Interaction of Fat Cell Size, Protein Intake & Co. W/ Fat Gain + Insulin Res. in Overfed Men + Women in Metabolic Ward

That's rather the low protein variety of overfeeding... but wait, was the high protein diet even "high" in protein? Well high enough to affect liver fat, for sure.

You will probably remember José Antonio's high protein overfeeding study series (learn more) from the articles here and on the SuppVersity Facebook page. The results were quite impressive, but the number of controlled covariates were small and the dietary control was limited to food logs.

In a more recent study, George A. Bray and colleagues from the Pennington Biomedical Research Center of the Louisiana State University System, the George Mason University, and the FL Hospital & Sanford-Burnham Prebys Discovery Research Institute (Bray. 2016) determined the effect of overfeeding diets with 5%, 15% or 25% energy from protein on glycemia + body fat distribution in healthy men and women with add. covariates and in a metabolic ward.
In total, 15 men and 5 women were overfed by 40% (extra calories above maintenance) for 56 days with diets containing

• 5% (LP) of the total energy as protein, 
• 15% (NP) of the total energy as protein, or 
• 25% (HP) of the total energy as protein

Insulin sensitivity was measured using a two-step insulin clamp at baseline and at 8 weeks. Body composition and fat distribution were measured by DXA and multi-slice CT scan ... so far not so different, but the subjects were contained in a metabolic ward, cheating on the diet was thus as impossible, as taking supplements or working out like maniacs.

Figure 1: Diagram that illustrates the 8-weekstudy design; N = 10 male, 5 female subjects (Bray. 2016).

In conjunction with the scientists' analysis of the subjects abdominal subcutaneous fat cell size, which was determined on osmium fixed fat cells, these are two strengths of a study, of which it is yet quite obvious that it also had its disadvantages:

• Review the effects of different macronutrients in overfeeding studies | more

  the protein content of the diet is simply hilarious - that's not just because eating 5% protein, only is nothing but idiotic, but also because 25% of protein is far away from what can be considered "high protein" these days;
• the lack of exercise limits the significance of the results - at least for the majority of SuppVersity readers overeating in phases in which you don't exercise is probably nothing they would even remotely consider.

The scientists observations that neither the subjects' insulin sensitivity and free fatty acids during low and high levels of insulin infusion did not differ after 8 weeks of overfeeding.

Figure 2: Effect of 8 weeks of overfeeding on abdominal fat distribution, ectopic lipid; rel. changes (Bray. 2016).

What did differ, however, were the changes in body fat distribution according to DXA and how the latter depended on the protein content on fat cell size before the overfeeding period. More specifically, ...

• the fat free mass (FFM) and intrahepatic lipid increased more on the high protein, whereas 
• % BF and fasting free fatty acids (FFA) increased more on the low protein diet, while

In addition, the scientists observed that a high initial fat cell size predicted increased visceral fat gains and the FFA suppression during the high-dose insulin clamp.

Figure 3: Relation of Baseline Fat Cell Size to Change in Visceral Adipose Tissue Mass with Eight Weeks of Overfeeding in heathy volunteers (VAT 0.040 +/- 0.70(FCS); P < .0063 | Bray. 2016)

The subjects' insulin levels at baseline, on the other hand, predicted the increase in subcutaneous but not visceral fat accumulation (see Figure 3) - most intriguingly with low fasting insulin
at baseline correlated predicting higher changes in % fat (for insulin the scientists observed a correlation with r = –0.43; P < .034), but not with other variables. It is thus not surprising that the most insulin sensitive subjects also gained the most subcutaneous fat... or, as the scientists put it: "HOMA IR predicted the increase in DSAT (r = 0.50; P <.016), but not other variables" (Bray. 2016).

Those are important insights of which the authors rightly point out that they clearly indicate that "an induction of insulin resistance with overfeeding is related to fat cell size and requires more than an expansion of adipose tissue stores" (Bray. 2016).
References:

• Bray, George A., et al. "Effect of three levels of dietary protein on metabolic phenotype of healthy individuals with 8 weeks of overfeeding." The Journal of Clinical Endocrinology & Metabolism (2016): jc-2016.