Rapamycin-mediated lifespan increase in mice is dose and sex dependent and metabolically distinct from dietary restriction

Abstract

Rapamycin, an inhibitor of mTOR kinase, increased median lifespan of genetically heterogeneous mice by 23% (males) to 26% (females) when tested at a dose threefold higher than that used in our previous studies; maximal longevity was also increased in both sexes. Rapamycin increased lifespan more in females than in males at each dose evaluated, perhaps reflecting sexual dimorphism in blood levels of this drug. Some of the endocrine and metabolic changes seen in diet-restricted mice are not seen in mice exposed to rapamycin, and the pattern of expression of hepatic genes involved in xenobiotic metabolism is also quite distinct in rapamycin-treated and diet-restricted mice, suggesting that these two interventions for extending mouse lifespan differ in many respects.

Keywords: IGF-1; aging; caloric restriction; glucose; insulin; longevity; mTOR; mouse; rapamycin; xenobiotic metabolism.

Figure 1

Survival curves at varying doses of rapamycin. Survival curves for male (top) and female (bottom) mice exposed to varying doses of rapamycin from 9 months of age. Data are pooled across the three test sites. At the time of analysis, fewer than 1% of the mice were still alive. Significance tests, median age, age of 90% mortality, and numbers of mice are shown in Table 1.

Figure 2

Weight changes in control and rapamycin-treated mice. Weights of control and rapamycin-treated mice at varying ages. Rapamycin treatment was initiated at 9 months of age. Points indicate mean values, pooled across the three test sites, for groups of approximately 150 mice in each rapamycin group and 300 control mice.

Figure 3

Metabolic status of dietary restriction (DR) and rapamycin-treated mice. Endocrine and metabolic endpoints in control, rapamycin-treated (at 14 ppm), and DR mice evaluated in plasma at 9 months of age after an overnight fast. DR and rapamycin exposure were begun at 4 months of age. Each bar shows mean and standard error of the mean for 8–10 animals of each sex. For insulin, IGF-1 and FGF-21, anova indicated a significant difference between males and females, and thus, each sex was analyzed separately. Sexes are pooled for the analyses of glucose, T4, and leptin. An asterisk indicates significant difference from control mice at P < 0.05. DR and Rapa mice differed at P < 0.006 for each endpoint, except IGF-1, for which there was no significant difference.

Figure 4

One-month exposure to rapamycin impairs glucose tolerance. UM-HET3 mice, at 4 months of age, were placed onto diet containing the indicated doses of rapamycin at UT and evaluated 1 month later for responses to intraperitoneal glucose injection. N = 5 of each treatment group for each sex. The upper panels show mean ± SEM for plasma glucose at the indicated times. The bottom panels show integrated area under the glucose curve as mean ± SEM; bars that share a letter code are not statistically significant by t-test.

Figure 5

Expression of mRNA for liver genes involved in xenobiotic metabolism. Effects of rapamycin (top panel) and DR diet (middle panel) on hepatic mRNA levels for 52 genes related to xenobiotic metabolism. Top and middle panels: mice were placed on rapamycin (14 ppm) or DR diet at 4 months of age and euthanized at 12 months of age. There were six males and six females in the control group and in each of the two treatment groups. All mice were housed at UM. Data from treated female mice are shown in the top of each panel, bars pointing upwards, and data from treated male mice are shown in the bottom half, with bars pointing downwards. The length of each bar is shown on a log2 scale, as the ratio of treated mice divided by control mice. Bars shown in blue are increases compared with control, and bars shown in red are decreases compared with control mice. Thus, a blue bar with a value of eight represents an increase of 2⁸=256-fold above control levels, and a red bar with a value of eight represents a decline of 256-fold below control levels. Dark blue and dark red bars show genes for which the nominal (unadjusted) P-value is P < 0.05. Lighter blue and red bars shown genes that do not reach this arbitrary significance threshold. The first gene, for example, Sult2a2 (see Table S3), is increased by DR, slightly but significantly, in both males and females, is dramatically increased (by 2^11.6 = 3100-fold) by rapamycin in males, but shows a large (2^12.4 = 5400-fold), significant decline in rapamycin-treated females. The cyan arrowheads point out the three mRNA that are elevated by both DR and rapamycin in both sexes.The bottom panel shows female/male ratios, for untreated UM-HET3 controls; genes at the left are expressed at levels approximately 2¹²-fold higher in female controls, and genes at the right are expressed roughly 2¹¹-fold higher in male controls.