The present study investigated sex differences in body compositional changes in response to a PRO weight loss diet compared to an isocaloric CARB weight loss diet in middle-aged adults. The main finding of this study was that diet and sex impacted changes in body composition independently and additively for the whole body, trunk and leg FM. No interactions of diet and sex were found on either whole body or regional body composition suggesting that males and females respond similarly to caloric restriction diets differing in protein content. To our knowledge, this is the first study to examine whether a higher protein diet can confer differences between men and women in degree or location of FM and LM reductions.
Some studies suggest that higher levels of protein in the diet may have a direct influence on degree of lean mass lost during energy restriction [9, 10, 14]. Because men generally have more LM, their absolute protein needs may be greater than in women; therefore, providing adequate protein during weight loss may be important in preserving LM. Contrary to our present results, Farnsworth et al., found that LM is preserved in women, but not in men on a higher protein diet . The study by Farnsworth  was limited by a small number of men and a shorter duration (16 weeks total) than in the present study. Also, different protein levels between studies may partly explain the different findings. The higher protein diet in the Farnsworth study provided approximately 110 g protein/d for both men (~1.0 g kg-1 d-1) and women (~1.2 g kg-1 d-1) .
Conversely, protein intakes in the current study were slightly higher in men (PRO: 130 g d-1, 1.3 g kg -1 d-1. CARB: 75 g d-1, 0.8 g kg-1 d-1) than in women (PRO: 100 g d-1 or 1.2 g kg-1 d-1. CARB: 64 g d-1, 0.7 g kg-1 d-1). One possible explanation for this discrepancy is that a threshold for LM maintenance was met in the current study, but not by the slightly lower protein intake relative to body weight reported by Farnsworth . A threshold effect for maintaining LM during negative energy balance has not been characterized; however, a previous report in post-menopausal women demonstrated that for every additional 0.1 g kg -1 d-1 of dietary protein intake, 0.62 kg LM was preserved during a 20-wk weight loss intervention . The range of protein intakes was below the current RDA for protein intake . Notably, a subsequent study demonstrates that protein intake of 1.5 g kg-1 d-1, which is above the current RDA, suppresses proteolysis thus inhibiting loss of lean mass [24, 25].
Although men lost more total weight than women, when expressed relative to baseline body weight both sexes lost similar amounts of weight (~10 %). Importantly, we found that more of the total weight loss was derived from fat relative to LM in men (63 % and 77 % for CARB and PRO, respectively) than women (57 % and 67 %), and that more fat relative to lean was lost in PRO participants of both sexes. This is similar to previous findings , whereas other studies report no such effect of diet . Although some studies show that sex does not influence the composition of weight loss from energy restriction [27, 28], this finding is inconsistent [29, 30]. In men, our results are similar to those reported in the literature, with the expectation that ~70 % of weight loss is comprised of FM during dieting alone . Women in the PRO group also compared similarly to what was expected for FM loss, while the women in the CARB group lost less FM. However, we found no significant impact of the diets on sex differences in whole body weight, FM or LM loss.
Given the well-known association between central adiposity and CVD and metabolic complications [32–35], identifying changes in body fat distribution with weight loss is of particular interest. It is well established that women have greater levels of adiposity than men , and the distribution of fat differs, with men storing a greater proportion centrally and women in the gluteofemoral region . However, with regard to the fat patterning, sex differences in changes in regional adipose depots through energy restriction are not well characterized. It appears that while both men and women lose FM from the abdominal area [7, 36] and the femoral region , men may lose more abdominal fat [9, 37] and women lose more femoral FM , even when matched on total weight and total fat loss . Our data support the sex disparity in region of fat loss. Although both men and women reduced the relative fat of the whole body, trunk and leg, and no significant sex difference in the effect of the diets was found, these improvements are more pronounced in men, with the greatest change found in the PRO male group. Further assessment of the changes in the distribution of fat loss, indicated men experienced a reduction in the ratio of trunk fat to leg fat, suggesting that a greater degree of fat loss was derived from the central region.
Little evidence is available in the literature on potential underlying mechanisms to explain regional differences in body composition changes between men and women undergoing weight loss. However, one possible explanation for differences between men and women in their response to a higher protein weight loss diet could be related to a greater post-meal Diet Induced Thermogenesis (DIT) found in men compared to women [38, 39]. Another explanation could be related to recent findings from an animal study, which showed that the protein content of a meal is the trigger for muscle protein synthesis, and this was shown to significantly increase energy expenditure, as measured by changes in adenosine 5’-triphosphate (ATP) and the signaling molecule adenosine monophosphate-activated protein kinase (AMPK) . Importantly, the content of the branch chained amino acid leucine in a higher protein diet plays a key role in mechanisms that support the preservation of lean mass . Other mechanisms, such as hormonal differences between sexes, may be responsible for the observed differences in body composition changes between men and women. Unfortunately, data from the present study are insufficient to examine underlying mechanisms.
Even though the sample size in the present study is larger than in most other studies, a diet by gender interaction may be only detectable with a larger baseline sample, which suggests that the effect of diet on sex differences in composition may be small. Men may benefit more from a greater protein intake during weight loss due to greater LM compared to women. A study designed to provide protein per kg of LM rather than in absolute terms or per kg body weight could elucidate this issue. Another consideration for our results is that the age range in the current study spans across menopausal status in women, potentially allowing hormonal status to influence the distribution of fat loss. Although our study was not designed to evaluate the impact of hormonal status or age per se on treatment effects, weight loss treatments that maximize FM loss while maintaining LM, especially in the legs are critical to combat the rising incidence of sarcopenic obesity. Indeed, statements in the literature regarding body composition and older adults indicate that this is a high research priority [11, 42]. Lastly, more accurate measures of regional body composition, such as with computed tomography or magnetic resonance imaging, could elucidate differences in loss of FM from the abdominal vs. gluteofemoral regions between men and women and perhaps differences in changes in LM, especially in the lower body which is critical for physical function.
In summary, although we found no significant sex differences in the effects of diet on how much and where FM and LM are reduced during weight loss states, there is some evidence that protein intake levels may impact the amount of LM preserved during weight loss, perhaps regardless of sex. Clearly, further intervention studies, designed to assess interactive effects of sex and macronutrient content of the diet, are required to determine whether protein intake recommendations under energy restriction need to be adjusted to help maintain LM while losing FM, especially in populations such as older adults who are at higher risk for sarcopenia.