QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.106.118018v1 322/1/70    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Coates, S. S. A. Articles by Gary, R. K. Search for Related Content PubMed PubMed Citation Articles by Coates, S. S. A. Articles by Gary, R. K.

CELLULAR AND MOLECULAR

Beryllium Induces Premature Senescence in Human Fibroblasts

Department of Chemistry, University of Nevada, Las Vegas, Las Vegas, Nevada (S.S.A.C., S.M.K., R.K.G.); Bioscience Division, Los Alamos National Laboratory, Los Alamos, New Mexico (B.E.L.); and Nevada Cancer Institute, Las Vegas, Nevada (S.S.)

Received November 30, 2006; accepted March 28, 2007.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
After cells have completed a sufficient number of cell divisions, they exit the cell cycle and enter replicative senescence. Here, we report that beryllium causes proliferation arrest with premature expression of the principal markers of senescence. After young presenescent human fibroblasts were treated with 3 µM BeSO₄ for 24 h, p21 cyclin-dependent kinase inhibitor mRNA increased by >200%. Longer periods of exposure caused mRNA and protein levels to increase for both p21 and p16(Ink4a), a senescence regulator that prevents pRb-mediated cell cycle progression. BeSO₄ also caused dose-dependent induction of senescence-associated -galactosidase activity (SA--gal). Untreated cells had 48 relative fluorescence units (RFU)/µg/h of SA--gal, whereas 3 µM BeSO₄ caused activity to increase to 84 RFU/µg/h. In chromatin immunoprecipitation experiments, BeSO₄ caused p53 protein to associate with its DNA binding site in the promoter region of the p21 gene, indicating that p53 transcriptional activity is responsible for the large increase in p21 mRNA elicited by beryllium. Forced expression of human telomerase reverse transcriptase (hTERT) rendered HFL-1 cells incapable of normal replicative senescence. However, there was no difference in the responsiveness of normal HFL-1 fibroblasts (IC₅₀ = 1.9 µM) and hTERT-immortalized cells (IC₅₀ = 1.7 µM) to BeSO₄ in a 9-day proliferation assay. The effects of beryllium resemble those of histone deacetylase-inhibiting drugs, which also cause large increases in p21. However, beryllium produced no changes in histone acetylation, suggesting that Be²⁺ acts as a novel and potent pharmacological inducer of premature senescence.

The major cell cycle regulatory proteins that are associated with replicative senescence are p53, p21, and p16. The p53 tumor suppressor is a transcription factor that coordinates senescence and other cellular responses such as apoptosis. p21(Waf1/Cip1/Sdi1) is a powerful broad spectrum cyclin-dependent kinase (CDK) inhibitor that blocks the activity of cyclin-CDK2 and cyclin-CDK4/6 complexes that are required for cell cycle progression. The expression of p21 is controlled by the transcriptional activity of p53, although it can also be regulated by p53-independent processes. An increase in the abundance of p21 is a central feature of senescence (Noda et al., 1994). The other major effector of cell cycle arrest in human senescent cells is p16(Ink4a). p16 inhibits cyclin D1-CDK4/6, which is the kinase complex responsible for pRb phosphorylation. In its hypophosphorylated form, pRb sequesters E2F transcription factors that are required for initiation of S phase. p16 siRNA knockdown studies indicate that p16 is needed to maintain senescence in human fibroblasts (Bond et al., 2004). In addition to these cell cycle regulatory proteins, senescence-associated -galactosidase (SA--gal) activity has become widely recognized as an enzymatic biomarker that correlates with senescence (Dimri et al., 1995). The increase in p21 occurs early in senescence, whereas increases in p16 and SA--gal develop somewhat later (Stein et al., 1999).

Several treatments can induce a senescence-like state prematurely in young cells. Ectopic expression of the Ras oncogene causes premature senescence in fibroblasts via kinase cascade-dependent activation of the p53/p21 and pRb/p16 response pathways (Serrano et al., 1997; Lin et al., 1998). Smurf2, an E3 ubiquitin ligase, is produced by normal fibroblasts upon telomere shortening, and expression of Smurf2 from a retroviral vector causes premature senescence through p53 and pRb pathways (Zhang and Cohen, 2004). Premature senescence can also be elicited by forced expression of the Leo1-like domain of the RDL protein (Zhao et al., 2005). An alternative approach that does not require transfection is to transiently expose cells to hydrogen peroxide (Chen and Ames, 1994). This method has become widely used to induce premature senescence, but a near-lethal dose (100–500 µM) is needed for good induction, and peroxide can cause appreciable apoptosis in this dose range (Chen et al., 2000).

Drugs that trigger senescence could potentially be therapeutically useful, but small molecule pharmacological agents that induce premature senescence are limited. Histone deacetylase (HDAC) inhibitors are the major drug class with this property. Sodium butyrate and trichostatin A, two structurally diverse HDAC inhibitors, produce a senescent-like state in human fibroblasts (Ogryzko et al., 1996). HDAC inhibitors cause global increases in histone acetylation, resulting in changes in chromatin structure that may enhance transcription of some genes. The p21 promoter is particularly sensitive to the effects of these drugs (Richon et al., 2000), but microarray and differential display studies have estimated that HDAC inhibitors affect the expression of 2 to 4% of all genes (Van Lint et al., 1996; Moore et al., 2004), including many that are irrelevant to cytostasis.

Here, we report a new kind of senescence-inducing drug, beryllium sulfate. The efficacy of beryllium salts in inhibiting cell growth was first noted for cultured chick embryo fibroblasts and skeletal muscle cells (Chevremont and Firket, 1951). This simple agent is surprisingly potent because the growth of IMR-90 human lung fibroblasts can be inhibited by concentrations as low as 10 ng/ml Be (approximately 1 µM), with either BeSO₄ or BeCl₂ being equally effective (Hart et al., 1982). Subsequent studies of rat liver cells showed that BeSO₄ arrests the cell cycle at the G₁ phase (Skilleter et al., 1983). Similar cell cycle effects were seen in primary cultures of human skin fibroblasts and in human lung fibroblasts (Lehnert et al., 2001). The only molecular effect of Be²⁺ reported thus far is an increase in p53 and p21 protein levels after treatment with relatively high doses (10 and 100 µM) of BeSO₄ (Lehnert et al., 2001). Because p53 and p21 are important regulators of senescence, in the present study, we sought to investigate whether exposing cells to low concentrations of beryllium can elicit pharmacologically induced senescence.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Cells and Reagents. HFL-1 cells, a normal diploid human lung fibroblast line, were received from ATCC (Manassas, VA) at passage number 10 and population doubling (PD) 19 and propagated in RPMI 1640 with GlutaMAX and 25 mM HEPES (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT), 100 U/ml penicillin G sodium, 100 µg/ml streptomycin sulfate, and 0.25 µg/ml amphotericin B (Invitrogen). Exponentially growing cells at passage number 15 or 16 (PD = 32–34) were used at the time of initial treatment for all experiments on young, presenescent cells, except for one experiment designed to test the effects of advanced passage number, which used HFL-1 cells at passage number 26 (PD = 52). Cells were seeded from a common cell stock, then assigned randomly as either test (BeSO₄ added) or control (no drug) and maintained in parallel until harvesting for analysis. In long-term studies, cells were split as needed to maintain subconfluency, and fresh regular or Be²⁺-containing media were provided every 3 days. Replicatively senescent HFL-1 cells were prepared by serial propagation until complete cessation of proliferation at PD = 62 to 63, then these cells were maintained in the senescent state with periodic media replacement for at least 1 additional week before use in experiments. BeSO₄·4H₂O was from Fluka (a division of Sigma-Aldrich, St. Louis, MO). Sodium butyrate, trichostatin A, 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-gal), and 4-methylumbelliferyl--D-galactopyranoside (MUG) were from Sigma-Aldrich.

Assay of Senescence-Associated -Galactosidase Activity. Histochemical staining of intact cultured HFL-1 cells using X-gal to show SA--gal activity was done as described previously (Dimri et al., 1995). Quantitative assay of protein-normalized SA--gal activity in HFL-1 cell lysates using MUG was done as described previously (Gary and Kindell, 2005).

mRNA Quantification. Unamplified gene-specific mRNAs were quantified by cooperative hybridization (Canales et al., 2006) according to the QuantiGene protocol (Genospectra, Fremont, CA). In this method, total cell lysates are mixed with chimeric capture oligos. One half of the capture oligo is complementary to a segment of the target mRNA, and the other half is complementary to an oligo that has been coated into a 96-well plate. A pool of five to six distinct capture oligos is used to target each mRNA of interest, so that the captured mRNA binds to the plate in a polyvalent manner. The attachment of up to six capture oligos per target transcript molecule permits a stringent wash condition that removes nontarget mRNA. Selectivity is further enhanced by including additional target-specific oligos that hybridize to the mRNA regions in between the capture oligos, so that a contiguous 300- to 500-nucleotide region of the mRNA target becomes heteroduplex. This cooperative hybridization promotes the coil-helix transition and serves to inhibit the formation of competing secondary structures and to stabilize the specific associations between the mRNA and the DNA oligo probe set. Some of the noncapture oligos serve as points of attachment for branched oligos bearing alkaline phosphatase, which is used to catalyze a luminescence reaction. Commercially available oligo probe sets were used to quantify human -actin (accession no. NM_001101 [GenBank] ), human p53 (accession no. NM_000546 [GenBank] ), human bax (accession no. NM_004324), and human p21 (accession no. NM_000389 [GenBank] ) mRNA (GenoSpectra). For human p16/INK4a (accession no. NM_000077), a custom probe set was designed. p16 has three exons. Exons 2 and 3 are shared with ARF, an alternatively spliced mRNA that uses a different reading frame than p16. Therefore, capture oligos must target exon 1 to ensure that only p16 can be captured. Within exon 1, five capture oligo annealing sites were selected on the basis of sequence uniqueness with respect to p15/INK4b, a homolog of p16/INK4a. A probe set specific for p16/INK4a and incapable of hybridizing to ARF or p15/INK4b was created by incorporating the probe sequences 5'-GGACCGCGGTATCTTTCCA-3', 5'-GCCAGCCCCTCCTCTTTC-3', 5'-CCGCTGCCTGCTCTCCC-3', 5'-GGCTCCATGCTGCTCCCC-3', or 5'-TAACTATTCGGTGCGTTGGG-3' into the five capture oligos. These capture oligos are complementary to nucleotides 36 to 54, 112 to 129, 181 to 197, 227 to 244, and 324 to 343 of p16/INK4a mRNA, respectively. QuantiGene reagents (Genospectra) were used for cell lysis, cooperative hybridization, and enzymatic detection, then luminescence was measured using a GENios plate reader (Tecan, San Jose, CA) set to 37°C.

Western Blots. Treated HFL-1 cells were collected by trypsinization and centrifugation, washed twice with PBS, and extracted with radioimmunoprecipitation assay buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 0.5 mM benzamidine, 0.25 mM phenylmethanesulfonyl fluoride). Total protein concentrations in the radioimmunoprecipitation assay extracts were measured by BCA assay (Pierce, Rockford, IL). For actin and p53 Westerns, samples were loaded onto 10% SDS-PAGE gels at 3 and 6 µg/lane, respectively. For p21 and p16 Westerns, samples were loaded onto 11 and 12% SDS-PAGE gels at 30 (24-h experiment) or 45 µg/lane (3-week experiment). Proteins were transferred to polyvinylidene difluoride membranes (Invitrogen) and probed with the following primary and secondary antibody pairs: mouse IgG mAb anti-human p53 (sc-126; Santa Cruz Biotechnology, Santa Cruz, CA) and HRP-conjugated goat anti-mouse IgG (A-9917; Sigma-Aldrich), goat pAb raised against human/rat/mouse C-terminal actin peptide (sc-1615; Santa Cruz Biotechnology) and HRP-conjugated bovine anti-goat IgG (sc-2350; Santa Cruz Biotechnology), mouse anti-mouse/human p21 mAb (sc-6246; Santa Cruz Biotechnology) and HRP-conjugated goat anti-mouse IgG, and goat anti-human p16(INK4a) pAb (sc-468-G; Santa Cruz Biotechnology) and HRP-conjugated bovine anti-goat IgG. Blots were developed with ECL-Plus fluorescent/chemiluminescent reagent (GE Healthcare, Piscataway, NJ) and imaged by laser scanning (fluorescence mode, excitation = 457 nm; emission = 520 nm) on a Typhoon Variable Mode Imager (GE Healthcare) or by direct exposure to BioMax X-ray film. For histone acetylation assays, HFL-1 cells were treated with drugs for 24 h, trypsinized, washed in PBS, pelleted at 5 x 10⁶ cells/sample, and stored frozen at –70°C. Cell pellets were resuspended in Triton extraction buffer (PBS containing 0.5% Triton X-100, 2 mM phenylmethanesulfonyl fluoride, 0.02% NaN₃) at a cell density of 1 x 10⁷ cells/ml. Cells were lysed on ice for 10 min with gentle mixing, then centrifuged at 4800g for 10 min at 4°C. The supernatant was discarded, and the cell pellet was washed with Triton extraction buffer a second time. The pellet was resuspended in 0.2 N HCl at a cell density of 4 x 10⁷ cells/ml, and the histones were acid extracted overnight at 4°C. The supernatant containing the histones was collected at 4800g for 10 min at 4°C, and protein concentration was determined using the BCA assay (Pierce). Ten micrograms of the histone extract was run on a 4 to 20% SDS-PAGE gel, transferred to 0.2-µm nitrocellulose, probed with anti-acetyl histone H3 antibody (06-599 from Upstate division of Millipore, Billerica, MA) and goat anti-rabbit IgG HRP secondary antibody (SAB-300I; Stressgen, Ann Arbor, MI), and detected using SuperSignal West Dura Chemiluminescent substrate (Pierce). The membrane was reprobed with rabbit polyclonal Total Histone H3 antibody (9715; Cell Signaling Technology, Danvers, MA).

Fibroblast Immortalization. A retroviral vector was used to introduce human telomerase reverse transcriptase (hTERT) into young presenescent HFL-1 cells. Retrovirus was produced by the PA317 packaging cell line carrying the pBABEpuro-hTERT construct (Ouellette et al., 1999). PA317 cell supernatants containing the virus were passed through a 0.45-µm filter and used to infect HFL-1 cells at passage 13 in the presence of 4 µg/ml polybrene. At 48 h postinfection, 500 ng/ml puromycin was added to select for cells carrying the expression construct. Cells were grown for 2 months in the continuous presence of 500 ng/ml puromycin before use in experiments, which were conducted in the absence of puromycin. Kill curves using uninfected HFL-1 cells established that 500 ng/ml puromycin was a sufficient dose to kill 100% of nonresistant cells within 4 days (data not shown). The retroviral vector became stably integrated because the selected cells could be grown in the absence of puromycin without loss of puromycin resistance or immortalized phenotype (data not shown).

Telomerase Assay. Telomerase enzymatic activity was determined by telomere repeat amplification protocol (TRAP) assay using the TRAPeze XL Telomerase Detection Kit (Chemicon International, Temecula, CA). Cell lysates were incubated with a telomere-like single-stranded oligo substrate. Telomerase activity adds a series of telomeric repeat sequences, resulting in variable length extensions of the oligo. The extended product was amplified with PCR primers that anneal at six-nucleotide intervals along telomeric repeat sequences, and the resulting PCR product ladder was analyzed by gel electrophoresis. Heat sensitivity of the activity was evaluated by heating the cell extract to 85°C for 10 min. TRAPeze PCR products were run on 10% polyacrylamide Tris borate-EDTA nondenaturing gels, stained with SYBR Green dye, and photographed.

Chromatin Immunoprecipitation. Control and beryllium-treated cells were fixed with 1% formaldehyde to cross-link DNA-associated proteins to DNA, then chromatin was sheared enzymatically using ChIP-IT Express (Active Motif, Carlsbad, CA). Chromatin immunoprecipitations used protein G-coated magnetic beads and either p53 DO-1 Ab (sc-126; Santa Cruz Biotechnology) or RNA pol II Ab (Active Motif). After a 12-h incubation at 4°C, beads were washed with ChIP-IT Express reagents, supplemented by additional washes with 100 mM Tris, pH 8.0, 500 mM LiCl, 1% Nonidet P-40, and 1% sodium deoxycholate. Cross-links were reversed by 4-h incubation at 65°C, the protein in the chromatin immunoprecipitation (ChIP) samples was digested with proteinase K, and then the DNA in the ChIP samples was used as template for quantitative real-time PCR.

ChIP PCR was done using the forward primer 5'-GTGGCTCTGATTGGCTTTCTG-3' and the reverse primer 5'-CTGAAAACAGGCAGCCCAAG-3', which flank a p53-binding site in the promoter region of the human p21 gene (Kaeser and Iggo, 2002). The primer binding sites are located at nucleotides 2290 to 2270 and 2205 to 2186 to the 5' side of the p21 gene transcription start site, respectively. Real-time PCR was conducted using a BioRad iCycler with HotStar Taq polymerase (QIAGEN, Valencia, CA) and SYBR Green fluorescence quantitation. All PCR reactions yielded a single 105-bp product with a melting temperature of 83.0 ± 0.5°C, corresponding to the p21 gene amplicon. Initial experiments were also done using a TaqMan probe 5'-[6-FAM]-CTCTGGCATAGAAGAGGCTGGTGGCTATTT-[BHQ1]-3' that anneals to the region of nucleotides 2236 to 2208 to the 5' side of the transcription start. This TaqMan probe gave similar results to those obtained by SYBR Green quantitation, but the latter was slightly more sensitive, so the data produced with SYBR Green are shown. The p21 PCR primers and the p21 TaqMan probe were custom synthesized (Operon Biotechnologies, Huntsville, AL). Some PCR assays used a glyceraldehyde-3-phosphate dehydrogenase (gapdh) primer set (Active Motif).

The amount of p21 gene (p53-response element) sequence present in ChIP immunoprecipitates was expressed relative to the amount of input DNA, which was also measured by real-time PCR using the same p21 primers. Input DNA represents the total amount initially added to each immunoprecipitation, before antibody capture and washing. PCR efficiency was determined from the slope of the standard curve made by serial dilution of template, using the equation: efficiency (E) = [10^(–1/slope)]–1. Relative amounts of p21 gene sequence in the input and ChIP samples were determined from real-time PCR threshold cycle (Ct) values according to the equation: relative starting quantity = (1 + E)^{(mean input Ct – ChIP Ct)}. This result expresses the starting quantity of template in the ChIP PCR reaction relative to that in the input PCR reactions. Volume corrections were applied to account for the fact that each PCR reaction used a 5-µl aliquot of template solution, which represents only a portion of the total input and total ChIP volumes.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Development of SA--Gal in Beryllium-Treated Fibroblasts. Beryllium sulfate caused a gradual increase in SA--gal activity over time, as judged by histochemical staining. Young presenescent HFL-1 cells treated with 3 µM BeSO₄ for 1 week displayed a perceptible increase in SA--gal staining (Fig. 1C), whereas untreated cells grown in parallel were negative (Fig. 1B). Staining increased considerably by 2 weeks (Fig. 1D). After 3 weeks of exposure to BeSO₄ (Fig. 1E), the staining intensity appeared to be comparable with that of authentic replicatively senescent cells, which had been cultured continuously for 8 weeks to attain senescence (Fig. 1A). Positive staining was not evident in young HFL-1 cells grown for 3 weeks in the absence of BeSO₄ (data not shown). The majority of cells treated with BeSO₄ displayed light staining after 1 week, moderate staining after 2 weeks, and intense staining by 3 weeks. To quantify the effect of BeSO₄, SA--gal activity in cell extracts was measured using a fluorescence-based assay (Gary and Kindell, 2005). A dose-dependent increase in SA--gal activity was observed in HFL-1 cells after 12 days of treatment with BeSO₄ (Fig. 2). The initial reaction rate was determined for each treatment group using data from the linear phase of the reaction. In senescent human fibroblasts, SA--gal activity was 135 ± 4 relative fluorescence units (RFU)/µg total protein/h ± S.D. The basal activity of early passage fibroblasts was 48 ± 1 RFU/µg/h. This basal activity was not affected by 1 µMBe²⁺ (51 ± 1 RFU/µg/h), but 3 and 10 µMBe²⁺ caused SA--gal to increase to 84 ± 2 and 85 ± 3 RFU/µg/h, respectively.

View larger version (44K): [in this window] [in a new window]   Fig. 1. Beryllium causes expression of SA--gal. X-gal staining of senescent fibroblasts (A), presenescent fibroblasts (B), or presenescent fibroblasts treated with 3 µMBe2+ for 1 (C), 2 (D), or 3 weeks (E).

View larger version (21K): [in this window] [in a new window]   Fig. 2. Quantitative SA--gal assay. Presenescent HFL-1 fibroblasts at passage 16 (PD34) were treated with 0, 1, 3, or 10 µM BeSO4 for 12 days, splitting as needed to avoid confluency. Replicatively senescent fibroblasts were generated by propagating cells until PD63. SA--gal activity was measured by incubating cell extracts with the fluorogenic substrate MUG at 37°C and monitoring the production of the fluorescent hydrolysis product 4-methylumbelliferone (4-MU) at pH 6.0.

Effects of Be²⁺ on the Expression of p53-Regulated Genes. HFL-1 cells were used to examine the effects of Be²⁺ on transcriptional targets of p53. p21 is a p53-regulated gene that promotes cell cycle arrest and senescence, and bax is a p53-regulated gene that promotes apoptosis. After 24 h of treatment with 10 µM BeSO₄, p21 mRNA increased to 262% of control levels, whereas bax mRNA was unaffected by exposure to the drug (Fig. 3). There was no change in p53 mRNA as a result of BeSO₄ treatment. This is a typical result because p53-inducing agents generally act by preventing p53 protein degradation without affecting the rate of p53 mRNA synthesis. The amount of -actin mRNA was the same in the treated and untreated samples. Actin is a housekeeping gene used for sample normalization whose expression is expected to remain relatively constant on a per cell basis.

View larger version (19K): [in this window] [in a new window]   Fig. 3. Up-regulation of p21 mRNA by BeSO4 is a gene-specific effect. Young HFL-1 cells (PD33) were grown in the presence (gray bars) or absence (black bars) of 10 µM BeSO4 for 24 h; then, the cells were collected and counted. Cell lysate equivalent to 10,000 (p21, p53, and bax) or 1000 (-actin) cells was analyzed for specific mRNA content by cooperative hybridization and luminescence. mRNA values are expressed as percent of control for n = 3 plates per group.

Changes in Cell Cycle Regulators Caused by Low Doses of Beryllium. Early passage HFL-1 cells (PD32) grow rapidly, with a doubling time of 24 h. The doubling time increases to 48 h by about passage 26 (PD52), and the cells cease dividing entirely after an additional 10 population doublings with the arrival of replicative senescence. We investigated whether the basal level of p21 mRNA in untreated cells changes as HFL-1 cells age and also whether the ability of BeSO₄ to induce p21 expression is affected by the age of the presenescent cells. Slower growing PD52 cells had 42% more p21 mRNA, normalized relative to actin mRNA, than the younger, faster growing PD32 cells (Fig. 4A). However, both kinds of cells showed large inductions of p21 in response to BeSO₄ (Fig. 4B). After 24-h treatment with 3 µM BeSO₄, young fibroblasts increased p21 mRNA content by more than 200% (to 337% of control), with similar increases seen for 10 and 30 µM BeSO₄. Even though untreated PD52 cells contain higher basal levels of p21 mRNA, these older cells were nearly as responsive to BeSO₄ as the younger cells.

View larger version (30K): [in this window] [in a new window]   Fig. 4. Be2+ stimulates p21 expression in fast-growing young fibroblasts and in slow-growing older fibroblasts. A, basal level of p21 mRNA in young fibroblasts at PD32 and older fibroblasts at PD52. B, expression of p21 mRNA in young and old fibroblasts after 24-h treatment with 0, 3, 10, or 30 µM BeSO4.

In contrast to the results for p21, BeSO₄ at 1, 3, or 10 µM produced no change in p16 mRNA at 24 h (data not shown). Because p16 accumulates gradually over time as cells become senescent, we examined expression after longer periods of BeSO₄ treatment. After 14 days of culture in the presence of 10 µM BeSO₄, young fibroblasts showed a 7.6-fold increase in actin-normalized p21 mRNA compared with untreated control cells, attaining p21 mRNA levels very similar to those of replicatively senescent cells (Fig. 5). The treatment also caused a 71% increase in p16 mRNA, indicating that BeSO₄ can cause up-regulation of p16 with longer exposure time. However, BeSO₄-induced p16 mRNA levels did not approach the high levels seen in late-stage senescent cells. In contrast to p21 and p16, actin-normalized p53 mRNA levels were virtually identical in untreated, Be²⁺-treated, and senescent cells.

View larger version (23K): [in this window] [in a new window]   Fig. 5. Transcription of endogenous cell cycle blockers (p21 and p16) is increased in senescent cells and Be2+-treated cells. Young fibroblasts were grown for 14 days in the presence or absence of 10 µM BeSO4 and compared with senescent cells. Cells were kept subconfluent by splitting as necessary. p21, p16, and p53 mRNA levels were measured by cooperative hybridization and normalized relative to actin mRNA. n = 4 ± S.D.

View larger version (22K): [in this window] [in a new window]   Fig. 6. Up-regulation of p21 and p16 mRNA by 3 µM BeSO4. HFL-1 cells were grown in the presence of 0, 1, or 3 µM BeSO4 for 1, 2, or 3 weeks, then p21 (A), p16 (B), and p53 (C) mRNA was measured and normalized relative to actin mRNA. n = 3to4 ± S.D.

View larger version (63K): [in this window] [in a new window]   Fig. 7. Short- and long-term induction of cell cycle regulatory proteins by BeSO4. A, presenescent HFL-1 cells were grown in the presence of 0, 1, 3, or 10 µMBe2+ (lanes 1–4) for 24 h, then p53, p21, p16, and actin proteins were analyzed by Western immunoblot. B, senescent (lane 1) or presenescent (lanes 2–4) HFL-1 cells were grown in the presence of 0, 1, or 3 µMBe2+ (lanes 2–4) for 3 weeks, then p53, p21, p16, and actin proteins were analyzed. Ponceau S-stained blots are shown as loading controls.

BeSO₄ Is an Effective Cytostatic Agent in Telomerase-Positive Cells. Normal human fibroblasts express little or no telomerase activity. Ectopic expression of hTERT, the catalytic subunit of human telomerase, is sufficient to immortalize WI-38, BJ, and other human fibroblast cell types. The potential of hTERT to immortalize HFL-1 cells has not been reported previously. Stable transfection with hTERT allowed HFL-1 cells to be propagated indefinitely, and no end of replicative lifespan has yet been observed (data not shown). The hTERT-HFL cells can be regarded as immortal, as defined by established lifespan extension criteria (Morales et al., 1999). Lysate prepared from the hTERT-transfected cells confirmed the presence of telomerase activity (Fig. 8A). The telomerase enzyme is heat-labile. PCR products arising from telomerase activity were abolished when the reactions were done using heat-treated lysate. The HFL-1 and hTERT-immortalized HFL-1 cells were grown in the presence of various concentrations of BeSO₄ for 9 days, with passage every 3 days to maintain subconfluent cell density. Cell counts after 9 days showed that regular HFL-1 cells and their immortalized counterparts were equally sensitive to BeSO₄ (Fig. 8B). The IC₅₀ was 1.9 µM for the normal fibroblasts and 1.7 µM for the hTERT-immortalized fibroblasts.

View larger version (25K): [in this window] [in a new window]   Fig. 8. Be2+ arrests proliferation of hTERT-immortalized fibroblasts. A, telomerase activity was analyzed as described under Materials and Methods. The TRAP assay was conducted on lysate from normal HFL-1 fibroblasts (lane 1) from HFL cells that had been stably transfected with hTERT (lane 2), heat-treated HFL lysate (lane 3), heat-treated hTERT cell lysate (lane 4), and a synthetic telomere-like oligo that was used as a positive control template for PCR (lane C). B, growth curves for normal HFL cells (circles) and hTERT-HFL cells (squares) after 9 days in the presence of BeSO4. The IC50 was 1.9 µM for the normal fibroblasts and 1.7 µM for the hTERT-immortalized fibroblasts.

View larger version (51K): [in this window] [in a new window]   Fig. 9. Analysis of histone acetylation in treated cells. HFL-1 fibroblasts were treated for 24 h with vehicle (lane 1), trichostatin A (lanes 2–3), sodium butyrate (lane 4), or beryllium sulfate (lanes 5–9) at the indicated concentrations. Histones were extracted from 5 x 106 cells from each group and immunoblotted for acetylated histone H3 (top) or total histone H3 (bottom).

BeSO₄ Causes p53 to Bind to the p21 Promoter. ChIP was used to assess whether p21 induction in response to beryllium is due to the action of p53 as a transcription factor. After 24-h exposure to 10 µM BeSO4, HFL-1 fibroblasts displayed a 4.1-fold increase in the occupancy of the p53-binding site of the p21 promoter, as measured by ChIP with real-time PCR quantitation based on cycle thresholds (Fig. 10A). As an additional measure, gel electrophoresis of the PCR product after a limited number of cycles gave an independent visual confirmation that there was indeed 4 times as much product in the ChIPs from Be-treated cells as in ChIPs from untreated cells (Fig. 10B). As shown by PCR of the input DNA, the treated and control chromatin samples were well matched for total DNA, and 2-fold differences in template were easily differentiated after PCR.

View larger version (27K): [in this window] [in a new window]   Fig. 10. Be2+ causes p53 to bind to the p21 promoter. A, HFL-1 fibroblasts were treated with 0 or 10 µM BeSO4 for 24 h, then ChIP was performed using a p53 antibody. The amount of p21 promoter DNA that had been bound to p53 protein in the ChIP samples was measured by real-time PCR. Nonspecific DNA recovery was shown by omitting the antibody (No Ab) in parallel ChIP assays. B, 2-fold serial dilutions of input DNA and single aliquots from three independent ChIP assays from the experiments in A were amplified by endpoint PCR (30 cycles), run on a polyacrylamide gel, stained, and photographed. C, hTERT-immortalized cells were treated with 0 or 10 µM BeSO4 for 24 h, then chromatin was analyzed by ChIP using a p53 Ab and p21 PCR primers. The same chromatin samples were also analyzed by ChIP using an RNAP antibody and PCR primers for gapdh, a gene that is not regulated by p53. Each value is the mean ± S.D. for n = 3 ChIPs.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Beryllium (atomic number 4) possesses distinctive physicochemical properties. Be²⁺ is the only divalent cation that is smaller than the physiological cations Mg²⁺ and Ca²⁺, and its small ionic radius gives Be²⁺ exceptional charge density. Because of its unique properties, Be²⁺ could bind tightly to clefts in proteins that are inaccessible to other cations. These considerations make it possible for beryllium to exert biological effects that are unlike those of any other metal salt.

Three distinct biological effects of beryllium have been described in the literature: low-dose antiproliferative effects, high-dose carcinogenic effects, and immune system effects relevant to chronic beryllium disease, a lung affliction caused by occupational exposure to airborne material. Very low concentrations of beryllium salt (i.e., 1–3 µM) are sufficient to inhibit cell proliferation. At high concentrations, beryllium is somewhat carcinogenic to mammalian cells (Gordon and Bowser, 2003). For example, 560 µM BeSO₄ causes a 14-fold increase in BALB/c-3T3 cell transformation frequency compared with the spontaneous transformation rate (Keshava et al., 2001). The effects of Be²⁺ on the immune system are complex. BeSO₄ at 1 to 10 µM is mitogenic in human lymphocytes isolated from people who have beryllium hypersensitivity (Rossman et al., 1988). Antigen-presenting cells from individuals with uncommon HLA-DP alleles can display Be²⁺-containing class II complexes that may stimulate T cell proliferation (Fontenot et al., 2000). This immune reaction has been studied intensively because workers in the beryllium industry who have this particular MHC polymorphism are at greatest risk of developing chronic beryllium disease.

The most potent general effect of Be²⁺ is production of cell cycle arrest. As shown here, this arrest appears to result from activation of some facet of the senescence signaling pathway. Decreased proliferation was evident at 1 µMBe²⁺ in long-term growth assays (Fig. 8), and Be²⁺ at 1–3 µM and higher produced increases in senescence effectors and biomarkers. All changes in p21 and p16 were reflected in their levels of mRNA and protein. In contrast, actin-normalized levels of p53 mRNA always remained constant (Figs. 3, 5, and 6), even when increases in p53 protein levels were observed (Fig. 7). In other systems, p53 is usually regulated by protein stabilization and not via changes in transcription and that appears to be the case with beryllium as well.

BeSO₄ induced rapid increases of p53 and p21, and both remained elevated over 3 weeks of treatment. Up-regulation of p16 and SA--gal were late events that developed only after 2 to 3 weeks of treatment. The kinetics of Be²⁺-inducible changes parallel those seen in telomere-initiated senescence. Newly senescent cells that have just ceased dividing have high levels of p21 but little p16 and SA--gal, which require several additional weeks to reach maximum levels (Stein et al., 1999).

The effects of beryllium resemble those of histone deacetylase inhibitors, which stimulate large increases in p21 that spark premature senescence (Ogryzko et al., 1996; Munro et al., 2004). However, 1 to 100 µM BeSO₄ produced no changes in histone acetylation (Fig. 9), showing that beryllium exerts its effects by a different mechanism than HDAC-inhibiting drugs. The ChIP data indicate that Be²⁺ causes p53 to associate with the p21 promoter (Fig. 10). The recruitment of p53 to the p21 promoter also occurs during replicative senescence, as seen by ChIP (Jackson and Pereira-Smith, 2006). Immortalizing cells with hTERT did not affect their sensitivity to Be²⁺ (Fig. 8), indicating that beryllium acts later in the senescence pathway than the signals that originate from telomeres.

Based on its potency, Be²⁺ may bind to a senescence signaling molecule with a dissociation constant in the low micromolar range. Presumably, such interaction would alter the function of the target, either inhibiting or activating it, leading to up-regulation of senescence effectors. The identity of this putative signaling molecule is not known, but in vitro studies have shown Be²⁺ to be a potent competitive inhibitor of two enzymes: myoinositol monophosphatase (Faraci et al., 1993) and glycogen synthase kinase 3 (Ryves et al., 2002; S. R. Mudireddy and R. K. Gary, unpublished data). It remains to be seen whether either of these actions is important in senescence. A model comparing replicative senescence with pharmacologically induced senescence is presented (Fig. 11). Replicative senescence occurs in telomerase-negative cells that have lost telomere repeat or overhang length (Stewart et al., 2003). Signals produced at the telomeres are transmitted to p53, which in turn coordinates entry into senescence. Certain drugs that activate p21 or p53 can produce a similar cytostatic state in young cells that have not yet experienced critical telomere shortening. Sodium butyrate and trichostatin A achieve this effect by inhibiting histone deacetylase, which causes aberrant transcription of p21. Be²⁺ is proposed to bind via a high-affinity interaction to a regulator that is upstream of p53 in senescence signaling. The resultant activation of p53 stimulates transcription of p21 and causes expression of the molecular markers of senescence. Telomerase expression does not block premature senescence induced by HDAC inhibitors (Munro et al., 2004) or BeSO₄ (Fig. 8). If beryllium proves to be more selective in its actions than H₂O₂ or HDAC inhibitors, it may represent a useful new addition to the set of pharmacological tools available to study inducible senescence.

View larger version (28K): [in this window] [in a new window]   Fig. 11. Pharmacologically induced senescence. Replicative senescence is triggered by telomere shortening, which causes up-regulation of p53, p21, p16, and SA--galactosidase activity. This process does not occur when telomerase activity (hTERT) is present. Cell cycle arrest is enforced by p21, which blocks CDKs, and p16, which inhibits phosphorylation of pRb. HDAC inhibitors, such as sodium butyrate and trichostatin A (labeled as I, for inhibitor) can induce a similar state by enhancing transcription of p21. Be2+ targets an unknown receptor molecule (R) to cause cell cycle arrest with increases in p53, p21, p16, and SA--gal. hTERT expression cannot prevent pharmacologically induced senescence caused by either class of agent.

	Acknowledgements

 
We thank Pamela Dino for technical assistance.

	Footnotes

 
This study was supported by National Institutes of Health Grant P20 RR-016464 from the Idea Networks of Biomedical Research Excellence (INBRE) Program of the National Center for Research Resources.

Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.106.118018.

ABBREVIATIONS: CDK, cyclin-dependent kinase; SA--gal, senescence-associated -galactosidase; HDAC, histone deacetylase; PD, population doubling; X-gal, 5-bromo-4-chloro-3-indolyl--D-galactopyranoside; MUG, 4-methylumbelliferyl--D-galactopyranoside; PBS, phosphate-buffered saline; PAGE, polyacrylamide gel electrophoresis; HRP, horseradish peroxidase; TRAP, telomere repeat amplification protocol; PCR, polymerase chain reaction; ChIP, chromatin immunoprecipitation; gapdh, glyceraldehyde-3-phosphate dehydrogenase; Ct, threshold cycle; RFU, relative fluorescence unit(s); RNAP, RNA polymerase II; hTERT, human telomerase reverse transcriptase.

Address correspondence to: Dr. Ronald K. Gary, Department of Chemistry, University of Nevada, Las Vegas, 4505 Maryland Parkway, Las Vegas, NV 89154-4003. E-mail: ronald.gary{at}unlv.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, Harley CB, Shay JW, Lichtsteiner S, and Wright WE (1998) Extension of life-span by introduction of telomerase into normal human cells. Science 279: 349 –352.[Abstract/Free Full Text]

Bond J, Jones C, Haughton M, DeMicco C, Kipling D, and Wynford-Thomas D (2004) Direct evidence from siRNA-directed "knock down" that p16(INK4a) is required for human fibroblast senescence and for limiting ras-induced epithelial cell proliferation. Exp Cell Res 292: 151–156.[CrossRef][Medline]

Canales RD, Luo Y, Willey JC, Austermiller B, Barbacioru CC, Boysen C, Hunkapiller K, Jensen RV, Knight CR, Lee KY, et al. (2006) Evaluation of DNA microarray results with quantitative gene expression platforms. Nat Biotechnol 24: 1115–1122.[CrossRef][Medline]

Chen Q and Ames BN (1994) Senescence-like growth arrest induced by hydrogen peroxide in human diploid fibroblast F65 cells. Proc Natl Acad SciUSA 91: 4130–4134.[Abstract/Free Full Text]

Chen QM, Liu J, and Merrett JB (2000) Apoptosis or senescence-like growth arrest: influence of cell-cycle position, p53, p21 and bax in H2O2 response of normal human fibroblasts. Biochem J 347: 543–551.[CrossRef][Medline]

Chevremont M and Firket H (1951) Action of beryllium on cells cultivated in vitro; effect on mitosis. Nature 167: 772.[Medline]

de Magalhães JP, Chainiaux F, de Longueville F, Mainfroid V, Migeot V, Marcq L, Remacle J, Salmon M, and Toussaint O (2004) Gene expression and regulation in H₂O₂-induced premature senescence of human foreskin fibroblasts expressing or not telomerase. Exp Gerontol 39: 1379–1389.[CrossRef][Medline]

Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano EE, Linskens M, Rubelj I, Pereira-Smith O, et al. (1995) A biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc Natl Acad SciUSA 92: 9363–9367.[Abstract/Free Full Text]

Faraci WS, Zorn SH, Bakker AV, Jackson E, and Pratt K (1993) Beryllium competitively inhibits brain myo-inositol monophosphatase, but unlike lithium does not enhance agonist-induced inositol phosphate accumulation. Biochem J 291: 369–374.[Medline]

Fontenot AP, Torres M, Marshall WH, Newman LS, and Kotzin BL (2000) Beryllium presentation to CD4+ T cells underlies disease-susceptibility HLA-DP alleles in chronic beryllium disease. Proc Natl Acad SciUSA 97: 12717–12722.[Abstract/Free Full Text]

Gary RK and Kindell SM (2005) Quantitative assay of senescence-associated -galactosidase activity in mammalian cell extracts. Anal Biochem 343: 329–334.[CrossRef][Medline]

Gordon T and Bowser D (2003) Beryllium: genotoxicity and carcinogenicity. Mutat Res 533: 99–105.[Medline]

Hart BA, Absher M, and Sylwester D (1982) The effect of beryllium on the growth of human lung fibroblasts. Environ Res 27: 150–158.[Medline]

Herbig U, Jobling WA, Chen BP, Chen DJ, and Sedivy JM (2004) Telomere shortening triggers senescence of human cells through a pathway involving ATM, p53, and p21(CIP1), but not p16(INK4a). Mol Cell 14: 501–513.[CrossRef][Medline]

Itahana K, Dimri G, and Campisi J (2001) Regulation of cellular senescence by p53. Eur J Biochem 268: 2784–2791.[Medline]

Jackson JG and Pereira-Smith OM (2006) p53 is preferentially recruited to the promoters of growth arrest genes p21 and GADD45 during replicative senescence of normal human fibroblasts. Cancer Res 66: 8356–8360.[Abstract/Free Full Text]

Kaeser MD and Iggo RD (2002) Chromatin immunoprecipitation analysis fails to support the latency model for regulation of p53 DNA binding activity in vivo. Proc Natl Acad SciUSA 99: 95–100.[Abstract/Free Full Text]

Keshava N, Zhou G, Spruill M, Ensell M, and Ong TM (2001) Carcinogenic potential and genomic instability of beryllium sulphate in BALB/c-3T3 cells. Mol Cell Biochem 222: 69–76.[CrossRef][Medline]

Lehnert NM, Gary RK, Marrone BL, and Lehnert BE (2001) Inhibition of normal human lung fibroblast growth by beryllium. Toxicology 160: 119–127.[CrossRef][Medline]

Lin AW, Barradas M, Stone JC, van Aelst L, Serrano M, and Lowe SW (1998) Premature senescence involving p53 and p16 is activated in response to constitutive MEK/MAPK mitogenic signaling. Genes Dev 12: 3008–3019.[Abstract/Free Full Text]

Moore PS, Barbi S, Donadelli M, Costanzo C, Bassi C, Palmieri M, and Scarpa A (2004) Gene expression profiling after treatment with the histone deacetylase inhibitor trichostatin A reveals altered expression of both pro- and anti-apoptotic genes in pancreatic adenocarcinoma cells. Biochim Biophys Acta 1693: 167–176.[Medline]

Morales CP, Holt SE, Ouellette M, Kaur KJ, Yan Y, Wilson KS, White MA, Wright WE, and Shay JW (1999) Absence of cancer-associated changes in human fibroblasts immortalized with telomerase. Nat Genet 21: 115–118.[CrossRef][Medline]

Munro J, Barr NI, Ireland H, Morrison V, and Parkinson EK (2004) Histone deacetylase inhibitors induce a senescence-like state in human cells by a p16-dependent mechanism that is independent of a mitotic clock. Exp Cell Res 295: 525–538.[CrossRef][Medline]

Noda A, Ning Y, Venable SF, Pereira-Smith OM, and Smith JR (1994) Cloning of senescent cell-derived inhibitors of DNA synthesis using an expression screen. Exp Cell Res 211: 90–98.[CrossRef][Medline]

Ogryzko VV, Hirai TH, Russanova VR, Barbie DA, and Howard BH (1996) Human fibroblast commitment to a senescence-like state in response to histone deacetylase inhibitors is cell cycle dependent. Mol Cell Biol 16: 5210–5218.[Abstract]

Ouellette MM, Aisner DL, Savre-Train I, Wright WE, and Shay JW (1999) Telomerase activity does not always imply telomere maintenance. Biochem Biophys Res Commun 254: 795–803.[CrossRef][Medline]

Richon VM, Sandhoff TW, Rifkind RA, and Marks PA (2000) Histone deacetylase inhibitor selectively induces p21WAF1 expression and gene-associated histone acetylation. Proc Natl Acad SciUSA 97: 10014–10019.[Abstract/Free Full Text]

Rossman MD, Kern JA, Elias JA, Cullen MR, Epstein PE, Preuss OP, Markham TN, and Daniele RP (1988) Proliferative response of bronchoalveolar lymphocytes to beryllium: a test for chronic beryllium disease. Ann Intern Med 108: 687–693.[CrossRef][Medline]

Ryves WJ, Dajani R, Pearl L, and Harwood AJ (2002) Glycogen synthase kinase-3 inhibition by lithium and beryllium suggests the presence of two magnesium binding sites. Biochem Biophys Res Commun 290: 967–972.[CrossRef][Medline]

Serrano M, Lin AW, McCurrach ME, Beach D, and Lowe SW (1997) Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 88: 593–602.[CrossRef][Medline]

Skilleter DN, Price RJ, and Legg RF (1983) Specific G1-S phase cell cycle block by beryllium as demonstrated by cytofluorimetric analysis. Biochem J 216: 773–776.[Medline]

Stein GH, Drullinger LF, Soulard A, and Dulic V (1999) Differential roles for cyclin-dependent kinase inhibitors p21 and p16 in the mechanisms of senescence and differentiation in human fibroblasts. Mol Cell Biol 19: 2109–2117.[Abstract/Free Full Text]

Stewart SA, Ben-Porath I, Carey VJ, O'Connor BF, Hahn WC, and Weinberg RA (2003) Erosion of the telomeric single-strand overhang at replicative senescence. Nat Genet 33: 492–496.[CrossRef][Medline]

Van Lint C, Emiliani S, and Verdin E (1996) The expression of a small fraction of cellular genes is changed in response to histone hyperacetylation. Gene Expr 5: 245–253.[Medline]

Zhang H and Cohen SN (2004) Smurf2 up-regulation activates telomere-dependent senescence. Genes Dev 18: 3028–3040.[Abstract/Free Full Text]

Zhao L, Tong T, and Zhang Z (2005) Expression of the Leo1-like domain of replicative senescence down-regulated Leo1-like (RDL) protein promotes senescence of 2BS fibroblasts. FASEB J 19: 521–532.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.106.118018v1 322/1/70    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Coates, S. S. A. Articles by Gary, R. K. Search for Related Content PubMed PubMed Citation Articles by Coates, S. S. A. Articles by Gary, R. K.