Fasting levels of hepatic p-S6 are increased in old mice

Abstract

TOR is involved in aging in a wide range of species from yeast to mammals. Here we show that, after overnight fasting, mTOR activity is higher in the livers of 28 months old female mice compared with middle-aged mice. Taken together with previous reports, our data predict that the life-extending effect of calorie restriction (CR) may be diminished, if CR is started in very old age. In contrast, rapamycin is known to be effective, even when started late in life.

mTOR and Aging

The nutrient-sensing TOR pathway stimulates protein synthesis, cellular growth and functions.^1-4 mTOR drives geroconversion from cellular quiescence to senescence⁵ and is involved in organismal aging and age-related diseases.^6-10 In agreement, rapamycin extends life span in mice.^11-21 However, this does not mean that the activity of mTOR is necessarily increased in old mice compared with young mice. In fact, even unchanged mTOR activity would be excessive in old animals compared with young, growing animals. In theory the activity may even be decreased but still be inappropriately high for an aging organism.²² In analogy, consider a car moving at 65 mph on highway. When a car exits highway to a small street, then even 50 mph will be too fast. As a more relevant example, protein synthesis is decreased (but not enough) in old yeast, worm and fly. Still, inhibition of protein synthesis further prolongs life span.^23-26

Thus on pure theoretical grounds, the effect of age on mTOR activity is difficult to predict. First, we will discuss mTOR activity in cellular senescence.

mTOR in Cellular Aging

In the absence of mitogenic stimulation, normal cells become quiescent. In culture, cellular quiescence is usually caused by withdrawal of growth factors (serum) and/or nutrients. In quiescent cells, mTOR is not activated. The condition is reversible: growth stimulation causes cell proliferation. In proliferating cells, mTOR is active, which is manifested by phosphorylated S6 (p-S6). When the cell cycle is forcefully arrested, by DNA damaging agents and CDK inhibitors, then still active mTOR converts cycle arrest to senescence.^27-38 Any conditions that inhibit mTOR suppress geroconversion.^33,34,39,40 During geroconversion, levels of p-S6 are the same as its levels in proliferating cells. So both proliferating and senescent cells are characterized by activated mTOR. It is important to emphasize that chronically active mTOR renders cells resistant to acute mTOR activation.⁴¹ For example, chronic activation of mTOR by glucose decreases responsiveness to insulin.⁴¹ The response is diminished upstream of mTOR and absent downstream of mTOR (post-mTOR resistance). In the presence of active mTOR, insulin did not increase translation of proteins such as HIF-1a., Thus, basal activation of mTOR is a characteristic of senescent cells.

In summary, in vitro model of cellular senescence suggests that in the organism:

1. mTOR activity should be measured in the organs with minimal number of proliferating cells. In fact, proliferating cells are expected to have high levels of p-S6, mimicking senescent cells;

2. mTOR activity should be measured at “basal” levels. Food is a physiological activator of mTOR. Therefore, we choose to measure mTOR activity after fasting.

Fasting Levels of Hepatic pS6

We measured p-S6 and p-Akt in the liver in mice fasting overnight, comparing mice at the age of 28 months (old) and 9 months (adult, middle-aged). As shown in Figure 1, fasting p-S6 levels were higher in old mice than in middle-aged mice. Noteworthy, fasting levels of p-S6 were lower compared with non-fasting levels in middle-aged mice (Fig. 1). However, we did not have sufficient material to compare fasting and non-fasting levels in old mice. Careful analysis of the literature allowed us to find missing data. In the paper by Sengupta et al,⁴² Figure 4 C shows that fasting strongly decreased levels of p-S6 in young mice, while only marginally affected levels of p-S6 in the old mice. This publication and our data complement each other. Levels of p-Akt-S473 were not changed in old mice (Fig. 1). Noteworthy, total body weight and insulin levels were slightly yet statistically significantly elevated in old mice (Fig. 2).

Figure 1.

Hepatic p-S6 and p-Akt in old and adult (middle-aged) mice. (A) Immunoblot analysis. Protein lysates from the liver of 28 and 10 month old (old vs middle aged) female C57BL/6NCr mice were prepared as described.^56,57 Fasted: mice were fasted overnight. Non-fasted: regular conditions. Numbers indicate individual mice. (B) Quantitative analysis of p-S6 signal shown in panel A. Data presented as mean ± SE.

Figure 2.

Comparison of weight and metabolic parameters in old (28 months old) and middle aged (10 months old) female mice of C57BL/6NCr strain. Blood was collected after overnight fasting and concentrations of insulin and triglycerides were determined in blood sera using, using Insulin (Mouse) Ultrasensitive ELISA kit (Alpco Diagnostics, Salem, NH) and Triglyceride Colorimetric Assay kit (Cayman Chemical Company, Ann Arbor, MI), respectively. Glucose was measured directly in blood using Accu-Chek Aviva strips (MaKesson, Atlanta, GA). Data presented as mean ± SE.

Implications for Health- and Life-Extension

In rodents, life-long CR prolongs life span by 40% in some strains.⁴³ The life extending effect of CR can be explained by inhibition of mTOR.^44-49 Fasting only minimally downregulated p-S6 in old mice, compared with young animals.⁴² As we showed here, fasting levels of p-S6 were higher in old mice compared with middle-aged animals. This predicts that fasting is less effective in old animals than in young/middle-aged animals. Therefore, late-life CR should be less effective than life-long CR. Indeed, the later in life the onset of CR, the less extension of life span was observed.^50,43 Furthermore, neither CR restriction nor amino-acid restriction had any longevity benefits in very old humans.^43,51 In contrast, rapamycin extends life span of mice, when started late in life.^11,52 Rapamycin also extends mouse lifespan when given as a brief (6 week) treatment to aged C57BL/6 mice.⁵² Thus in old humans, rapamycin rather than CR should be considered to extend health- and life span.

Noteworthy, physical exercise and IGF-I/nutrients cause mTOR-dependent increase of protein synthesis and muscle hypertrophy. In the aging muscle, this response is diminished.^53,54 This may be caused by basal mTOR activity in aged mice,⁵⁵ which precludes appropriate response to stimulation.⁵³ Therefore, in order to prevent sarcopenia, one can consider an intermittent rapamycin treatment. This may decrease basal levels of mTOR activity (during rapamycin treatment), while allowing physical exercise and other physiological stimuli to induce mTOR (during rapamycin withdrawal). This prediction needs to be tested.