GRSF1 suppresses cell senescence

GRSF1 levels decline in senescent cells

The computational merging of proteins less abundant in the membrane fraction of senescent cells (Dn_Sen) [9], mitochondria-resident proteins (mitoCarta) [10], and RNA-binding proteins (RBPs) (http://rbpdb.ccbr.utoronto.ca/) [11] revealed that GRSF1 and tRNA methyltransferase 1 (TRMT1) were the only mitochondria-residing RBPs reduced in senescent fibroblasts (Fig. 1A). We confirmed the decline in GRSF1 levels in a model of replicative senescence: human WI-38 fibroblasts that were either proliferating (‘P’, early population doubling level, PDL23) or were rendered senescent (‘S’) by replicative exhaustion (PDL58). The levels of GRSF1 were assayed in whole-cell lysates (WCL), membrane preparations (MP), and cytosolic preparations (CP) obtained from P and S WI-38 cells by Western blot analysis (Fig. 1B). Immunofluorescence analysis confirmed that GRSF1 levels declined in senescent cells (Fig. 1C) and Western blot analysis further established that GRSF1 levels decreased early in senescence (Fig. 1D). In addition, we measured GRSF1 in cells rendered senescent by exposure to DNA-damaging agents, doxorubicin (Doxo, 2 μg/mL for 24 h) or ionizing radiation (IR, 10 Gy), and assayed 7 or 10 days later, respectively [9]. As shown, GRSF1 levels declined markedly in these senescence models (Fig. 1E,F).

GRSF1 levels decline with senescence due to reduced protein stability

To address whether GRSF1 mRNA levels change during cellular senescence, we measured GRSF1 mRNA abundance by reverse transcription (RT) followed by real-time, quantitative (q) PCR analysis after exposure to ionizing radiation (IR) with 10 Gy and additional culture for 10 or 17 days (D10, D17). Unexpectedly, only a modest reduction in the steady-state levels of GRSF1 mRNA was observed in senescent cells (Fig. 2A). Analysis of GRSF1 mRNA stability by treating P and S cells with actinomycin D to block de novo RNA synthesis, and then assessing the rate of clearance of GRSF1 mRNA (or a stable control transcript, GAPDH mRNA) confirmed that GRSF1 mRNA is highly stable in both proliferating and senescent WI-38 cells (Fig. 2B).

Therefore, we asked whether GRSF1 protein production or degradation might instead be regulated during senescence. GRSF1 protein translation was studied by fractionating the polysomal compartment of the cell on sucrose gradients using standard procedures [12]. We isolated fractions that were devoid of ribosome particles (fractions 1 and 2), fractions containing ribosomal subunits and single ribosomes (fractions 3-5), low-molecular weight polysomes (fractions 6-8), and high-molecular-weight polysomes (fractions 9-12). RNA was prepared from each fraction, and the levels of GRSF1 mRNA in each fraction were calculated as a percentage of total GRSF1 mRNA in the gradient. This analysis revealed that the association of GRSF1 mRNA on polysomes of proliferating and senescent cells was rather similar, suggesting that the rate of GRSF1 translation was not strongly affected by senescence (Fig. 2C). We thus examined whether GRSF1 protein stability might differ between proliferating and senescent cells by measuring GRSF1 protein half-life in WI-38 fibroblasts following treatment with cycloheximide (CHX), an inhibitor of de novo translation. Interestingly, the stability of GRSF1 was markedly reduced in senescent cells relative to proliferating cells, as determined by Western blot analysis (Fig. 2D). In sum, GRSF1 levels decline in senescent WI-38 fibroblasts due in large part to the decreased stability of GRSF1 protein in senescence.

Loss of GRSF1 increases DNA damage and suppresses cell proliferation

Given that GRSF1 is needed to maintain mitochondrial oxidative phosphorylation, particularly the function of complex I [3], we asked if the senescence-associated reduction of GRSF1 might increase oxidative stress in the cell by elevating mitochondria-generated reactive oxygen species (ROS). To test this possibility, we transduced WI-38 cells with lentiviruses that expressed a GRSF1-directed (shGRSF1) shRNA to reduce GRSF1 levels constitutively or a control shRNA (shCTRL) (Fig. 3A, top left). The levels of mitochondria-derived superoxide, as assessed by staining with MitoSOX dye, revealed that lowering GRSF1 increased MitoSOX signals, indicating that reducing GRSF1 increased oxidative stress (Fig. 3A). To address whether the rise in oxidative damage might eventually lead to DNA damage, we asked if GRSF1 levels influenced two measures of DNA damage; both the quantification of γ-H2A.X-positive signals by flow cytometry (MFI, mean fluorescence intensity) (Fig. 3B) and the counting of immunofluorescent foci per cell that were reactive to the DNA damage response factor 53BP1 (Fig. 3C) revealed that GRSF1-silenced cells displayed increased DNA damage. Exposure to IR indicated that silencing GRSF1 further sensitized cells to DNA damage (Fig. 3C). Furthermore, WI-38 fibroblasts that were synchronized by double thymidine block and were stimulated to re-enter the cell division cycle showed a delay in cell cycle progression and transition through G2/M if GRSF1 levels were reduced (Fig. 3D). Finally, GRSF1-silenced WI-38 fibroblasts displayed lower ³H-thymidine incorporation (Fig. 3E). Collectively, these data indicate that loss of GRSF1 triggered mitochondria stress associated with key senescence traits, including elevated DNA damage (risen γ-H2A.X levels and 53BP1-positive foci) and reduced cell division.

Loss of GRSF1 increases IL6 production

To investigate if the senescence-associated loss of GRSF1 also elevated the production and secretion of the major SASP factor interleukin 6 (IL6), we tested the effect of GRSF1 on IL6 abundance in several conditions. First, lentiviruses expressing shCTRL or shGRSF1 were used to silence GRSF1 in WI-38 fibroblasts Fig. 3A) and the levels of secreted IL6, as measured in conditioned medium by ELISA, were found to increase after GRSF1 silencing (Fig. 4A). Second, WI-38 fibroblasts at PDL30 were transfected with either control siRNA (siCTRL) or GRSF1-directed siRNA (siGRSF1) on days 1 and 5 and cultured for an additional 8 and 15 days (D8, D15) to allow GRSF1 levels to change. SA-β-gal analysis revealed greater numbers of positive (blue) senescent cells after GRSF1 depletion, supporting the notion that lowering GRSF1 promoted cell senescence (Fig. 4B). Third, exposure to IR triggered WI-38 fibroblast senescence regardless of GRSF1 abundance, but SA-β-gal-positive senescent cells increased after depletion of GRSF1 (Fig. 4C). Interestingly, increases in senescent marker p53 (Fig. 4D, left) were dependent on GRSF1 silencing. The levels of IL6 mRNA and secreted IL6 were significantly higher in GRSF1-silenced WI-38 fibroblasts, as determined by RT-qPCR analysis (Fig. 4D) and AlphaLISA (Fig. 4E). Together, these data indicate that silencing GRSF1 accelerates senescence and induces IL6 secretion.

Paracrine IL6 signaling following GRSF1 depletion

Activated monocytes are the primary source of IL6 in blood, particularly during inflammation [13]. Silencing GRSF1 by stable expression of lentivirus-delivered shRNA in human THP-1 monocytes significantly elevated IL6 secretion (Fig. 5A,B). THP-1 cells treated with the IL6 inducer lipopolysaccharide (LPS) were included as a positive control (Fig. 5B). To test if IL6 produced by THP-1 cells might influence neighboring cells, we co-cultured proliferating or senescent WI-38 fibroblasts (bottom chamber) with either shGRSF1 or shCTRL THP-1 cells (top chamber) (Fig. 5C). As shown above, expression of IL6 mRNA was significantly induced in GRSF1-depleted THP-1 cells regardless of the type of co-cultured fibroblasts (Fig. 5D). Interestingly, IL6 mRNA levels were higher in WI-38 fibroblasts (proliferating or senescent) co-cultured with the shGRSF1 population of THP-1 cells (Fig. 5E). Collectively, these findings indicate that the increase in IL6 caused by lowering GRSF1 can promote the local production of IL6 in neighboring cells.

Cell culture, establishment of senescence, transfection, treatments, and SA-β-gal activity

THP-1 cells were cultured in RPMI-1640 supplemented with 10% FBS, antibiotics, antimycotics (Invitrogen). WI-38 human diploid fibroblasts (Coriell Cell Repositories) were cultured in DMEM supplemented with 10% FBS (Gibco), antibiotics, antimycotics, and non-essential amino acids (Invitrogen). Cells were counted using a TC-20 cell counter (Bio-Rad). Proliferating WI-38 fibroblasts were used at population doubling levels (PDLs) ranging between 20-25, and senescent WI-38 fibroblasts were used after additional culture (PDL50-PDL59). WI-38 cells (PDL25) were treated with Doxorubicin (2 µg/mL) for 24 h and were harvested for analysis 7 days later. Proliferating WI-38 cells (~PDL25) were rendered senescent by exposure to 10 Gy of ionizing radiation (IR); cells were harvested 10 days later unless otherwise indicated. Cells were transfected with 50 nM of control or GRSF1-directed siRNAs (siCTRL and siGRSF1, respectively) using Lipofectamine 2000 (Thermo Fisher Scientific).

Lentiviral particles delivering shGRSF1 or shCTRL (Origene) were transduced at multiplicity of infections (MOIs) of 10-20 in the presence of 8 μg/mL of Polybrene (Santa Cruz Biotechnology). To generate THP-1 cells expressing constitutively lower levels of GRSF1, cells were transduced with lentiviral particles expressing shGRSF1 or shCTRL (Origene) one week later they were treated with puromycin (0.5 μg/mL) to select for cells stably expressing integrated vectors. Pooled GRSF1-silenced populations were used for experiments. Senescence-associated-β-galactosidase (SA-β-gal) activity was assessed using a kit (Cell Signaling).

Reverse transcription (RT) and real-time quantitative (q)PCR analysis

Total RNA was extracted using TriPure Isolation Reagent (Roche), and cDNA was synthesized using random hexamers and reverse transcriptase (Invitrogen). Gene-specific primers were then used for qPCR amplification by employing SYBR Green master mix (Kapa Biosystems) and an Applied Biosystems 7300 instrument. The primer sequences for regular RT-qPCR analysis (each forward and reverse) were:

GCCCAGCTATGAACTCCTTCT and GAAGGCAGCAGGCAACAC for IL6 mRNA,

GGGAAACGAAGGGGTGATG and GCGGTGCTTCTCTAAGGCTTT for GRSF1 mRNA,

TGCACCACCAACTGCTTAGC and GGCATGGACTGTGGTCATGAG for GAPDH mRNA,

CATGTACGTTGCTATCCAGGC and CTCCTTAATGTCACGCACGAT for ACTB (β-Actin) mRNA,

CGAACGTCTGCCCTATCAACTT and ACCCGTGGTCACCATGGTA for 18S rRNA.

Protein preparation and Western blot analysis

Cells were lysed in modified lysis buffer (MEB) (50 mM Tris-Cl [pH 7.5], 150 mM NaCl, 1 mM MgCl₂, and 1% NP-40) supplemented with 1 mM PMSF and a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific). Membrane proteins were isolated using the Mem-PER™ Plus Membrane Protein Extraction Kit (Thermo Scientific). To measure protein stability, WI-38 proliferating (PDL25) and senescent cells (PDL50-51) were incubated with 100 µg/mL of cycloheximide (Calbiochem) for the times indicated, then lysed for analysis. Proteins were size-fractionated by SDS-PAGE using 4-12% NuPAGE gels, transferred onto nitrocellulose membranes (iBlot Stack, Thermo Fisher Scientific). The primary antibodies used recognized were used recognizing these proteins: GRSF1 (AV40382) from Sigma; β-Actin (sc-47778), HSP90 (sc-1055), p53 (sc-126), and GAPDH (sc-32233), from Santa Cruz Biotechnology; p21 (05-345) from Millipore; TOMM20 (ab186735) from Abcam; and TOMM70 (14528-1-AP) from Proteintech. After incubation with secondary antibodies (GE Healthcare), Western blotting signals were detected by enhanced chemiluminescence (Millipore) using a KwikQuant Imager (Kindle Biosciences, LLC).

Immunofluorescence analysis

Fixed or live cells were imaged using a confocal microscope LSM710 (Carl Zeiss, Inc.) or KEYENCE fluorescence microscope (BZ-X700). Immunofluorescence was performed as previously described [9]. Briefly, WI-38 fibroblasts were fixed with 100% methanol (Fisher Scientific) or 3.7% formaldehyde (methanol-free, Polysciences) for 10 min at -20 °C or RT, respectively, and permeabilized in 0.2% Triton X-100 for 5 min. After washes with PBS and incubation with 10% normal goat serum (Thermo Fisher Scientific) for 16 h at 4 °C, anti-GRSF1 (Sigma) or anti-53BP1 (BD Biosciences) antibodies were added to the cells and incubated at 37°C for 1 h. Washed cells were then incubated at 37 °C for an additional 30 min with Alexa Fluor 568-conjugated secondary antibodies (Thermo Fisher Scientific). Nuclei were stained in ProLong® Gold Antifade reagent with DAPI (for fixed cells, Thermo Fisher Scientific) or Hoechst 33342 dye (for live cells, Invitrogen). To measure the formation of mitochondrial superoxide, WI-38 fibroblasts were stained with a fluoroprobe (MitoSOX Red, Thermo Scientific) and assessed by the Keyence Fluorescence Microscope (Keyence).

Flow cytometry

To monitor DNA double-strand breaks, fluorescently-labeled γH2AX was detected by flow cytometry. Briefly, cells were fixed for 1 h at 25 °C, washed, and permeabilized (00-8333-56, eBioscience). After one wash with FACS buffer (0.5% BSA in PBS), cells were incubated with mouse control IgG-PE (BioLegend) or anti-H2AX(pS139)-PE (BD Biosciences) in the dark for 1 h at 25 °C. Washed cells were transferred to FACS tubes for analysis by flow cytometry (FACSCanto II, BD Biosciences).

[³H]-Thymidine incorporation assay

WI-38 fibroblasts expressing lentiviral shRNAs (shGRSF1 or shCTRL) were cultured in 6-well plates and incubated with 1 µCi/mL of [³H]-thymidine (Perkin Elmer) for 16 h at 37°C, whereupon cells were washed twice with ice-cold PBS, washed twice with 5% TCA, and solubilized in 0.8 mL lysis buffer (0.5 M NaOH, 2% SDS). The solubilized cell suspension (0.4 mL) was collected into vials containing 5 mL of scintillation cocktail and the radioactivity was measured using LS6500 scintillation counter (Beckman). Data were normalized to total protein amounts used for each sample.

Measurement of secreted IL6

The AlphaLISA® Human Interleukin-6 (IL6) Detection Kit (Perkin Elmer Inc.) was used to measure IL6 in CM following the manufacturer's instructions. To prepare conditioned media (CM), cells were cultured in serum-depleted media overnight, unless indicated. IL6 levels were normalized the number of cells in the plate. For ELISA, CM was prepared as follows: cells were washed with 1× PBS twice, then incubated with fresh DMEM with 10% FBS. After incubation for an additional 24-48 h, the CM was collected for further analysis. IL6 secreted into the culture medium were measured using a Human IL6 Quantikine ELISA Kit (R&D systems).

Polysome analysis

Polysome analysis was performed as previously described [26]. Briefly, cells were treated with cycloheximide (Calbiochem; 100 µg/mL) for 10 min and lysed in PEB (polysome extraction buffer; 20 mM Tris-HCl [pH 7.5], 100 mM NaCl, 5 mM MgCl₂ and 0.3% NP-40). After the lysate was separated through 10% to 50% sucrose gradients, 12 fractions were collected for further analysis. The distribution of mRNAs was quantified by RT-qPCR analysis and plotted as a percentage of the specific mRNA in each fraction relative to the total amount of that mRNA in the gradient.

Cell synchronization by double thymidine block and FACS analysis

To block cell cycle progression, GRSF1-silenced (shGRSF1) and control (shCTRL) WI-38 fibroblasts were incubated with 2 mM thymidine (Sigma Aldrich) for 18 h at 37 °C and then washed with PBS twice and incubated in standard culture media for an additional 9 h; cells were then cultured in thymidine-containing media for an additional 15 h to achieve synchronization. Synchronized cells were washed twice with PBS, and after adding fresh media to release cells into the division cycle, cells were collected after varying time periods. For collection, cells were trypsinized, resuspended and fixed in 70% ethanol, and kept at -20°C for 16 h. One day later, cells were pelleted and resuspended in PBS, propidium iodide (PI) was added to stain the DNA, and cell cycle analysis was performed using a Canto II flow cytometer (BD Biosciences) and FlowJo Software (FlowJo v10.2).