Suppression of replicative senescence by rapamycin in rodent embryonic cells

Abstract

The TOR (target of rapamycin) pathway is involved in aging in diverse organisms from yeast to mammals. We have previously demonstrated in human and rodent cells that mTOR converts stress-induced cell cycle arrest to irreversible senescence (geroconversion), whereas rapamycin decelerates or suppresses geroconversion during cell cycle arrest. Here, we investigated whether rapamycin can suppress replicative senescence of rodent cells. Mouse embryonic fibroblasts (MEFs) gradually acquired senescent morphology and ceased proliferation. Rapamycin decreased cellular hypertrophy, and SA-beta-Gal staining otherwise developed by 4-6 passages, but it blocked cell proliferation, masking its effects on replicative lifespan. We determined that rapamycin inhibited pS6 at 100-300 pM and inhibited proliferation with IC50 around 30 pM. At 30 pM, rapamycin partially suppressed senescence. However, the gerosuppressive effect was balanced by the cytostatic effect, making it difficult to suppress senescence without causing quiescence. We also investigated rat embryonic fibroblasts (REFs), which exhibited markers of senescence at passage 7, yet were able to slowly proliferate until 12–14 passages. REFs grew in size, acquired a large, flat cell morphology, SA-beta-Gal staining and components of DNA damage response (DDR), in particular, γH2AX/53BP1 foci. Incubation of REFs with rapamycin (from passage 7 to passage 10) allowed REFs to overcome the replicative senescence crisis. Following rapamycin treatment and removal, a fraction of proliferating REFs gradually increased and senescent phenotype disappeared completely by passage 24.

Keywords: :

• MTOR
• aging
• hypertrophy
• rapamycin
• senescence

This article is referred to by:

A critical role for TORC1 in cellular senescence

Dissecting the role of mTOR complexes in cellular senescence

Introduction

In cell culture, mammalian cells undergo permanent cell cycle arrest (replicative senescence) after a limited number of divisions.^1,2 In human cells, progressive telomere shortening causes cell cycle arrest.³ Mouse embryonic fibroblasts (MEFs) possess sufficiently long telomeres and telomerase activity.⁴ Yet in MEFs, replicative senescence occurs after 5–10 divisions by telomere-length-independent mechanisms.⁵ These mechanisms may involve stresses and cellular over-stimulation.⁶ The onset of cellular senescence does not simply reflect the accumulation of cell divisions, but can be prematurely activated in response to oncogenic stimuli.^7,8 Senescence can be accelerated by expression of Ras, MEK, Akt^7-14 as well as depletion of TSC2,^15,16 all manipulations that activate mTOR (target of rapamycin).^17-23

During cell cycle arrest, the mTOR pathway accelerates senescence (geroconversion). Thus, active mTOR converts cell cycle arrest caused by p21, p16 and DNA damage as well as quiescence (caused by serum starvation) to irreversible senescence.^24-35 When the cell cycle is blocked and proliferation is impossible, then active mTOR promotes cell-size growth (hypertrophy) and cellular senescence. Rapamycin suppresses geroconversion, maintaining reversible cell cycle arrest and preventing such markers of senescence as hypertrophy, hypersecretory phenotype and pseudo-DDR.^28,29 Also, stimulation of mTOR in normal stem cells causes hyper-proliferation, pro-gerogenic conversion and cell exhaustion.^36-39 Not co-incidentally, the TOR pathway is involved in organismal longevity from yeast to mammals.^40-51

It is noteworthy that inhibition of TOR decelerates chronological senescence in yeast.^42,52-55 In amazing analogy, rapamycin also inhibits “yeast-like chronological senescence” due to metabolic self-destruction in mammalian cells.^56-58 Most importantly, inhibition of TOR decelerates replicative aging in yeast.^43,59-63 Yeast replicative lifespan is a measure of the number of divisions a mother yeast cell can undergo. When the mother cell has divided many times, it begins to enlarge, and its capacity to produce progeny decreases. Yeast replicative aging is comparable with telomere-independent replicative senescence observed in rodent cells in culture. During replicative senescence, cells become larger and eventually stop proliferating. In yeast, genetic deactivation of the TOR pathway decelerates replicative aging.⁴³ Here, we investigate whether rapamycin can decelerate replicative senescence of rodent cells in culture.

Results and Discussion

We used several preparations of MEFs from different mice. After 4–5 passages, cells become large, flat and SA-β-Gal positive. Although rapamycin does not block proliferation in most cell lines, it inhibited proliferation of MEFs cells (Fig. 1). We attempted to select concentrations of rapamycin that partially decrease mTOR activity without blocking proliferation. MEFs were very sensitive to rapamycin, which inhibited phosphorylation of S6 (pS6 is a standard marker of mTORC1 activity) at concentrations as low as 100–300 pM (Fig. 1A). However, at these concentrations, rapamycin also blocked cell proliferation (Fig. 1B). Therefore, it was impossible to dissociate effects of rapamycin on mTOR and proliferation. For further experiments, we used 30 pM rapamycin. As shown in Figure 2, rapamycin prevented senescent morphology: cells remained small and mostly SA-β-Gal-negative. However, over time, MEFs were losing the sensitivity to rapamycin, requiring an increase in concentrations of rapamycin, which then blocked proliferation. This made it difficult to prevent both quiescence (when rapamycin levels were too high) and senescence (when rapamycin levels were too low).

Figure 1. Dose response to rapamycin in MEFs. (A) Immunoblot. MEFs were treated with indicated concentrations of rapamycin for 24 h and then pS6 (using 2 different antibodies) and S6 were measured by immunoblot (B) Cytostatic assay. MEFs were treated with indicated concentrations of rapamycin for 3 d and then cells were counted. Results are shown as percent of control. The effect of nutlin-3a is shown for comparison.

Figure 2. The effect of rapamycin on morphology of aging MEFs. Cells were cultured with (Rapa) or without (control) 30 pM rapamycin. At passages 4–5 cells were stained for SA-β-Gal and microphotographed. Scale bar = 50 μm.

Therefore, in rat embryonic fibroblasts (REFs), we employed a different strategy: rapamycin was added when REFs were becoming senescent (passages 7–10). By passage 7, REFs became large and SA-β-Gal-positive (Fig. 3) and also acquired markers of DNA damage response (DDR), as evidenced by γH2AX and 53BP1 (Fig. 4). However, DDR does not necessarily indicate actual DNA damage. In senescent and mitotic cells, DDR may be activated in the absence of DNA damage. In senescence caused by non-DNA damaging agents, DDR was partially inhibited by rapamycin.²⁹ In starved senescent cells, DDR can be induced by serum stimulation,⁶⁴ which is known to activate mTOR.

Figure 3. Rapamycin decelerates replicative senescence of REF cells. REFs were cultured as described in Methods. After seven passages, rapamycin was added. After three additional passages, rapamycin was removed. Left column: Cellular morphology by Romanovsky-Giemsa. Center column: SA-β-Gal staining. Right: Immunofluorescence for pS6.

Figure 4. Immunofluorescence for γH2AX and 53BP1 in REF cells. REFs were cultured as described in Methods. After seven passages, rapamycin was added. After three additional passages, rapamycin was removed. Passage - p. REFs stained with DAPI (for DNA), for γH2AX (Ser139) and for 53BP1 at various passages (p) 1, 3, 7 before treatment with rapamycin and at passages 16 and 21 (after rapamycin treatment at p7–10). Immunofluorescence of γH2AX (red) and 53BP1 (green). Nuclei were stained with DAPI (blue), scale 20 μM.

After seven passages, we added rapamycin to aging REFs. At that time point, REFs proliferated very slowly, consistent with a marked decrease in a number of cells in S phase (Fig. 5). Noteworthy, REFs were hypertrophic (Fig. 5, right part). This is consistent with a large, flat cell morphology (Fig. 3). Although rapamycin completely blocked the mTOR activity, as measured by pS6 immunofluorescence (Fig. 3), it did not completely inhibit cell proliferation. After REFs grew for three additional passages, rapamycin was removed. During further passages without rapamycin, the cells became progressively smaller (Fig. 3), losing DDR markers (Fig. 4). Furthermore, by passage 24, a fraction of S phase cells achieved the basal levels of the early passage (Fig. 5). The size of REFs returned to normal (Figs. 5 and 3). Noteworthy, mTORC1 inhibition by rapamycin activated autophagy (Fig. S1).

Figure 5. Cell cycle distribution and analysis of cell size by flow cytometric light scattering. REF cells at different passages were prepared as described in Methods and analyzed for cell cycle distribution (left column) as well as cell volume by forward scattering (right column). Cells were fixed and stained with propidium iodide for DNA content (absciss axis). Forward light scattering (right column). (A) passage 1; (B) passage 7 and (C) passage 24 after rapamycin treatment during passages 7–10.

Although the mechanism of rapamycin-induced immortalization needs to be further investigated (Pospelova et al., manuscript in preparation), this may be a reminiscence of immortalization of MEFs by hypoxia. As shown by Campisi and coworkers, MEFs did not senesce in physiological (3%) oxygen levels, but underwent a spontaneous event that allowed indefinite proliferation.⁶⁵ Although different interpretations are possible,⁶⁵ we suggest that this effect may be due to inhibition of the mTOR pathway by hypoxia. In fact, hypoxia is known to inhibit the mTOR pathway.^66-73 Also, hypoxia suppresses geroconversion during cell cycle arrest (Leontieva et al. submitted).

Finally, our data support the outstanding study by Kolesnichenko et al. By using primary human fibroblasts undergoing either replicative or oncogenic RAS-induced senescence, Kolesnichenko et al. demonstrated that senescence can be delayed, and some aspects of senescence can be reversed by inhibition of mTOR using either rapamycin or overexpression of REDD1, a negative regulator of TORC1, or by depletion of TORC1.⁷⁴ It is noteworthy that replicative senescence of human fibroblasts is telomere-dependent. Thus, rapamycin suppresses both types of replicative senescence in mammalian cells.

Materials and Methods

Cell culture and reagents.

Primary REF cells (second passage) were cultured in DMEM with 10% FBS (HyClone) without antibiotics in 5% CO₂/95% air atmosphere. The early passage cells were split 1:3 twice a week by plating 5 x 10⁵ cells per 60 mm dish. Mid-passage REFs (7–10 passages) were split 1:2 weekly, and late-passage REFs were split 1:2 after confluence. In this way, several subsequent passages were performed until the cells failed to undergo population doublings. Under the given conditions of cultivation, REFs usually reached this state after 14–15 passages. Primary REFs were treated with 200 nM rapamycin from 7–10 or 7–11 pp. After three passages, rapamycin was removed. Primary mouse embryonic fibroblasts (MEFs, second passage) were cultured in DMEM supplemented with 10% FBS in 5% CO₂/95% air atmosphere. Cells were split 1:3–4. Stock of rapamycin (LC Laboratories) was prepared as 5 mM in DMSO. Nutlin-3a was purchased from Sigma-Aldrich and dissolved in DMSO as 10 mM stock.

Immunoblot analysis.

Whole-cell lysates were prepared using boiling lysis buffer (1% SDS, 10 mM Tris.HCl, pH 7.4). Equal amounts of proteins were separated on 10% or gradient polyacrylamide gels and transferred to nitrocellulose membranes. The following antibodies were used: rabbit anti-phospho-S6 (Ser235/236) and rabbit anti-phospho-S6 (Ser240/244) and mouse anti-S6 from Cell Signaling Biotechnology. Secondary anti-rabbit and anti-mouse HRP conjugated antibodies were from Cell Signaling Biotechnology. Signals were visualized using ECL chemilumenescence kit from Pierce.

Immunofluorescence microscopy.

REF cells were seeded and grown on coverslips, washed in PBS, fixed in fresh 4% paraformaldehyde for 15 min at room temperature and washed three times in PBS for 10 min each time. The cells were permeabilized with 0.2% Triton X-100 for 20 min, and blocked by incubation with 5% bovine serum albumin (V fraction, Sigma) in PBS for 30 min. The coverslips were incubated with the following primary antibodies: anti-S6 mouse monoclonal antibody (Cell Signaling), LC3 rabbit polyclonal (Cell Signaling) or pS6 rabbit polyclonal antibodies and rabbit policlonal anti γH2AX (Ser 139) (Cell Signaling), anti 53BP1 rabbit polyclonal (Santa Cruz) for 1 h. The coverslips were washed three times in PBS-1% BSA for 10 min each time and incubated with the secondary antibodies: rabbit anti-mouse fluorochrome-conjugated Alexa Fluor 488-conjugated or goat anti-rabbit Alexa Fluor 543-conjugated (Invitrogen) and DAPI or To-Pro III, (Molecular Probes) to stain nuclei for 1 h at room temperature. The images were obtained using Leica TCS SL confocal microscope (Leica Microsystems).

SA-β-Gal staining.

For Figure 2, SA-β-Gal staining was performed using senescence-galactosidase staining kit (Cell Signaling Technology) according to manufacturer’s protocol. For Figure 3, cells were fixed for 5 min in β-galactosidase fixative (2% formaldehyde; 0.2% glutaraldehyde in PBS) and washed in PBS and stained in β-galactosidase solution (1 mg/ml 5-bromo-4-chloro-3-indolyl-β-Gal (X-gal) in 5 mM potassium ferricyamide, 5 mM potassium ferrocyamide, 2 mM MgCl₂ in PBS) at 37°C until β-Gal staining become visible in either experiment or control plates. Thereafter, cells were washed in PBS, and the number of β-galactosidase-positive cells (blue staining) was counted under bright field illumination.

Romanovsky-Giemsa staining.

Cells on coverslips were washed several times in PBS, fixed in cold methanol and stained for 4 min with a dye prepared from a mixture of Azur, eosin and methylene blue. After staining, slides were washed several times with tap water, dried and embedded in Canada balsam. Analysis was performed on Axioscope at magnification 40 x 10.

Flow cytometry.

Cells were trypsinized, washed twice with the phosphate buffered saline (PBS) and treated with 0.01% saponin for 30 min at room temperature. The cells were then washed with PBS and incubated with RNase A (Sigma, 0.1 mg/ml) and propidium iodide (Sigma, 50 mg/ml) for 15 min at 37°C. An analysis of cell cycle distribution was performed by using an Odam cytofluorimeter (Brucker). Cell size of control and rapamycin-treated cells were compared by means of cytometric light scatering of propidium iodide stained cells by using Win MDI program version 2.8.

Acknowledgments

The work was supported by Russian Fund for Basic Research (grants 10–04–01152 and 12–04–01393), Russian Academy of Sciences Program “Molecular and Cell Biology,” and grant from the St. Petersburg State University, contract #1.37.122.2011.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Note

Supplemental materials can be found at: www.landesbioscience.com/journals/cc/article/20882