SEARCH

SEARCH BY CITATION

Keywords:

  • aging;
  • cellular senescence;
  • nuclear pore complex;
  • reactive oxygen species;
  • Sp1

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Hyporesponsiveness to external signals, such as growth factors and apoptotic stimuli, is a cardinal feature of cellular senescence. We previously reported that an aging-dependent marked reduction in nucleocytoplasmic trafficking (NCT)-related genes could be responsible for this phenomenon. In searching for the mechanism, we identified the transcription factor, Sp1, as a common regulator of NCT genes, including various nucleoporins, importins, exportins, and Ran GTPase cycle-related genes. Sp1 knockdown led to a reduction of those genes in young human diploid fibroblast cells (HDF); Sp1 overexpression induced those genes in senescent cells. In addition, epidermal growth factor stimulation–induced p-ERK1/2 nuclear translocation and Elk-1 phosphorylation were severely impaired by Sp1 depletion in young HDFs; Sp1 overexpression restored the nuclear translocation of p-ERK1/2 in senescent HDFs. Furthermore, we observed that Sp1 protein levels were decreased in senescent cells, and H2O2 treatment decreased Sp1 levels in a proteasome-dependent manner. In addition, O-GlcNAcylation of Sp1 was decreased in senescent cells as well as in H2O2-treated cells. Taken together, these results suggest that Sp1 could be a key regulator in the control of NCT genes and that reactive oxygen species-mediated alteration in Sp1 stability may be responsible for the generalized repression of those genes, leading to formation of the senescence-dependent functional nuclear barrier, resulting in subsequent hyporesponsiveness to external signals.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Senescent cells do not proliferate upon growth factor stimulation and are resistant to apoptotic stimuli, which are fundamental features of senescent cells that make them distinct from cells in the G0 state (Lim et al., 2000; Ryu et al., 2006). However, its underlying mechanism has not yet been clarified. We previously reported that p-ERK1/2 and NF-κB fail to translocate into the nucleus upon epidermal growth factor (EGF) stimulation in senescent human diploid fibroblasts (HDFs) (Lim et al., 2000; Kim et al., 2010). In addition, a significant amount of actin and gelsolin accumulates in the nuclear fraction of senescent cells (Lim et al., 2000; Ahn et al., 2003). Furthermore, nuclear import of Rev-GFP, a reporter protein, is impaired in senescent HDFs and in cells from old donors (Pujol et al., 2002). These data imply senescence-dependent operation of the functional nuclear barrier, which blocks nucleocytoplasmic trafficking of biomolecules. Recently, we showed that the expression of most NCT-related genes (NCT genes) are markedly repressed in senescent HDFs when compared with those of young HDFs, including various nucleoporins, karyopherin α, karyopherin β, and Ran GTPase cycle-related factors (Kim et al., 2010). These results led us to hypothesize that there are some common factor(s) controlling the expression of these NCT genes and that reduced activities of these common factors might result in impairment of the nucleocytoplasmic trafficking of signaling molecules in senescent cells, ultimately responsible for senescence-associated hyporesponsiveness to external stresses.

In this study, we identified the transcription factor, Sp1, as one of the master factors involved in controlling NCT gene expression. We present evidence that modulation of the Sp1 level may be responsible for senescence-associated hyporesponsiveness to either growth factors or apoptotic stress via its effects on NCT gene expression.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Identification of Sp1 as a putative common regulator of nucleocytoplasmic trafficking-related gene expression

We previously showed a marked reduction in NCT gene expression in senescent HDFs as compared to young cells through a DNA microarray analysis (Kim et al., 2010). In addition, by using semi-quantitative reverse transcription–polymerase chain reaction (RT–PCR), we demonstrated that three principal groups of NCT genes were significantly down-regulated in senescent HDFs, including nucleoporin genes (e.g., Nup50, Nup88, Nup107, and Nup155), transport receptor genes [KPNA2 (karyopherin α2 or importin α1), and KPNB1 (karyopherin β1 or importin β1)], and Ran GTPase cycle-related genes (RAN, RanBP1, and RanGAP1) (Kim et al., 2010). This generalized suppression of NCT genes led us to believe that there must be a common mechanism controlling NCT genes during the aging process. To identify common factor(s) involved in NCT gene control, we searched common transcriptional factors for those genes. For this purpose, we obtained sequence information from the PubMed genomic database on the promoter regions of 41 NCT genes, including various NUPs (nucleoporins), IPOs (importins), XPOs (exportins), KPNs (karyopherins), as well as RAN and RANBPs (Fig. 1A). Then, using the AliBaba2 program, we obtained 131 transcription factors with putative binding sites on the promoter regions of those 41 genes. Among the 131 transcription factors, 8 had a frequency over 50, and the Sp1 transcription factor showed the highest and most overwhelming frequency (Fig. 1B). These results implicate Sp1 as the most common and important regulator of NCT genes.

image

Figure 1. Identification of Sp1 as a common putative transcription factor for nucleocytoplasmic trafficking-related (NCT) genes. (A) Consensus binding sequences for putative transcription factors were analyzed on the promoters of 41 NCT genes (see Experimental procedures for details). (B) Eight common transcription factors showed frequencies over 50. (C) Basal expression levels of several transcription factors. Cell lysates were prepared from young (Y) and senescent (O) human diploid fibroblast (HDFs), and Western blot analysis was performed using indicated antibodies as described in the Experimental procedures section (left panel). The Sp1 signals were quantified by densitometry (right panel). The data are mean ± SD of three independent experiments. **P < 0.01 vs. young HDFs. (D) Tissues obtained from young (Y) and old (O) mice were prepared and subjected to Western blotting for Sp1 and tubulin (upper panel). The lower panel represents data quantified by densitometric analysis. All the data are means ± SD of three independent experiments. *P < 0.05, **P < 0.01 vs. young mice.

Download figure to PowerPoint

We then determined the basal levels of several transcription factors, including Sp1, in young and senescent HDFs. Sp1 protein levels were markedly reduced in senescent cells as compared to young cells, while there was no significant change in the protein levels of C/EBPα, Oct-1, or STAT1 (Fig. 1C). In addition, Sp1 protein levels were decreased in various tissues, including liver, spleen, and kidney from aged mice when compared with those from young mice (Fig. 1D). These results suggest that Sp1 might be a good candidate as a regulator for control of NCT genes and that decreased Sp1 protein levels in senescent cells and tissues could be a causal factor for aging-dependent generalized suppression of NCT genes.

Functional role of Sp1 as a common regulator for control of nucleocytoplasmic trafficking-related gene expression

To test whether Sp1 was a common regulator involved in NCT gene control, we determined changes in NCT gene expression after modulation of Sp1 expression. When we transfected si-Sp1 into young HDFs, we observed a global down-regulation of NCT genes. Specifically, nucleoporins (Nup50, Nup88, Nup107, and Nup155), transport receptors (KPNA2 and KPNB1), and Ran GTPase cycle-related genes (RanGAP) were significantly down-regulated by Sp1 depletion (Fig. 2A). In addition, protein levels of Nup50, Nup88, Nup107, Nup155, karyopherin α2, and RCC1 were decreased in Sp1-depleted HDFs as compared to control siRNA-transfected cells (Fig. 2B).

image

Figure 2. Sp1 regulates the expression of nucleocytoplasmic trafficking-related genes. (A) Young human diploid fibroblasts (HDFs) were transfected with siRNA against Sp1 (si-Sp1) or the control siRNA for 72 h. After transfection, total RNAs were extracted and the respective gene expression was determined by real-time RT–PCR. (B) Young HDFs were transfected with siRNA against Sp1 (si-Sp1) or the control siRNA. At 72 h after transfection, cell lysates were prepared, and Western blot analyses were carried out using antibodies against Sp1, nucleoporins, karyopherins, Ran, and RCC1 (left panel). The right panel represents data quantified by densitometric analysis. (C) Senescent HDFs were transfected with Sp1 cDNAs or empty vectors. Thereafter, total RNAs were extracted and the respective gene expression was determined by real-time RT–PCR. (D) Senescent HDFs were transiently transfected with Sp1 cDNAs or empty vectors (control). After 48 h, cell lysates were prepared and Western blot analyses with the respective antibody were carried out (left panel). The right panel represents data quantified by densitometric analysis. All the data are means ± SD of three independent experiments. *P < 0.05, **P < 0.01 vs. control vector-treated cells.

Download figure to PowerPoint

To confirm the effect of Sp1 on NCT gene control, we analyzed NCT gene expression after Sp1 overexpression. When we transfected Sp1 cDNA into senescent HDFs, we observed a global up-regulation of NCT genes. Specifically, nucleoporins (Nup50, 88, 107), transport receptors (KPNA2 and KPNB1), and Ran GTPase cycle-related genes (RanBP1, RanGAP, and RCC1) significantly increased as compared to those in control vector-transfected HDFs (Fig. 2C). In addition, protein levels of Nup50, Nup88, Nup107, Nup155, karyopherin α2, and RCC1 were up-regulated by Sp1 overexpression (Fig. 2D). These results support the hypothesis that Sp1 is a common functionally active regulator involved in the control of NCT gene expression.

Effect of Sp1 modulation on nucleocytoplasmic trafficking of p-ERK and its downstream signals

To examine the role of Sp1 in nucleocytoplasmic trafficking of signaling molecules, we monitored the nuclear translocation of p-ERK1/2 in response to EGF stimulation in Sp1-depleted or Sp1-overexpressing cells. As shown in Fig. 3A, the majority of p-ERK1/2 was detected in the nuclear fraction after EGF stimulation in control siRNA-transfected young HDFs; the nuclear localization of p-ERK1/2 was significantly attenuated in si-Sp1-transfected cells. Confocal microscopy data also illustrated that p-ERK1/2 moved from the cytosol to the nucleus upon EGF stimulation in young control siRNA-transfected HDFs, while the majority of p-ERK1/2 was localized in the cytosol in si-Sp1-transfected cells, even in the presence of EGF (Fig. 3C). On the other hand, the majority of p-ERK1/2 was detected in the cytoplasmic fraction of senescent control cells, but Sp1 overexpression significantly increased the nuclear localization of p-ERK1/2 after EGF treatment in these cells (Fig. 3B). Confocal microscopy data also confirmed the cytoplasmic sequestration of p-ERK1/2 in senescent cells and the recovery of its nuclear translocation by Sp1 overexpression (Fig. 3D).

image

Figure 3. Sp1 regulates the nuclear translocation of p-ERK and its downstream signals. (A) Young human diploid fibroblasts (HDFs) were transfected with si-Sp1 or the control siRNA. At 72 h after transfection, cells were incubated with 10 ng mL−1 epidermal growth factor (EGF) for 15 min. The cytoplasmic (C) and nuclear (N) fractions were prepared, and Western blotting was carried out with a p-ERK antibody. Tubulin and lamin B1 were used as contamination markers for the cytoplasmic and nuclear fractions, respectively (upper panel). Cytosolic and nuclear p-ERK band intensities were quantified by densitometry. Data are normalized for protein loading and represent the fold difference of p-ERK expression in nucleus relative to the value of cytosol (lower panel). (B) Senescent HDFs were transiently transfected with Sp1 cDNAs or empty vectors (control). At 48 h after transfection, cells were incubated with EGF for 15 min. The nuclear localization of p-ERK was analyzed in the cytoplasmic and nuclear fractions (upper panel). Cytosolic and nuclear p-ERK band intensities were quantified by densitometry. Data are normalized for protein loading and represent a fold difference of p-ERK expression in nucleus relative to the value of cytosol (lower panel). (C) Young HDFs were treated as described in A and confocal microscopic analysis was carried out. Methods described in detail in the Experimental procedures section. The scale bar represents 20 μm. (D) Senescent HDFs were treated as described in B, and confocal microscopic analysis was performed. (E) Young HDFs were treated as described in A, and Western blot analysis was performed with an antibody against phospho-Ser383-Elk-1. The right panel represents data quantified by densitometric analysis. The data are mean ± SD of three independent experiments. **P < 0.01 vs. control siRNA-treated cells. (F) Young HDFs were treated as described in A, and quantitative real-time RT–PCR was performed. All the data are means ± SD of three independent experiments. *P < 0.05, **P < 0.01 vs. control siRNA-treated cells.

Download figure to PowerPoint

To further confirm the role of Sp1 in modulation of nuclear translocation of the signaling molecules, ERK1/2 downstream signaling was monitored. Thereby, we analyzed the phosphorylation level of Elk, a known substrate of ERK1/2, as well as c-fos mRNA, a p-Elk target gene. As expected, Elk phosphorylation at serine 383 was markedly decreased in Sp1-depleted cells compared with that in the young control cells (Fig. 3E). In addition, Sp1 ablation significantly reduced c-fos mRNA (Fig. 3F). These results strongly suggest that Sp1 influences the nucleocytoplasmic trafficking of signaling molecules, as well as their downstream signaling, through its effect on the functional efficiency of NCT gene expression, and that decreased level of Sp1 protein may be a causal factor for aging-dependent hyporesponsiveness to external stresses.

Role of reactive oxygen species in aging-related down-regulation of Sp1 protein level

We next investigated the mechanism by which Sp1 protein levels are down-regulated under senescent conditions. We first monitored Sp1 mRNA levels, but detected no differences between young and senescent HDFs (Fig. 4A), indicating that Sp1 expression is regulated not at the transcriptional level but at the post-translational level.

image

Figure 4. H2O2 induces proteasome-dependent degradation of Sp1. (A) Total RNAs were extracted from young and old HDFs and performed real-time RT–PCR analysis. The data are mean ± SD of three independent experiments. NS = not significant vs. young HDF cells (B), HDFs were pulsed for 1 h with 1 mm H2O2 without or with 5 mm NAC 16 h pre-incubation. Cells were harvested at the indicated time point and subjected to Western blot analysis (left panel). The right panel represents the data quantified by densitometric analysis. (C, D) Young and senescent HDFs were pulsed with H2O2 without or with NAC for 16 h pre-incubation or 25 μm N-Ac-Leu-Leu-norleucinal for 24 h pretreatment. Cells were harvested, and Western blot analysis was performed using an anti-Sp1 antibody (left panel). The right panel represents data quantified by densitometric analysis. All the data are means ± SD of three independent experiments. NS = not significant vs. control cells. *P < 0.05, **P < 0.01 vs. control cells.

Download figure to PowerPoint

As levels of reactive oxygen species (ROS) are elevated in senescent HDFs (Fig. S1, Supporting information) and ROS play a key role in cellular senescence and in vivo aging (Sohal & Weindruch, 1996; Beckman & Ames, 1998), we monitored Sp1 protein levels in response to H2O2 treatment. As shown in Fig. 4B, the Sp1 protein level was strikingly decreased by H2O2 treatment, while that of STAT1 was unchanged. In addition, the H2O2-mediated reduction in Sp1 was blocked by treatment with N-acetylcysteine (NAC), a radical scavenger. These data led us to postulate that an aging-related increase in ROS may to a certain degree be responsible for Sp1 protein down-regulation. In the next step, we attempted to confirm whether ROS-induced modification of Sp1 protein was associated with proteasome-mediated degradation. Thus, we examined Sp1 protein levels in the presence or absence of the proteasome inhibitor N-Ac-Leu-Leu-norleucinal (ALLN). We found that the H2O2-dependent reduction in the Sp1 protein level was prevented by ALLN in both young and senescent HDFs (Fig. 4C,D). These results provide evidence that Sp1 protein levels are down-regulated by ROS through proteasome-mediated degradation during the aging process.

Reduced O-GlcNAcylation of Sp1 in senescent and H2O2-treated cells

As O-glycation (GlcNAc) of Sp1 has been suggested as a mechanism by which protein stability is determined (Han & Kudlow, 1997), we examined O-GlcNAcylation levels of Sp1 in young and senescent HDFs. As shown in Fig. 5, O-GlcNAcylation levels of Sp1 were markedly decreased in senescent cells when compared with young cells.

image

Figure 5. Hypo-O-glycosylation of Sp1 in senescent cells is associated with its proteasome-dependent degradation. (A) Cell lysates prepared from young (Y) and senescent (O) HDFs were immunoprecipitated with an anti-Sp1 antibody, and Western blotting was performed with an anti-Sp1 antibody or RL2 antibody. Methods were described in detail in the Experimental procedures section. The right panel: O-glycosylation levels of Sp1 was quantified by densitometry. The data are mean ± SD of three independent experiments. *< 0.05 vs. young HDF cells. (B) HDFs were treated with 1 mm H2O2 and 25 μm N-Ac-Leu-Leu-norleucinal. Cells were collected at the indicated time points, and the cell lysates were immunoprecipitated with an anti-Sp1 antibody. Western blotting was carried out using antibodies for Sp1 and RL2. The right panel: O-glycosylation levels of immunoprecipitated Sp1 were quantified by densitometry. The data are mean ± SD of three independent experiments. *P < 0.05 vs. control cells.

Download figure to PowerPoint

To test whether ROS were involved in the hypo-O-GlcNAcylation of Sp1, we treated HDFs with H2O2 and monitored O-GlcNAcylated Sp1 levels by immunoprecipitation. As shown in Fig. 5, the level of O-GlcNAcylated Sp1 was significantly decreased by H2O2 treatment in the presence of ALLN. These data suggest that ROS can decrease the level of O-GlcNAcylation of Sp1, which might facilitate its proteasome-mediated degradation in the aging process.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

The transcription factor, Spl, is a DNA-binding protein present in most types of mammalian cells. It has three zinc finger motifs that bind to the GC box in the genome, through which it activates a variety of target genes, including PDGF, PDGFR, EGFR, c-Src, Cyclin D1, Cyclin E, Cyclin B1, Cdk2, Cdc25, E2F-1, c-Myc, c-Jun, and c-Fos (Kadonaga et al., 1987). We report here a novel class of Sp1 target genes: NCT-related genes.

Nucleocytoplasmic trafficking, the exchange of matter between the nucleus and cytoplasm, requires many specific proteins: (i) nucleoporins, important constituents of nuclear pore complexes (NPCs), (ii) transport receptors, which bind to cargo molecules and assist them in passing through the NPCs, and (iii) Ran, which binds to transport receptors with GTP gradients (Allen et al., 2000). In the present study, we analyzed the putative promoter regions of 41 NCT genes with bioinformatics tools and found that Sp1 was the most common and dominant transcription factor responsible for regulation of NCT gene expression (Fig. 1B). Notably, most NCT gene promoter regions have Sp1 binding sites with multiple binding sites for Sp1. The subsequent Sp1 knockdown experiments in young HDFs and Sp1 overexpression in senescent cells provide clear evidence that Sp1 is actively involved in regulation of NCT gene expression (Fig. 2).

Our previous study showed that most NCT genes are markedly repressed in senescent HDFs as compared to young cells. Furthermore, the number of NPCs seems to be decreased in senescent HDFs (D'Angelo et al., 2009; Kim et al., 2010). These findings imply that reduced Sp1 activity in senescent cells might to a certain degree be responsible for the vast repression of NCT gene expression. In fact, it was reported that Sp1 stability is decreased in senescent HDFs (Oh et al., 2007). In addition, Sp1 activity decreases in aged brain and liver tissue (Ammendola et al., 1992; Adrian et al., 1996). Consistent with these reports, our data also showed that Sp1 was down-regulated in senescent cells and tissues from aged mice (Fig. 1C,D). Furthermore, we found that HDFs become growth-arrested for a prolonged period following Sp1 depletion. The majority of Sp1-depleted cells had a senescence-like phenotype: enlarged morphology, elevated expression of senescence-associated β-galactosidase (Fig. S2A, Supporting information), and increased ROS levels, which were approximately 3- and 4-fold higher than those of control cells, based on DHR123 for hydroxyl radicals and hydrogen peroxides and MitoSOX for superoxide anions, respectively (Fig. S2B,C, Supporting information). These results provide strong support for the concept that Sp1 is itself an intracellular regulator of cellular senescence. However, we detected no significant changes in senescence-associated β-galactosidase expression in Sp1-overexpressing senescent HDF cells. These data suggest that a more complex mechanism is necessary to restore the senescence phenotype in HDF cells.

Sp1 activity is readily regulated by post-translational modifications, such as phosphorylation, glycosylation, acetylation, ubiquitination, or poly(ADP-ribosyl)ation. For example, Bcl-2 overexpression in M14 human melanoma cells activates ERK, thereby increasing Sp1 phosphorylation, stability, and transcriptional activity on the uPAR promoter (Trisciuoglio et al., 2004). Mechanical stress activates p38 MAPK in Rat-2 cells, which in turn phosphorylates Sp1, thereby increasing its transcriptional activity on the filamin A promoter (D'Addario et al., 2006). H2O2 treatment activates JNK1, leading to Sp1 phosphorylation and reduced Sp1 activity in human alveolar epithelial cells (Chu & Ferro, 2006). With regard to glycosylation, serine and threonine residues of Sp1 are O-glycosylated by N-acetylglucosamine (GlcNAc), and hypo-O-glycosylation of Sp1 is associated with proteasome-mediated degradation of Sp1 protein in NRK cells (Han & Kudlow, 1997). These reports suggest that Sp1 levels are down-regulated in senescent cells or aged tissues through alterations in post-translational modifications, activated by senescence-associated signaling pathways.

To clarify the mechanism involved in Sp1 down-regulation in senescent HDFs and aged tissues, we focused on the role of ROS and O-glycosylation in its proteasomal degradation. We observed that Sp1 was hypo-O-glycosylated in senescent HDFs (Fig. 5). We also noted that Sp1 became hypo-O-glycosylated when cells were exposed to H2O2 (Fig. 5). Concordant with this hypo-O-glycosylated state, Sp1 was rapidly degraded by H2O2 treatment, which was prevented by the proteasome inhibitor ALLN (Fig. 4). Therefore, we assume that an aging-related ROS increase could aggravate its hypo-O-glycosylation, resulting in the consequent degradation of Sp1. Furthermore, the altered glucose metabolism in senescent cells (Zwerschke et al., 2003) might affect O-GlcNAcylation of Sp1, probably at the hexosamine turnover status, requiring high energy.

The Ras–Raf–ERK signaling pathway is the best-characterized cascade involved in cellular proliferation. When healthy young cells are exposed to mitogenic stimuli like EGF, the intracellular signaling molecules, Ras, Raf, and ERK, are serially activated. The activated/phosphorylated ERK enters into the nucleus, where it phosphorylates its substrate, Elk-1, at residues T353, T363, T368, S383, S389, and T417. Elk-1 is an Ets family transcription factor that binds to the serum-responsive element (SRE) on the promoter, through which it activates the expression of the immediate early genes, such as c-fos, egr-1, and JunB (Cruzalegui et al., 1999; Buchwalter et al., 2004; Zandi et al., 2007). Previously, we reported that the nuclear translocation of p-ERK and NF-κB is severely impaired in senescent cells, accompanied by a generalized reduction in NCT gene expression (Kim et al., 2010). Here, we present evidence that Sp1 is a key regulator of NCT gene expression, which is responsible for the aging-dependent down-regulation of the ERK signaling cascade under mitogenic stimuli (Figs 1-3). All these data support our hypothesis that senescent cells operate a functional nuclear barrier leading to hyporesponsiveness to external stimuli, which is attributable to down-regulation of the common transcription factor, Sp1. The Sp1 status-modulated functional nuclear barrier contributes to the fundamental nature of aging: a trade-off between growth arrest and apoptosis resistance. Moreover, as Sp1 status can be readily modified by a variety of post-translational modifications in response to nutrient and stress signaling, we presume that Sp1 is one of the ultimate senescent phenotype determinants.

Collectively, our data suggest that the general decrease in NCT gene expression detected in senescent HDFs is, at least in part, due to the ROS-mediated proteasomal degradation of the common transcription factor, Sp1, which is hypoglycosylated in its senescent status. Therefore, we propose Sp1 to be a master regulator of the senescence-associated functional nuclear barrier and responsible for the senescence-dependent hyporesponsiveness toward external signals.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Cell culture and senescence-associated β-galactosidase staining

Human diploid fibroblasts were isolated from the foreskin of a healthy 6-year-old boy. Cells were grown in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and 1% penicillin/streptomycin in a 5% CO2 incubator. Cells were subcultured serially at a ratio of 1:4. We defined young cells as those <25 population doublings and senescent (old) cells as those >66 population doublings. Cellular senescence was confirmed by their delayed population doubling times (over 3 weeks) and by the senescence-associated β-galactosidase (SA-β-gal) staining. SA-β-gal assays were performed as previously described (Dimri et al., 1995).

Search for common transcription factors

For each gene, the accession number was entered into the PubMed gene bank, and its mRNA sequence was obtained. Using the nucleotide blast program (http://blast.ncbi.nlm.nih.gov/), the promoter region of each mRNA sequence was screened. Finally, common transcription factors were identified by the alibaba2 program (http://www.gene-regulation.com/pub/programs.html).

Western blotting and immunoprecipitation

For immunoblot analysis, cells and tissues were solubilized with a lysis buffer containing 50 mm Tris pH 7.4, 150 mm NaCl, 1 mm EDTA, pH 8.0, 1% SDS, protease inhibitor cocktail (Roche, Inc., Nutley, NJ, USA), 1 mm PMSF, 1 mm NaF, and 1 mm sodium orthovanadate. Protein contents were determined using Bradford assay. Protein (30 μg) of each sample was resolved by SDS–PAGE, transferred onto nitrocellulose membranes (Schleicher & Schuell Bioscience Inc., Keene, NH, USA), and blocked with TBS containing Tween-20 in 2.5% nonfat dry milk. The membranes were incubated with the primary antibodies at 4 °C overnight. Secondary antibodies were added for 1 h at RT. The antibody–antigen complexes were detected using the Pierce ECL detection system (Thermo Fisher Scientific Inc., Rockford, IL USA). For immunoprecipitation, an equal amount of each protein lysates (800 μg) were gently mixed with Sp1 antibodies (4 μg) and incubated overnight at 4 °C, followed by overnight incubation with 50 μL protein A magnetic beads (Millipore Inc., Bedford, MA, USA). The beads were washed three times with the lysis buffer, eluted with SDS sample buffer, and subjected to reducing SDS–PAGE. For reprobing, membranes were stripped by incubation in a solution containing 2% SDS, 100 mM mercaptoethanol, and 62.5 mm Tris–HCl (pH 6.8) for 45 min at 50 °C. After extensive washing and blocking, the membranes were reprobed with the appropriate antibody and developed. Antibodies against actin (A5441, dilution 1:20 000), Nup88 (sc-98351, dilution 1:1000), Nup155 (sc-133858, dilution 1:1000), p-Erk (sc-7383, dilution 1:1000 for Western blotting, dilution 1:150 for immunofluorescence), p-Elk (sc-8406, dilution 1:500), and Sp1 (sc-14027, dilution 1:1000) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Polyclonal antibodies against Nup50 (A301-782A, dilution 1:1000) and Nup107 (A301-9579, dilution 1:2000) were purchased from Bethyl Laboratories (Montgomery, TX, USA). Monoclonal antibodies specific for karyopherin α2 (610485, dilution 1:2000), karyopherin β (610559, dilution 1:1000), and Ran (610340, dilution 1:1000) were obtained from BD Biosciences (San Jose, CA, USA). Monoclonal Antibody for O-linked N-acetylglucosamine (RL2) (MA1-072, dilution 1:1000) was purchased from Thermo Fisher Scientific (Rockford, IL, USA). Quantification of the band signals (from three independent experiments) was carried out by using gel-pro analyzer version 3.1 (Media Cybernetics Inc., Sliver Spring, MD, USA).

Subcellular fractionation

To prepare nuclear and cytosol extracts, 6 × 106 cells were harvested by trypsin, washed three times with phosphate-buffered saline, and resuspended in TM-2 buffer containing 10 mm Tris, pH 7.4, 2 mm MgCl2, 1 mm phenylmethylsulfonyl fluoride, and protease inhibitors. Resuspended cells were incubated at room temperature for 1 min and then transferred into a tube in ice for 5 min with Triton X-100 added to a final concentration of 0.5% and incubated on ice for 10 min. Cell lysates were separated by 70–80 passages through a 26-gauge needle. The nuclei were isolated from the cytosol by centrifugation at 3300  g at 4 °C for 10 min. Isolated nuclei were washed with TM-2 buffer and resuspended in a nuclear extraction buffer containing 20 mm of HEPES, pH 7.9, 0.42 m NaCl, 1.5 mm MgCl2, 25% (v/v) glycerol, 0.2 mm EDTA, 0.5 mm dithiothreitol, and protease inhibitors. Resuspended nuclei were incubated on ice for 30 min with occasional shaking and sonicated to extract the nuclear proteins and finally were spun down in a microcentrifuge for 5 min.

Confocal microscopy

Cells were plated on coverslips in 24-well plates and treated with EGF. The cells were washed with ice-cold PBS, fixed with a solution of 4% paraformaldehyde/PBS for 15 min at room temperature, permeabilized with a solution of 0.5% Triton X-100/PBS for 10 min, and blocked with a solution of 2% BSA/PBS. After incubation with the primary antibody overnight at 4 °C (diluted 1:150 for the p-Erk), the cells were washed with ice-cold PBS three times and incubated with FITC-conjugated secondary antibody (diluted 1:200) for 30 min at RT. Then, the nuclei were stained with DAPI. Coverslips were washed with ice-cold PBS four times and then mounted on glass slides. Stained cells were viewed on a confocal imaging system.

Animals and tissue preparation

C57Bl/6J male mice of two different age groups, young (4 months) and old (24 months), were sacrificed by cervical dislocation. Spleen, liver, kidney, and skin were frozen in liquid nitrogen and kept at −80 °C. Tissues were homogenized with a polytron tissue homogenizer and a teflon–glass homogenizer (GlasCol, Terre Haute, IN, USA) and then sonicated with a VCX 400 sonicator (Sonics & Materials Inc., Danbury, CT, USA) in a reagent containing 10 mm Tris–HCl, pH 7.6, 5 mm EDTA, 5 mm EGTA, 0.5 μg mL−1 antipain, 0.1 mm PMSF, and 140 μg mL−1 trypsin inhibitor. The homogenate was centrifuged at 16 000 g for 15 min, and the supernatant was collected and used in the subsequent experiments.

RNA isolation and real-time PCR

Total RNA was extracted from cultured HDFs using TRIZOL reagent (Invitrogen Corp., Carlsbad, CA, USA) and reverse transcribed by Superscript II reverse transcriptase (Invitrogen Corp., Carlsbad, CA, USA). Then, real-time PCR was performed with the SYBR Green PCR master kit (Qiagen Inc., Valencia, CA, USA) in an ABI 7000 PCR machine (Applied Biosystems, Foster City, CA, USA). The forward and reverse primers used in these experiments are listed in the Table S1 (Supporting information).

RNA interference and overexpression

The Sp1-specific siRNA pools (ON-TARGETplus SMARTpool Sp1, L-026959-00-0020) and control-scrambled siRNA were purchased from Dharmacon Co (Thermo Fisher Scientific Inc., Rockford, IL, USA). The sequences are listed in the Table S2 (Supporting information). The transfection of siRNA duplexes was performed as described previously (Elbashir et al., 2001). The Sp1 cDNA was purchased from Cosmo Genetech Co (Seoul, South Korea), and a mammalian expression plasmid was generated by inserting the full-length PCR fragment of Sp1 into the pcDNA3.1 vector. Cells were transfected with the Sp1 plasmid using lipofectamine reagent (Invitrogen Corp., Carlsbad, CA, USA).

Measurement of ROS

For quantitation of ROS, the cells were incubated with 5 μm DHR123 (Anaspec Corp., San Jose, CA, USA) and 0.2 μm MitoSOX (Invitrogen Corp., Carlsbad, CA, USA) for 30 min at 37 °C, washed with PBS, trypsinized, collected in PBS, and analyzed on a FACSCaliber (Becton Dickinson and Company, Franklin Lakes, NJ, USA). The results were analyzed by using cell quest 3.2 software (Becton Dickinson and Company, Franklin Lakes, NJ, USA) for analysis.

Statistics

Data were presented as means ± SD. Data were either analyzed by Student's t-test or one-way analysis of variance (anova) followed by the Tukey's or Dunnett's post hoc test for multiple comparisons to determine statistical differences between groups using graphpad prism analysis software. All data presented are the representative of at least three separate experiments. Results were considered significant at P < 0.05.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

This work was supported by the National Research Foundation (NRF) grants funded by the Korean Government (MEST) (No. 2010-0029150), Research Program of Cancer and Aging from KOSEF, KRIBB Research Institute Program, and Health Fellowship Foundation. Authors appreciate greatly the initial contributions of Hong Ju Ahn and Dr.Sung Jin Ryu to this research.

Author contributions

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

S.Y.K., H.T.K., and S.C.P. conceived and planned the study; S.Y.K. and H.T.K. carried out experimental work; S.Y.K., H.T.K., and S.C.P performed data analysis; S.Y.K., H.T.K., J.A.H., and S.C.P. involved in manuscript composition; all authors discussed the results and commented on the manuscript; and S.C.P. supervised the entire project.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
acel12012-sup-0001-FigS1.tifimage/tif115KFig. S1 Senescent HDF cells have increased levels of ROS.
acel12012-sup-0002-FigS2.tifimage/tif7117KFig. S2 Premature senescence induced by Sp1 siRNA has similar features as replicative senescence.
acel12012-sup-0003-TableS1.docWord document38KTable S1 Primer sequences for real-time RT–PCR.
acel12012-sup-0004-TableS2.docWord document28KTable S2 Sequences of Sp1 siRNA.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.