I'm always amused when a rodent study that seemingly counters the calorie-based theories on obesity comes down the pike. In predictable fashion, someone in the LC webosphere will pick up on it and use it as an example to counter calorie-based theories on weight/obesity. A recent example of this is brought to us by low carbers' favorite comedian edutainer, Tom Naughton: Fat Mice & The Laws of Thermodynamics that referenced the following study: Mild calorie Restriction Induces Fat Accumulation in Female c57BL/6J Mice.
Now Tom starts out misrepresenting things by picturing a fat mouse, likely an ob/ob. This study actually involved females of the strain pictured below:
Kind of a cute little critter, no? The study was also conducted in the context of calorie restriction and its demonstrated increase in longevity of mice. They are not a mutant strain genetically predisposed to obesity.
These mice can live almost 3 years and average about half that. So the two experiments were conducted during normal growth (4 weeks for 8 week old mice in one study, 3 weeks for 10 week old mice in the other). The mice were not made obese then put on a calorie restricted diet. Both groups of mice grew slightly or not at all during the studies. So this study was NOT looking at weight loss in the obese, it was looking at the effect of restricting calories in a normal mouse and testing the response. Now reproductively these mice were adults, but these mice continue to grow throughout their lives. So this might be somewhat akin to putting normal weight young children on calorie restricted diets and seeing what happens. The results may well hold a clue to those impoverished cultures with high rates of true undernutrition in childhood leading paradoxically to obesity in adulthood, but that's not what CDS sufferers see.
Now, before I continue on to discuss this study and its implications, I simply can't help but point out that the last link in the previous paragraph is to a study entitled The High-Fat Diet–Fed Mouse. In this study they fed this very same strain of mouse the usual (11%) fat chow and a high (58%) fat diet. The regular diet was quite high carb (63%) while the high fat diet was fairly low carb (~25%). The results: "Body weight was higher in mice fed the high-fat diet already after the ﬁrst week, due to higher dietary intake in combination with lower metabolic efﬁciency." Why did these mice ACTUALLY *get fat*? They ate more! Someone ought to write a book on this stuff. (I guess Marta forgot to turn up their thermostats and get those futile cycles spinnin, but I digress ...) Something else happened to the fattening mice, they became hyperglycemic, hyperinsulinemic, and had impaired glucose tolerance. This sounds to me like mice overeating on a high fat diet getting obese and diabetic while those eating low fat/high carb diets maintain proper energy balance for normal growth and metabolic regulation to me. Oh wait ... that can't be right! Yeah, I forgot ... this is a mouse study and therefore has no relevance to humans after all. Phew! Explained that one away nicely. Nothing to see here folks, let's move on ...
Sorry, but before we do, let's summarize: This mouse grows relatively insignificantly in mass over the span of 3-4 weeks when fed a low fat - high carb diet in either ad libitum or calorie restricted fashion. With CR, the metabolism slows so that these mice divert more intake towards fat stores than lean mass. The very same strain of mouse fed a relatively low carb high fat diet gets fatter and heavier because the mice eat more.
OK ... So back to the original study, thermodynamics, etc. All of the mice were acclimated to eating ad libitum (AL) throughout their lives up to the start of the study. There were two study "legs":
I: 8 week old mice divided into 2 groups AL v. CR for four weeks. Lean and fat mass were assessed.
II: 10 week old mice divided into 2 groups AL v. CR for three weeks. Total fat mass, distribution between various fat depots, resting energy and total energy expenditures (REE,TEE) were measured.
In Study I, both groups increased body weight slightly and similarly. But the AL added more lean mass while the CR added more fat mass. The result reporting as percents is a bit misleading IMO. Note that we're talking mice that weighed around 20g. After 4 weeks the CR mice had 3.37 ± 0.23 g fat mass vs. 2.00 ± 0.09 g. So while the CR group added 68% more fat mass, we're talking 1.37 g of a 20 g mouse - e.g. a redistribution of <7% bodyweight in a period. I also note the variability for the CR mice is quite a bit higher than that for the AL mice indicating that some of the CR mice faired differently than others.
OK, so the CR mice got fatt-I-er, but not "fatter", hence the title of this blog post. The CR mice accumulated more fat tissue compared to the AL mice, and added less lean mass, but, they didn't gain more weight. In other words, they did not become obese. This is important because if we're talking "Why We Get Fat" and saying that it's all about dysregulated fat that accumulates lipid, this study simply does not support such a thesis.
For the overweight ladies in the audience (since this study was in females) who became so in adulthood, think now about how you got fat for a moment. Did you start out a 130 lb woman (or whatever "normal" is for your height/build) with 20-25% body fat and start noticing that despite weighing 130 lb week in and week out for a period of years you started turning to goo and found yourself with 50% body fat? Of course not. When we get fatter, we see the weight on the scale increase week in and week out as we become, say, a 180 lb woman. Does that mean each of us gained 50 lbs of fat? Of course not. I discussed this HERE.
Back to the study, in Part II, they measured energy expenditure and lo and behold TFLOT holds up pretty well. In this part of the study, both groups of mice were relatively weight stable for three weeks. The 5% CR corresponded to a 5% reduction in TEE and both groups were thus in energy balance. The resting energy expenditure (REE) dropped 20% (from 6.09 to 4.83 kcal/day, a difference of 1.26 kcal/day) while total energy expenditure dropped 5% (from 8.38 to 7.97 kcal/day, a difference of 0.41 kcal/day). For the AL mice REE/TEE was ~73% while it was only ~61% of the CR group. Doing the math, the other energy expenditures in the CR group must have gone up even though it wasn't noticeable with the locomotive activity or BAT (brown adipose tissue) thermogenesis (although they didn't measure body temp, rather UCP activity so it is possible that there was temperature compensation not related to BAT). Seems the mice were conserving energy by lowering the "idle" on the engine, all the while expending a bit more on something - could that be the mouse equivalent of fidgeting or other nervous behavior resulting from sensing minor starvation? Who knows. But, the mice that accumulated the most fat also expended a greater proportion of their intake on something other than basal metabolism.
So, where does this leave us in the end? Well, we do know that a certain level of body fat does seem to improve survival rates for disease and injury (sorry no handy reference at the moment). Perhaps CR, by causing the mice to be more efficient at storing energy and putting a bit less into bone/muscle/organ growth end up with bodies that require fewer calories to maintain later in life and allow for sufficient reserves to get through harder times. Makes sense to me. I'm being repetitive, but it is important to note: they did NOT become obese or even fat. The mouse's organs and systems were not strained by managing a significantly heavier body. Even on a small mouse, about one and a third gram is just not a lot of fat spread around several depots - that's like 1/4th of one teaspoon of butter.
OK, so how does this translate to humans, weight gain and weight loss? Well, firstly, when we get fat we GAIN weight. We don't just start redistributing lean and fat mass. And what of losing weight. Isn't that 5% change in TEE in response to a 5% CR proof that the body is adaptable so, that whole 100 cal/day = 10 lbs in a year is totally wrong? Well, for this we could go on and on with the rodent studies, OR, we could acknowledge that we have a crap load of data for what happens to actual humans who are already over-fat and look at that to see if there's really much to be learned here. From the discussion in this very paper we have the following:
A growing body of literature has demonstrated that moderate or severe CR (30–60%) leads to significant changes in body composition including reduced body fat mass and/or lean mass, with or without weight loss (4–6,22,23). For example, 25% CR in humans resulted in a clear decline of body fat mass and fat-free mass after 6 months (5). In 6-month-old mice, 55% CR resulted in a 71% reduction in total fat mass after 6 months (22).
Let us also keep in mind the metabolic rates vs. body masses for various mammals as well. The energy expenditure:body weight ratio is substantially higher for mice than humans. See, for example, the graphic below:
Considering that body temp maintenance is a major component of REE, the mice seem to have a disproportionately large ability to conserve energy as compared to, for example, us humans, because it's a bigger "cost" per unit mass for them.
Here is just one study that quantified both REE and TDEE for actual free-living humans (as luck would have it obese women), that I've blogged on previously. The study looked at LC vs. LF diets and EX (3X 45 min moderate cardio) vs. NX (no exercise) by comparing 4 groups of dieters. The changes in body composition, REE and even TEF did not differ between the two diet groups. For both diets, REE was reduced by ~130 cal/day. This is GOING to happen folks! IF you are losing weight without changing activity, you are in caloric deficit from eating less. Even TEF wasn't different - I've long suspected that this is because the Atwater factors represent metabolizable energy that seems to have been partially if not fully adjusted for the "cost" of extracting the energy.
What influenced body composition most? The much maligned slogging cardio. Yeah, 3X 45 minutes of 15 minute sessions on one of three cardio machines. From the discussion:
More important than total weight losses [of which LC were slightly more than LF], however, are the relative changes in FFM and FM. Exercise training was a major determinant of the changes in body composition, with FM comprising 89.4% of the weight loss in the Ex group, compared with only 71.3% in the Nx group.
So much for exercise being useless for fat loss, but I digress ... In any case, the exercisers had their intake matched to approximate the caloric cost of the exercise itself so that the "calculated" caloric deficit would be the same for both groups: e.g. they got to eat a bit more. But let's look at TDEE:
Directly due to Exercise: EX: +0.07 ± 1.23 NX: -1.46 ± 1.04 MJ/d
Not attributable to Exercise: EX: +0.75 ± 1.06 NX: -0.61 ± 1.03 MJ/d
As I've stated before, I'm not sure these are additive, but +0.82 MJ vs -2.07 MJ between the two. This is a swing of almost 3 MJ or ~ 700 calories!! The result is that the exercisers lost 2.7 kg more on average (that's almost 6 pounds) of FAT. And why? Because they actually increased TDEE from baseline rather than reducing it. And not just from the exercise, but because NEAT (non-exercise activity thermogenesis) increased. So much for the theory that exercisers just lounge around on couches more the rest of the day to compensate. But again, I digress ...
If there is a real lesson to be learned here folks, it is that we should strive for our children to eat a diet that allows them to grow normally and remain lean eating ad libitum, because once we get fat, it's a minefield out there what to do about it. And if we find ourselves as overweight adults, there's no easy fix. We gain and lose both lean and fat mass and we have limited control over it. When we lose weight, TDEE goes down in part because REE decreases (and stays there for the most part) but NEAT goes down too. Exercising can help counter this.
Eat Less, Move More ... Move More you can eat a little more.