Before tamoxifen administration, the mean body mass of Mstn[f/f] mice was slightly greater than that of Mstn[w/w] mice, 27.6 vs. 26.1 g (P<0.01). Mice of both genotypes lost some body mass during 2 weeks of tamoxifen feeding (mean 2.2 g in Mstn[f/f] and 1.8 g in Mstn[w/w], P>0.3 for genotype difference). During the first 2 weeks after tamoxifen feeding, before the mice started consuming the high-fat diet, Mstn[f/f] mice gained more weight than Mstn[w/w] mice (4.3 vs. 2.5 g, P<0.001), presumably because of a rapid increase in muscle mass (in a recent unpublished study, we determined that gastrocnemius and quadriceps muscles were 25–30% larger in Mstn[f/f]/CreER+ mice than in Mstn[w/w]/CreER+ mice 2 weeks after the cessation of tamoxifen feeding). Thus, body mass was greater (P<0.001) in myostatin-depleted mice at the onset of high-fat feeding (29.1 vs. 26.5 g) and at the same time point in mice that continued on the low-fat diet (30.1 vs. 27.0 g). After the more rapid weight gain in Mstn[f/f] mice during the first 2 weeks post-tamoxifen, there was no further effect of myostatin depletion on the increment in total body mass in mice in which the low-fat diet was continued during the final 22 weeks of the study (3.8 g in myostatin-deficient mice; 3.6 g in normal mice).
Over the first 8 weeks of high-fat feeding, the cumulative mean weight gain of the myostatin-deficient mice was only half that of the mice with normal myostatin expression (, upper panel). Beyond 8 weeks, the rate of weight gain was similar in myostatin-deficient and control mice, but cumulative weight gain remained lower in the myostatin-deficient mice. By ANOVA, the pattern of biweekly weight changes (, lower panel) was significantly different in control and myostatin-deficient mice (P<0.05 for myostatin status×time interaction). The only individual time interval with significantly lower weight gain (P<0.05) in myostatin-deficient mice was the interval between 2 and 4 weeks after starting the high-fat diet. Myostatin-deficient mice had a greater body mass when the high-fat diet was started (because of increased muscle mass), so their smaller weight gains did not result in significantly lower body mass at any time point (P>0.1, not shown except for final weights in ).
Mean (±SEM) change in total body mass.
Effects of myostatin depletion and high-fat feeding on body and organ mass.
At the end of the study, the mass of fat within the abdominal cavity was more than 5-fold greater in the high-fat groups than in low-fat groups (). Myostatin depletion did not significantly affect the mass of epididymal or retroperitoneal fat at the end of the study in either the low-fat or the high-fat groups. In mice with normal myostatin genes, the high-fat diet increased hepatic mass by 30% (). In myostatin-deficient mice, the high-fat diet did not consistently increase hepatic mass (mean+5%, P
0.4). Myocardial mass was increased by the high-fat diet, 14% in control mice and 10% in myostatin-deficient mice. Kidney mass was not significantly affected by the high-fat diet in either normal or myostatin-deficient mice (P
>0.1, ). We did not weigh the gastrointestinal tract or other organs, but by visual inspection we did not notice any major effects of the high-fat diet or myostatin depletion on the size of internal organs. Skeletal muscle mass (gastrocnemius, quadriceps, and triceps muscles) was about 30% greater than normal in myostatin-deficient mice, as expected, regardless of dietary condition (, ).
Mean (+SEM) muscle and intra-abdominal adipose tissue mass.
Of the total difference in mean body mass of 19.9 g between low-fat and high-fat groups with normal myostatin expression (), less than 5 g can be accounted for by the increased mass of intra-abdominal fat, liver, and heart. Skeletal muscle mass was not increased to any significant extent by the high-fat diet (), meaning that we cannot account for ~15 g of the extra body mass. It was very clear that subcutaneous fat mass was markedly increased after 5 months of high-fat feeding. This compartment might explain most of the unaccounted weight gain, but this was not quantified. The difference in body mass between the low-fat and high-fat myostatin-deficient groups was 14.1 g, and only ~4 g of this extra weight can be explained by the increased mass of abdominal fat, liver, and heart. Thus, the unexplained portion of the weight difference between myostatin-deficient mice fed the high-fat diet versus those fed the low-fat diet was ~10 g.
Myostatin depletion did not significantly affect blood glucose levels in mice fed a low-fat diet (). As expected, the high-fat diet induced hyperglycemia, both 5 hr after food was withdrawn and 90 min after ip glucose injection (P<0.001, ). In mice fed the high-fat diet for 5 months, myostatin depletion did not affect fasting glycemia but reduced glucose levels 90 min after ip glucose (P<0.01). The same pattern of mean glucose levels was observed after 3 months of high-fat feeding, but 90 min data were available for only 3 myostatin-deficient mice and therefore there was limited power to assess statistical significance.
Mean (+SEM) blood glucose concentrations.
The high-fat diet induced an increase of ~4-fold in the amount of fat in triceps muscles in mice with normal myostatin levels (). In contrast, the high-fat diet increased muscle fat content only 60% in myostatin-deficient mice. The method used to assess muscle fat content did not differentiate between fat within the muscle fibers and fat that accumulated in adipocytes between muscle fibers. Fat deposition in the liver also was attenuated in myostatin-deficient mice (). All 12 livers from control mice fed the high-fat diet received the highest steatosis score, whereas only 5 of 13 livers from myostatin-deficient mice received the highest score. The difference between control and myostatin-deficient mice in the distribution of steatosis scores was significant by Fisher's exact probability test (P<0.01).
Mean (+SEM) lipid mass in triceps brachii muscles.
Hepatic steatosis scores and representative micrographs of liver sections.