Metabolic profiles of male meat eaters, fish eaters, vegetarians, and vegans

It isn’t far-fetched to think that individuals with difference dietary habits might have different metabolic profiles. However, this issue has not been well investigated, leading Schmidt et al to investigate the differences in concentrations of circulating metabolites in middle-aged healthy men from the Oxford component of the European Prospective Investigation into Cancer and Nutrition (EPIC-Oxford). Using data from a food-frequency questionnaire, the researchers randomly selected 91-97 men per group who could be classified as meat eaters, fish eaters (who do not eat meat but do eat fish), vegetarians (who eat neither meat nor fish), and vegans (who do not eat meat, fish, dairy products, or eggs).

Of the 118 total metabolites analyzed, concentrations of 12 sphingolipids (86%), 61 glycerophospholipids (79%), 6 acylcarnitines (67%), 2 biogenic amines (67%), and 12 amino acids (63%) varied across the 4 diet groups. In the vast majority of these, vegans had the lowest concentrations and the meat eaters had the highest. When the 4 groups were re-analyzed as vegan vs non-vegan, most of these differences were maintained (86 metabolite differences), but the differences within the non-vegan group (that is, between the meat, fish, and vegetarian) dropped markedly to 39 of 118 metabolites.

Of all these differences, the most extreme was between meat eaters and vegans for shingolipids and glycerophospholipids, which are obtained in two ways: dietary animal products and de novo synthesis. Clearly vegans need to rely on the latter. The problem with this is that of the two main pathways for de novo synthesis, one requires choline, which is mainly found in animal products, and the other involves methylation, which is inhibited when methionine availability is low (such is the case with plant proteins). It is therefore not entirely surprising that the lowest concentrations were seen in vegans, and this is particularly concerning in light of these metabolites’ importance in cell membrane fluidity and integrity.

Metabolic profiles of low- and high-carbohydrate ultra-endurance athletes

Another area where metabolic profile data is lacking is on athletes, particularly elite athletes. With an increasing number of ultra-endurance athletes pursuing a low-carbohydrate diet, Volek et al felt it was only necessary to perform a cross-sectional study comparing male ultra-endurance runners 21-45 years of age consuming a low-carbohydrate (LC; n=10) or high-carbohydrate (HC; n=10) diet who were in the top 10% of finalists competing in sanctioned running events >50 km and/or triathlons of at least half iron-man distance (113 km). In other words, these guys were pros.

So what constitutes a LC and HC athlete? According to the 3-day food logs each one completed before coming to the research laboratory for metabolic testing, the LC athletes were consuming an average of 82 grams of carbohydrates (10% of kcal) per day and 226 grams of fat (70%). The HC athletes were at the other extreme, consuming 486 g carbs (59%) and 91 g fat (25%). Caloric intake was similar between groups (~3000 kcal) and protein intake was non-significantly higher in the LC group (2.1 g/kg vs 1.7 g/kg HC).

The athletes traveled to the research laboratory for two consecutive days of testing. On day 1 they performed a maximal oxygen consumption test, and on day 2 they ran on a treadmill at 65% of their maximal oxygen consumption for 3 hours to determine metabolic response before, during, and after exercise.

Overall, peak fat oxidation was an average of 2.3-fold greater in the LC group, with every subject exceeding the highest value achieved in the HC group. This is despite a similar percent of maximal oxygen consumption (as per study design), similar ratings of perceived exertion, and similar energy expenditure between the groups. Moreover, fat and carbohydrate oxidation were stable throughout exercise in the LC group, whereas the HC group showed an initial reliance on carbohydrate oxidation and slow transition towards increased fat burning over time.

As could be reasonably expected, blood ketone and glycerol levels were significantly greater in the LC group before, during, and after exercise compared to the HC group, indicating increased fat breakdown and oxidation. However, plasma triglycerides were never difference between the groups, and free fatty acid concentrations were similar by the end of exercise in both groups. Blood glucose and insulin were also not different before or during exercise. Finally, muscle glycogen was significantly decreased by 62-66% in both groups after exercise, with no differences between groups at this time or in pre-exercise levels.

It is worth remembering that all these differences are despite the fact that the athletes were all professionals of a similar performance level in competitions (although no direct performance testing was done in this current study, which would have been cool).

Putting your brain a leg above the rest

Despite an abundance of literature documenting acute benefits of exercise on cognition, no study has yet to show a long-term benefit. This isn’t too surprising considering that cognition changes over years and decades, making exercise interventions impractical. As such, longitudinal research can help fill in the gaps by showing how certain baseline characteristics affect outcomes of interest down the road. Of course, this too has limitations: cognitive change may be explained by confounding through genetics and early life factors.

Fortunately, Steves et al recently published a paper that overcomes most of these barriers to forming solid conclusions. They did this by testing the leg power (proxy for physical activity levels) of 324 female twins in 1999 and comparing it to cognitive function 12 years later as assessed by a battery of computerized tests. Moreover, they performed MRI scanning of the brain in a small group of these women to see how leg power impacts brain volume.

Leg power had the most consistent and largest effect size of all tested variables, whereby every 40W increase in leg power at baseline led to roughly 3.3 years difference in cognitive vs real age. This effect persisted through adjustment for age, developmental, psychosocial, health, and disease factors. Also, neither grip strength, lung function, nor leg lean mass were predictive of cognitive function when exchanged for leg power, and neither frailty nor telomere length modified the relationship. Similarly, more leg power was positively associated with total grey matter volume 12 years later.

So basically, increased leg power at baseline was associated with improved cognitive aging over the following 10-12 years, and these effects cannot be explained by genetic or environmental factors thanks to the use of twins. Better get to the gym and make a deposit today.

The effects of weight regain in females with anorexia

It isn’t often you read about individuals with anorexia (AN), and they do truly provide some insightful information on human physiology given their circumstances. Most recently, Ghoch et al investigated the body composition of 90 adult female patients with a clinical diagnosis of AN before and after weight-regain treatment and compared it to health female controls carefully matched for age and post-treatment BMI.

The treatment lasted for 20 weeks and involved dietitian-assisted eating that provided a steady weight gain of 1-1.5 kg (2-3.5 lbs) per week until a BMI of at least 18.5 was reached. The average before and after BMIs of the AN females were 15.3 and 19.8, respectively, and before beginning weight regain they had a DXA-assessed body fat percentage of 9.3% that shot up to 24% after refeeding. Clearly these women were starved and showed a successful weight rebound.

What is interesting, however, is that after this substantial and very quick weight restoration, the AN females did not differ from the controls in any measure of body composition except for a statistically significant 0.87% lower lean trunk mass. Total skeletal muscle mass, extremity mass, lean-body mass, and fat mass were all similar. Moreover, further analyses revealed that the lower lean trunk mass only occurred in the AN females who had a BMI below 16.5 before treatment.

The takeaway is simple – the biological regulation of weight gain is conserved in anorexia and thus only modestly affected by the severity of malnutrition.