missing. Chapter 1 ends with delineating the thesis aim, which is providing insight into whether and to what degree nutritional factors affect the bile acid metabolism in rainbow trout and how disturbances of the bile acid metabolism reflect on fat digestion.

Firstly, the study of Chapter 2 investigated whether dietary NSP level and feeding level affect faecal bile acid loss in rainbow trout (Oncorhynchus mykiss). For this, fish were fed, consecutively restrictively and to satiation, diets that differed in the inclusion level of an NSP-rich ingredient mixture. The combination of a high dietary NSP level and satiation feeding resulted in significantly enhanced faecal bile acid loss, and the latter was ascribed to a wash-out of bile acids with enhanced faeces production. Enhanced faecal bile acid loss in this study coincided with a decrease in fat digestibility, which suggest that enhanced faecal bile acid loss can cause a lack of bile acids available for digestion in the intestine.

Secondly, the study of Chapter 2 investigated whether the relationship between faecal bile acid loss and fat digestion is dependent on the dietary levels of bile acid precursors (i.e., cholesterol, taurine, methionine and cysteine). These precursors are involved in bile acid synthesis, and a lower supply of these compound in plant-based diets was hypothesised to reduce the potential of bile acid synthesis to compensate faecal bile acid loss, and thus maintain a sufficiently large bile acid pool size needed for proper fat digestion. A contrast in dietary supply of bile acid precursors was realised by creating 2 diets for each NSP level, either using fishmeal or a plant protein mixture as the main protein source. In contrast to the fishmeal-based diets, satiation feeding did not result in hampered fat digestion for any of the plant-based diets. This suggests that a lower supply of bile acid precursors in plant-based diets are not limiting for fat digestion in rainbow trout. However, the absence of an effect on fat digestion for the plant-based diets might also have been masked by the much lower feed intake and consequently faecal bile acid loss compared to the fishmeal-based diets during satiating feeding.

Based on the inverse relationship of Chapter 2 between faecal bile acid loss and fat digestion, Chapter 3 investigated a possible cause-effect relationship between conditions that enhance faecal bile acid loss and hampered fat digestion in rainbow trout. A possible cause-effect relationship was investigated by looking at the remedial potential of dietary bile acid supplementation (0% vs. 0.2% taurocholic acid) for hampered fat digestion in trout with enhanced faecal bile acid loss. To create a contrast in faecal bile acid loss, two fishmeal-based diets with different NSP levels (Low-NSP vs. High-NSP) from the study of Chapter 2 were fed consecutively restrictively and to satiation. Similar as for the study of Chapter 2, a combination of satiation feeding and High-NSP diets resulted in the highest faecal bile acid losses. Increasing faecal bile acid loss occurred alongside a decrease in fat digestibility, but only for the diets without bile acid supplementation. In fish fed those diets, fat digestibility correlated with a negative bile acid balance (i.e., bile acid intake minus faecal bile acid loss). Fat digestibility of diets with bile acid supplementation was not adversely affected by enhanced faecal bile acid loss and did not correlate with the bile acid balance. The outcome of this study shows that bile acid supplementation is an effective way to restore hampered fat digestion related to conditions that enhance faecal bile acid loss in rainbow trout. Furthermore, the beneficial effects of bile acid supplementation on fat digestion in the fish with enhanced faecal bile acid loss suggests that hampered fat digestion is indeed related to a lack of bile acids available in the intestine.

Fat digestion in the studies of Chapter 2 & 3 was measured in the last week of the experiment. Therefore, it is not known if hampered fat digestion under conditions of enhanced faecal bile acid loss occurs instantly or rather develops over time. Based om mammalian studies and the suggestion of Chapter 3 that hampered fat digestion in trout with enhanced faecal bile acid loss is caused by a lack of bile acids in the intestine, it was hypothesised that enhanced faecal bile acid loss causes a gradual depletion of the total bile acid pool size that is critical for proper fat digestion. Therefore, Chapter 4 investigated time-related changes (at 3 time points; week 2, 4 and 6) in both faecal bile acid loss and fat digestibility of rainbow trout. Low- and High-NSP diets were fed to satiation with the aim to create a contrast in faecal bile acid loss. Both NSP levels were tested for a diet low and a diet high in fat (Low-Fat vs. High-Fat). Dietary fat level was included in this study to investigate its effect on the relationship between faecal bile acid loss and fat digestibility. Overall, the High-NSP diets resulted in the highest faecal bile acid loss, although the difference between diets was less strong compared to the previous studies. Furthermore, faecal bile acid loss showed a time-related decrease for all diets, which was related to a strong time-related decrease in feed intake and consequently faeces production. Fat digestibility improved over time regardless of dietary NSP level, which does not support the hypothesis that enhanced faecal bile acid loss causes a depletion of the total body bile acid pool size that is critical for proper fat digestion. However, the absence of a time-related decrease in fat digestibility might be connected to the strong time-related decrease in faecal bile acid loss in this experiment. The High-Fat diets consistently resulted in the lowest fat digestibility, which might be (partly) related to a lower availability of bile acids relative to the level of dietary fat and thus less efficient emulsification and digestion.

The first three experimental chapters (Chapter 2, 3 & 4) focussed on faecal bile acid loss in relation to fat digestion. The last two experimental chapters (Chapter 5 & 6) additionally studied the effects of several dietary factors on bile acid synthesis and the total body bile acid pool.

Mammalian studies show that diet macronutrient composition can affect the bile acid metabolism. Such information is lacking for fish but could be of importance as the ongoing replacement of fishmeal by plant ingredients increases the level of carbohydrates in aquafeeds. Chapter 5 set out to investigate how changes in the type of main non-protein energy source affect the size and composition of the total body bile acid pool, faecal bile acid loss and bile acid synthesis. Two diets were formulated with similar digestible protein to digestible energy ratios but differing in inclusion of either maize starch (Starch) or rapeseed oil (Fat) as the main non-protein source. Rainbow trout were fed to satiation and the initial and final body bile acid pool and daily faecal bile acid loss were determined. Daily bile acid synthesis was quantified for the first time in fish through means of a bile acid balance, taking into account bile acid intake, faecal bile acid loss and the change in bile acid pool size over time. Using both data from the experiment of Chapter 5 and data from literature, a linear relationship was demonstrated between the bile acid pool size and body weight of rainbow trout. The identical initial body bile acid pool size, but different final body bile acid pool size despite comparable body weight of fish fed either one of the experimental diets, shows that the increase in total body bile acid pool size with body weight is dependent on diet composition. The type of non-protein energy source did not substantially affect the final body bile acid pool composition. However, feeding the Starch diet resulted in a larger final total body bile acid pool size compared to the Fat diet, and this despite enhanced faecal bile acid loss when feeding the Starch diet that was related to more faeces being produced. Bile acid synthesis in fish fed the Starch diets was more than twice as high as the synthesis in fish fed the Fat diet, showing that diet can quantitatively affect bile acid synthesis. The underlying mechanisms for this difference in synthesis need to be further investigated, but most likely both differences in the kinetics of enterohepatic circulation and/or direct effects of macronutrient on bile acid synthesis played a role. Additionally, data of this study show that the levels of conjugated and secondary bile acids in both body and faeces of rainbow trout are much lower compared to mammalian species, and this is independent from diet composition. Secondary bile acids do not seem to play an important role in rainbow trout, but this could be different in other fish species (e.g., warm-water species) and needs further investigation. Finally, Chapter 5 shows a discrepancy between the bile acid profile of the body and faeces, with the share of cholic acid in the faeces lower and the share of chenodeoxycholic acid larger compared to those in the body. Together with the lower fractional turnover rates of cholic acid for both diets compared to chenodeoxycholic acid, this suggest better intestinal conservation of cholic acid in rainbow trout.

Although Chapters 2 & 3 showed that faecal bile acid loss and fat digestion can be inversely related, those chapters and Chapter 4 did not show if hampered fat digestion under conditions of enhanced faecal bile acid loss is related to a decrease of the total body bile acid pool size. Furthermore, it was not quantified how bile acid synthesis alters in response to changes in faecal bile acid loss. Therefore, Chapter 6 aimed at quantifying changes in the bile acid pool size and bile acid synthesis of rainbow trout fed the same diets as in Chapter 3. The Low- and High-NSP diet was fed to satiation with the aim of creating a contrast in faecal bile acid loss. Both diets were tested with and without bile acid supplementation (0% BAS vs. 0.3% BAS). Contrary to Chapter 3, part of the bile acid mixture supplemented to the diets in Chapter 6 contained a bile acid which does not naturally occur in rainbow trout (i.e., glycocholic acid). The latter was used to study if supplemented bile acids efficiently enter the enterohepatic circulation. Opposite to previous chapters, dietary NSP level did not enhance faecal bile acid loss in Chapter 6, and the reason for this remains unclear. However, also fat digestion was unaffected by dietary NSP level in the study of Chapter 6, thus the outcome does not contradict the earlier observed inverse relationship between faecal bile acid loss and fat digestion. As there was no contrast in faecal bile acid loss between treatments, Chapter 6 focussed more on the effects of BAS on the bile acid metabolism. Total bile acid synthesis was a 4-fold lower in fish fed the diets with BAS, showing that BAS reduced the need for de novo bile acid synthesis. Higher levels of bile acids in the chyme and liver + gallbladder of fish fed the diets with BAS most likely caused a negative feedback inhibition on bile acid synthesis. Negative synthesis was observed for the body foreign glycocholic acid in fish fed the BAS diets, which suggest that this bile acid is catabolised or converted into other bile acids in rainbow trout. The enlarged pool of taurocholic acid, but especially of the body foreign glycocholic acid, with BAS in both the whole-body samples and liver + gallbladder samples shows that dietary bile acids are effectively taking part in EHC. The relative absorption of T-CA and G-CA over the different intestinal compartments was comparable, which shows that EHC does not differentiate between T-CA and G-CA. In accordance with mammalians, it is generally assumed that the majority of bile acids in fish is actively absorbed in the distal part of the intestine. While relative absorption of bile acids in the study of Chapter 6 was highest between the distal intestine and the faeces, the absolute decrease in chymal bile acid content was highest between the pyloric region and the mid intestine. The latter shows that significant bile acid absorption occurs more proximal in the intestine of rainbow trout than previously thought.

In Chapter 7, the main findings of the different studies of this thesis are compared and put into a broader context. Based on the inverse relationship between faecal bile acid loss and fat digestion that was observed in several of the studies, the present state of insight in possible relationships between faecal bile acid loss, bile acid synthesis and the total body bile acid pool size are presented. The chapter ends with a of the main conclusions reached and a recommendation for future research.