Mitochondria and Obesity Revisited

Several months ago I blogged about the results from a Finnish twin study that found lower mitochondria numbers and disturbed mitochondrial energy metabolism activity in fat cells from identical twins who were leaner than their genetically identical co-twins. These impairments correlated with critical clinical measures of obesity including liver fat accumulation, reduced whole-body insulin sensitivity, hyperinsulinemia, hypoadiponectinemia and adipocyte hypertrophy.

In this month's issue of OBESITY, Tomas Gianotti and colleagues from the University of Buenos Aires, Argentina, report a significantly lower mitochondrial-to-nuclear DNA ratio (mtDNA/nDNA) in insulin resistant (IR) adolescents recruited out of a subset (n=175) of a cross-sectional, population-based study of 934 high school students. In this study, the mtDNA/nDNA ratio was also inversely correlated with HOMA index, a crude but simple measure of insulin resistance.

This study is very much in line with the notion that obesity-prone individuals may have impaired mitochondrial number and/or function resulting in increased risk for obesity.

From the aforementioned twin study, we know that the decreased number and function is not corrected by weight loss.

Indeed the question is whether or not mitochondrial number and function can be increased by prescribing higher activity levels? If yes, how much activity will be needed to reverse these changes? And most importantly, will people with impaired mitochondrial function actually be able to enjoy exercise enough to actually stick with this prescription?

Perhaps it is not obesity that causes impaired mitochondrial function but rather impaired mitochondrial number and/or function that predisposes to obesity. This impairment could be genetic but also due to intrauterine programing or perhaps simply luck of the draw (remember - all mtDNA comes from your mom).

Of course this is not an "excuse" for obesity as is often misinterpreted when data on the genetics and biology are presented. However, it is clear that if you have impaired mitochondrial number and/or function you are much more likely to become obese in an environment that promotes sedentariness than if you were dependent on physical activity to meet your basic needs for survival.

Remember, there were times, not too long ago, when people were actually paid to be physically active. Today, choosing to be physically active actually costs money (not to mention time).

AMS
Edmonton, Alberta

Proconvertase Gene Linked to Obesity

As unequivocally documented by twin studies, obesity is one of the most heritable complex traits. However, so far only a handful of genes that may play a role in the common "garden variety" of obesity (i.e. not just the rare monogenic forms) have been identified .

This week, a large research team led by Philippe Froguel (Picture) that included scientists from France, Denmark, Sweden, Germany and the UK report in Nature Genetics that relatively common variants of the PCSK1 gene, which codes for the proconvertase enzyme are associated with higher BMI. This enzyme is responsible for producing fully functioning versions of hormones such as insulin and glugagon that play important roles in carbohydrate metabolism but also for melanocortin, a key regulator of satiety.

Although the identified variants of the PCSK1 gene cause only relative minor functional changes in this enzyme, the effect on the relevant hormones is quite significant.

Obviously, no single gene or variant thereof can account for all of obesity and there is no reason to believe that all obesity is the same. Rather, it is clear that the heritability of obesity must be due to the distribution of a large number of variants of numerous genes in the population.

So when do we begin using genetic testing in our obesity clinics - not for a while I am guessing, i.e. till we can actually show that specific genes also predict better (or worse) outcomes with specific treatments.

In the dark, all cats are grey.

AMS
Edmonton, Alberta

Genes for Weight and Weight Gain are Different

We know from twin studies that measures of weight (e.g. BMI) tend to be highly heritable - i.e. monozygotic twins are far more likely to resemble each other in terms of weight than dizygotic twins.

We also know that the ability to gain (and lose) weight is very much determined by genetic factors - i.e. for the same degree of excess energy (or energy restriction), monozygotic twins tend to resemble each other in weight gain (or loss) more than dizygotic twins. For e.g. identical twins lose virtually the same amount of weight following obesity surgery, when surgery is performed in the same setting (Hagedorn et. al).

Given this relationship, one may easily assume that genes that control body weight are the same that control weight gain.

A new study by Jacob Hjelmborg from the University of Southern Denmark, Odense together with colleagues from Finland, Italy and the US, just published in OBESITY, suggest that this may not be the case.

Hjelmsberg and colleagues anlaysed data from the longitudinal twin study of the cohort of Finnish twins (N = 10,556 twin individuals) aged 20-46 years at baseline followed up for 15 years.

Simply stated, they found a high level of heritability of BMI levels at baseline (we knew this) and a relatively high heritability of weight gain over the observation period (this is also not new).

However, in their models, it turns out that the two phenomena only show rather modest correlation at the genetic level.

What this means is that while both baseline BMI and weight gain are genetically determined, they are probably each regulated by a different set of genes.

So, while one set of genes may determine how big you are, other genes may determine how large you can get.

We know that some people are large and just stay that way all their life without losing or gaining much. Others may start out at a given size and end up gaining a lot of weight. All depends on your genetic background (and of course how this interacts with your lifestyle and other environmental factors).

Just a reminder how complex genetics actually is.

AMS
Edmonton, Alberta

Obesity: It's all in Your Cells?

Yesterday I blogged about a remarkable Finnish twin study, in which the investigators went to the considerable trouble of finding monozygotic twin pairs who showed marked differences in body weight. The biggest predictor of weight gain in these genetically identical but weight-discordant co-twins was a markedly lower physical activity level, which in turn, declined even further as the obese co-twins packed on the pounds.

Assuming that this was not just a bunch of "lazy" co-twins, I wondered about what biological factors could possibly be causing these co-twins to be less physically activity. The answer to this question may lie in the results of another study by the same investigators in the same set of twins published in the open access journal PLoS Medicine.

In this study, Kirsi Pietiläinen and colleagues compared the genetic expression profiles in fat cells and macrophages between the obese and non-obese co-twins. Because, by design, the twins were genetically identical, they were able to normalise expression patterns for differences in genetic background, gender and age - thereby cutting through the considerable noise generally associated with expression studies.

In short, the authors found that the fat tissue from the obese co-twins showed a significant up-regulation of inflammatory pathways, significantly reduced mitochondrial DNA copy number, and disturbed mitochondrial energy metabolism—statistically most significantly, the decreased catabolism of branched-chain amino acids (BCAA). These impairments correlated with critical clinical measures of obesity including liver fat accumulation, reduced whole-body insulin sensitivity, hyperinsulinemia, hypoadiponectinemia and adipocyte hypertrophy.

In one individual, who the investigators were able to study before and after an additional weight gain of around 11 Kg over 3 years, mtDNA copy number was further reduced while serum BCAA concentrations and inflammatory activity increased even further.

Although the authors acknowledge that correlations do not prove causality, it is clear from the tone of their discussion that they believe that the metabolic derangements and low mtDNA copy count are a consequence of the obesity and are thus amenable to treatment by diet and exercise (the "politically correct" conclusion).

This is where I wonder if not the reverse may be true. I am no expert on mitochondrial biology, but I would assume that a key consequences of a reduction in mtDNA copy number is a decreased maximal capacity for oxidative phosphorylation, i.e., utilization of fat for energy production.

Assuming for a moment that these findings are also present in skeletal muscle, it would not be hard to imagine that these individuals are likely to find exercise more difficult and tiring than their co-twins with a normal mitochondrial population - less exercise means further weight gain and further decline in mitochondrial function - a nice little vicious cycle, if I ever saw one.

It is hard for me to image that in all 14 obese co-twins lack of physical activity alone was able to bring about the reduced mtDNA copy number, increased inflammation and reduced BCAA metabolism - somehow I find it easier to imagine that it was rather a malfunction in their mitochondria which significantly affected their ability to be (and enjoy being) physically active in the first place.

But of course, this is a chicken-or-egg question that cannot be resolved by the present study.

So the obvious questions now are: Can these co-twins be "rescued" by prescribing higher activity levels? How much activity will be needed to reverse these changes? And most importantly, will these co-twins stick with this prescription?

It's probably hard to enjoy exercise when there's a problem with your fuel cell.

AMS

Activity Determines Discordant Obesity in One-Egg Twins

So for Easter I wanted a posting with "Egg" in the title. What better paper to choose than the recent publication by Kirsi Pietiläinen and colleagues from Helsinki University published in last month's OBESITY?

In this remarkable study, Pietiläinen and colleagues went to the considerable effort of screening 1,870 young adult twin pairs to find 658 monozygotic (MZ) pairs, of whom only 14 (!) pairs reported a BMI difference of at least 4 kg/m2, with one twin being non-obese (BMI approximately 25 kg/m2) and the other obese (BMI approximately 30 kg/m2).

This effort alone shows how rare it is to find MZ twins that are discordant for weight, clear evidence for the well-known fact that body weight is one of the most heritable complex traits found in man, in fact, only marginally less heritable than height.

Nevertheless, the investigators were able to further characterize at least 10 of the 14 BMI-discordant twin pairs and readily identified the difference in physical activity as the key determinant of the discordance. Virtually all of the co-twins who ultimately became obese reported having been less physically active in adolescence than their non-obese co-twins. Furthermore, as they grew heavier, the obese co-twins' physical activity levels declined even further as did their self-perceived physical fitness. Based on accelerometer recordings, the obese co-twins had less than half the daily activity of their non-obese co-twins

In contrast, in the weight-concordant reference MZ twin pairs, physical activity patterns and fitness changed little during adolescence and were similar in the co-twins.

Thus, this paper appears to suggest that even in genetically identical individuals, large amounts of physical activity (or lack thereof) can "override" the genetic determinants of BMI.

What the paper, however, fails to tell us is why in these rare instances (only 14 out of 658 or 2% of MZ pairs) the obese co-twins were less active than their non-obese counterparts.

Why, as discussed in a previous posting on this blog, if the disposition to be physically active is such an 'innate" trait, did these 14 co-twins behave so differently from their siblings? Was it lack of interest, competing hobbies, sibling rivalry, injury?

Or was it epigenetics - i.e. post-conceptional modification of DNA that may involve paramutations, bookmarking, imprinting, gene silencing, X chromosome inactivation, position effects, reprogramming, transvection or maternal (intrauterine) effects - all of which could impact character traits or behaviour in later life?

So, while the paper shows that even in genetically identical individuals the heritability of BMI can be overridden by marked differences in physical activity, it does not provide the answer to whether or not this level of activity can be cognitively induced or, rather, happens as a result of rare quirks of nature.

Clearly, there are a few more eggs to crack before we fully understand why some people "chose" to be physically active and others don't.

Happy Easter!

AMS