Dietary long-chain, but not medium-chain, triglycerides impair exercise performance and uncouple cardiac mitochondria in rats

Metabolic efficiency is a major determinant of endurance performance, determining the speed or power that can be generated for a given level of oxygen consumption [1]. Diet can alter metabolic efficiency, and thereby affect performance, even over a relatively short time. Indeed, we demonstrated that consumption of a high fat diet for 9 days impaired exercise capacity in rats [2] and in similar studies, found that short-term high fat feeding impaired exercise efficiency [3] and cardiac energetics [4] in men, although interestingly the detrimental effects on exercise performance were not seen in highly-trained individuals [5].

We proposed that the mechanism underlying the effects of high fat feeding involve altered mitochondrial function, mediated by changes in transcription of metabolic proteins. Plasma free fatty acid (FFA) levels increase during high fat feeding, and increase fatty acid uptake by cardiac and skeletal muscle [6] activating the nuclear transcription factor, peroxisome proliferator-activated receptor α (PPARα). Expression of fatty acid oxidation enzymes, such as carnitine palmitoyl-transferase 1 (CPT1) and medium-chain acyl-CoA dehydrogenase (MCAD) [7, 8] is thereby increased. PPARα activation, however, also upregulates levels of the mitochondrial uncoupling protein UCP3 [9, 10]. When activated, UCP3 allows protons to cross the inner mitochondrial membrane independent of ATP synthesis, thereby uncoupling oxidative phosphorylation and decreasing metabolic efficiency (ATP produced/O₂ consumed). We found that short-term high fat feeding increased cardiac and skeletal muscle UCP3 levels, and uncoupled skeletal muscle mitochondria [2]. Moreover, a similar regime of high-fat feeding in rats increased mitochondrial fatty acid oxidation and uncoupling, resulting in decreasing cardiac efficiency [11].

Activation of PPARα may, however, be influenced by the type of fatty acid available. Long-chain fatty acids act as more potent ligands for PPARα than medium-chain fatty acids, and dietary supplementation of long-chain triglycerides increased muscle UCP3 expression, whereas medium-chain triglyceride feeding did not [12]. As such, medium-chain triglycerides (MCTs), unlike long-chain triglycerides (LCTs), may not impair exercise performance. Moreover, it has been suggested that MCTs might in fact enhance exercise performance [13], because a high MCT load elicits a ketogenic response [14], and ketone bodies improve metabolic efficiency in the isolated, perfused heart [15] and other tissues [16]. There is, however, little conclusive evidence for this [13].

In this study, we investigated the effects of short-term high-fat diets containing either MCTs or LCTs on physical performance and mitochondrial function in rats. Male Wistar rats were fed either standard laboratory chow (7.5% kcal from fat) or one of two high-fat diets containing 46% kcal from either MCTs or LCTs. Endurance capacity was assessed using a motorised treadmill and the rates and efficiency of fatty acid oxidation were measured in isolated cardiac mitochondria, along with levels of UCP3.

In heart and skeletal muscle, metabolic gene expression and thus mitochondrial function and efficiency (ATP produced/O₂ consumed) can be influenced by dietary composition [21, 22]. In particular, high fat feeding activates the transcription factor, PPARα, promoting the expression of genes associated with relatively inefficient fatty acid oxidation including the mitochondrial uncoupling proteins [21, 22]. Metabolic efficiency is a key determinant of endurance performance [1], and we have shown previously that short-term consumption of a high-fat diet impaired exercise capacity in rats [2] and decreased exercise efficiency in humans [3]. Unlike long-chain fatty acids however, medium-chain fatty acids do not activate PPARα [12], and therefore in this study we investigated the effects of short-term high-fat feeding on exercise capacity and cardiac mitochondrial function using diets rich in LCTs and MCTs.

In sedentary rats, LCT-feeding increased body weights and adiposity compared with chow-fed or MCT-fed rats, though not in exercising rats, and heart weight to body weight ratios were the same for all animals. As expected, both MCT and LCT diets increased circulating triglyceride levels, though only in sedentary animals, and LCT-feeding increased cholesterol in both sedentary and exercising rats. In agreement with other studies, consumption of a LCT-rich diet increased cardiac UCP3 expression in sedentary rats, whereas the MCT-rich diet did not [12], and here we have extended this finding to exercised rats. Moreover, whilst no diet altered ADP-stimulated fatty acid oxidation rates or State IV respiration rates in cardiac mitochondria, the mitochondria from LCT-fed rats had a lower oxidative phosphorylation efficiency than those from either chow-fed or MCT-fed rats as shown by a decreased ADP/O ratio, and this effect was ameliorated by the UCP3 inhibitor, GDP. ADP/O ratios are calculated independently of respiration rates, and give a more sensitive measure of the coupling of fuel oxidation and ADP phosphorylation than can be inferred from respiration rates alone [20]. Unfortunately, due to the time commitment of carrying out exercise tests on 18 rats, we were unable to isolate mitochondria from the hearts of exercised rats and instead used a sedentary group. This additional data would have been interesting, and whilst the effects of diet on UCP3 levels were similar in the two groups, it cannot be excluded that respiration might have been different in the exercised and sedentary rats.

Curiously, unlike our findings in the heart we did not find that consumption of the LCT-rich diet increased UCP3 levels in soleus. This is perhaps in contrast with our finding that a 55% fat diet increased UCP3 in soleus [2], although it should be noted that the diets used in this study had a lower total fat content than used previously and a higher carbohydrate content. It should also be noted that as, with our previous study using a 55% fat diet [2], we measured plasma metabolite levels in the fed state, and although we saw slightly higher plasma FFA levels in sedentary rats fed the MCT and LCT diets compared with their chow-fed counterparts this did not reach significance. Differences in plasma FFA levels between animals fed the two diets would likely be more apparent in the fasted state and could underlie the increased cardiac UCP3 levels in the MCT and LCT-fed rats via PPARα-activation [9], whilst not being sufficiently elevated to elicit activation in skeletal muscle, which has a lower fatty acid oxidation capacity than heart.

The lack of an effect of the MCT and LCT diets on skeletal muscle UCP3, compared with our previous high fat diet [2], may also explain another discrepancy. Previously, we found that the combination of a high fat diet and sustained exercise stress resulted in a 20% increase in heart/body weights [2]. We speculated that this may have represented an adaptive response to increase cardiac output and thereby enhance oxygen delivery to muscles in which there was notable uncoupling. We did not find an increase in heart/body weights here in rats fed either MCT or LCT diets, and perhaps this is in agreement with unchanged skeletal muscle UCP3 levels, although again we did not measure mitochondrial function directly in skeletal muscle mitochondria in this investigation.

An intensive exercise challenge over 5 days resulted in the exercised rats having lower body weights than their sedentary counterparts, along with decreased blood glucose, lactate, β-hydroxybutyrate and triglyceride levels, but increased free fatty acid levels, perhaps resulting from exercise-induced lipolysis. Over the five days of treadmill running, the exercise performance of chow-fed and MCT-fed rats was maintained at baseline levels, whereas LCT-fed rats were unable to sustain their performance level, running on average 55% less far than they did at baseline. This decrement is remarkably similar to the degree to which performance fell in high-fat fed rats in our previous study [2], despite a somewhat lower dietary fat content in this study (47% compared with 55% previously). Taken together with the elevated cardiac UCP3 levels in these rats and altered mitochondrial function in the sedentary animals, this might suggest that the detrimental effect of short-term high fat feeding resulted from long chain fatty acid activation of PPARα, and consequent loss of metabolic efficiency.

Notably, whilst the MCT-fed rats did not exhibit a significant loss of exercise capacity over five days of treadmill running, neither did they improve, either compared with their own baseline or compared to chow-fed rats. It has been theorised that MCT-rich diets might improve energy utilisation during exercise[13], perhaps through increased ketogenesis [14], however there is little conclusive evidence for this. One study reported an increased swimming endurance capacity in mice fed an MCT-rich diet compared with LCT-fed mice over 6 weeks, yet in this study both groups of mice were initially untrained at the task and both improved throughout the trial [23]. As MCT-fed mice improved swim capacity more than LCT-fed counterparts [23], this study, like ours, in fact demonstrated a detrimental effect of LCT-feeding compared with MCT, rather than a compelling beneficial effect of MCTs on performance. Interestingly, MCT-feeding did not significantly increase plasma ketone body levels in mice in this study [23], nor in sedentary or exercised rats in our study, perhaps because carbohydrate levels in the diets used were not particularly low.

In clinical trials, there is also little support for a training benefit from MCT-rich diets. One study demonstrated no difference in cycling performance amongst subjects fed 400 kcal as MCTs, LCTs or carbohydrates, despite increased ketogenesis in MCT-fed subjects [24]. Meanwhile others have reported a drop in cycling performance when dietary carbohydrates are replaced by MCTs [25, 26], although in one case this was possibly due to gastrointestinal upset [25].

Finally, it is likely that the upregulation of UCP3 in response to excess dietary fat forms part of an adaptive response. There has been some controversy concerning the physiological function of UCP3 with many possible roles suggested. There is compelling evidence to support a mild uncoupling function [27‐29] of UCP3. This uncoupling probably plays a role in lowering reactive oxygen species (ROS) production [30], with both the mitochondrial membrane potential and ROS generation increased in UCP3^-/- mice and following inhibition of UCP3 with GDP [30]. Alternatively, it has been suggested that UCP3 is involved in the export of fatty acid anions or lipid peroxides from the mitochondrial matrix [31]. Symmetry analysis of the structure of UCP3, alongside that of mitochondrial carriers known to transport fatty acid derivatives, however, does not support this hypothesis, with the UCPs lacking hydrophobic residues in their binding pocket on the matrix side [32]. The authors of this study conclude that the substrate of the UCPs is more likely to be small carboxylic or keto acids, transported in symport with protons [32]. None of these alternative functions however, exclude a mild uncoupling effect, whereby protons re-enter the matrix independently of ATP-synthesis, decreasing the efficiency of oxidative phosphorylation (ATP synthesised per O₂ consumed).

In conclusion, our study demonstrates that consumption of an LCT-rich diet impaired exercise capacity in rats over just 15 days of feeding. The LCT diet also increased cardiac UCP3 levels, which decreased efficiency of oxidative phosphorylation. Cardiac mitochondrial uncoupling may underlie the decrement in performance that we report here, although curiously we did not observe any changes in skeletal muscle UCP3 levels. Consumption of an MCT-rich diet did not alter cardiac UCP3 expression or oxidative phosphorylation efficiency at the mitochondria, and neither worsened nor improved exercise performance, compared with a chow diet. Whilst there remains no compelling case for an ergonomic benefit from MCT-rich diets, there is a strong case against the consumption of LCT-rich diets during endurance training.