Restriction of protein synthesis abolishes senescence features at cellular and organismal levels

Abstract

Cellular senescence or its equivalence is induced by treatment of cells with an appropriate inducer of senescence in various cell types. Mild restriction of cytoplasmic protein synthesis prevented induction of all aspects of cellular senescence in normal and tumor-derived human cells. It allowed the cells to continuously grow with no sign of senescent features in the presence of various inducers. It also delayed replicative senescence in normal human fibroblasts. Moreover, it allowed for growth of the cells that had entered a senescent state. When adult worms of the nematode C. elegans were grown under protein-restricted conditions, their average and maximal lifespans were significantly extended. These results suggest that accumulation of cytoplasmic proteins due to imbalance in macromolecule synthesis is a fundamental cause of cellular senescence.

Figure 1. Effect of inhibition of protein synthesis on the loss of colony forming ability induced by several agents in HeLa cells.

(a) Cells were cultured in the presence of the agents together with the inhibitors of protein synthesis indicated for 2 weeks, and colonies formed were stained with Coomassie Brilliant Blue. None, no addition; EdT, 2 mM excess thymidine; CPT, 7.5 nM camptothecin; AC, 0.3 μM aphidicolin; + CHX, 0.15 μM cycloheximide; + Aniso, 0.057 μM anisomycin; + RAP, 0.055 μM rapamycin; + Eme, 0.033 μM emetine; + ActD, 0.8 nM actinomycin D. (b) Cells were cultured under methionine-restricted conditions for 2 weeks, and processed as in (a). None, no addition; Met3-Met100, 3–100 mg methionine/L. (c) Quantification of the colony numbers in (a) (n = 3). (d) Quantification of the colony sizes in (a) (n = 50). Colony sizes were measured for 50 colonies. Bars denote means with a standard error of the mean (SEM). *P < 0.05, **P < 0.01 and ***P < 0.001 in comparison with the cells added with the agent.

Figure 2. Effect of cycloheximide on senescence features induced by several agents in human cells.

(a) HeLa cells were cultured in the presence of the agent indicated for 2 weeks, stained with the senescence-associated ß-galactosidase, and subjected to photography. None, no addition; CHX, 0.15 μM cycloheximide; EdT, 2 mM thymidine; CPT, 7.5 nM camptothecin; AC, 0.3 μM aphidicolin. (b) TIG-7 cells were cultured in the presence of the agent indicated for 3 weeks, and processed as in (a). None, no addition; CHX, 0.15 μM cycloheximide; EdT, excess 1 mM thymidine; CPT, 2 nM camptothecin; AC, 0.2 μM aphidicolin. (c) Numbers of HeLa cells stained with the senescence-associated ß-galactosidase were calculated (n = 3). (d) Numbers of TIG-7 cells stained with the senescence-associated ß-galactosidase were calculated (n = 3) as in (c). (e) Colonies of HeLa cells were stained with Coomassie Brilliant Blue. (f) Colonies of TIG-7 cells were stained with Coomassie Brilliant Blue. Bars denote means with a standard error of the mean (SEM). **P < 0.01 and ***P < 0.001, respectively, in comparison between -CHX and + CHX. Scale bars, 100 μm.

Figure 3. Effect of cycloheximide on growth of human cells cultured with inducers of senescence.

(a) HeLa were cultured as in Fig. 2, and cell numbers were counted at intervals (n = 4). (b) TIG-7 cells were cultured as in Fig. 2, and processed as in (a) (n = 4). Bars denote means with a standard error of the mean (SEM). ***P < 0.001 in comparison with the cells added with the agent alone.

Figure 4. Effect of cycloheximide on protein content and cell volume in human cells.

(a) HeLa cells were cultured for 7 days as in Fig. 2, and total protein content per cell (n = 3-7) were determined. (b) TIG-7 cells were cultured for 7 days as in Fig. 2, and processed as in (a) (n = 3). (c) Mean volume of HeLa cells (n = 50) were determined. (d) Mean volume of TIG-7 cells (n = 50) were determined, and processed as in (c). Bars denote means with a standard error of the mean (SEM). *P < 0.05, **P < 0.01 and ***P < 0.001 in comparison between -CHX and + CHX.

Figure 5. Western blot analysis of ERK1/2 in HeLa cells.

(a) Cells were cultured in the presence of the agents indicated for 1 day, and cell extracts were subjected to western blot analysis using specific antibodies to total and phosphorylated ERK1/2 as described previously⁴⁴. None, no addition; CHX, 0.15 μM cycloheximide; EdT, 2 mM excess thymidine; CPT, 7.5 nM camptothecin; AC, 0.3 μM aphidicolin. (b) Cells were cultured for 5 days, and processed as in (a). The full-length images are shown in Supplementary Fig. 7.

Figure 6. Effect of cycloheximide on replicative senescence in normal human fibroblasts.

(a) TIG-7 cells at 67 PDLs were cultured for 1 month to form colonies in the presence and absence of 0.075 μM cycloheximide. Colonies were stained with the senescence-associated ß-galactosidase and Coomassie Brilliant Blue. (b) TIG-7 cells at various PDLs were cultured by serial passages in the presence and absence of 0.075 μM cycloheximide to determine the maximum lifespan. Arrows indicate the time of addition of cycloheximide (n = 4). (c) TIG-7 cells at fully senescent state were cultured in the presence and absence of 0.075 μM cycloheximide for 1 month, and the cells were stained with the senescence-associated ß-galactosidase and Coomassie Brilliant Blue. (d) Fully senescent TIG-7 cells were cultured as in (c), and cell growth was monitored (n = 3). (e) Numbers of TIG-7 cells stained with the senescence-associated ß-galactosidase were calculated (n = 3) as in (c). ***P < 0.001 in comparison with the control cells. (f) Fully senescent TIG-7 cells were cultured for 1 month as in (c), and total protein content per cell was determined (n = 3). (g) Mean cell volume for the cells used for (f) (n = 50) was determined. Bars denote means with a standard error of the mean (SEM). *P < 0.05, ***P < 0.001 in comparison between -CHX and + CHX. Scale bars, 100 μm.

Figure 7. Effect of cycloheximide on longevity in C. elegans.

(a) Indicated numbers of adult N2 worms were placed on NGM agarose medium containing various concentrations (0.1 to 10 μM) of cycloheximide. At intervals the worms were picked up and inoculated on new plates and viable worms were counted every day. Three independent experiments gave similar results. (b) L1 worms were cultured for 11 days as in (a), and stained with the senescence-associated ß-galactosidase. (c) Numbers of worms whose body was totally stained with the senescence-associated ß-galactosidase were counted (n = 3) as in (b). Bars denote means with a standard error of the mean (SEM). *P < 0.05, ***P < 0.001 in comparison with the control worms.